WO2017189951A1 - Methods for screening and diagnosing glaucoma - Google Patents

Methods for screening and diagnosing glaucoma Download PDF

Info

Publication number
WO2017189951A1
WO2017189951A1 PCT/US2017/030034 US2017030034W WO2017189951A1 WO 2017189951 A1 WO2017189951 A1 WO 2017189951A1 US 2017030034 W US2017030034 W US 2017030034W WO 2017189951 A1 WO2017189951 A1 WO 2017189951A1
Authority
WO
WIPO (PCT)
Prior art keywords
mitochondrial
variant
poag
variants
glaucoma
Prior art date
Application number
PCT/US2017/030034
Other languages
French (fr)
Inventor
David W. COLLINS
Venkata H. GUDISEVA
Joan O'BRIEN
Venkata R.M. CHAVALI
Original Assignee
The Trustees Of The University Of Pennsylvania
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by The Trustees Of The University Of Pennsylvania filed Critical The Trustees Of The University Of Pennsylvania
Publication of WO2017189951A1 publication Critical patent/WO2017189951A1/en

Links

Classifications

    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6876—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes
    • C12Q1/6883—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material
    • C12Q1/6886—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material for cancer
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61P—SPECIFIC THERAPEUTIC ACTIVITY OF CHEMICAL COMPOUNDS OR MEDICINAL PREPARATIONS
    • A61P25/00—Drugs for disorders of the nervous system
    • A61P25/28—Drugs for disorders of the nervous system for treating neurodegenerative disorders of the central nervous system, e.g. nootropic agents, cognition enhancers, drugs for treating Alzheimer's disease or other forms of dementia
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/156—Polymorphic or mutational markers

Definitions

  • POAG Primary open-angle glaucoma
  • a method for evaluating a subject's risk of developing glaucoma comprises determining from a biological sample from a subject whether the patient has a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
  • a method for evaluating a subject's risk of developing glaucoma comprises determining from a biological sample from a subject the existence of one or more mutational variant within a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
  • a method for evaluating the risk of progression of disease in a glaucoma patient comprises determining from a biological sample from the patient whether the patient has a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
  • a method for evaluating the risk of progression of disease in a glaucoma patient comprises determining from a biological sample of the patient the existence of one or more mutational variant within a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
  • a method for evaluating the risk of progression of disease in a glaucoma patient comprises determining from a biological sample from the patient whether the patient has a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
  • a method for diagnosing glaucoma in a patient comprises determining from a biological sample of the patient the existence of one or more mutational variant within a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
  • these methods further involve both diagnosing or evaluating the risk or progression of glaucoma and treating the patient, so diagnosed with therapeutic or dietary regimens to support mitochondrial health.
  • FIG. 1 is a graph showing complete coverage of mtDNA obtained by Pool-seq. A representative IGV coverage map from one of the 43 DNA pools is shown, with the positions of the major mitochondrial genes, and numbering corresponding to the rCRS mtDNA reference sequence. Positions with a variance frequency higher than 3% are indicated by darker bars to indicate the base composition of the variable, with coverage depth indicated by the height of each bar.
  • FIG. 2A is a graph showing that Pool-seq identified hundreds of variable positions on mtDNA, mostly associated with one or more African haplogroups.
  • the mean variance frequency at each of the 16,549 positions on mtDNA is plotted as a percentage for all 43 pools, representing a total of 1,999 individuals.
  • Variant positions that are associated the most common African-American mitochondrial macro haplogroups are indicated with solid circles. Open circles represent variants associated with multiple haplogroups, or subgroups of the major groups listed in the legend.
  • FIG. 2B is a graph comparing individual Sanger sequencing results on 1,999 individuals for a 605 bp region within the MT-COl gene to Pool-seq results on the same group.
  • the location and population frequencies from Sanger sequencing (solid diamonds) are plotted with those inferred by Pool-seq (open circles), and numeric labels indicate the position on the mtDNA reference sequence.
  • "FP" denotes two of the positions, 6420 and 6442, that were categorized as noisy and excluded from analysis as false positives.
  • FIG. 3 is a bar graph showing that the distribution of major mtDNA haplogroups inferred for the POAAGG study population is consistent with African- American ancestry.
  • the frequency of each haplogroup was estimated by the mean variance frequency from a set of associated positions on mtDNA by Pool-seq.
  • the estimates for POAG cases and controls are compared to those from an independent multicenter study of African- American haplogroups.
  • Above each haplotype indication are a set of 3 bars. Each set is organized from left to right as African American, Controls and POAG cases. Error bars represent 95% confidence intervals.
  • FIG. 4 is a schematic plotting data generated from Pool-seq on POAG cases versus controls that implicated the Llc2 and L2 branches of the mitochondrial phylogenetic tree.
  • the relationships among haplogroups detected in the POAAGG study population are represented in schematic form, with branch lengths representing approximate divergence times from the most recent common ancestor (MRCA), with branching order adapted from Behar et al. and Schlebush et al. 9 ' 67 Numbering from 1 to 3 adjacent the lines indicates the degree of imbalance observed in the POAG versus case pools, expressed as odds ratios (OR) from estimated population frequencies for one or more variants associated with that lineage.
  • MRCA most recent common ancestor
  • Haplogroup Llc2 is defined in part by the disease-associated missense variants m.6150G>A and 6253T>C in. MT-COl, with subgroups Llc2bla'b also having the m.2220T>C variant in MT-RNR2.
  • the L2 lineages are associated with variant m.2416T>C in M7-RNR2 and other variants.
  • FIG. 5 is a schematic showing the association of 381 variable positions on mtDNA with POAG in African- Americans.
  • Variable positions annotated as having a disease association in MITOMAP or as "pathogenic" or “likely pathogenic" by the NCBI variant viewer are indicated by markers with solid light gray fill with dark rim. Selected markers are identified by their coordinates on the rCRS reference sequence.
  • FIG. 6 is a schematic plot showing Sanger sequencing on MT-COl implicated the Llc2 haplogroup. Odd ratios (OR) from sequencing 1,308 POAG cases vs. 849 controls, and phylogenetic relationship of the observed common variants and associated haplogroups are shown. Evolutionary relationship and POAG associations of common MT-COl variants are also shown. The relative frequencies of the variants in cases :
  • FIG. 7 is a graph showing the affinity of Wt-MTCOl (Val83) peptide for ⁇ 1-42 confirmed by ELISA, whereas Mut-MTCOl (Val83Ile) decreased interaction by 92%.
  • FIG. 8 is a graph showing the distribution of the haplogroups which are common separated by the case and control frequencies. Each haplogroup labels two bars, the leftmost being the case bar and the rightmost being the control bar for each set.
  • the haplogroups which are significantly associated with POAG are denoted by arrows below the label, the protective groups have light gray and risk has dark gray arrows.
  • the macro haplogroups represented by grey bars below the chart.
  • FIG. 9 is a phylogenetic charts showing the relationship of the 29 common haplogroups represented with the divergence time.
  • the odds ratio for each haplogroup was represented using the numbers represented the Odds Ratios 1-3 alongside the appropriate line, reported as number for the broken line for male / number for the solid line for female, i.e., OR (l)/OR(3), etc.
  • the haplogroups that were significant were identified by *.
  • FIG. 10 is a schematic of the predicted domain structure of the COl protein and subcellular localization of three POAG-associated missense variants.
  • FIG. 11 is a graph showing the interaction of mutant and wild type COl peptides with ⁇ or scrambled ⁇ in ELISA assay.
  • APOE protein was the positive control
  • TBX3 protein and scrambled Abeta peptide (ScrAB) were negative controls.
  • compositions and methods described herein provide means for diagnosing or detecting the existence or risk of, or monitoring the progress of, glaucoma in a patient by identifying the mtDNA haplogroups or haplotypes and/or the presence of certain genetic mutations defining certain mtDNA haplogroups or haplotypes in the patient.
  • One such method for evaluating a subject's risk of developing glaucoma or having progressively severe glaucoma or diagnosing glaucoma in a subject comprises determining from a biological sample from a subject whether the patient has a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
  • the method involves determining from said sample the existence of one or more mutational variants that defined the suspect mitochondrial haplogroup. As demonstrated herein, these mitochondrial haplogroups are indicative of greater risk of developing glaucoma and/or of progression to more severe glaucoma.
  • the term "patient” or “subject” as used herein means a male or female human, or an animal, e.g., a domestic animal or pet, and animals normally used for clinical research.
  • the subject of these methods and compositions is a human.
  • the subject of these methods and compositions is a human of African ancestry.
  • Glaucoma refers to a neurodegenerative illness of the eye. Glaucoma pathogenesis likely involves mitochondrial dysfunction 55 . It is possible that a subgroup of patients with glaucoma may demonstrate a mitochondrial optic neuropathy but one that has more complex genetics than the mitochondrial disease Leber hereditary optic neuropathy (LHON).
  • LHON mitochondrial disease
  • POAG Primary open-angle glaucoma
  • POAG lymphoblasts have been shown to exhibit a complex I defect in the mitochondrial oxidative phosphorylation pathway, with decreased rates of respiration, which could confer an increased susceptibility to cell death on retinal ganglion cells 48 .
  • mtDNA means mitochondrial DNA. All living humans share a most recent common ancestor (MRCA), an African woman who is estimated to have lived approximately 194,000 ⁇ 33,000 years ago 23 . The mitochondrial genome is extremely informative regarding maternal ancestry because of the genome's matrilineal inheritance and lack of genetic recombination.
  • the human mitochondrial family tree has been characterized extensively and subdivided into thousands of branches (hyperlink: PhyloTree).
  • the human mtDNA sequence and nucleic acid positions are identified herein using the Revised Cambridge Reference Sequence (rCRS) of the Human Mitochondrial DNA (Genbank NC_012920 gi:251831106). This 16,569 bp sequence is also reproduced as SEQ ID NO: 1.
  • mtDNA haplogroups or “haplogroups” as used herein is meant the major ancestral groupings of mitrochondrial DNA defined by collections of variants scattered throughout the mitochondrial genome that are inherited together and have accumulated over evolutionary time. The oldest mutations define the major macrohaplogroups and the deepest roots of the tree.
  • the human mtDNA phylogeny is divided into major haplogroups designated L0, LI, L2, L3, L4, L5, and L6, which are further divided into smaller subgroups.
  • Contemporary non- African haplogroups are derived from L3, with major branches designated M and N, and associated with ancient diasporas from Africa that occurred approximately 60,000 to 70,000 years ago 23 ' 9 .
  • a prior survey of 44 subjects determined that L3, L2, LI, and L0 groups are all present in the POAAGG study population, as are non-African haplogroups 13 .
  • a study of a Saudi Arabian population 3 found a positive association of POAG with African (L) haplogroups and evidence for a protective effect of the Eurasian haplogroup N, but studies of Arab 4 , Ghanaian 2 , and northern European 5 populations failed to find significant associations with mitochondrial haplogroups and POAG.
  • mutation refers to any change in a nucleotide or amino acid sequence from a reference nucleotide or reference sequence.
  • variant as used herein is mean the occurrence of a different nucleic acid at a position in the mtDNA which is different from that nucleic acid that occurs at that position in healthy subjects or those without glaucoma. These variants are identified below in Table 1 and about 381 such significant variants are identified in Table 5. These mtDNA variants can result in amino acid changes when they occur in certain
  • nucleic acid variants can be silent, i.e., not result in any amino acid change.
  • the variants can be identified in Table 5 as having an Odds Ratio greater than 1.0 and/or a Chi test of less than 0.05.
  • polymorphism means any sequence variant present at a frequency of >1% in a population.
  • the sequence variant may be present at a frequency significantly greater than 1% such as 5% or 10 % or more.
  • the term may be used to refer to the sequence variation observed in an individual at a polymorphic site.
  • Polymorphisms include nucleotide substitutions, insertions, deletions and microsatellites and may, but need not, result in detectable differences in gene expression or protein function.
  • Single nucleotide polymorphisms SNP are the most common type of genetic variation and involve substitution of a single nucleotide in a sequence.
  • missense mutation is meant is a mutation in which a single nucleotide change or variant results in a codon that codes for a different amino acid. It is a type of nonsynonymous substitution.
  • Pool-seq is meant a strategy of sequencing pooled DNAs used in the examples of this specification.
  • Next-generation sequencing methods now permit large regions of DNA to be investigated economically, but population-scale whole genome analysis is still prohibitively expensive, due to the complexity of next-generation sequencing library construction and the incremental costs of enriching samples for regions of interest.
  • The, “Pool-seq,” method has emerged as a cost-effective alternative to sequencing individuals separately 68 .
  • Population allele frequencies determined by Pool-seq have been shown to correlate well with actual allele frequencies 64 , and a pooled sequencing approach, using pools of 20 individual DNAs, has been successfully implemented on the Ion Torrent 66 PGM non-optical semiconductor platform 21 .
  • POAAGG Primary Open-Angle Glaucoma Genetics
  • the identification and frequency of occurrence of such mutations permits the identification of subsets of the African American population who are at increased risk from mitochondrial genetic dysfunction and/or geographical ancestries that may be linked to mitochondrial sequence variation.
  • Methods of detecting this information in a patient helps target screening efforts and/or directs clinical interventions such as bioenergetic therapies that may slow vision loss in glaucoma by supporting mitochondrial function 55,31, 49
  • the diagnostic mitochondrial haplogroup is Llc2.
  • the suspect mitochondrial haplogroup is Llc2blbl .
  • Still other haplogroups and risks factors are indicated in FIGs. 4 and 6 and in the Examples and the Tables 1-4B below.
  • the investigators identified certain missense variants or nucleic acid variants at positions in the mtDNA from the naturally occurring, i.e., most frequently naturally occurring nucleic acid in subjects without glaucoma, which are useful in evaluating a patient's likelihood or risk of progressing to glaucoma.
  • the variant is a mutation within the MT-COl gene of the patient's mtDNA.
  • the variant occurs at one or more mtDNA positions 6150, 6253 or 6480.
  • certain variants can occur alone or with other variants.
  • a mtDNA variant that is indicative of developing/progressing to severe disease is a G to A variant at mtDNA position 6150.
  • the variant is a T to C variant at mtDNA position 6253.
  • the variant is a G to A variant at mtDNA position 6480.
  • the variant is a variant identified in Table 1 or 2 below.
  • the variant is a A to G variant at mtDNA position 2220.
  • the variant is a C to T variant at mtDNA position 2332.
  • the variant is a T to C variant at mtDNA position 2416.
  • the variant is identified from among the positions in Table 5, as having an Odds Ratios of greater than at least 1.0. In other embodiments, the Odds Ratios defining the variants are greater than 1.2. In other embodiments, the Odds Ratios defining the variants are greater than 1.4. In other embodiments, the Odds Ratios defining the variants are greater than 1.6. In other embodiments, the Odds Ratios defining the variants are greater than 1.8. In other embodiments, the Odds Ratios defining the variants are greater than 2.0, 2.5 or 3.0.
  • variants can be identified from among those rCRS nucleotide positions listed in Table 5 by Chi test numbers less than 0.1 or less than 0.5 or less than 0.3. These variants can be identified from the list and information provided in FIG. 9 of the priority application No. 62/330133 and in the Appendix 1 provided in the on-line version of Collins, et al, 2016 14 .
  • the variant alters an amino acid residue encoded by a mitochondrial gene. See for example the example below describing the variant that changed an amino acid in a mitochondrial protein which affected the interaction of MT- COl with amyloid beta ( ⁇ ). See, e.g., FIG. 7.
  • the altered amino acid residue may alter the function or interaction of amyloid beta or another gene or protein pathway in the subject. In still other embodiments the variant may be silent.
  • the methods of detecting or determining the haplogroup or presence of a variant in a subject's mitochondrial DNA can be determined by evaluating the subject's biological sample to one of many types of DNA sequence analysis techniques.
  • the sample can be any biological fluid or tissue from which mtDNA may be evaluated.
  • the most suitable samples for use in the methods and with the compositions are samples which require minimal invasion for testing.
  • the sample is saliva.
  • the sample is a blood samples, including serum, plasma, or whole blood.
  • the sample is urine.
  • Still other biological samples are known for use in obtaining mtDNA, e.g., secretions, ascites fluids or peritoneal fluid and the like. Such samples may further be diluted with saline, buffer or a physiologically acceptable diluent. Alternatively, such samples are concentrated by conventional means.
  • Suitable assay methods for DNA analysis and sequencing can include next gen sequencing techniques including the Illumina (Solexa) sequencing, Roche 454 sequencing, Ion torrent: Proton / PGM sequencing methods, and PCR-based methods, such as reverse transcription polymerase chain reaction (RT-PCR) or qPCR, among many others.
  • next gen sequencing techniques including the Illumina (Solexa) sequencing, Roche 454 sequencing, Ion torrent: Proton / PGM sequencing methods, and PCR-based methods, such as reverse transcription polymerase chain reaction (RT-PCR) or qPCR, among many others.
  • RT-PCR reverse transcription polymerase chain reaction
  • qPCR reverse transcription polymerase chain reaction
  • ILLUMINA assay mass spectrometry quantitative assays, PCR, RT-PCR, QPCR or next generation sequencing techniques.
  • Other commercial assays or reagents include TRI- REAGENT, Qiagen RNeasy mini-columns, MASTERPURE Complete DNA and RNA Purification Kit (EPICENTRE®, Madison, Wis.), Paraffin Block RNA Isolation Kit
  • RNA Stat-60 Tel-Test
  • MassARRAY-based method Sequenom, Inc., San Diego, CA
  • differential display equenom, Inc., San Diego, CA
  • amplified fragment length polymorphism iAFLP
  • BeadArrayTM technology Illumina, San Diego, CA
  • the sample may be genotyped or subjected to whole genome sequencing (WGS) and computational analysis of the WGS. It may be useful if the variants results in amino acid changes, for other known techniques to be used to detect changes in the amino acid sequences of any peptides encoded by the mtDNA. Such methods are also well known in the art. The methods described herein are not limited by particular detection methods employed.
  • the method of evaluation and diagnosis may be further accompanied by evaluation of the subject clinically.
  • the subject in addition to detecting the suspect haplotype and/or variants defining the suspect haplogroups identified herein as indicative of glaucoma or risk of glaucoma, the subject may also receive additional evaluations.
  • Such clinical evaluations include evaluating said patient for clinical abnormalities in cup- to-disc ratio (CDR), or evaluating the subject's visual field or acuity, and/or evaluating the subject's intraocular pressure (IOP).
  • CDR cup- to-disc ratio
  • IOP intraocular pressure
  • the methods described herein include not only detecting the haplogroups and/or variants defining it as indicative of a diagnosis of, or increased risk of developing, glaucoma, but also further treating the subject so diagnosed.
  • suitable treatments are those therapeutic agents, exercise regimens or dietary supplements or regimens indicated to preserve or enhance mitochondrial function.
  • dietary supplements include coenzyme-QlO, a ketogenic diet, L-Arginine, lipoic acid, and the like.
  • certain clinical trials for other mitochondrial diseases may be useful for a subject diagnosed as having an increased risk of developing glaucoma by the methods described herein.
  • Such clinical trials and therapeutic agents include those described in, e.g., Ravina, BM et al, Neurology, 60(8): 1234-40 (April 2003); Schults, CW et al, Arch Neurol. 59(10): 1541-50 (Oct 2002) and Kerr, DS, Mol Genet Metab. 99(3):246-55(Mar 2010) among others, incorporated by reference herein.
  • POAAGG Primary Open-Angle African American Glaucoma Genetics
  • POAG primary open-angle glaucoma
  • MT-COl cytochrome c oxidase subunit 1
  • L2 as risk factors for POAG. Approximately one in four African Americans have these mitochondrial ancestries, which may contribute to their elevated glaucoma risk. These haplogroups are defined in part by ancestral variants in iheMT-RNR2 and/ or MT-COl genes, several of which have prior disease associations, such as MT-COl missense variants that have been implicated in prostate cancer.
  • Example 9 we investigated the role of the m.6150G>A (V83I) polymorphism in the mitochondrial cytochrome c oxidase subunit 1 gene (MT-COl) in African American (AA) primary open-angle glaucoma (POAG). Sanger sequencing and phenotypic characterization of Primary Open- Angle African American Glaucoma Genetics
  • Yeast 2-hybrid (Y2H) library screen quantification of protein-protein interactions by Y2H and ELISA.
  • UBQLN1 involved in Alzheimer's disease (AD).
  • EXAMPLE 1 Subject Recruitment, Phenotyping, And Specimen Collection
  • Cases were defined by the demonstration of characteristic optic nerve defects as well as corresponding visual field loss. Subjects classified as glaucoma suspects were deliberately excluded from this study, as were those few who did not describe their ancestry as Black, African American, African, or mixed race including African ancestry. Approximately 30 ml of peripheral blood was obtained from each subject in EDTA tubes. Blood specimens were frozen at -20 °C before DNA extraction. Alternatively, 2 ml of saliva was collected.
  • the baseline demographics of the POAAGG cohort including phenotyping methods, and criteria for classification as POAG case or control, have been described previously 13 ' 11 .
  • a total of 1,999 African American patients with POAAGG were included in the mitochondrial Pool-seq analysis based on the availability of genomic DNA extracted from whole blood and status as glaucoma case or control (glaucoma suspects were excluded).
  • the majority of cases and controls were female, with a larger proportion of men (44.5%) in the case group than in the control group (33.2%).
  • the control group was significantly younger (61 ⁇ 12 years) than the case group (71 ⁇ 11 years).
  • Elevated IOP is a maj or risk factor for POAG, and as expected, the maximum recorded IOP was significantly higher for the POAG cases (mean 25 mm Hg in cases versus 16 mm in controls, p ⁇ 0.0001), as was neurodegeneration, measured by the cup:disc ratio of optic nerves (mean 0.7 in cases versus 0.3 in controls, pO.0001).
  • DNA was extracted from whole blood using PureGene Gentra kits (Qiagen, Valencia, CA), and the optional RNase digestion step was included. DNA concentrations were measured using the fluorescence-based Quant iT dsDNA Broad-Range (BR) assay kit (#Q-33130, Life Technologies, Foster City, CA). Fluorescence was measured with a Tecan Infinite 200 Pro multimode microplate reader (Morrisville, NC). An aliquot of each sample was normalized to a concentration of approximately 1-2 ng/ ⁇ using SequalPrep Normalization plate kits (Cat # A10510-01 , Life Technologies). The concentrations of the normalized DNAs were verified using Quant iT dsDNA High sensitivity (HS) assay kits (Cat # Q-33120, Life Technologies).
  • HS Quant iT dsDNA High sensitivity
  • mitochondrial genomes had previously been sequenced and were known to correspond to the phylogenetically distant mitochondrial haplogroups (Lie and H3b). These unpooled DNAs were used as controls for the Ion Torrent sequence analysis.
  • the replication group for Sanger sequencing included DNAs collected from saliva samples using OraGene-500 kits (DNA GenoTek, Ottawa, Canada, Cat # OGR-500) and extracted with prepIT reagents (#prepIT-L2P) according to the manufacturer's instructions.
  • EXAMPLE 3 Enrichment For Mitochondrial DNA And Ion Torrent Sequencing Library Construction
  • the PCR reactions contained 1.2 ⁇ 10X reaction buffer, 0.24 ⁇ of 10 mM dNTP mix, 0.48 ⁇ of 50 mM MgCl 2 , 2.4 ⁇ of primer mix with each primer at 2 pmol/ ⁇ , and 3 ⁇ of template DNA (from 1 to 10 ng), 0.1 ⁇ of Platinum Taq, 3.6 ⁇ of 5M betaine, and 1 ⁇ of nuclease-free water.
  • the PCR reactions contained 1.2 ⁇ 10X reaction buffer, 0.24 ⁇ of 10 mM dNTP mix, 0.96 ⁇ of 50 mM MgCl 2 , 2.4 ⁇ of primer mix with each primer at 2 pmol/ ⁇ , and 3 ⁇ of template DNA (from 1 to 10 ng), 0.1 ⁇ of Platinum Taq, 3.6 ⁇ of 5M betaine, and 0.5 ⁇ of nuclease-free water.
  • Sequencing libraries were quantified with quantitative polymerase chain reaction (qPCR) with the Ion Library Quantitation Kit (Cat # 4,468,802, Life Technologies) and diluted to 100 pM.
  • the diluted sequencing libraries (10 pM) were then amplified on Ion SphereTM Particles (ISPs) and enriched for template-positive ISPs using the Ion PGM Template OT2 200 Kit (Cat # 4,480,974), Ion One Touch 2 instrument, and Ion One
  • Variance frequencies estimated with next-generation sequencing on DNA pools were evaluated with Sanger sequencing targeting a region of the N-terminal part of the MT-COl gene (NCBI database, Gene ID 4512, OMIM 516030; see also SEQ ID NO: 2), performed individually for the 1,999 subjects whose DNAs were pooled, as previously described 13 .
  • the actual mean variance frequency at each mtDNA position in this interval for all DNA pools was determined with Sanger sequencing of individual DNAs, and these means were compared to the Pool-seq estimates.
  • a total of 2,157 subjects, 1,308 cases and 849 controls were analyzed; that is, an additional 158 subjects were included.
  • Sanger sequencing targeting the MT-RNR2 gene was performed to ascertain whether the frequencies of several variants detected in this region differed significantly, as predicted by sequencing on DNA pools. Additional Sanger sequencing for a replication group consisting of 1,389 subjects (510 cases, 879 controls) was performed to confirm the enrichment of L2- and Llc2b- associated variants, using POAG cases that were not included in the pooled sequencing group.
  • EXAMPLE 6 DNA Sequence Analysis, Estimation Of Population Frequencies, Data Filtering, And Variant Annotation
  • Counts of the four nucleotides at each position of the alignments were obtained using the count utility of the IGV Tools package from the command line. Output files were compiled in an Excel spreadsheet, and the nucleotide counts were converted to percentages to normalize for read depth. For each position on mtDNA, we computed the frequency of adenine (A), cytosine (C), guanine (G), and thymine (T) and the total frequency of all minor alleles.
  • A adenine
  • C cytosine
  • G guanine
  • T thymine
  • noisy was defined as having a mean background variance frequency of > ⁇ % (on libraries corresponding to the two unpooled individuals' samples).
  • sequencing accuracy As the fraction of concordant calls (agreements/total), where “agreement” means the same genotype was called by both methods above a given threshold, and total refers to the total number of base pairs that were sequenced.
  • the false negative rate on variants was calculated as (disagreements) / (total variants called by Sanger) above a given variance threshold.
  • the false positive rate on variants was defined as (disagreements) / (total variants called by Ion Torrent pooled sequencing).
  • the mitochondrial genome was annotated for disease association using data from the MITOMAP compendium 52 and annotated for haplogroup associations using Build 16 of PhyloTree 79 .
  • Haplogroup population frequencies were estimated using the following positions on mtDNA:
  • Variance frequencies were averaged when more than one position was associated with a particular haplogroup. Variant positions harboring back mutations per PhyloTree, those common to more than one of any of these haplogroups, and those that did not correspond to high-confidence variable positions in the Pool-seq data were excluded from the frequency estimations and estimates of case: control odds ratios (ORs). We did not attempt to estimate the frequency of the common L2a haplogroup due to a lack of variable positions that met these criteria.
  • the TTEST function (two-tailed, unequal variance) in Microsoft Excel was used to compare mean variance frequencies in the pooled versus unpooled samples and used as a filter to differentiate low-frequency population sequence variation from background sequencing noise.
  • the CHITEST function in Excel was used to compare the estimated numbers of cases and controls that have variants at the 381 positions that were selected for the association study, to evaluate the relative likelihood that each difference in the estimated number of cases and controls that have a particular genotype was explainable by chance.
  • the confidence intervals on proportions, for reported and estimated frequencies of African American haplogroups, were determined using the modified Wald method and an online calculator (GraphPad).
  • Tables 4A-4D Masked genotype/phenotype analysis of POAG patients was performed with MTCOl missense mutations (Tables 4A-4D).
  • Table 4A shows the Llc2 (MT-COl double missense) patents had more severe disease vs. matched Lib POAG case controls (no missense).
  • Table 4B shows data demonstrating worse visual field loss for Llc2 MT-COl double missense patients vs. matched Lib POAG case controls (no missense).
  • Table 4C shows data demonstrating higher CDR for Llc2 MT-COl double missense patients vs. matched Lib POAG case controls (no missense).
  • Table 4D shows data demonstrating lower IOP for Llc2 MT-COl double patients vs. matched Lib POAG case controls (no missense).
  • the POAG case group was significantly enriched for three disease-associated missense variants, which define a common African mitochondrial haplogroup and previously implicated in prostate cancer. 59
  • the African mitochondrial haplogroups so associated are Llc2 or a sublineage, Llc2blb.
  • POAG cases harboring these variants had worse disease at significantly lower intraocular pressure than nono-missense cases, after matching for age, gender, family history of POAG and mitochondrial ancestry.
  • mitochondrial lineages that harbor missense mutations in MT-COl e.g., haplotypes Llc2 and Llc2blbl, appear to further elevate African-Americans' existing high risk for glaucoma and result in worse disease.
  • the 43 DNA pools were constructed with DNAs from 1,999 subjects, with 1,217 POAG cases represented in 26 pools, and 782 controls represented in 17 pools.
  • the Ion Torrent sequencing coverage depth was variable across the mitochondrial genome, with the highest coverage depth within the MT-CYB gene and regions of lower coverage in the regions of the MT-ND1, MT-ATP6, and MT-ND6 genes.
  • a representative coverage map from sequencing on one of the pools is shown in FIG. 1.
  • the 43 pools were sequenced to a mean depth of l,931x, with coverage depth on individual libraries ranging from 718x to 3,189*. Assuming that just one individual in a pool of 48 has a homoplasmic mutation at a particular position on mtDNA and proportionate representation of individuals in that sequencing library, the expected variance frequency is 1/48 (2.1%), which would generate approximately 40 mutant base calls at the mean coverage depth. In other words, there was sufficient depth to detect the presence of single rare variants within each pool.
  • haplogroup L3 gray circles, upper left corner, FIG. 2A.
  • FIG. 4 summarizes the phylogenetic relationships among haplogroups that were inferred based on defining sets of variants, with numbering to indicate which groups were enriched in the POAG case or control pools, with potential high-risk groups (OR 1.4-3.3) numbered as (3), i.e, L2, L2c, L2B, L2alc, L2al '2'3'4 and Llc2.
  • the Llc2 lineage is defined in part by a pair of missense mutations in the MT-COl gene, and the pooled sequencing predicted that the subgroups Llc2bla'b, defined by m.2220A>G in MT-RNR2, and Llc2blb, defined by a third missense MT- COl, m.6480G>A, were also present.
  • Llcl lineage which did not emerge as a potential risk factor (FIG. 4).
  • the ten other variants that define Llc2 are either synonymous or within non-coding regions.
  • SNPs single nucleotide polymorphisms
  • the Llc2 lineage is defined in part by the pair of missense variants, m.6150G>A (Val83Ile) and m.62530T (Metl 17Thr), in the MT-COl gene.
  • DNA samples used for pooled sequencing were drawn from the first group of 2,090 POAAGG subjects (1,258 cases, 832 controls). The replication group consisted of an additional 1,389 POAAGG subjects (510 cases, 879 controls).
  • Table 2 shows disease-associated positions on mtDNA harboring common variants in the POAAGG population.
  • the top seven associations have imputed p values of less than 1 ⁇ 10 ⁇ 6 , and they are highly significant assuming the population frequencies inferred from the pooled sequencing are correct.
  • the following seven variable positions are primarily associated with L2 haplogroups, with the exception of m.2220: m.8206G>A (L2 + (non- L)), m.1 1944T>C (L2a'b'c'd + (non-L)), m.2416T>C (L2, L3d3al a, + (non-L)), m. l3590G>A (L2, LOk + (non-L)), m.7274C>T (L2al'2'3'4), m.2220A>T,G (A>T:
  • L4b2bl + (non-L); A>G: Llc2bla'b), and m. l6390G>A (L2, L0a2b + (non-L)).
  • the highest odds ratio among the top hits was estimated for position m.2220 within MT-RNR2, which was predicted to harbor two variants: an Llc2bla'b-associated A>G transition and an L4b2bl -associated A>T transversion.
  • the variance frequency at m.2220 estimated by Pool-seq was 5.2% in the POAG pools versus 0.3% in the control pools.
  • Sanger sequencing on 2,090 individual unpooled DNAs confirmed the existence of both variants but determined that m.2220A>G was about fourfold more common than m.2220A>T in both cases and controls; that is, the predicted association of m.2220 with POAG stemmed primarily from Llc2bl haplogroups, not L4b2bl.
  • the most common disease-associated variant was m.195T>C (rs2857291), located in the non-coding D-Loop and associated with bipolar disease in MITOMAP and with numerous African haplogroups, but there was no evidence for POAG association; the minor allele was observed at 30% frequency in the case and control pools.
  • the missense mutation m. l0086A>G, associated with the L3b haplogroup, has been linked to hypertension-associated end stage renal disease in African Americans 82 ; however, the risk allele was estimated here to be more common in controls (7.9%) than in POAG cases (4.5%), consistent with L3b ancestry not conferring elevated POAG risk among African Americans or being mildly protective (FIG. 4).
  • LHON is caused directly by mutations in mitochondrial DNA (mtDNA), albeit with highly variable expression and penetrance that is modified by the mtDNA genetic background 22 ' 19 .
  • LHON is characterized by the selective degeneration of the retinal ganglion cell layer and the optic nerve that develops in young adults, and approximately 90% of LHON is attributed to three mitochondrial mutations (m.3460G>A, m. l l778G>A, and m.l4484T>C) in the MT-ND4, MT-ND6, and MT-ND1 genes, respectively 84 .
  • LHON cases do not harbor one of these mutations, and this disease has variable penetrance and expressivity.
  • the primary variants that cause LHON have not been linked to POAG, and these three mutations were not detected in either the case or control pools.
  • Other mutations have been described as causing LHON, but their status is less certain, lacking evidence of clear segregation with cases.
  • l0398A>G missense variant has also been proposed as an LHON secondary mutation that may act synergistically to increase the penetrance of LHON when coupled with one of the four primary mutations 70 ' 71 .
  • the m. l0398A>G polymorphism is associated with African haplogroups Llclal, LOdlbl, and L3ela3.
  • TheMT-RNR2 gene which encodes the 16S subunit of the mitochondrial ribosome, harbored seven variants that have associations with POAG that were potentially significant (FIG. 5).
  • RNA variant which had a prior disease association
  • pool-seq was estimated by Pool-seq as 7% in the POAG pools versus 4% in the control pools.
  • Sanger sequencing confirmed that m.3010G>A was common (5.2% in cases), but this proportion was nearly identical in the controls (5.1%; Table 1).
  • a minor allele, m.3010A is associated with cyclic vomiting syndrome with migraine (MITOMAP).
  • MITOMAP cyclic vomiting syndrome with migraine
  • the m.3010G>A variant has also been implicated in sensitivity to linezolid, an antibiotic that can cause optic neuropathy.
  • m.2706 and m.3010 positions were of particular interest because they are reportedly close to the ribosomal peptidyl transfer center, which binds antibiotics such as chloramphenicol and linezolid, and these positions define some mtDNA haplogroups implicated in LHON penetrance.
  • mitochondrial ribosomes may be unintended targets of antibiotics that disrupt bacterial protein synthesis, e.g., linezolid.
  • Pacheu-Grau et al. 57 reported that cybrids harboring the m.301 OA allele had significantly lower amounts of mitochondrial translation products, lower ratios of p. MT-COl /succinate dehydrogenase subunit A, and lower ratios of complex IV quantity to citrate synthase activity after treatment with linezolid.
  • the m.3010G>A variant has arisen repeatedly; it is present in some western European populations and is associated with multiple phylogenetically distant African haplogroups (L0a4, L2alc, and L4bla).
  • the D4 haplogroup found in Asian and Native American populations, is defined by m.3010G>A and two other variants and has been associated with the clinical expression of LHON in Chinese patients 85 .
  • m.3010G>A is one of two variants that define Jl haplogroups, which have been solidly established as increasing the penetrance of the primary LHON mutations 22 ' 19 .
  • this variant appears to stem from African haplogroup L2alc and European haplogroup H65a.
  • our results do not suggest that m.3010G>A confers elevated susceptibility to POAG.
  • m.2706A allele is one of two variants that define the European haplogroup H, reported to be an LHON resistance factor 5 .
  • Most African Americans (97.5% of cases versus 96.6% of controls) were found to possess the m.2706G allele, which appears to be ancestral, but the association with POAG was insignificant.
  • Mitochondrial J cybrids associated with LHON severity, reportedly have significantly decreased mitochondrial protein synthesis, and it was proposed that this might be explained by m.2706G 11 .
  • m.2706G may contribute to African Americans' elevated susceptibility to POAG relative to populations who have European ancestry, in addition to the LHON risk of some J European haplogroups.
  • the MT-RNR2 gene also encodes humanin, a peptide that has been shown to have protective effects in age-related diseases, particularly Alzheimer 24 . Pool-seq did not detect any common variants within the humanin-encoding region, but bidirectional Sanger sequencing on individual DNAs detected an unusual rare mutation, m.2639C>A, in a single glaucoma patient; therefore, it is not impossible that variation in this
  • neuroprotective peptide might be relevant to glaucoma pathogenesis in isolated cases.
  • the patient's mutation is predicted to cause the third amino acid in the humanin peptide, proline ("P3"), to be replaced by threonine.
  • P3 is known to be an essential amino acid for humanin function, and replacement of P3 by alanine abrogates function 8 .
  • This particular patient was enrolled in the POAAGG study at age 89 with bilateral POAG, which was severe stage in the right eye and moderate stage in the left eye. Before cataract surgery, the patient had high myopia (-7.50 sph right eye, -8.00 + 0.75x055 left eye). The axial length of her right eye was 26.68 mm and that of her left eye was 26.72 mm.
  • Heritable and somatic mutations in the MT-COl gene are found at disproportionate rates in patients with prostate cancer 6 .
  • prostate cancer as for POAG, African
  • the m.7389T>C MT-COl missense variant was estimated by Pool-seq to be slightly more common in the cases (21%) than in the controls (17%), but this variant was outside the regions that were also Sanger sequenced; therefore, this result could not be confirmed.
  • missense variants implicated in prostate cancer 59 were determined to be common in the POAAGG population with minor allele frequencies estimated as 2%-4%: m.6150G>A (Val83Ile, absent from dbSNP, haplogroups Llc2, HVlalb), m.6253T>C (rs200165736, Metl l7Thr, H15, D5bl, Mla3bl, M13'46'61, A2am, Llc2), and m.6663A>G (rs200784106, Ile254Val, L2alc). The existence of all four missense variants was confirmed with Sanger sequencing on individual DNAs (FIG.
  • Mitochondrial Variants Linked To The Microbiome the human microbiome has been linked to several common diseases, including an ocular disease, uveitis, which is triggered by microbiota-dependent autoimmunity 34 . It has been proposed that manipulation of the microbiome could be beneficial for uveitis and for glaucoma, possibly by modulating brain-derived neurotropic factor to promote survival of retinal ganglion cells 28 .
  • the oral microbiome has recently been implicated in glaucoma; patients with glaucoma were found to have significantly higher oral bacteria counts than controls 7 .
  • m.16390G>A HVS 1 hypervariable region
  • MT-CYPB m.15784T>C
  • the Pool-seq methodology enabled the rapid and economic integration of whole mitochondrial genome data from a large study population but imposed several limitations. As shown in FIG. 2B, the minor allele frequencies inferred by Pool-seq were subject to experimental error. This is likely to result from variable normalization of mitochondrial DNA within each pool, resulting in disproportionate representation of individuals. The concentration of DNA eluted from the normalization plates was somewhat variable, and total DNA concentration was used as a proxy for mitochondrial DNA concentration, whereas the ratio of mitochondrial to nuclear DNA may have differed from sample to sample, and at least one study has reported that haplogroup background affects mtDNA content 45 .
  • the effective population size of the case and control pools is likely to be smaller than assumed in the calculations that were used to rank potential variant associations (FIG. 5), and the frequencies of variants and their ancestral haplogroups may be over- or underestimated.
  • Another limitation stems from sequencing pools of enzymatically sheared DNAs in conjunction with a short read (about 200-300 bp) technology.
  • the analysis used reduction of the read alignments to simple nucleotide counts to infer population frequencies. This means that potential information about the phase of nearby variants within individual reads was not considered; instead, the presence of individual haplogroups was inferred by the presence of a particular set of defining variants, not the complete haplotypes that might have been obtained from complete mtDNA genome sequencing on individuals. Because many polymorphisms have arisen independently on multiple lineages, using the frequencies of individual variants as a proxy for haplogroup frequency necessarily yields approximations that are most likely to overestimate, but as shown in FIG. 3, the estimates accorded reasonably well with those from an independent study of haplogroup prevalence in other populations of African Americans.
  • the Pool-seq of 1,999 individuals demonstrated that the mitochondrial haplotype distribution of the POAAGG study population is broadly similar to that of other African American populations, with all major African macrohaplogroups represented and with L3, L2, LI, and L0 in decreasing order of frequency, consistent with a preliminary survey of this population 13 .
  • Significant differences were found between haplogroup prevalence in the POAG case and control subgroups.
  • the case group was confirmed to be enriched in Llc2 lineages, which harbor up to three disease-associated missense variants in the MT-COl gene and an MT-RNR2 variant.
  • L2 haplogroups were also implicated as risk factors for POAG; these lineages contain a common variant MT- RNR2 and ones recently linked to the microbiome, whereas the Lib haplogroup was confirmed not to be a risk factor for POAG among our subjects.
  • EXAMPLE 9 The MT-C01 V83I polymorphism is a risk factor for primary open-angle glaucoma in African American men. Introduction
  • POAG Primary open-angle glaucoma
  • POAG African Americans
  • AA African Americans
  • AD Alzheimer's disease
  • PD Parkinson's disease
  • mtDNA small circular genome
  • OXPHOS oxidative phosphorylation pathway
  • COl protein is localized to the mitochondrial inner membrane, where it is an essential component of Complex IV, the final enzyme complex in the electron transport chain, which transfers electrons from reduced cytochrome c to molecular oxygen, producing water.
  • Complex IV is essential for all oxygen-based life, and is the target of some potential neuroprotective interventions that may protect RGCs from death as a result of glaucoma. For example, methylene blue has been shown to protect rat RGCs from toxic insults to mitochondria, such as rotenone, which can cause optic neuropathy 16 .
  • V83I COl mutation is of particular interest, because it lies within a region of
  • COl that interacts with amyloid beta 2 the product of the APP gene which figures prominently in the pathology of AD. COl has also been reported to interact with alpha- synuclein, which is found in Lewy bodies and associated with PD and Lewy body dementia. Substantial evidence indicates that ⁇ may be involved in POAG pathology, and that AD and POAG may share etiologies 15 20 . There is overlap between AD and POAG pathologies, for example glaucoma is associated with ⁇ accumulation in the retina with decreased ⁇ in the vitreous, and AD is associated with retinal nerve fiber layer and visual cortex pathology 29 .
  • Subjects were age 35 or older and self- identified as Black, African American or African ancestry and were classified as glaucoma case or control by glaucoma specialists. Subjects classified as glaucoma suspects were excluded from all analyses. Research was approved by an institutional review board at the University of Pennsylvania, and was conducted in accordance with the Declaration of Helsinki 1 .
  • Ml 17T double missense mutations, associated with mtDNA haplogroup Llc2.
  • the coding sequence of the human Coxl fragment (aa. 41-101 of GenBank accession number gi: 251831109 and NCBI Ref YP_003024028.1 and SEQ ID NO: 3) was PCR-amplified and cloned in frame with the LexA DNA binding domain (DBD) into plasmid pB27 (N-LexA-Coxl-C), derived from the original pBTMl 16 80 . The DBD construct was checked by sequencing the entire insert. Hybrigenics' reference for this "bait" is hg3681vl_pB27. Codon usage of the Coxl insert was optimized for yeast expression and adjusted to prevent amino acid changes from differences in the human mitochondrial and yeast nuclear genetic codes.
  • the "prey" fragment for the human APP was aa. 672-713, GenBank accession number gi: 228008403.
  • the prey fragment was cloned in frame with the Gal4 Activation Domain (AD) into plasmid pP7, derived from the original pGADGH 8 .
  • the AD construct was also checked by sequencing.
  • the prey Hybrigenics reference is hgx3682vl_pP7.
  • Bait and prey constructs were transformed in the yeast haploid cells: L40 Gal4 (mata) strains.
  • the diploid yeast cells were obtained using a mating protocol with both yeast strains 18 .
  • These assays are based on the YHGX13 (Y187 ade2-101 ::loxP-kanMX- loxP, mat HIS3 reporter gene (growth assay without histidine).
  • the following interaction pairs were tested: Smad3 / Smurfl (Hybrigenics' positive control, empty LexA bait vector / empty prey vector (negative control) , empty LexA bait vector / AD-APP (negative control) , LexA-Coxl/ empty prey vector (negative control), LexA-Coxl / AD-APP.
  • the coding sequence of the human Coxl wt and Cox-1 mut fragments (aa. 41-101 of SEQ ID NO: 3) were synthesized and cloned in frame with the Gal4 DNA binding domain (DBD) into plasmid pB66 (N-Gal4- Bait-C), derived from pAS2 18 .
  • the DBD constructs were checked by sequencing the entire insert.
  • Hybrigenics reference for those baits are hgx3681vl_pB66 (for Coxl wt) and hgx3915_pB66 (for Cox-lmut).
  • the prey fragments for the human SCN1A (aa.
  • GenBank accession number gi:260166632 and UBQLN1 were extracted from the ULTImate Y2HTM screening of Coxlwt ((aa.41-101 of SEQ ID NO: 3) against the Human Adult Brain cDNA library).
  • the prey fragments are cloned in frame with the Gal4 Activation Domain (AD) into plasmid pP6, derived from the original pGADGH 8 .
  • the AD construct was checked by sequencing.
  • the prey Hybrigenics references are hgx3681vl_pB27_B7 (SCN1A) and hgx3681vl_pB27_A346 (UBQLN1).
  • the pP7 prey plasmid used in the control assay was derived from the pP6 plasmid.
  • Bait and prey constructs were transformed in the yeast haploid cells, respectively
  • CG1945 (mata) and YHGX13 (Y187 ade2-101 ::loxP-kanMX-loxP, mat) strains.
  • the diploid yeast cells were obtained using a mating protocol with both yeast strains 18 . These assays are based on the HIS3 reporter gene (growth assay without histidine).
  • Interaction 6) Gal4- Cox-1 mut / empty prey vector (negative control), 7) Gal4- Cox-1 mut / AD- SCN1A, 8) empty Gal4 bait vector / AD- UBQLN1 (negative control) , 9) Gal4- Coxl wt / AD- UBQLN1 (interaction), 10) Gal4- Cox-1 mut / AD- UBQLNl(interaction) .
  • Interaction pairs were tested in triplicate in form of streaks as three independent clones from each diploid were picked for the growth assay.
  • the DO-2 selective medium lacking tryptophan and leucine was used as a growth control and to verify the presence of the bait and prey plasmids.
  • the DO-3 selective medium without tryptophan, leucine and histidine selects for the interaction between bait and prey.
  • Sandwich ELISA assays were performed with biotinylated synthetic peptides to confirm the previously reported affinity of ⁇ (1-42) peptide for an N-terminal region of COl, and to quantify the effect of the V83I amino acid replacement on this interaction.
  • Purified APOE protein was used as a positive control for interaction with ⁇
  • TBX3 protein OraGene
  • scramble ⁇ peptide were used as negative controls for evaluating interaction with ⁇ .
  • POAAGG cohort 1 found to have nominal association with POAG 14 (m.6150G>A (V83I), m.6253T>C (Ml 17T), m.6480G>A (V193I)). Seven of these twenty-four variants have been implicated in prostate cancer 59 , in addition to the three targeted variants (Table 6).
  • SIFT analysis classified six of the missense variants (L20M, N46D, N46T, T146A, F148L, L215F) as deleterious mutations, however with a total of five observations in POAG cases and four in controls. Only two of the variants classified as deleterious by SIFT, L20M and T146A, were not observed in controls, but each was observed in only one case.
  • PC prostate-cancer associated i MITOMAP
  • MRCA most recent common ancestor
  • FIG. 6 depicts the mtDNA family tree, and indicates the approximate locations of common synonymous and non-synonymous variants MT-COl variants in evolutionary time.
  • the pair of missense mutations, V83I and Ml 17T defines the Llc2 haplogroup, which arose more than 50,000 years ago.
  • the third POAG-associated missense, V193I occur on an offshoot of Llc2, Llc2blb.
  • V83I also is found in a subset of POAAGG subjects belonging to the L2c haplogroup. All three variants may occur on non- African haplogroups, although none of these are common in the POAAGG cohort.
  • the non-synonymous variant, m.6548C>T is uniquely associated with Lib haplogroups, which share LI ancestry, but lack the three missense variants that occur on Llc2 lineages.
  • the V83I patients had significantly lower IOP.
  • the finding that V83I patients had worse visual function was surprising because IOP is a major risk factor, suggesting that V83I patients might be more vulnerable to elevated pressure.
  • Mean central comeal thickness (CCT) was higher in the V83I group, but this difference was not significant.
  • V83I patients mostly having the Ll c2 African haplogroup, had significantly worse visual function than V83 WT patients despite lower IOP.
  • the two groups were also matched for age, gender, and family history of POAG (yes/no) in order to control for these potential confounders.
  • the chart review was conducted with a trained glaucoma specialist.
  • the Ll c2 group had worse disease, as assessed by ICD9 codes for glaucoma severity that had been previously entered in EPIC, with 52% of Ll c2 having "severe” disease as opposed to 9.5% of Lib (Table 9).
  • Table 9 Glaucoma severity codes (ICD9) of matched Llc2 vs. Lib POAG patients.
  • Yeast 2-hybrid (Y2H) test of COl / ⁇ interaction and cDNA library screen for COl interactors.
  • V83I V83I / Ubqll interaction was equally strong, so V83I did not affect this interaction. Both interactions resisted up to 10 mM of 3 AT (data not shown).
  • V83I reduced the interaction by 92%, to a level that was comparable to the negative control with TBX3 protein.
  • wild type and mutant (V83I) COl peptides had similar lack of affinity for the scrambled ⁇ negative control peptides, suggesting that the interaction of COl and ⁇ is specific, and is disrupted by V83I.
  • V83I patients had worse visual function and degeneration of the optic nerve, despite significantly lower IOP, even after controlling for age, gender, family history of POAG, African-American ancestry and LI mtDNA haplogroup. This evidence is consistent with the proposal that this missense mutation may contribute to POAG pathogenesis, and these patients may more vulnerable to POAG in general, and at lower IOP than other African Americans.
  • the potential involvement of AD genes APP, and UBQLNl, implies that AD / dementia or other intellectual disabilities could be more likely to be co-morbidities with POAG in these patients, although this can't be concluded from our results.
  • the survey of ICD9 codes related to AD / dementia found the proportion of V83I POAG cases with intellectual disabilities was higher than non-V83I cases, however this was based on only one observation in V83I cases.
  • the three missense mutations may occur either alone, for example V83I on an L2c mtDNA background, or in combination, for example V83I and M117T on Llc2-related haplogroups, which may also carry VI 931 (FIG. 6). All three of the POAG-associated missense mutations have been proposed as potentially pathogenic in studies of other diseases, and might relate directly to POAG pathogenesis, either alone or in combination. a) V83I (m.6150G>A)
  • V83I is the strongest candidate for being causal in addition to a marker for POAG susceptibility.
  • the association of V83I with POAG in African American men was stronger than for Ml 17T or V193I. This is because the association of V83I with POAG stemmed not only from Llc2 subjects, who also have M117T, but was some L2c subjects who have V83I, but lack the other two missense mutations.
  • V83I is also the only mutation within the ⁇ binding region, and we showed that this replacement greatly abrogated this interaction in the ELISA assay, which suggests V83 may be critical for ⁇ interaction.
  • Met74Thr is located only 9 residues from V83. This mutation was not observed in POAAGG patients, however it also within the ⁇ -binding region, and was shown to be functional, causing resistance to statin induced apoptosis 72 , increase reactive oxygen production, and enhanced cellular proliferation 6 . Met74Thr cybrid cells had a significantly faster doubling time than wild type cells. So it is reasonable to speculate that V83I or other nearby mutations might also directly affect proliferation, albeit negatively, thereby promoting neurodegeneration.
  • the retrieval of UBQLN1, an AD associated gene with neuroprotective function, by Y2H cDNA library screen is more evidence that this region may function, perhaps in autophagy. However, we found no evidence that V83I residue is involved in the interaction with UBQLN1.
  • M117T was observed only on Llc2 haplogroups , which also carry V83I, the (weaker) association of Ml 17T with POAG implicates V83I.
  • Cytochrome c oxidase is a "bigenomic" protein machine with components encoded by both the nuclear and mitochondrial genomes, and the mtDNA-encoded proteins, e.g. COl, are functionally constrained by the requirement to interact with the nuclear proteins.
  • Ml 17T has been predicted to be pathogenic based on 3D structural modeling that suggests it participates in these protein-protein interactions, and might influence the stability of Complex IV 51 .
  • V193I (m.6480G>A)
  • the VI 931 variant was originally detected in a patient having cytochrome c oxidase deficiency, and was proposed to be pathogenic, based on its absence in 300 controls 9 .
  • VI 931 is now known to be a common polymorphism associated with the Llc2blb African haplogroup, which also carries V83I and M117T 79 and with non-African haplogroups, but it is still possible that it may be deleterious.
  • V193I has been proposed as a "helper" variant that acts in synergy with the primary LHON mutation m.11778G>A in a Chinese family 10 .
  • V193I like Ml 17T, has also been predicted to be pathogenic by 3D modeling and may affect interactions with Complex IV proteins encoded by nuclear genes 51 .
  • UBQLNl ubiquilin 1 gene
  • UBQLNl has not been linked to POAG, but was among a small number of anti-apoptotic genes found to be up-regulated by platelet derived growth-factor CC, which protects retinal ganglion cells from death, in optic nerve crush-injured mouse retinae 76 .
  • platelet derived growth-factor CC which protects retinal ganglion cells from death, in optic nerve crush-injured mouse retinae 76 .
  • the ability of ⁇ to interact 1) with itself, 2) with COl and 3) with UBQLNl suggests that all three interactions might occur simultaneously, with ⁇ as either a soluble monomer or neurotoxic oligomer.
  • V83I did not disrupt the strong interaction of COl with UBQLNl in the Y2H system (data not shown), but did disrupt the interaction of COl with ⁇ (FIG. 11) in the ELISA assays. This suggests that V83I is not directly involved the UBQLNl interaction, and that COl might interact with UBQLNl and ⁇ cooperatively, as opposed to competitively. We hypothesis that ⁇ may bridge and facilitate this interaction, whereas V83I may prevent the binding of UBQLNl with monomeric ⁇ bound. It is possible that oligomerization of ⁇ might also be disruptive to the interaction of UBQLNl and COl, if this requires ⁇ to be in soluble monomeric form.
  • Glaucoma and Alzheimer's Disease have several features in common, including increasing incidence with age, and loss of specific neuronal subpopulations.
  • Down's syndrome patients have an extra copy and elevated expression of the APP gene, and frequently develop Alzheimer's at a young age, and glaucoma in their teens. This is consistent with overlapping molecular etiologies that involve amyloid ⁇ ( ⁇ ).
  • ⁇ is known to be present in inside mitochondria, and intramitochondrial ⁇ may directly cause neurotoxicity and mitochondrial dysfunction, including impairment of OXPHOS and interaction with mitochondrial proteins 58 .
  • ⁇ is known to localize to the mitochondrial matrix and inner mitochondrial membrane so there is a potential for ⁇ to interact directly with Complex IV.
  • ⁇ -induced mitochondrial function involves the forkhead box 03 a FOX03A gene, which is present in neuronal mitochondrial binds mtDNA and leads to decreased MT-COl expression, so the influence of ⁇ on Complex IV function may be indirect. However the mechanism(s) of ⁇ -induced neurotoxicity are not well understood.
  • Glaucomatous changes might arise from multiple mechanisms of mitochondrial damage, some of which might involve ⁇ . These factors might act alone, or in concert with elevated IOP. For example, elevated pressure causes release of cytochrome c and OPA1 release in RGC-5 retinal ganglion cells, resulting in apoptotic cell death 4 .
  • Mitochondrial damage from IOP was also demonstrated in glaucomatous DBA/2J mice, where elevated IOP resulted in mitochondrial fission, matrix swelling, cytochrome c release, and a moderate reduction of expression of MT-COl mRNA 44 .
  • Oxidative stress caused by ozone exposure has been shown to increase ⁇ -42 production and accumulation of ⁇ -42 in mitochondria, with colocalization of OPA1, ⁇ and COl .
  • Another potential mechanism for mitochondrial damage in AD and POAG is an increased rate of somatic mutation, and this might involve ⁇ and possibly estrogen.
  • V83I, Ml 17T and V193I with POAG in African American men may be explained by unknown factors associated with the corresponding mtDNA ancestries (primarily Llc2 and L2c lineages). Alternatively, other mtDNA or nuclear variants or factors linked to these ancestries may explain the strong associations with POAG in AA men.
  • mtDNA ancestries primarily Llc2 and L2c lineages.
  • other mtDNA or nuclear variants or factors linked to these ancestries may explain the strong associations with POAG in AA men.
  • V83I common germline mtDNA polymorphisms, such as the V83I mutation in COl, play a role in susceptibility to POAG, particularly in men, and impact disease severity.
  • V83I is predominately associated with African mitochondrial ancestries (Llc2, L2c and others), this variant contributes to African Americans higher POAG risk relative to Americans with European ancestries.
  • the m.6150G>A (V83I) variant in COl was present in 6.5% of male AA POAG patients, but only 1.1% of male AA controls, and this difference was extremely significant.
  • V83I disrupts a reported protein-protein interaction with ⁇ , which could be significant in light of proposals that the etiology of POAG may overlap with other neurodegenerations, particularly AD, with the APP gene common to both.
  • a second gene, UBQLNl interacts with COl, and that this gene, which is a potential therapeutic target for AD 75 , is relevant to POAG too.
  • Interventions to treat optic neuropathies by supporting mitochondrial function are under development 27 .
  • Patients having V83I or other COl missense mutations may benefit from emerging neuroprotective therapies that directly support mitochondrial respiration by donating electrons to cytochrome c oxidase, e.g. exposure to near infrared light, or treatment with methylene blue, which preferentially enters neuronal mitochondria after systemic administration 25 .
  • emerging neuroprotective therapies that directly support mitochondrial respiration by donating electrons to cytochrome c oxidase, e.g. exposure to near infrared light, or treatment with methylene blue, which preferentially enters neuronal mitochondria after systemic administration 25 .
  • Recently metformin and the antioxidant mitotempo have been shown to be protect human neural stems and cultured mouse neurons, respectively, against ⁇ -induced mitochondrial dysfunction 12 7 .
  • emerging gene-based therapies now offer the ability to correct deleterious germline alterations in mtDNA.
  • POAAGG The POAAGG study population consists of self-identified Blacks (African Americans, African descent, or African Caribbean), aged 35 years or older. POAAGG subjects were recruited from the University of Pennsylvania from the Scheie Eye Institute and its satellite locations in Philadelphia. All subjects provided informed written consent, in accordance with the tenets of the Declaration of Helsinki, under an IRB-approved protocol. Following visual examination by a glaucoma specialist, subjects were classified as case, suspect, or control (POAAGG). Cases were defined by demonstrating
  • Plasma samples were collected by venipuncture in 10 ml purple top tubes with EDTA anticoagulant. These samples were frozen at -20 degrees prior to DNA isolation.
  • subjects were asked to refrain from drinking or eating prior to donating specimens.
  • Two milliliters of saliva per subject were collected in Oragene DISCOVER (OGR-500) self-collection kits (DNA Genotek, Canada).
  • the saliva specimens were mixed with stabilizing reagent within the collection tubes per manufacturer's instructions, and these were stored at room temperature until DNA extraction.
  • Genotek, Canada and precipitated with ethanol according to manufacturer's instructions.
  • the saliva DNA samples were RNAse treated by double digestion with RNaseA and RNase T and re-precipitated using ethanol according to manufacturer's instructions.
  • the concentrations of DNA from blood and saliva samples were determined using the fluorescence-based Quant iT dsDNA Board-Range (BR) assay kit (cat # Q-33130, Life Technologies, CA). Fluorescence was measured with a Tecan Infinite M 200 Pro multimode microplate reader (Tecan, NC). Genotyping
  • a 25 ⁇ aliquot of each sample was plated for array-based high throughput genotyping.
  • the 3,684 DNA samples were genotyped using MEGA (Multi ethnic genotyping array) V2 (EX) consortium chip on the Infinium iSelect platform by Illumina FastTrack Services (Illumina, San Diego, CA).
  • the MEGA chip has 1.8 million successfully genotyped SNPs and 1294 of those are Mitochondrial SNPs.
  • the genotype calls were generated using the Genome Studio genotyping module (GT). Cluster optimization, reproducibility analysis for paired samples were performed as per standard practices at Illumina FastTrack services.
  • the p-values for the odds ratios were calculated using chi-square tests using r software package and Graph pad prism V6. All the phenotype data was analyzed using linear regression, with generalized estimating equations (GEE) used to account for the correlation between the eyes of the same person. All the p-values for multiple comparisons (in both the chi-squares and linear regressions) were adjusted for with the Hochberg step-up method.
  • GOE generalized estimating equations
  • the genotypes from the 1293 mitochondrial probes were extracted using Genome Studio software. Of the 1293 probes, 234 were added as a part of the custom content based on the preliminary pooled sequencing of 2000 subjects Mitochondria (MolVis). The frequency of each of the 1293 positions was calculated and classified to different bins based on the criteria below.
  • a total of 275 positions that have 1 % and above variance frequency (data not shown) are used for further disease association analysis.
  • Haplogrep2 is a public software that generates haplogroups automatically using a clustering algorithm based on phylotree classification.
  • the haplogroup classification was primarily based on the Kulczynksi distance, a dissimilarity metric feature which implements 3 other algorithms based on Jaccard index, Hamming distance and kimura 2-parameter distance also runs in back ground to estimate the distance concordance.
  • Table 12 The 29 haplogroups which are common in POAAGG cohort along with the rare haplogroups collapsed under the macro haplogroups. Significant groups denoted by **.
  • H5a3a 0.85 (0.55, 0.44 1.17 (0.51, 0.71 0.77 (0.46, 0.32
  • FIG. 8 The distribution of these common haplogroups (FIG. 8) shows the distribution of the 29 haplotypes and 6 combined groups where the remaining rare haplotypes are grouped based on macrohaplogroup.
  • FIG. 9 the associated risk based on the odds ratio is represented as protective, moderate and potential high risks groups.
  • the gender distribution is with overlapping lines, one solid and one broken line.
  • 275 positions are selected for POAG association analysis.
  • the odds ratio and p values are calculated separated by gender as we have seen a gender effect with the haplogroup analysis above.
  • Table 13 shows the top 20 mtDNA positions with the significant p value.
  • the haplogroup assignment for the top positions show the same significant bins from the haplogroup association above.
  • Mitochondrial variants may influence the phenotypic manifestation of Leber's hereditary optic neuropathy-associated ND4 Gl 1778A mutation. J Genet Genomics. 2008;35:649-655.
  • Mitochondrial DNA haplogroup Y is associated to leigh syndrome in Chinese population. Gene 2013; 512:460-3. 23111160 He Y, et al. Mitochondrial complex I defect induces ROS release and degeneration in trabecular meshwork cells of POAG patients: Protection by antioxidants. Invest Ophthalmol Vis Sci 2008; 49: 1447-58. 18385062 Hernandez-Zimbron LF, et al. Amyloid-beta peptide binds to cytochrome C oxidase subunit 1. PLoS One. 2012;7:e42344. Hernandez-Zimbron LF, Rivas-Arancibia S. Oxidative stress caused by ozone exposure induces beta-amyloid 1-42 overproduction and mitochondrial accumulation by activating the amyloidogenic pathway. Neuroscience.
  • Mitochondrial polymorphism A10398G and haplogroup I are associated with fuchs' endothelial corneal dystrophy. Invest Ophthalmol Vis Sci 2014; 55:4577-84. 24917144 Lloyd RE, McGeehan JE. Structural analysis of mitochondrial mutations reveals a role for bigenomic protein interactions in human disease. PLoS One.

Landscapes

  • Chemical & Material Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Organic Chemistry (AREA)
  • Engineering & Computer Science (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Zoology (AREA)
  • Wood Science & Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Genetics & Genomics (AREA)
  • Immunology (AREA)
  • Neurology (AREA)
  • Pathology (AREA)
  • Biochemistry (AREA)
  • Biotechnology (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Molecular Biology (AREA)
  • Biomedical Technology (AREA)
  • Hospice & Palliative Care (AREA)
  • Microbiology (AREA)
  • Neurosurgery (AREA)
  • Biophysics (AREA)
  • Psychiatry (AREA)
  • General Chemical & Material Sciences (AREA)
  • Medicinal Chemistry (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Pharmacology & Pharmacy (AREA)
  • Animal Behavior & Ethology (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Oncology (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)

Abstract

A method for evaluating a subject's risk of developing glaucoma, or having glaucoma, or having glaucoma that will progress in severity comprising determining from a biological sample from a subject whether the patient has a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl. In another embodiment, the method involves determining from a biological sample the occurrence of one or more mutational nucleotide variants in the mtDNA that define a suspect mitochondrial haplogroup. In still another embodiment, the method involves treating the subject having the indicated mitochondrial haplogroup and/or mutational variants with a therapeutic or dietary regimen designed to support mitochondrial function

Description

METHODS FOR SCREENING AND DIAGNOSING GLAUCOMA
INCORPORATION-BY-REFERENCE OF MATERIAL SUBMITTED IN
ELECTRONIC FORM
Applicant hereby incorporates by reference the Sequence Listing material filed in electronic form herewith. This file is labeled " 16-7895PCT_ST25.txt " , has a size of 28Kb and is dated April 19, 2017.
STATEMENT OF GOVERNMENT INTEREST
This invention was made with government support under Grant No.
RO1EY023557, Contract Nos HHSN260220700001C and HHSN263210200001C awarded by the National Institutes of Health. The government has certain rights in the invention.
BACKGROUND OF THE INVENTION
Primary open-angle glaucoma (POAG) is a progressive degeneration of the optic nerve with characteristic clinical findings that correspond to patterns of visual field loss. POAG is the major cause of irreversible blindness worldwide. POAG is especially prevalent in African Americans, who are four times more likely than Caucasians to develop the disease and succumb at an earlier age77.
Current methodologies for diagnosing or determining the risk of a patient for having glaucoma is limited to observation of primarily eye pressure by an
ophthalmologist. However, certain forms of glaucoma are characterized surprisingly by a decrease in eye pressure, particularly in persons of African descent. Therefore there is a need in the art for methods that would enable easy diagnosis of the risk of, or presence of, glaucoma.
SUMMARY OF THE INVENTION
In one aspect, a method for evaluating a subject's risk of developing glaucoma comprises determining from a biological sample from a subject whether the patient has a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
In another aspect, a method for evaluating a subject's risk of developing glaucoma comprises determining from a biological sample from a subject the existence of one or more mutational variant within a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl. In another aspect, a method for evaluating the risk of progression of disease in a glaucoma patient comprises determining from a biological sample from the patient whether the patient has a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
In another aspect, a method for evaluating the risk of progression of disease in a glaucoma patient comprises determining from a biological sample of the patient the existence of one or more mutational variant within a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
In another aspect, a method for evaluating the risk of progression of disease in a glaucoma patient comprises determining from a biological sample from the patient whether the patient has a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
In another aspect, a method for diagnosing glaucoma in a patient comprises determining from a biological sample of the patient the existence of one or more mutational variant within a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
In yet another aspect, these methods further involve both diagnosing or evaluating the risk or progression of glaucoma and treating the patient, so diagnosed with therapeutic or dietary regimens to support mitochondrial health.
These and other embodiments and advantages of the invention are described in more detail below.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a graph showing complete coverage of mtDNA obtained by Pool-seq. A representative IGV coverage map from one of the 43 DNA pools is shown, with the positions of the major mitochondrial genes, and numbering corresponding to the rCRS mtDNA reference sequence. Positions with a variance frequency higher than 3% are indicated by darker bars to indicate the base composition of the variable, with coverage depth indicated by the height of each bar.
FIG. 2A is a graph showing that Pool-seq identified hundreds of variable positions on mtDNA, mostly associated with one or more African haplogroups. The mean variance frequency at each of the 16,549 positions on mtDNA is plotted as a percentage for all 43 pools, representing a total of 1,999 individuals. Variant positions that are associated the most common African-American mitochondrial macro haplogroups are indicated with solid circles. Open circles represent variants associated with multiple haplogroups, or subgroups of the major groups listed in the legend.
FIG. 2B is a graph comparing individual Sanger sequencing results on 1,999 individuals for a 605 bp region within the MT-COl gene to Pool-seq results on the same group. The location and population frequencies from Sanger sequencing (solid diamonds) are plotted with those inferred by Pool-seq (open circles), and numeric labels indicate the position on the mtDNA reference sequence. "FP" denotes two of the positions, 6420 and 6442, that were categorized as noisy and excluded from analysis as false positives.
FIG. 3 is a bar graph showing that the distribution of major mtDNA haplogroups inferred for the POAAGG study population is consistent with African- American ancestry. For POAAGG subjects, the frequency of each haplogroup was estimated by the mean variance frequency from a set of associated positions on mtDNA by Pool-seq. The estimates for POAG cases and controls are compared to those from an independent multicenter study of African- American haplogroups. Above each haplotype indication are a set of 3 bars. Each set is organized from left to right as African American, Controls and POAG cases. Error bars represent 95% confidence intervals.
FIG. 4 is a schematic plotting data generated from Pool-seq on POAG cases versus controls that implicated the Llc2 and L2 branches of the mitochondrial phylogenetic tree. The relationships among haplogroups detected in the POAAGG study population are represented in schematic form, with branch lengths representing approximate divergence times from the most recent common ancestor (MRCA), with branching order adapted from Behar et al. and Schlebush et al.9'67 Numbering from 1 to 3 adjacent the lines indicates the degree of imbalance observed in the POAG versus case pools, expressed as odds ratios (OR) from estimated population frequencies for one or more variants associated with that lineage. Mitochondrial ancestry implicated in higher POAG risk (case: control OR >1.4) is indicated by the number 3, moderate risk is indicated by the number 2, and mitochondrial ancestry that may be protective is indicated by the number 1. All depicted haplogroups are African with the exception of branches R, M and N, which are ancestral to Asians and European populations. Haplogroup Llc2, is defined in part by the disease-associated missense variants m.6150G>A and 6253T>C in. MT-COl, with subgroups Llc2bla'b also having the m.2220T>C variant in MT-RNR2. The L2 lineages are associated with variant m.2416T>C in M7-RNR2 and other variants. The m.3010G>A variant in MT-RNR2, linked to linezolid sensitivity, has arisen on branch L2alc, and a recent European haplogroup, H65a (not shown).
FIG. 5 is a schematic showing the association of 381 variable positions on mtDNA with POAG in African- Americans. The number of individuals estimated to differ from the rCRS reference sequence in the POAG case pool (n=l,217) was compared to the number of estimated variant individuals in the control pool (n=782) for each variable position, the significance of the predicted difference was assessed using a Chi-square test, and the negative logio of the resulting p value is plotted. Variable positions annotated as having a disease association in MITOMAP or as "pathogenic" or "likely pathogenic" by the NCBI variant viewer are indicated by markers with solid light gray fill with dark rim. Selected markers are identified by their coordinates on the rCRS reference sequence.
FIG. 6 is a schematic plot showing Sanger sequencing on MT-COl implicated the Llc2 haplogroup. Odd ratios (OR) from sequencing 1,308 POAG cases vs. 849 controls, and phylogenetic relationship of the observed common variants and associated haplogroups are shown. Evolutionary relationship and POAG associations of common MT-COl variants are also shown. The relative frequencies of the variants in cases :
controls are indicated by numbering of the OR from 1 to 3 with the number placed adjacent the appropriate relationship line as in FIG. 4. Amino acid changes are indicated for missense variants, and the three missense variants with POAG association are highlighted with bold font. The m.6150G>A (V83I) variant also occurs on additional lineages, e.g. L2c, (not shown).
FIG. 7 is a graph showing the affinity of Wt-MTCOl (Val83) peptide for Αβ 1-42 confirmed by ELISA, whereas Mut-MTCOl (Val83Ile) decreased interaction by 92%.
FIG. 8 is a graph showing the distribution of the haplogroups which are common separated by the case and control frequencies. Each haplogroup labels two bars, the leftmost being the case bar and the rightmost being the control bar for each set. The haplogroups which are significantly associated with POAG are denoted by arrows below the label, the protective groups have light gray and risk has dark gray arrows. The macro haplogroups represented by grey bars below the chart.
FIG. 9 is a phylogenetic charts showing the relationship of the 29 common haplogroups represented with the divergence time. The odds ratio for each haplogroup was represented using the numbers represented the Odds Ratios 1-3 alongside the appropriate line, reported as number for the broken line for male / number for the solid line for female, i.e., OR (l)/OR(3), etc. The haplogroups that were significant were identified by *.
FIG. 10 is a schematic of the predicted domain structure of the COl protein and subcellular localization of three POAG-associated missense variants.
FIG. 11 is a graph showing the interaction of mutant and wild type COl peptides with Αβ or scrambled Αβ in ELISA assay. APOE protein was the positive control, and TBX3 protein and scrambled Abeta peptide (ScrAB) were negative controls.
DETAILED DESCRIPTION OF THE INVENTION
The compositions and methods described herein provide means for diagnosing or detecting the existence or risk of, or monitoring the progress of, glaucoma in a patient by identifying the mtDNA haplogroups or haplotypes and/or the presence of certain genetic mutations defining certain mtDNA haplogroups or haplotypes in the patient.
One such method for evaluating a subject's risk of developing glaucoma or having progressively severe glaucoma or diagnosing glaucoma in a subject comprises determining from a biological sample from a subject whether the patient has a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl. In another embodiment, the method involves determining from said sample the existence of one or more mutational variants that defined the suspect mitochondrial haplogroup. As demonstrated herein, these mitochondrial haplogroups are indicative of greater risk of developing glaucoma and/or of progression to more severe glaucoma.
Unless defined otherwise in this specification, all scientific and technical terms used herein have the same meaning as commonly understood by to a person of skill in the fields of biology, biotechnology and molecular biology and by reference to published texts, which provide one skilled in the art with a general guide to many of the terms used in the present application. However, for clarity, the following terms are defined as follows.
As used herein, the term "patient" or "subject" as used herein means a male or female human, or an animal, e.g., a domestic animal or pet, and animals normally used for clinical research. In one embodiment, the subject of these methods and compositions is a human. In one embodiment, the subject of these methods and compositions is a human of African ancestry.
As used herein, the term "glaucoma" refers to a neurodegenerative illness of the eye. Glaucoma pathogenesis likely involves mitochondrial dysfunction55 . It is possible that a subgroup of patients with glaucoma may demonstrate a mitochondrial optic neuropathy but one that has more complex genetics than the mitochondrial disease Leber hereditary optic neuropathy (LHON). Primary open-angle glaucoma (POAG) has been associated with a large number of nuclear genes, many of which are involved in mitochondrial function46. POAG lymphoblasts have been shown to exhibit a complex I defect in the mitochondrial oxidative phosphorylation pathway, with decreased rates of respiration, which could confer an increased susceptibility to cell death on retinal ganglion cells48. Highly efficient systemic mitochondrial function was recently shown to be protective against glaucomatous optic neuropathy in those with elevated intraocular pressure (IOP)47. Mutations in the 16,569 bp mtDNA are well established as the primary cause of LHON. However, to date, the role of heritable mitochondrial mutations in POAG, if any, is unclear.
As used herein, the abbreviation "mtDNA" means mitochondrial DNA. All living humans share a most recent common ancestor (MRCA), an African woman who is estimated to have lived approximately 194,000±33,000 years ago23. The mitochondrial genome is extremely informative regarding maternal ancestry because of the genome's matrilineal inheritance and lack of genetic recombination. The human mitochondrial family tree has been characterized extensively and subdivided into thousands of branches (hyperlink: PhyloTree). The human mtDNA sequence and nucleic acid positions are identified herein using the Revised Cambridge Reference Sequence (rCRS) of the Human Mitochondrial DNA (Genbank NC_012920 gi:251831106). This 16,569 bp sequence is also reproduced as SEQ ID NO: 1.
By "mtDNA haplogroups" or "haplogroups" as used herein is meant the major ancestral groupings of mitrochondrial DNA defined by collections of variants scattered throughout the mitochondrial genome that are inherited together and have accumulated over evolutionary time. The oldest mutations define the major macrohaplogroups and the deepest roots of the tree. The human mtDNA phylogeny is divided into major haplogroups designated L0, LI, L2, L3, L4, L5, and L6, which are further divided into smaller subgroups. Contemporary non- African haplogroups are derived from L3, with major branches designated M and N, and associated with ancient diasporas from Africa that occurred approximately 60,000 to 70,000 years ago23'9. A prior survey of 44 subjects determined that L3, L2, LI, and L0 groups are all present in the POAAGG study population, as are non-African haplogroups13. A study of a Saudi Arabian population3 found a positive association of POAG with African (L) haplogroups and evidence for a protective effect of the Eurasian haplogroup N, but studies of Arab4, Ghanaian2, and northern European5 populations failed to find significant associations with mitochondrial haplogroups and POAG.
As used herein, the term "mutation" refers to any change in a nucleotide or amino acid sequence from a reference nucleotide or reference sequence.
By the term "variant" as used herein is mean the occurrence of a different nucleic acid at a position in the mtDNA which is different from that nucleic acid that occurs at that position in healthy subjects or those without glaucoma. These variants are identified below in Table 1 and about 381 such significant variants are identified in Table 5. These mtDNA variants can result in amino acid changes when they occur in certain
mitochondrial genes. In other embodiments, the nucleic acid variants can be silent, i.e., not result in any amino acid change. In one embodiment, the variants can be identified in Table 5 as having an Odds Ratio greater than 1.0 and/or a Chi test of less than 0.05.
As used herein, the term "polymorphism" means any sequence variant present at a frequency of >1% in a population. The sequence variant may be present at a frequency significantly greater than 1% such as 5% or 10 % or more. Also, the term may be used to refer to the sequence variation observed in an individual at a polymorphic site.
Polymorphisms include nucleotide substitutions, insertions, deletions and microsatellites and may, but need not, result in detectable differences in gene expression or protein function. Single nucleotide polymorphisms (SNP) are the most common type of genetic variation and involve substitution of a single nucleotide in a sequence.
By "missense mutation" is meant is a mutation in which a single nucleotide change or variant results in a codon that codes for a different amino acid. It is a type of nonsynonymous substitution.
By "Pool-seq" is meant a strategy of sequencing pooled DNAs used in the examples of this specification. Next-generation sequencing methods now permit large regions of DNA to be investigated economically, but population-scale whole genome analysis is still prohibitively expensive, due to the complexity of next-generation sequencing library construction and the incremental costs of enriching samples for regions of interest. The, "Pool-seq," method has emerged as a cost-effective alternative to sequencing individuals separately68. Population allele frequencies determined by Pool-seq have been shown to correlate well with actual allele frequencies64, and a pooled sequencing approach, using pools of 20 individual DNAs, has been successfully implemented on the Ion Torrent66 PGM non-optical semiconductor platform21.
It should be understood that while various embodiments in the specification are presented using "comprising" language, under various circumstances, a related embodiment is also be described using "consisting of or "consisting essentially of language. Throughout this specification, the words "comprise", "comprises", and
"comprising" are to be interpreted inclusively rather than exclusively. The words
"consist", "consisting", and its variants, are to be interpreted exclusively, rather than inclusively.
The terms "a" or "an" refers to one or more, for example, "an immunogenic composition" is understood to represent one or more such compositions. As such, the terms "a" (or "an"), "one or more," and "at least one" are used interchangeably herein.
As used herein, the term "about" is defined as a variability of 10 % from the reference given, unless otherwise specified.
One skilled in the art may readily reproduce the compositions and methods described herein by use of the teachings provided herein and conventional techniques and products, assays, etc, which are publicly available from conventional sources.
Applicants identified certain mitochondrial variation in a population consisting of 1,999 subjects, including 1,217 African American POAG cases and 782 African American controls, recruited in Philadelphia, Pennsylvania, for the Primary Open-Angle Glaucoma Genetics (POAAGG) Study11. Using the Pool-seq approach, we sought to simultaneously query all 16,569 positions of mtDNA for the presence of known or potentially pathogenic point mutations, infer the relative frequencies of all common mitochondrial haplogroups, and estimate the potential for the association of haplogroups and each position on mtDNA with POAG. The identification and frequency of occurrence of such mutations permits the identification of subsets of the African American population who are at increased risk from mitochondrial genetic dysfunction and/or geographical ancestries that may be linked to mitochondrial sequence variation. Methods of detecting this information in a patient helps target screening efforts and/or directs clinical interventions such as bioenergetic therapies that may slow vision loss in glaucoma by supporting mitochondrial function 55,31, 49
As described in detail in the examples below, Pool-seq of 1,999 individuals demonstrated that significant differences were found between haplogroup prevalence in the POAG case and control subgroups. Specifically, the POAG group was confirmed to be enriched in Llc2 lineages, which harbor up to three disease-associated missense variants in the MT-COl gene and an MT-RNR2 variant.
L2 haplogroups were also implicated as risk factors for POAG. L2 haplogroups were predicted to be overrepresented in the POAG case population by Pool-seq, and the difference was confirmed to be significant with Sanger sequencing, that targeted the L2- associated variants m.2416T>C (rs28358580, OR 1.2, p=0.02) and m.2332C>T (OR 1.2, p=.02) in MT-RNR2. Another variant within MT-RNR2, m.3010G>A (rs3928306), previously implicated in sensitivity to the optic neuropathy-associated antibiotic linezolid, and arising on D4 and Jl lineages, associated with Leber hereditary optic neuropathy (LHON) severity, was confirmed to be common (>5%), but was not significantly enriched in the POAG cases. Two variants linked to the composition of the gut microbiome, m. l5784T>C (rs527236194, haplogroup L2al) and m. l6390G>A (rs41378955, L2 haplogroups), were also enriched in the POAG case DNA pools.
These lineages contain a common variant in MT-RNR2 and ones recently linked to the microbiome. These results implicate African mtDNA haplogroups Llc2, Llc2b, and L2 as risk factors for POAG. Approximately one in four African Americans have these mitochondrial ancestries, which may contribute to their elevated glaucoma risk, the detection of these haplogroups is a useful screening tool to identify POAG risk factors. Once identified by the methods described herein, those subpopulations in one embodiment are prioritized for screening efforts and for inclusion in clinical trials designed to test therapies or dietary interventions intended to preserve or enhance mitochondrial function.
Thus, in one embodiment of the methods described herein, the diagnostic mitochondrial haplogroup is Llc2. In another embodiment, the suspect mitochondrial haplogroup is Llc2blbl . Still other haplogroups and risks factors are indicated in FIGs. 4 and 6 and in the Examples and the Tables 1-4B below.
As demonstrated in the examples, below, the investigators identified certain missense variants or nucleic acid variants at positions in the mtDNA from the naturally occurring, i.e., most frequently naturally occurring nucleic acid in subjects without glaucoma, which are useful in evaluating a patient's likelihood or risk of progressing to glaucoma. In one embodiment, the variant is a mutation within the MT-COl gene of the patient's mtDNA. In one embodiment, the variant occurs at one or more mtDNA positions 6150, 6253 or 6480. For example, as shown in the studies of the examples below, certain variants can occur alone or with other variants. In one embodiment, a mtDNA variant that is indicative of developing/progressing to severe disease is a G to A variant at mtDNA position 6150. In another embodiment, the variant is a T to C variant at mtDNA position 6253. In still another embodiment, the variant is a G to A variant at mtDNA position 6480.
In yet a further embodiment, the variant is a variant identified in Table 1 or 2 below. In one embodiment, the variant is a A to G variant at mtDNA position 2220. In another embodiment, the variant is a C to T variant at mtDNA position 2332. In another embodiment, the variant is a T to C variant at mtDNA position 2416.
In yet a further embodiment, the variant is identified from among the positions in Table 5, as having an Odds Ratios of greater than at least 1.0. In other embodiments, the Odds Ratios defining the variants are greater than 1.2. In other embodiments, the Odds Ratios defining the variants are greater than 1.4. In other embodiments, the Odds Ratios defining the variants are greater than 1.6. In other embodiments, the Odds Ratios defining the variants are greater than 1.8. In other embodiments, the Odds Ratios defining the variants are greater than 2.0, 2.5 or 3.0. In still other embodiments, variants can be identified from among those rCRS nucleotide positions listed in Table 5 by Chi test numbers less than 0.1 or less than 0.5 or less than 0.3. These variants can be identified from the list and information provided in FIG. 9 of the priority application No. 62/330133 and in the Appendix 1 provided in the on-line version of Collins, et al, 2016 14.
In certain aspects, the variant alters an amino acid residue encoded by a mitochondrial gene. See for example the example below describing the variant that changed an amino acid in a mitochondrial protein which affected the interaction of MT- COl with amyloid beta (Αβ). See, e.g., FIG. 7. The altered amino acid residue may alter the function or interaction of amyloid beta or another gene or protein pathway in the subject. In still other embodiments the variant may be silent.
The methods of detecting or determining the haplogroup or presence of a variant in a subject's mitochondrial DNA can be determined by evaluating the subject's biological sample to one of many types of DNA sequence analysis techniques. The sample can be any biological fluid or tissue from which mtDNA may be evaluated. The most suitable samples for use in the methods and with the compositions are samples which require minimal invasion for testing. In one embodiment, the sample is saliva. In other embodiments, the sample is a blood samples, including serum, plasma, or whole blood. In another embodiment, the sample is urine. Still other biological samples are known for use in obtaining mtDNA, e.g., secretions, ascites fluids or peritoneal fluid and the like. Such samples may further be diluted with saline, buffer or a physiologically acceptable diluent. Alternatively, such samples are concentrated by conventional means.
Suitable assay methods for DNA analysis and sequencing can include next gen sequencing techniques including the Illumina (Solexa) sequencing, Roche 454 sequencing, Ion torrent: Proton / PGM sequencing methods, and PCR-based methods, such as reverse transcription polymerase chain reaction (RT-PCR) or qPCR, among many others. The methods described herein are not limited by the particular techniques selected to perform them. Exemplary commercial products for generation of reagents or performance of assays include ELISAs, such as a platform multiplex ELISA, TAQMAN assay,
ILLUMINA assay, mass spectrometry quantitative assays, PCR, RT-PCR, QPCR or next generation sequencing techniques. Other commercial assays or reagents include TRI- REAGENT, Qiagen RNeasy mini-columns, MASTERPURE Complete DNA and RNA Purification Kit (EPICENTRE®, Madison, Wis.), Paraffin Block RNA Isolation Kit
(Ambion, Inc.) and RNA Stat-60 (Tel-Test), the MassARRAY-based method (Sequenom, Inc., San Diego, CA), differential display, amplified fragment length polymorphism (iAFLP), and BeadArray™ technology (Illumina, San Diego, CA) using the commercially available LuminexlOO LabMAP system.
Such assays include those described in the examples below. See, also, Metsker,
ML, Sequencing Technologies - The Next Generation, Nature Reviews Genetics 11, 31- 46 (January 010) doi: 10.1038/nrg2626, incorporated herein by reference. In another embodiment, the sample may be genotyped or subjected to whole genome sequencing (WGS) and computational analysis of the WGS. It may be useful if the variants results in amino acid changes, for other known techniques to be used to detect changes in the amino acid sequences of any peptides encoded by the mtDNA. Such methods are also well known in the art. The methods described herein are not limited by particular detection methods employed.
The method of evaluation and diagnosis may be further accompanied by evaluation of the subject clinically. In one embodiment, in addition to detecting the suspect haplotype and/or variants defining the suspect haplogroups identified herein as indicative of glaucoma or risk of glaucoma, the subject may also receive additional evaluations.
Such clinical evaluations include evaluating said patient for clinical abnormalities in cup- to-disc ratio (CDR), or evaluating the subject's visual field or acuity, and/or evaluating the subject's intraocular pressure (IOP).
In yet a further aspect, the methods described herein include not only detecting the haplogroups and/or variants defining it as indicative of a diagnosis of, or increased risk of developing, glaucoma, but also further treating the subject so diagnosed. Among suitable treatments are those therapeutic agents, exercise regimens or dietary supplements or regimens indicated to preserve or enhance mitochondrial function. Examples of dietary supplements include coenzyme-QlO, a ketogenic diet, L-Arginine, lipoic acid, and the like. Similarly, certain clinical trials for other mitochondrial diseases may be useful for a subject diagnosed as having an increased risk of developing glaucoma by the methods described herein. Such clinical trials and therapeutic agents include those described in, e.g., Ravina, BM et al, Neurology, 60(8): 1234-40 (April 2003); Schults, CW et al, Arch Neurol. 59(10): 1541-50 (Oct 2002) and Kerr, DS, Mol Genet Metab. 99(3):246-55(Mar 2010) among others, incorporated by reference herein.
The following examples were performed to estimate the population frequencies of all common mitochondrial variants and ancestral haplogroups among 1,999 subjects recruited for the Primary Open-Angle African American Glaucoma Genetics (POAAGG) Study, including 1,217 primary open-angle glaucoma (POAG) cases and 782 controls. These examples also identified ancestral subpopulations and mitochondrial mutations as potential risk factors for POAG susceptibility. Briefly subjects were classified by characteristic glaucomatous optic nerve findings and corresponding visual field defects, as defined by enrolling glaucoma specialists, stereo disc photography, phlebotomy, extraction of total DNA from peripheral blood or saliva, DNA quantification and normalization, PCR amplification of whole mitochondrial genomes, Ion Torrent deep semiconductor DNA sequencing on DNA pools ("Pool-seq"), Sanger sequencing of 3,479 individual mitochondrial DNAs, and bioinformatic analysis. The distribution of common African haplogroups within the POAAGG study population was broadly similar to prior surveys of African Americans. However, the POAG case population was found to be enriched in Llc2 haplogroups, which are defined in part by missense mutations m.6150G>A (Val83Ile, odds ratio [OR] 1.8, p=0.01), m.62530T (Metl l7Thr, rs200165736, OR 1.6, p=0.04), and m.6480G>A (Vall93Ile, rsl99476128, OR 4.6, p=0.04) in the cytochrome c oxidase subunit 1 (MT-COl) gene and by a variant, m.2220A>G (OR 2.0, p=0.01), in MT-RNR2, which encodes the mitochondrial ribosomal 16s RNA gene. L2 haplogroups were predicted to be overrepresented in the POAG case population by Pool-seq, and the difference was confirmed to be significant with Sanger sequencing, that targeted the L2-associated variants m.2416T>C (rs28358580, OR 1.2, p=0.02) and m.23320T (OR 1.2, p= 02) in MT-RNR2.
Another variant within MT-RNR2, m.3010G>A (rs3928306), previously implicated in sensitivity to the optic neuropathy-associated antibiotic linezolid, and arising on D4 and Jl lineages, associated with Leber hereditary optic neuropathy (LHON) severity, was confirmed to be common (>5%) but was not significantly enriched in the POAG cases. Two variants linked to the composition of the gut microbiome, m. l5784T>C
(rs527236194, haplogroup L2al) and m. l6390G>A (rs41378955, L2 haplogroups), were also enriched in the case DNA pools.
These examples demonstrate that African mtDNA haplogroups Llc2, Llc2b, and
L2 as risk factors for POAG. Approximately one in four African Americans have these mitochondrial ancestries, which may contribute to their elevated glaucoma risk. These haplogroups are defined in part by ancestral variants in iheMT-RNR2 and/ or MT-COl genes, several of which have prior disease associations, such as MT-COl missense variants that have been implicated in prostate cancer.
In Example 9, we investigated the role of the m.6150G>A (V83I) polymorphism in the mitochondrial cytochrome c oxidase subunit 1 gene (MT-COl) in African American (AA) primary open-angle glaucoma (POAG). Sanger sequencing and phenotypic characterization of Primary Open- Angle African American Glaucoma Genetics
(POAAGG) study subjects (1308 cases, 846 controls), masked chart review of POAG patients having two COl missense mutations (V83I and Ml 17T, n=29) vs. a matched control group of POAG patients (n=29), controlling for LI mtDNA haplogroup, age, gender, and family history of POAG. Yeast 2-hybrid (Y2H) library screen, quantification of protein-protein interactions by Y2H and ELISA.
The association of the m.6150G>A (V83I) variant (rs879053914) with POAG in African Americans was extremely significant for men (OR 6.5, p=0.0001), but not for women (OR 1.1, p=0.78). Compared to matched POAG controls, cases having COl double missense mutation (V83I + Ml 17T) had higher cup-disc ratio (p=0.04), worse visual function (partem standard deviation: p=0.009; mean deviation: p=0.002), at a lower mean intraocular pressure. The known interaction of the V83I region of COl with amyloid beta peptide was confirmed by ELISA, and found to be abrogated by V83I. A Y2H screen of an adult brain cDNA library with COl bait retrieved the neuroprotective gene
UBQLN1, involved in Alzheimer's disease (AD).
The extreme gender bias observed for the association of m.6150G>A (V83I) with POAG is reminiscent of Leber's Hereditary Optic Neuropathy (LHON), with male carriers of mtDNA mutations at several-fold higher risk than females. The results potentially implicate two AD-associated genes in POAG: APP (amyloid precursor protein), and UBQLN1 (ubiquilin 1).
EXAMPLES
The following examples are illustrative only, and do not limit the scope of the present invention.
EXAMPLE 1 : Subject Recruitment, Phenotyping, And Specimen Collection
A total of 3,479 subjects were included in this study. All were recruited from the Scheie Eye institute and satellite locations in Philadelphia, USA and self-identified as American American. 50% of this total group had POAG and 50% were classified as normal controls at the time of recruitment. The subjects' mean age was 66 years and 72% were female and 28% were male. Subjects were recruited for the institutional review board (IRB)-approved POAAGG Study11 after informed consent was obtained. Consistent with ARVO's statement on human subjects, this study adhered to the tenets of the Declaration of Helsinki1. Following examination by a fellowship-trained glaucoma specialist, subjects were classified as glaucoma case, glaucoma suspect, or control11.
Cases were defined by the demonstration of characteristic optic nerve defects as well as corresponding visual field loss. Subjects classified as glaucoma suspects were deliberately excluded from this study, as were those few who did not describe their ancestry as Black, African American, African, or mixed race including African ancestry. Approximately 30 ml of peripheral blood was obtained from each subject in EDTA tubes. Blood specimens were frozen at -20 °C before DNA extraction. Alternatively, 2 ml of saliva was collected.
The baseline demographics of the POAAGG cohort, including phenotyping methods, and criteria for classification as POAG case or control, have been described previously13'11. A total of 1,999 African American patients with POAAGG were included in the mitochondrial Pool-seq analysis based on the availability of genomic DNA extracted from whole blood and status as glaucoma case or control (glaucoma suspects were excluded). The majority of cases and controls were female, with a larger proportion of men (44.5%) in the case group than in the control group (33.2%). The control group was significantly younger (61±12 years) than the case group (71±11 years). Elevated IOP is a maj or risk factor for POAG, and as expected, the maximum recorded IOP was significantly higher for the POAG cases (mean 25 mm Hg in cases versus 16 mm in controls, p<0.0001), as was neurodegeneration, measured by the cup:disc ratio of optic nerves (mean 0.7 in cases versus 0.3 in controls, pO.0001).
EXAMPLE 2: DNA Extraction Quantification And Normalization
DNA was extracted from whole blood using PureGene Gentra kits (Qiagen, Valencia, CA), and the optional RNase digestion step was included. DNA concentrations were measured using the fluorescence-based Quant iT dsDNA Broad-Range (BR) assay kit (#Q-33130, Life Technologies, Foster City, CA). Fluorescence was measured with a Tecan Infinite 200 Pro multimode microplate reader (Morrisville, NC). An aliquot of each sample was normalized to a concentration of approximately 1-2 ng/μΐ using SequalPrep Normalization plate kits (Cat # A10510-01 , Life Technologies). The concentrations of the normalized DNAs were verified using Quant iT dsDNA High sensitivity (HS) assay kits (Cat # Q-33120, Life Technologies). Equal volumes of up to 48 DNAs, corresponding to either POAG cases or controls, were combined to create a total of 43 DNA pools. Most pools contained DNA representing 48 people, but a small number contained fewer, because the samples were withdrawn from the original 96-well stock DNA plates, for example, if a subject's status had progressed from control to POAG suspect or case. Twenty-six of the pools were created from POAG case DNAs, with 1,217 cases represented, and the remaining 17 pools were created using DNAs from 782 control subjects.
Unpooled genomic DNAs were obtained from two individuals, whose
mitochondrial genomes had previously been sequenced and were known to correspond to the phylogenetically distant mitochondrial haplogroups (Lie and H3b). These unpooled DNAs were used as controls for the Ion Torrent sequence analysis. The replication group for Sanger sequencing (n=l,389) included DNAs collected from saliva samples using OraGene-500 kits (DNA GenoTek, Ottawa, Canada, Cat # OGR-500) and extracted with prepIT reagents (#prepIT-L2P) according to the manufacturer's instructions.
EXAMPLE 3: Enrichment For Mitochondrial DNA And Ion Torrent Sequencing Library Construction
Whole mitochondrial genomes were amplified as nine overlapping fragments with PCR, using DNA from each of the 43 pools or two individuals as templates, with a primer set designed to avoid coamplification of nuclear mitochondrial pseudogenes62. The two additional unpooled individual DNAs were amplified in parallel with the pooled DNAs, to estimate the background noise from sequencing each position on mtDNA in the absence of signals arising from population variation. All PCR reactions were performed in a total volume of 12 μΐ using the Platinum Taq DNA polymerase kit (Cat # 10,966-034, Thermo Fisher).
To obtain high yields on all products, two different PCR protocols were used, one optimized for six of the nine fragments, and the second optimized for the remaining three fragments. For mtDNA fragments designated "1,2,5,7,8,9" 62, the PCR reactions contained 1.2 μΐ 10X reaction buffer, 0.24 μΐ of 10 mM dNTP mix, 0.48 μΐ of 50 mM MgCl2, 2.4 μΐ of primer mix with each primer at 2 pmol/μΐ, and 3 μΐ of template DNA (from 1 to 10 ng), 0.1 μΐ of Platinum Taq, 3.6 μΐ of 5M betaine, and 1 μΐ of nuclease-free water. Cycling conditions were 95 °C for 5 min; 35 cycles (94 °C for 1 min, 64 °C for 40 s, 72 °C for 3 min) with 0.2 degrees of touchdown; 72 °C for 5 min; final hold 10 °C. For mtDNA fragments designated "3,4,6" , the PCR reactions contained 1.2 μΐ 10X reaction buffer, 0.24 μΐ of 10 mM dNTP mix, 0.96 μΐ of 50 mM MgCl2, 2.4 μΐ of primer mix with each primer at 2 pmol/μΐ, and 3 μΐ of template DNA (from 1 to 10 ng), 0.1 μΐ of Platinum Taq, 3.6 μΐ of 5M betaine, and 0.5 μΐ of nuclease-free water. Cycling conditions for fragments 3,4,6 were 95 °C for 2 min; 35 cycles (94 °C for 1 min, 58 °C for 40 s, 72 °C for 3 min) without touchdown; 72 °C for 3 min; final hold at 10 °C. Thermal cycling was performed on an Applied Biosy stems 9700 with dual 384-well block, with all case and control pools amplified from the same master mixes and in parallel within the same thermal cycling run. The presence of the expected bands and yields was confirmed with gel electrophoresis.
Unequal volumes of PCR products were pooled to help equalize sequencing depth on the nine fragments. Nine microliters each of the fragment 3,4,5,6 PCR products were combined with 6 μΐ each of fragments 1,2,7,8,9 to compensate for lower PCR yields in the former group. An aliquot of each nine-fragment PCR product pool was then used as the template to construct the Ion Torrent library. Forty-five barcoded sequencing libraries were constructed according to the manufacturer's instructions using 35 μΐ DNA containing approximately 100 ng of each PCR product pool as inputs (Ion Xpress™ Plus Fragment Library Kit Cat # 4,471,269, Ion Xpress Bar Code Adapters 1-16 kit, Cat # 4,471,250), the Ion Xpress™ Barcode Adapters 17-32 kit (Cat # 4,474,009), and the Ion Xpress™ Bar Code Adapters 81-96 kit (Cat # 4,474,521, Life Technologies).
EXAMPLE 4: Semiconductor DNA Sequencing On The Ion Torrent PGM
Sequencing libraries were quantified with quantitative polymerase chain reaction (qPCR) with the Ion Library Quantitation Kit (Cat # 4,468,802, Life Technologies) and diluted to 100 pM. The diluted sequencing libraries (10 pM) were then amplified on Ion Sphere™ Particles (ISPs) and enriched for template-positive ISPs using the Ion PGM Template OT2 200 Kit (Cat # 4,480,974), Ion One Touch 2 instrument, and Ion One
Touch ES. The Ion Sphere Quality Control Kit (Cat # 4,468,656, Life Technologies) was used to determine the fraction of template-positive ISPs. Enriched ISPs were then sequenced using the Ion PGM Hi-Q sequencing Kit (Cat # A25592, Life Technologies) and the 318 Chip Kit v2 (Cat # 4,484,354, Life Technologies). Ion Torrent sequencing results were deposited in the NCBI Sequence Read Archive with study accession number SRP067259. EXAMPLE 5: Validation Of POOL-SEQ With Sanger Sequencing Of Individual DNAs
Variance frequencies estimated with next-generation sequencing on DNA pools were evaluated with Sanger sequencing targeting a region of the N-terminal part of the MT-COl gene (NCBI database, Gene ID 4512, OMIM 516030; see also SEQ ID NO: 2), performed individually for the 1,999 subjects whose DNAs were pooled, as previously described13. The actual mean variance frequency at each mtDNA position in this interval for all DNA pools was determined with Sanger sequencing of individual DNAs, and these means were compared to the Pool-seq estimates. To determine the association of variants within the MT-COl region with POAG with Sanger sequencing, a total of 2,157 subjects, 1,308 cases and 849 controls, were analyzed; that is, an additional 158 subjects were included. Sanger sequencing targeting the MT-RNR2 gene, for 2,090 subjects (1,258 cases and 832 controls), was performed to ascertain whether the frequencies of several variants detected in this region differed significantly, as predicted by sequencing on DNA pools. Additional Sanger sequencing for a replication group consisting of 1,389 subjects (510 cases, 879 controls) was performed to confirm the enrichment of L2- and Llc2b- associated variants, using POAG cases that were not included in the pooled sequencing group.
EXAMPLE 6: DNA Sequence Analysis, Estimation Of Population Frequencies, Data Filtering, And Variant Annotation
Sanger sequencing electropherograms were scored and converted to variance tables with Sequencher version 5.1 software (Softgenetics, State College, PA). Torrent Suite (software) version 4.2 was used for alignment of Ion Torrent PGM reads to the hgl9 human genome reference sequence. The BAM files were then downloaded and reviewed using the Integrated Genomics Viewer (IGV)65.
Counts of the four nucleotides at each position of the alignments were obtained using the count utility of the IGV Tools package from the command line. Output files were compiled in an Excel spreadsheet, and the nucleotide counts were converted to percentages to normalize for read depth. For each position on mtDNA, we computed the frequency of adenine (A), cytosine (C), guanine (G), and thymine (T) and the total frequency of all minor alleles. That is, if the rCRS reference sequence base was T, and the A, C, G, and T counts were 1%, 4%, 0%, and 95%, respectively, we would infer a 1% variant frequency for T>A, a 4% variant frequency for T>C, and a total raw "variance frequency" of 5%. To correct for spurious background noise, we calculated the mean variance frequency observed on the two unpooled samples for each position on mtDNA. This background frequency was then subtracted from the mean variance frequencies that were observed in the case and control DNA pools.
To be considered a "high confidence" call of a common variant from pooled sequencing, the following three conditions had to be met:
1) The position on mtDNA was not classified as "noisy"; noisy was defined as having a mean background variance frequency of >\% (on libraries corresponding to the two unpooled individuals' samples).
2) The observed mean variance frequency on the pooled samples was significantly higher than on six replicates on unpooled samples. The filter was a t test, two-tailed, unequal variance, and p<0.05.
3) The residual estimated variance frequency was >1%, after subtraction of background noise at that position.
For comparing sequencing results from pooled sequencing to those from Sanger sequencing of individual samples, we defined "sequencing accuracy" as the fraction of concordant calls (agreements/total), where "agreement" means the same genotype was called by both methods above a given threshold, and total refers to the total number of base pairs that were sequenced. We defined a false negative variant as one for which Sanger sequencing detected a variant, but Ion Torrent sequencing did not
("disagreement"). The false negative rate on variants was calculated as (disagreements) / (total variants called by Sanger) above a given variance threshold. The false positive rate on variants was defined as (disagreements) / (total variants called by Ion Torrent pooled sequencing).
The mitochondrial genome was annotated for disease association using data from the MITOMAP compendium52 and annotated for haplogroup associations using Build 16 of PhyloTree79. Haplogroup population frequencies were estimated using the following positions on mtDNA:
L0 (1048, 3516, 5442, 9042, 9347, 10,589, 12,007, 12,720),
L0a (5231, 11 , 176, 14,308), LI (7055, 7389, 13,789, 14, 178, 14,560),
Lib (185, 710, 1438, 1738, 2768, 3308, 3693, 6548, 6827, 6989, 7867, 8248, 12,519, 14,769, 15,1 15, 16,126, 16,264, 16,270, 16,293), Li e (151 , 186, 5951 , 6071, 10,586, 12,810),
Ll cl (3796, 3843),
Ll c2 (6150, 6253, 7076, 8784, 8877, 10,792, 10,793, 11 ,654, 16,286, 16,527),
Ll c3 (195, 6221, 6917, 1 1,302),
L2 (2416, 8206, 9221 , 10, 1 15, 13,590, 16,390),
L2al234 (2789, 7274, 7771 , 13,803, 14,566),
L2b (1706, 2358, 4158, 4370, 4767, 5027, 5331 , 5814, 6713, 8080, 8387, 12,948,
14,059, 16,1 14, 16,213),
L2c (93, 325, 680, 3200, 13,928, 15,849),
L3 (769, 1018),
L3b (3450, 5773, 9449, 10,086, 13,914, 15,311 , 15,824),
L3d (7424, 8618, 13,886, 14,284),
L3e (14,212),
L3f (3396, 4218, 15,514, 16,209),
M (489, 10,400, 14,783, 15,043),
N (10,398),
R (12,705),
RO (73), and
U (1 1,467, 12,308, 12,372).
Variance frequencies were averaged when more than one position was associated with a particular haplogroup. Variant positions harboring back mutations per PhyloTree, those common to more than one of any of these haplogroups, and those that did not correspond to high-confidence variable positions in the Pool-seq data were excluded from the frequency estimations and estimates of case: control odds ratios (ORs). We did not attempt to estimate the frequency of the common L2a haplogroup due to a lack of variable positions that met these criteria.
EXAMPLE 7: Statistics
The TTEST function (two-tailed, unequal variance) in Microsoft Excel was used to compare mean variance frequencies in the pooled versus unpooled samples and used as a filter to differentiate low-frequency population sequence variation from background sequencing noise. The CHITEST function in Excel was used to compare the estimated numbers of cases and controls that have variants at the 381 positions that were selected for the association study, to evaluate the relative likelihood that each difference in the estimated number of cases and controls that have a particular genotype was explainable by chance. The confidence intervals on proportions, for reported and estimated frequencies of African American haplogroups, were determined using the modified Wald method and an online calculator (GraphPad). For confirmatory Sanger sequencing, the number of cases and controls who harbor each variant was analyzed as 2x2 contingency tables using a two- tailed Fisher's exact test or a chi-square test with Yates correction (if n>2000) with version 6 of GraphPad Prism (GraphPad Software, La Jolla, CA).
EXAMPLE 8: Mutations Of The MT-C01 Gene
To determine whether common missense mutations in the mitochondrial cytochrome c oxidase subunit I gene (MT-COl) are linked to glaucoma susceptibility and traits in African Americans, the following experiment was performed.
Subjects were recruited from the Scheie Eye Institute and research affiliates in Philadelphia for the POAAGG study11. Glaucoma cases were defined by demonstrating characteristic optic nerve defects with corresponding visual field loss. Genomic DNAs were extracted from whole blood26, an N-terminal fragment of the MT-COl gene was amplified from mitochondrial DNA (mtDNA) by PCR and Sanger sequenced, as previously described13. Masked genotype/phenotype analyses of POAG patients with and without MT-COl mutations were completed. We compared 1,308 African-American POAG cases and 849 African-American controls. The PhyloTree resource (Build 17, www.Phylotree.org) was used to infer mitochondrial haplogroups (FIG. 6, Table 3).
Figure imgf000023_0001
Figure imgf000024_0001
Masked genotype/phenotype analysis of POAG patients was performed with MTCOl missense mutations (Tables 4A-4D). Table 4A shows the Llc2 (MT-COl double missense) patents had more severe disease vs. matched Lib POAG case controls (no missense).
Figure imgf000024_0002
Table 4B shows data demonstrating worse visual field loss for Llc2 MT-COl double missense patients vs. matched Lib POAG case controls (no missense).
Figure imgf000024_0003
Table 4C shows data demonstrating higher CDR for Llc2 MT-COl double missense patients vs. matched Lib POAG case controls (no missense). Table 4C
CDR Llc2 cases (# eye) Lib cases (# eyes) p-value
0.6 4 2
0.6-0.8 25 34 0.03
>0.8 19 6
Mean (SD): 0.77 0.71 0.04
Table 4D shows data demonstrating lower IOP for Llc2 MT-COl double patients vs. matched Lib POAG case controls (no missense).
Figure imgf000025_0001
Two case groups matched for age, gender and family history of POAG, analyzed for differences in cup-to-disc ratio (CDR), visual fields, intraocular pressure (IOP), and disease severity. 29 cases were found having two missense mutations, m 6150G>A (Val83Ile) and m.6253T>C (Metl 17Thr ), that define African haplogroup Llc2. 29 cases were found having the synonymous MT-COl variant m.6548C>T, which defines African haplogroup Lib. p-values were calculated using logistic regression, with general estimating equations (GEE) to account for correlation between eyes.
Sandwich ELISA assays were performed with biotinylated synthetic peptides. Reported amyloid-β (Αβ) peptide interaction with MTCOl was examined59, and the effect of Val83Ile mutation observed. A 61 amino acid region of the mitochondrial cytochrome c oxidase subunit I gene {MT-COl, i.e., the amino acid stretch from aa 40-99 of the Cox 1 sequence of GenBank accession number gi: 251831109 and NCBI Ref YP_003024028.1 ) is conserved among the following species: Homo_100/1-60, Lophocebus_97/1-60, Trachypithecus_98/l-60, Papio_98/1-60, Colobus_98/1-60, Cercocebus_98/1-60, Semnopithecus_98/1-60, Theropithecus_98/1-60, and Nasalis_98/l-60, Rungwecebus_98/l-60, and Macaca_97/l-60. These region compares to the m.6150G>A (Val83Ile) variant. This region has been reported54 to interact with amyloid β (Αβ). See, also FIG. 7. Controls used were an ApoE positive control, TBX3 and scrambled Αβ negative controls.
The POAG case group was significantly enriched for three disease-associated missense variants, which define a common African mitochondrial haplogroup and previously implicated in prostate cancer. 59 The African mitochondrial haplogroups so associated are Llc2 or a sublineage, Llc2blb. POAG cases harboring these variants had worse disease at significantly lower intraocular pressure than nono-missense cases, after matching for age, gender, family history of POAG and mitochondrial ancestry. Thus mitochondrial lineages that harbor missense mutations in MT-COl, e.g., haplotypes Llc2 and Llc2blbl, appear to further elevate African-Americans' existing high risk for glaucoma and result in worse disease. Functional characterization of these MT-COl missense variants yield insight into the etiology of this subset of POAG patients, which may involve Αβ. Mitochondrial sequence analysis can inform screening efforts and personalize treatment by identifying those glaucoma patients who might benefit from dietary supplementation or from other interventions designed to support mitochondrial function.
The results of the examples are described in more detail below.
PCR Amplification And Deep Sequencing Of Mitochondrial Genomes
The 43 DNA pools were constructed with DNAs from 1,999 subjects, with 1,217 POAG cases represented in 26 pools, and 782 controls represented in 17 pools. The Ion Torrent sequencing coverage depth was variable across the mitochondrial genome, with the highest coverage depth within the MT-CYB gene and regions of lower coverage in the regions of the MT-ND1, MT-ATP6, and MT-ND6 genes. A representative coverage map from sequencing on one of the pools is shown in FIG. 1.
The 43 pools were sequenced to a mean depth of l,931x, with coverage depth on individual libraries ranging from 718x to 3,189*. Assuming that just one individual in a pool of 48 has a homoplasmic mutation at a particular position on mtDNA and proportionate representation of individuals in that sequencing library, the expected variance frequency is 1/48 (2.1%), which would generate approximately 40 mutant base calls at the mean coverage depth. In other words, there was sufficient depth to detect the presence of single rare variants within each pool.
Two individual subjects' (unpooled) DNA samples, known to have distant mitochondrial ancestry and non-overlapping sets of variants from prior sequencing, were sequenced in parallel with the pooled samples. This was done to estimate the background noise and the spectrum of spurious base calls at every position on mtDNA in the absence of low-level signals from true population variation. Neglecting sites with true variants, the mean variance frequency on these unpooled samples was 0.4±0.2%; however, this distribution was asymmetric with a long tail, with about 2% of sites having >\% variance frequency. We classified these sites as "noisy" and excluded them from subsequent analysis (data not shown). These signals could conceivably represent low-level heteroplasmy in the two unpooled subjects' DNA, but typically, these calls were observed in both subjects and are far more likely to represent spurious basecalls and potential false positives. We estimate that the analytical sensitivity of this sequencing protocol, defined as the fraction of mutant alleles that can be detected on a wild-type background, was generally less than 1 % for >98% of the positions on mtDNA.
Several hundred variants were inferred from Pool-seq, most corresponding to one or more common haplogroups. The population-scale whole genome analysis for all 16,569 positions on mtDNA is summarized in FIG. 2A. Results for each position were reported in Appendix 1 of Collins et al 14, incorporated herein by reference. As expected, no population variance above the sequencing background noise was observed at most positions; human mtDNA has been diverging for only about 200,000 years, albeit at a rapid rate relative to nuclear genes. There is a conspicuous absence of variance with minor allele frequency above 10% in the region corresponding approximately to positions 4000 to 7000, spanning the MT-ND2 gene and parts of the MT-ND1 and MT-COl genes, and another such region near position 2000 at the junction of the 12s and 16s RNA genes, MT- RNR1 and MT-RNR2. The relative scarcity of highly polymorphic sites in these regions suggests they may be selectively constrained at the nucleotide level. Consistent with FIG. 1 and expectations, the hypervariable regions, designated HVR1 and HVR2, contain the highest concentration of variants, many of which are highly polymorphic. Mitochondrial Haplogroups Inferred From POOL-SEQ
A recent independent multicenter study of mitochondrial haplogroup lineages in African Americans identified 13 major macrohaplotypes as common, with population frequencies >1% 42. Several of these groups overlap; that is, the LI haplogroup is composed of subclades Lib and Lie. The mtDNA positions associated with these particular haplogroups are highlighted in FIG. 2A. As expected, all 13 haplogroups all appear to be represented in the POAAGG cohort, with one or more of the associated variants detected. The population variance frequency observed at these positions ought to be proportional to the expected population frequency of the associated haplogroup. For example, L3 was reported by Johnson et al. to be the most common haplogroup in African Americans 42, and the two mtDNA positions with the highest estimated frequencies, 34% and 35%, among this subset of haplogroup-associated variants correspond to haplogroup L3 (gray circles, upper left corner, FIG. 2A).
Validation Of Ion Torrent Pool-Seq With Ce Sanger Sequencing On Individual Samples A 605 bp region of mtDNA was sequenced for the 1,999 subjects individually, using the Sanger method. This region spans approximately 3.7% of the mitochondrial genome and corresponds to an N-terminal part of the MT-COl gene. The Sanger sequencing was analyzed for each pool, to evaluate the accuracy of the Ion Torrent pooled sequencing. Genotyping efficiency >98% was achieved on the 605 bp interval, and the Sanger sequencing identified 11 positions on mtDNA as variable and with variance frequencies above 2%. These 11 positions corresponded to the following known common mitochondrial variants (with the associated L-haplogroups per PhyloTree in parentheses): m.6071T>C (Lie), m.6150G>A (Llc2), m.6185T>C (L0), m.6221T>C (L3b, L3el) and m.6221T>A (Llc3), m.6253T>C (Llc2), m.6260G>A (Llc3a and L4b2), m.6413T>C (L3e2alb), m.6524T>C (L3e3), m.6548C>T (Lib), m.65870T (L3el), and m.6663A>G (L2alc).
The variance frequencies determined from Sanger and estimated from Pool-seq are compared in FIG. 2B. Sanger sequencing also confirmed the presence of two minor alleles at m.6221: the more common OT transition that has arisen twice independently on different L3 lineages and the less common OA transversion, associated Llc3. False positives in the Ion Torrent Pool-seq data are also perceivable in FIG. 2B. A region of elevated background noise can be seen near rCRS position 6400; positions 6420 and 6442 are labeled "FP" and had predicted variance frequencies >2% but were not validated with Sanger sequencing. These are examples of positions that were characterized as "noisy" and filtered from subsequent analysis. A total of 381 variable nucleotide positions were identified on mtDNA that survived filtering and were considered high-confidence variable positions. The nucleotide positions are referenced to the above-cited rCRS sequence and include the positions referenced in Table 5 below:
Table 5
Nucleotide positions - rCRS sequence numbering - High confidence variables
64 71 72 73 93 95 143
150 151 182 183 185 186 189
195 198 199 200 207 236 247
297 325 356 358 418 489 499
513 680 709 710 750 769 825
847 921 1011 1018 1040 1048 1438
1442 1503 1534 1535 1706 1719 1738
1739 1789 1791 1794 1822 2000 2056
2220 2245 2332 2352 2358 2416 2472
2706 2758 2768 2789 3010 3200 3308
3396 3420 3434 3450 3495 3505 3516
3693 3796 3843 3866 3915 3918 4158
4216 4218 4312 4370 4436 4454 4506
4655 4767 4823 4917 5027 5036 5046
5096 5231 5262 5285 5331 5393 5442
5460 5581 5584 5601 5603 5655 5711
5773 5814 5951 6026 6071 6150 6152
6164 6221 6253 6257 6260 6378 6455
6524 6548 6587 6663 6680 6703 6704
6713 6806 6827 6917 6989 7055 7076
7158 7274 7333 7335 7336 7389 7424
7624 7648 7771 7789 7805 7816 7867
8080 8087 8206 8226 8228 8248 8251
8387 8410 8428 8527 8566 8618 8655
8784 8790 8856 8877 9042 9221 9347
9377 9449 9476 9477 9485 9554 9559
9755 9818 9950 9966 10031 10044 10047
10064 10070 10086 10115 10321 10373 10387
10388 10398 10400 10463 10586 10589 10640
10664 10667 10688 10792 10793 10809 10810
10816 10828 10869 10870 10886 10915 10920
11002 11029 11143 11164 11176 11251 11257
11302 11440 11467 11641 11654 11887 11899 11914 11944 12007 12019 12049 12091 12172
12236 12248 12299 12301 12303 12304 12308
12330 12334 12372 12406 12477 12501 12519
12609 12630 12693 12705 12720 12810 12948
13101 13105 13197 13240 13242 13276 13368
13470 13506 13590 13650 13651 13708 13749
13789 13803 13874 13880 13886 13898 13914
13924 13928 13934 14059 14070 14071 14072
14092 14152 14178 14182 14203 14212 14284
14308 14486 14487 14489 14526 14527 14528
14529 14532 14533 14560 14566 14755 14766
14769 14783 14869 14905 15043 15061 15110
15115 15136 15217 15236 15244 15301 15311
15431 15452 15514 15629 15658 15670 15758
15784 15812 15824 15849 15881 15883 15942
16048 16051 16086 16093 16111 16114 16124
16126 16129 16145 16148 16167 16168 16169
16172 16179 16185 16192 16209 16213 16218
16219 16230 16239 16261 16264 16270 16286
16292 16293 16295 16304 16309 16311 16319
16320 16327 16355 16362 16368 16390 16399
16519 16527 16554
These results confirmed the reliability of the Ion Torrent Pool-seq method for the detection of common mitochondrial variants. However, the estimated minor allele frequencies were clearly subject to experimental error, likely arising from imperfect normalization, variation in the ratio of nuclear to mtDNA, and/or variable PCR amplification of the individual mitochondrial genomes. The variance frequencies estimated from Pool-seq in some cases underestimated and in other cases overestimated true frequencies (FIG. 2B). Sanger sequencing confirmed the absence of extremely polymorphic variants, those that have minor allele frequencies >\5%, within the 605 bp interval (FIG. 2A), and confirmed that none of the variants that had been designated "high confidence" within this interval was a false positive. For high confidence variants, sequencing accuracy by Ion Torrent was 100%; the false negative rate for variants was 0%, and the false positive rate for variants was 0%.
Association Of Mitochondrial Haplogroups With POAG In African Americans We compared the prevalence of the most common haplogroups in the study's controls relative to reported frequencies in African Americans, as well as the haplogroup distribution in POAG cases versus controls. Using the 381 high confidence variable positions identified in Table 5 above, we computed the mean variance frequencies for sets of haplogroup-associated variants and compared these frequencies to estimates from an independent multicenter study 42. As depicted in FIG. 3, the frequencies inferred from the POAG controls mostly accorded well with those reported in the 2015 study 42, but L2, L2b, and non- African haplogroups (M, N, R, R0, and U) frequencies were lower for POAAGG subjects, and the proportion of L3e was higher. The POAG case pools appeared to be substantially enriched in Lie and L2 haplogroups relative to the controls.
To visualize the potential variation in POAG risk across the population in the context of their ancestral mitochondrial relationships and relative divergence times, FIG. 4 summarizes the phylogenetic relationships among haplogroups that were inferred based on defining sets of variants, with numbering to indicate which groups were enriched in the POAG case or control pools, with potential high-risk groups (OR 1.4-3.3) numbered as (3), i.e, L2, L2c, L2B, L2alc, L2al '2'3'4 and Llc2. The pooled sequencing results suggested that L0(a), Lib, Llc(l,3), and L3(b,e) and the non-African branches M, N, and R were associated with protection or lack of risk (estimated OR 0.7-1.0). The Llc2, L2, L2al'2'3'4, L2alc, L2b, and L2c lineages were implicated for potentially elevated POAG risk (OR >1.4). The Llc2 lineage is defined in part by a pair of missense mutations in the MT-COl gene, and the pooled sequencing predicted that the subgroups Llc2bla'b, defined by m.2220A>G in MT-RNR2, and Llc2blb, defined by a third missense MT- COl, m.6480G>A, were also present.
None of these missense variants are present in the Llcl lineage, which did not emerge as a potential risk factor (FIG. 4). The ten other variants that define Llc2 are either synonymous or within non-coding regions. With the exception of some ancient single nucleotide polymorphisms (SNPs) in the hypervariable regions, common to nearly all branches represented by PhyloTree, there is no overlap between the sets of variants that define these Llc2 and L2 sublineages; that is, there was no evidence of glaucoma risk arising from a variant common to both lineages.
Validation Of POAG Association With Haplogroups With Sanger Sequencing The 605 bp N-terminal region of the MT-COl gene that was Sanger sequenced encompassed several common variants associated with one or more common haplogroups, thus having the potential to validate the haplogroup associations suggested by the pooled sequencing results in FIG. 4 and FIG. 5. Sanger sequencing of 1,308 cases and 849 controls showed significant overrepresentation of three MT-COl missense variants in the POAG case group: m.6150G>A (OR 1.8, p=0.01), m.6253T>C (OR 1.6, p=0.04), and m.6480G>A (OR 4.6, p=0.04). Pool-seq correctly predicted the m.6150G>A missense variant as the "top hit" for POAG association in this region, although Pool-seq overestimated the odds ratio (data not shown). The Llc2 lineage is defined in part by the pair of missense variants, m.6150G>A (Val83Ile) and m.62530T (Metl 17Thr), in the MT-COl gene. A third less common missense variant MT-COl, m.6480G>A
(Vall92Ile), has arisen on the Llc2blb subgroup of Llc2. Because this variant was (correctly) classified as rare by Pool-seq, with <1% minor allele frequency, m.6480G>A was excluded from the association analysis that was limited to common variants (FIG.5). Sanger sequencing also confirmed the predicted lack of positive association of POAG with several other African haplogroups, consistent with FIG. 4: Lib (m.6548C>T, OR 1.0, p=0.82), L3b+L3el (m.6221T>C, OR 0.9, p=0.82), L3e2alb (m.6413T>C, OR 0.7, p=0.37), Llc3 (m.6221T>A, OR 0.7, p=0.26), L3e3 (m.6524T>C, OR=0.9, p=0.66), and L0 (m.6185T>C, OR 1.1, p=0.81).
Additional Sanger sequencing targeting t eMT-RNR2 region for 2,090 subjects was performed to confirm the predicted overrepresentation of the Llc2, L2, and L2alc haplogroups in cases (FIGs. 3 and 4) and the top association hits within this region (FIG. 5). Two variants, associated with L2 haplogroups, were confirmed to be significantly more prevalent in cases, and the expected lack of association of POAG with Lib and other haplogroups was also confirmed by sequencing the MT-RNR2 region. Table I shows the frequencies of common variants identified by Sanger sequencing within the mitochondrial 16s RNA (MT-RNR2).
Figure imgf000032_0001
m.2245A>G 0.023 0.021 1.1 0.913 0.625 0.75 LOa'b'fg m.2245A>C 0.017 0.016 1.1 0.989 0.18 0.423 L0ala2
L2b'c'd, m.2332C>T 0.122 0.102 1.2 0.131 0.271 0.020* L3bla8 m.2352T>C 0.246 0.294 0.8 0.017* 0.6 0.113 Lib, L3e m.2358A>G 0.047 0.048 1 0.999 0.6 0.7 L2b m.2395delA 0.109 0.09 1.2 0.19 1 0.624 Lie m.2416T>C 0.317 0.263 1.2 0.008* 0.767 0.017* L2 m.2483T>C 0.019 0.022 0.9 0.58 1 0.73 L3e2bla m.2706A>G 0.975 0.966 0.7 0.24 1 0.341 Many m.2758G>A 0.238 0.22 1.1 0.352 nd Many m.2768A>G 0.091 0.093 1 0.942 nd Lib m.2789C>T 0.188 0.159 1.2 0.106 nd L2al'2'3'4 m.2885T>C 0.241 0.223 1.1 0.375 nd Many
L2alc, m.3010G>A 0.052 0.051 1 0.953 nd H65a m.3200T>A 0.064 0.044 1.4 0.066 nd L2c
Sanger sequencing on individual samples confirmed the presence of variance at 15 positions with minor allele frequencies greater than 2%, in addition to rare variants (data not shown). DNA samples used for pooled sequencing were drawn from the first group of 2,090 POAAGG subjects (1,258 cases, 832 controls). The replication group consisted of an additional 1,389 POAAGG subjects (510 cases, 879 controls).
Table 2 shows disease-associated positions on mtDNA harboring common variants in the POAAGG population.
TABLE 2 - Disease Associated Variants from Pool-SEQ
Pool-seq
Positio Amino
variance Disease Disease Haplo- n Locus acid
frequenassociation* allele group(s)
(rCRS) change
cy
30% 195 DLOOP BD C noncoding Many
possibly L3e Llb +
21% 2352 RNR2 C 16S rRNA
LVNC other (non-L) L3hla2b Lib
Melanoma
% 16270 DLOOP T noncoding + other (non- patients
L)
SZ- LOd Lib +% 1438 RNR1 A 12S rRNA
associated other (non-L)
MELAS/
Llb L2alcl +% 3308 ND1 LVNC/DEA C Met>Thr
other (non-L) F enhancer
L3hla2al% 12236 TS2 DEAF A tRNA Ser L2b'c L5a + other (non-L)
DEAF Llbl + other% 5655 TA C tRNA Ala
enhancer (non-L)
PD Llclal protective LOdlbl% 10398 ND3 G Thr>Ala
factor, breast L3ela3 + cancer risk other (non-L)
CVS with L2alc, L4bla migraine, L0a4 + other% 3010 RNR2 A 16S rRNA
linezolid (non-L, e.g sensitivity H65a, Jl)
Hypertensive
L3b + other% 10086 ND3 end-stage G Asn>Asp
(non-L) renal disease
Possibly
% 921 RNR1 C 12S rRNA L3dr2'3'4'5'6
LVNC
FBC,
% 15784 CYTB microbiome C syn. Llc2a, L2al regulation
Possibly L3el + other% 15942 TT C tRNA Thr
LVNC (non-L)
L3e3'4'5 +% 750 RNR1 sz- A 12S rRNA
associated other (non-L)
L0a4 L3hlal% 16192 DLOOP Melanoma T noncoding + other (non- L)% 5460 ND2 AD / PD A Ala>Thr many
Mitochondria L2b% 5814 TC 1 C tRNA Cys
encephalopat hy
Prostate L2al c
4% 6663 COl G Ile>Val
cancer
L2a4a L3e3b
LHON
3% 15812 CYB A Val>Met + other (non- secondary
L)
MDD- L2c2a + other
3% 15043 CYB A syn.
associated (non-L)
Prostate Llc2 + other
3% 6150 COl A Val>Ile
cancer (non-L)
Adult-onset
3% 3796 ND1 G Thr>Ala other (non-L) dystonia
Prostate Llc2 + other
3% 6253 COl C Met>Thr
cancer (non-L)
When combined with Sanger data from a replication group consisting of an additional 1,389 subjects, not included in the Pool-seq group, association with POAG was significant for two L2-associated variants and one Ll c2b-associated variant.
Potential Association Of 381 Variable Positions On MtDNA With POAG
We also analyzed each of the 381 positions with "high confidence" variance individually for association with POAG. Many of these have arisen on multiple lineages and are associated with more than one haplogroup. For each position, the observed mean variance frequency in the POAG case and control pools was multiplied by the total number of individuals whose DNA was present in the pools, to estimate the number of individuals in the case and control pools that have sequence variation at that position (FIG. 9). These numbers were then compared to ask whether the estimated numbers of cases and controls differed significantly for each variable position, given the known population sample sizes, and assuming the Pool-seq population frequency estimates were correct. The results are summarized in FIG. 5, with the result for each position plotted as -lOlog
(imputed p value). The top seven associations have imputed p values of less than 1 χ 10~6, and they are highly significant assuming the population frequencies inferred from the pooled sequencing are correct. The following seven variable positions are primarily associated with L2 haplogroups, with the exception of m.2220: m.8206G>A (L2 + (non- L)), m.1 1944T>C (L2a'b'c'd + (non-L)), m.2416T>C (L2, L3d3al a, + (non-L)), m. l3590G>A (L2, LOk + (non-L)), m.7274C>T (L2al'2'3'4), m.2220A>T,G (A>T:
L4b2bl + (non-L); A>G: Llc2bla'b), and m. l6390G>A (L2, L0a2b + (non-L)).
The highest odds ratio among the top hits was estimated for position m.2220 within MT-RNR2, which was predicted to harbor two variants: an Llc2bla'b-associated A>G transition and an L4b2bl -associated A>T transversion. The variance frequency at m.2220 estimated by Pool-seq was 5.2% in the POAG pools versus 0.3% in the control pools. Sanger sequencing on 2,090 individual unpooled DNAs confirmed the existence of both variants but determined that m.2220A>G was about fourfold more common than m.2220A>T in both cases and controls; that is, the predicted association of m.2220 with POAG stemmed primarily from Llc2bl haplogroups, not L4b2bl. The association of the less common variant, m.2220A>T, with POAG in the Pool-seq group (OR 1.6) was positive but not significant, whereas the association of the m.2220A>G variant was significant (p=0.023) in the replication group of 1,389 additional samples and in the combined data set (p=0.013, Table 1). The association of the presence of either variant, G or T, at m.2220 was also significant (p=0.003) in the combined data set (n=3,479).
Positions that have been annotated as disease-associated by the MITOMAP resource 79 and/or with variants classified as "pathogenic" or "likely pathogenic" by the NCBI Variation Viewer are highlighted with red markers in FIG. 5, and disease-associated variants that have estimated POAAGG population frequencies >3% are listed in Table 2. Included among these are the variants at positions 2352, 3010, 6150, 6253, 10086, 10044, 15301 and 15784, the nt coordinates on the rCRS reference sequence. The status of most of these variants is "reported" and not "confirmed," with highly variable evidence for association with disease; accordingly the variants in Table 2 may be potentially deleterious (or protective) but not necessarily pathogenic. The most common disease-associated variant was m.195T>C (rs2857291), located in the non-coding D-Loop and associated with bipolar disease in MITOMAP and with numerous African haplogroups, but there was no evidence for POAG association; the minor allele was observed at 30% frequency in the case and control pools. The missense mutation m. l0086A>G, associated with the L3b haplogroup, has been linked to hypertension-associated end stage renal disease in African Americans 82; however, the risk allele was estimated here to be more common in controls (7.9%) than in POAG cases (4.5%), consistent with L3b ancestry not conferring elevated POAG risk among African Americans or being mildly protective (FIG. 4). LHON Variants
LHON is caused directly by mutations in mitochondrial DNA (mtDNA), albeit with highly variable expression and penetrance that is modified by the mtDNA genetic background 22'19. LHON is characterized by the selective degeneration of the retinal ganglion cell layer and the optic nerve that develops in young adults, and approximately 90% of LHON is attributed to three mitochondrial mutations (m.3460G>A, m. l l778G>A, and m.l4484T>C) in the MT-ND4, MT-ND6, and MT-ND1 genes, respectively 84.
However, approximately 10% of LHON cases do not harbor one of these mutations, and this disease has variable penetrance and expressivity. The primary variants that cause LHON have not been linked to POAG, and these three mutations were not detected in either the case or control pools. Other mutations have been described as causing LHON, but their status is less certain, lacking evidence of clear segregation with cases.
"Secondary mutations," common polymorphisms identified as more common in LHON cases, e.g., m.4216T>C, m. l3708G>A, and m. l5257G>A, have also been reported29, and multiple studies have found that mitochondrial haplogroup background affects the clinical presentation of LHON 22 ' 19 ' 31. The missense variant m.15812G>A was detected in the POAAGG population and has been proposed as an LHON helper variant (Table 2), but the estimated frequencies in the cases and controls did not differ substantially. The m. l0398A>G missense variant has also been proposed as an LHON secondary mutation that may act synergistically to increase the penetrance of LHON when coupled with one of the four primary mutations 70'71. The m. l0398A>G polymorphism is associated with African haplogroups Llclal, LOdlbl, and L3ela3.
However, we observed the minor allele more frequently in the POAG control pools (9.1%) relative to the case pools (6.1 %); thus, it does not appear to be a potential risk factor for African American POAG. This variant has been associated with Fuchs' endothelial corneal dystrophy in a study of Europeans, with the minor allele (G) associated with decreased risk 50. The minor allele has also been linked to decreased risk of Parkinson disease, albeit with weak and conflicting evidence 56. A study of Chinese patients with Leigh syndrome (LS) associated the G allele with increased risk for LS 0. LS is a mitochondrial illness with symptoms that may include optic neuropathy. rRNA Genes (MT-RNR1, MT-RNR2, And Humanin) TheMT-RNR2 gene, which encodes the 16S subunit of the mitochondrial ribosome, harbored seven variants that have associations with POAG that were potentially significant (FIG. 5). The m.2220 position in MT-RNR2 was a top hit, and Sanger sequencing confirmed that variance at m.2220 was significantly more common in POAG cases (OR=2.0, p=0.003), implicating Llc2bla'b haplogroups as risk factors for POAG. The top association in MT-RNR2 predicted by Pool-seq was for a much more common variant, m.2416T>C (rs28358580), which is associated with L2 and L3d3al a haplogroups, and had an estimated variance frequency of 31% in the case pools versus 18% in the control pools. Sanger sequencing of 2,090 individuals confirmed that m.2416T>C was significantly more common in cases than controls (OR 1.2, p=0.008) in the Pool-seq group; however, the difference was less significant (p=0.02) when data from the replication group of 1,389 samples was added (Table 1).
The apparent absence of m. l5208A>G, associated with haplogroup L3d3al, suggests that any association of m.2416T>C with POAG is due to the L2 haplogroups. We are not aware of the prior association of m.2416T>C with disease, but two other variants in MT-RNR2, m.2352T>C (rs28358579) and m.3010G>A (rs3928306), have been associated with pathologies and were predicted by Pool-seq to have potential association with POAG. In the case of m.2352T>C, the minor allele (C) appears to be protective for POAG; accordingly, the majority of the study subjects possess the potential risk allele. Sanger sequencing confirmed that the C allele was more common in the controls, and the difference was significant in the Pool-seq group of 2,090 samples (OR=0.8, p=0.017), but the difference was not statistically significant in the replication group (Table 1). The m.3200T>A variant was confirmed to be elevated in cases (OR 1.4), consistent with the prediction that haplogroup L2c confers risk (FIG. 4), but this difference was not statistically significant (p=0.07, Table 1).
Another 16s RNA variant, m.3010G>A, which had a prior disease association, was estimated by Pool-seq as 7% in the POAG pools versus 4% in the control pools. Sanger sequencing confirmed that m.3010G>A was common (5.2% in cases), but this proportion was nearly identical in the controls (5.1%; Table 1). A minor allele, m.3010A, is associated with cyclic vomiting syndrome with migraine (MITOMAP). Interestingly, the m.3010G>A variant has also been implicated in sensitivity to linezolid, an antibiotic that can cause optic neuropathy.
The m.2706 and m.3010 positions were of particular interest because they are reportedly close to the ribosomal peptidyl transfer center, which binds antibiotics such as chloramphenicol and linezolid, and these positions define some mtDNA haplogroups implicated in LHON penetrance. On account of their bacterial endosymbiotic origin, mitochondrial ribosomes may be unintended targets of antibiotics that disrupt bacterial protein synthesis, e.g., linezolid. Pacheu-Grau et al. 57 reported that cybrids harboring the m.301 OA allele had significantly lower amounts of mitochondrial translation products, lower ratios of p. MT-COl /succinate dehydrogenase subunit A, and lower ratios of complex IV quantity to citrate synthase activity after treatment with linezolid.
The m.3010G>A variant has arisen repeatedly; it is present in some western European populations and is associated with multiple phylogenetically distant African haplogroups (L0a4, L2alc, and L4bla). The D4 haplogroup, found in Asian and Native American populations, is defined by m.3010G>A and two other variants and has been associated with the clinical expression of LHON in Chinese patients 85. In Europeans, m.3010G>A is one of two variants that define Jl haplogroups, which have been solidly established as increasing the penetrance of the primary LHON mutations 22'19. In our study population, this variant appears to stem from African haplogroup L2alc and European haplogroup H65a. However, our results do not suggest that m.3010G>A confers elevated susceptibility to POAG.
In the case of m.2706, the m.2706A allele is one of two variants that define the European haplogroup H, reported to be an LHON resistance factor 5. Most African Americans (97.5% of cases versus 96.6% of controls) were found to possess the m.2706G allele, which appears to be ancestral, but the association with POAG was insignificant. Mitochondrial J cybrids, associated with LHON severity, reportedly have significantly decreased mitochondrial protein synthesis, and it was proposed that this might be explained by m.2706G 11. If variation at this position influences mitochondrial protein production, on account of proximity to the peptidyl transfer center, it is tempting to speculate that m.2706G may contribute to African Americans' elevated susceptibility to POAG relative to populations who have European ancestry, in addition to the LHON risk of some J European haplogroups. However, our results do not support this proposal. The MT-RNR2 gene also encodes humanin, a peptide that has been shown to have protective effects in age-related diseases, particularly Alzheimer 24. Pool-seq did not detect any common variants within the humanin-encoding region, but bidirectional Sanger sequencing on individual DNAs detected an unusual rare mutation, m.2639C>A, in a single glaucoma patient; therefore, it is not impossible that variation in this
neuroprotective peptide might be relevant to glaucoma pathogenesis in isolated cases. The patient's mutation is predicted to cause the third amino acid in the humanin peptide, proline ("P3"), to be replaced by threonine. P3 is known to be an essential amino acid for humanin function, and replacement of P3 by alanine abrogates function 8 .This particular patient was enrolled in the POAAGG study at age 89 with bilateral POAG, which was severe stage in the right eye and moderate stage in the left eye. Before cataract surgery, the patient had high myopia (-7.50 sph right eye, -8.00 + 0.75x055 left eye). The axial length of her right eye was 26.68 mm and that of her left eye was 26.72 mm. She was noted to have myopic appearing nerves at age 49, slight pallor of the right optic nerve at age 56, and bilateral pallor with optic disc cupping in the right eye at age 61, at which time IOP- lowering therapy was started. Disc photos taken after glaucoma diagnosis show peripapillary atrophy in both eyes with optic nerve cupping in the right eye greater than in the left. The patient's maximum intraocular pressure was 23 mm Hg in both eyes, and her pachymetry was normal (545 μηι in the right eye and 550 μηι in the left eye). The patient's past medical and surgical history was significant for arthritis, spinal disc surgery at age 49, and vascular surgery of the right leg at age 81.
Three common disease-associated variants (m.750A>G, m.921T>C, m. l438G>A) were detected MT-RNRl, encoding the 12S RNA gene of the mitochondrial ribosome, but their estimated prevalence differed little between the POAG case and control pools.
Sequence Variation In The MT-C01 Gene
Heritable and somatic mutations in the MT-COl gene are found at disproportionate rates in patients with prostate cancer 6. For prostate cancer, as for POAG, African
American ancestry, advanced age, and positive family history are recognized as important risk factors. Two variants that were found to be significantly associated with prostate cancer in African American men63 are the synonymous m.6221T>C polymorphism, associated with haplogroup Llc3, and m.7389T>C (rs9783095, associated with haplogroups LI, LOdlala, L3d3b B4f) that causes a Tyr496His missense change. Sanger sequencing showed that m.6221T>C was not elevated in POAG cases.
The m.7389T>C MT-COl missense variant was estimated by Pool-seq to be slightly more common in the cases (21%) than in the controls (17%), but this variant was outside the regions that were also Sanger sequenced; therefore, this result could not be confirmed. Three additional missense variants implicated in prostate cancer59 were determined to be common in the POAAGG population with minor allele frequencies estimated as 2%-4%: m.6150G>A (Val83Ile, absent from dbSNP, haplogroups Llc2, HVlalb), m.6253T>C (rs200165736, Metl l7Thr, H15, D5bl, Mla3bl, M13'46'61, A2am, Llc2), and m.6663A>G (rs200784106, Ile254Val, L2alc). The existence of all four missense variants was confirmed with Sanger sequencing on individual DNAs (FIG. 2B), and Sanger sequencing confirmed there was a significant enrichment of m.6150G>A (p=0.01), m.62530T (p=0.04), and m.6480G>A (p=0.04) variants among the POAG cases. These data suggest that MT-COl missense mutations might simultaneously elevate African Americans' risk for prostate cancer and POAG.
Mitochondrial Variants Linked To The Microbiome In recent years, the human microbiome has been linked to several common diseases, including an ocular disease, uveitis, which is triggered by microbiota-dependent autoimmunity34. It has been proposed that manipulation of the microbiome could be beneficial for uveitis and for glaucoma, possibly by modulating brain-derived neurotropic factor to promote survival of retinal ganglion cells28. The oral microbiome has recently been implicated in glaucoma; patients with glaucoma were found to have significantly higher oral bacteria counts than controls7. Interestingly, two of the variants with the strongest potential for association with POAG from the present study, m.16390G>A (HVS 1 hypervariable region), and m.15784T>C (MT-CYPB), were also top hits in an association study of mitochondrial variation with alterations in the gut microbiome53. The m.16390G>A variant was also found to be more common in type 2 diabetes cases than in controls in a Tunisian study population, but the nominally significant association (p=0.04) did not survive multivariate regression analysis36.
Use of Next-Generation Sequencing Of MtDNA
Most of the common variants described in this report were previously reported in a study of 22 African American POAG patients and 22 controls, based on whole mitochondrial genome next-generation sequencing of individual (unpooled) DNAs . That study was intended to detect novel pathogenic variants, but the African American POAG cases were found to closely resemble the controls in the proportion and likely
pathogenicity of novel variants, transversion rates, and mutational spectrum detectable in blood. Jeoung et al. sequenced mtDNA in a discovery cohort of 20 Korean patients with normal tension glaucoma (NTG), followed by Sanger sequencing on 196 patients with NTG and 202 controls41. Their association of m.4883C>T in the MT-ND2 gene with NTG survived correction for multiple testing, but this variant was not considered a high- confidence variant in the present study. Sundaresan et al. performed whole mitochondrial genome sequencing on the Ion Torrent platform for 32 POAG cases with Irish and Indian ancestry; the researchers identified 15 novel or known variants with pathogenic potential based on PolyPhen analysis73. Two of these variants, m. l 892A>G and m.2755A>G, were located in the MT-RNR2 16s RNA gene, but they were not detected in the African American POAAGG cohort.
A Mitochondrial Etiology For POAG
Recent human cybrid studies have found mitochondrial haplogroup background to be associated with multiple functional effects. As an example, a comparison of H
(European) versus L (African) haplogroup cybrids found differential expression of respiratory genes, differences in mtDNA copy number, ATP turnover rates, reactive oxygen species production, and differences in nuclear gene expression, including inflammation-related signaling genes and the canonical complement system50. A review by Coskun et al.15 outlined a mechanism for Alzheimer and Parkinson disease progression in which common mitochondrial variations may influence individuals' predisposition to neurodegeneration. The researchers proposed that some ancient functional mtDNA variants may now be maladaptive, causing mild defects in mitochondrial function.
Although these heritable variants are insufficient in themselves to reduce the threshold required for normal neurologic function, they might accelerate the rate of age- related accumulation of somatic mtDNA mutations, permitting a threshold to be crossed, beyond which sufficient mitochondrial energy production cannot be sustained. Such defects would manifest primarily in tissues with extremely high energy demands, such as retinal ganglion cells. This mechanism is consistent with the associations found between some common mitochondrial variants and POAG in our study, but functional studies are needed. Subjects harboring these variants may be more sensitive to elevated intraocular pressure or environmental insults, such as exposure to linezolid or other mitotoxic chemicals. Interventional studies, involving protective antioxidants and other approaches, have demonstrated that such therapies may promote neuroprotection by counteracting oxidative stress and mitochondrial dysfunction60. Accordingly, mitochondrial sequence analysis has the potential to personalize such treatments.
The Pool-seq methodology enabled the rapid and economic integration of whole mitochondrial genome data from a large study population but imposed several limitations. As shown in FIG. 2B, the minor allele frequencies inferred by Pool-seq were subject to experimental error. This is likely to result from variable normalization of mitochondrial DNA within each pool, resulting in disproportionate representation of individuals. The concentration of DNA eluted from the normalization plates was somewhat variable, and total DNA concentration was used as a proxy for mitochondrial DNA concentration, whereas the ratio of mitochondrial to nuclear DNA may have differed from sample to sample, and at least one study has reported that haplogroup background affects mtDNA content45. The latter errors might be minimized by quantifying mitochondrial DNA directly with a method such as qPCR, instead of using total DNA concentration as a proxy for mtDNA. Another possible source of error was unequal PCR amplification, caused by variation in initial template DNA quality, or during the downstream steps during sequencing library construction. As a result of these issues, it is possible that
representation of some individuals was lost before or during library construction, and particular individuals are necessarily under- or overrepresented within each pool.
Accordingly, the effective population size of the case and control pools is likely to be smaller than assumed in the calculations that were used to rank potential variant associations (FIG. 5), and the frequencies of variants and their ancestral haplogroups may be over- or underestimated.
Another limitation stems from sequencing pools of enzymatically sheared DNAs in conjunction with a short read (about 200-300 bp) technology. The analysis used reduction of the read alignments to simple nucleotide counts to infer population frequencies. This means that potential information about the phase of nearby variants within individual reads was not considered; instead, the presence of individual haplogroups was inferred by the presence of a particular set of defining variants, not the complete haplotypes that might have been obtained from complete mtDNA genome sequencing on individuals. Because many polymorphisms have arisen independently on multiple lineages, using the frequencies of individual variants as a proxy for haplogroup frequency necessarily yields approximations that are most likely to overestimate, but as shown in FIG. 3, the estimates accorded reasonably well with those from an independent study of haplogroup prevalence in other populations of African Americans.
We did not attempt to analyze the pooled sequencing data for the presence of insertion or deletion (indel) variants or complex changes; therefore, the data interpretation was necessarily restricted to point mutations. It is also possible that some of the point mutations that were inferred are artifacts of adjacent indels, although most correspond to known common variants, which are predominately point mutations. Finally, a subset of positions on mtDNA that were characterized as "noisy," corresponding to about 2% of the mitochondrial genome, were excluded from the association analysis. However, other ancestral variants from the same haplotype would have been included if they corresponded to high-confidence positions.
In summary, the Pool-seq of 1,999 individuals demonstrated that the mitochondrial haplotype distribution of the POAAGG study population is broadly similar to that of other African American populations, with all major African macrohaplogroups represented and with L3, L2, LI, and L0 in decreasing order of frequency, consistent with a preliminary survey of this population13. Significant differences were found between haplogroup prevalence in the POAG case and control subgroups. Specifically, the case group was confirmed to be enriched in Llc2 lineages, which harbor up to three disease-associated missense variants in the MT-COl gene and an MT-RNR2 variant. L2 haplogroups were also implicated as risk factors for POAG; these lineages contain a common variant MT- RNR2 and ones recently linked to the microbiome, whereas the Lib haplogroup was confirmed not to be a risk factor for POAG among our subjects. Approximately one in four African Americans possesses mitochondria that have Llc2 or L2 ancestries. If they are confirmed as POAG risk factors, those subpopulations might be prioritized for screening efforts and for inclusion in clinical trials designed to test therapies or dietary interventions intended to preserve or enhance mitochondrial function.
EXAMPLE 9: The MT-C01 V83I polymorphism is a risk factor for primary open-angle glaucoma in African American men. Introduction
Primary open-angle glaucoma (POAG) is characterized by chronic and progressive degeneration of the optic nerve and loss of retinal ganglion cells, coupled with
corresponding visual field defects. The genetics of POAG are complex, and many genes and biological pathways have implicated by candidate gene and genome wide-association studies 40 8 . Elevated intraocular pressure (IOP) is a major risk factor, but POAG may also occur at normal tension. Other well-known risk factors include positive family history and African ancestry. POAG is highly prevalent in African Americans (AA); a recent study found that 9.2% of Philadelphian AA over age 50 had POAG with an additional 20.9% classified as glaucoma suspects 81 .
Mitochondrial dysfunction figures prominently in POAG pathogensis as in other progressive neurodegenerative diseases, such as Alzheimer's disease (AD) and
Parkinson's disease (PD). The vast majority of mitochondrial proteins are expressed from nuclear genes, but the human mitochondrion has its own small circular genome (mtDNA), a relic of an ancient bacterial genome, which has retained some key proteins of the oxidative phosphorylation (OXPHOS) pathway. Mutations in nuclear mitochondrial genes can cause a wide spectrum of disease, often with serious neuorological symptoms that may include optic neuropathy. The first disease shown to be caused by mutations in mtDNA was Leber's hereditary optic neuropathy (LHON)
One of the genes retained by mtDNA MT-COl, encodes cytochrome c oxidase subunit 1, (COl). COl protein is localized to the mitochondrial inner membrane, where it is an essential component of Complex IV, the final enzyme complex in the electron transport chain, which transfers electrons from reduced cytochrome c to molecular oxygen, producing water. Complex IV is essential for all oxygen-based life, and is the target of some potential neuroprotective interventions that may protect RGCs from death as a result of glaucoma. For example, methylene blue has been shown to protect rat RGCs from toxic insults to mitochondria, such as rotenone, which can cause optic neuropathy 16 . Administration of near-infrared light is also neuroprotective, likely via the same mechanism as methylene blue: transfer of electrons directly to Complex IV25 . Mutations in MT-COl gene or Complex IV dysfunction may cause of spectrum of symptoms. Both germline and somatic missense variants in MT-COl have been implicated in prostate The Primary Open-Angle African American Glaucoma Genetics (POAAGG) study 11 previously reported disease-associated missense mutations 1 in the N-terminal region of MT-COl, and found three of these (V83I, M117T, V193I) to be associated with POAG 14 . The m.6150G>A (V83I) missense mutation (rs879053914) is the most common in POAAGG subjects, with a minor allele frequency of approximately 5%, and was also the most significantly enriched in AA POAG cases vs. AA controls (OR=1.8, p=0.01) 14 . All three COl amino acid changes are present in the Llc2blb African mitochondrial haplogroup, and two of these: V83I and Ml 17T, are among the variants that define Llc2 haplogroups.
The V83I COl mutation is of particular interest, because it lies within a region of
COl that interacts with amyloid beta 2 , the product of the APP gene which figures prominently in the pathology of AD. COl has also been reported to interact with alpha- synuclein, which is found in Lewy bodies and associated with PD and Lewy body dementia. Substantial evidence indicates that Αβ may be involved in POAG pathology, and that AD and POAG may share etiologies 15 20 . There is overlap between AD and POAG pathologies, for example glaucoma is associated with Αβ accumulation in the retina with decreased Αβ in the vitreous, and AD is associated with retinal nerve fiber layer and visual cortex pathology 29 .
To further explore the association of MT-COl missense mutations with POAG in African-Americans, we determined whether the previously observed POAG associations related to gender; screened the N-terminal region of MT-COl for rare missense variants, characterized the phenotypes of POAG patients having the V83I mutation relative to other POAG patients; tested whether the reported COl / Αβ interaction is affected by the V83I amino acid replacement, and screened for other potential COl protein interactors, in order to shed light on the biological function(s) of the N-terminal region of COl .
Methods - Study subject recruitment
The baseline demographics of the POAAGG study and inclusion and exclusion criteria have been described previously 11 . Subjects were age 35 or older and self- identified as Black, African American or African ancestry and were classified as glaucoma case or control by glaucoma specialists. Subjects classified as glaucoma suspects were excluded from all analyses. Research was approved by an institutional review board at the University of Pennsylvania, and was conducted in accordance with the Declaration of Helsinki 1 .
Methods - PCR, DNA sequencing and mutation analysis
An amplicon containing the N-terminal region of the MT-COl gene had been PCR amplified and Sanger sequenced, with methods and results reported previously 14 . The original Sanger sequencing cohort contained POAG cases, controls and some glaucoma suspects (excluded from analyses). Version 5.2 of Sequencher software was used to score sequencing chromatograms for rare missense variants. Because some subjects had progressed since the cohort was sequenced, e.g. from control to suspect or case, subject status was updated to be current as of August, 2016, prior to re-analysis for association with POAG by gender. The statistical significance of differences in MT-COl variant frequencies in females and male cases vs. controls was determined using Fisher's exact test, 2-tailed, with version 6 of GraphPad Prism software. Missense variants were annotated using the Mitlmpact resource and current version of SIFT to identify potentially deleterious mutations. The MITOMAP compendium was used to identify previously reported variants and reports of disease association.
Methods - Phylogeny, localization and conservation of amino acid replacements
Build 17 of the PhyloTree resource and the MITOMAP compendium were used to associate mitochondrial haplogroups with the observed mtDNA variants.. Predictions based on sequence analysis UniProt were used to infer the subcellular localization of MT- COl missense mutations relative to the mitochondrial inner membrane. Phastcon was used to assess the relative degrees of conservation of mtDNA variants in an alignment of 100 vertebrate sequences.
Methods - Phenotypic characterization of POAG patients
For the masked comparisons, 29 patients were selected who possessed V83I and
Ml 17T ("double missense") mutations, associated with mtDNA haplogroup Llc2.
"Triple missense" patients, possessing the third variant, VI 931, associated with closely related haplogroup Llc2blb, were excluded. The control group was 29 patients, having the m.6548C>T synonymous variant, which is diagnostic of haplogroup Lib, which lacks all three missense variants. Each Llc2 subject was paired with an Lib control patient having the same gender and reported family history of glaucoma (yes or no), and similar age at enrollment. Haplogroup was masked during chart review. P-values were calculated using logistic regression, with general estimating equations to account for differences between eyes.
Yeast 2-hybrid studies were performed by Hybrigenics Services, Paris, France as follows:
1) Construction of Y2H vectors
The coding sequence of the human Coxl fragment (aa. 41-101 of GenBank accession number gi: 251831109 and NCBI Ref YP_003024028.1 and SEQ ID NO: 3) was PCR-amplified and cloned in frame with the LexA DNA binding domain (DBD) into plasmid pB27 (N-LexA-Coxl-C), derived from the original pBTMl 1680 . The DBD construct was checked by sequencing the entire insert. Hybrigenics' reference for this "bait" is hg3681vl_pB27. Codon usage of the Coxl insert was optimized for yeast expression and adjusted to prevent amino acid changes from differences in the human mitochondrial and yeast nuclear genetic codes.
The "prey" fragment for the human APP was aa. 672-713, GenBank accession number gi: 228008403. The prey fragment was cloned in frame with the Gal4 Activation Domain (AD) into plasmid pP7, derived from the original pGADGH 8 . The AD construct was also checked by sequencing. The prey Hybrigenics reference is hgx3682vl_pP7.
2) Y2H test COl / Αβ interaction
Bait and prey constructs were transformed in the yeast haploid cells: L40 Gal4 (mata) strains. The diploid yeast cells were obtained using a mating protocol with both yeast strains 18 . These assays are based on the YHGX13 (Y187 ade2-101 ::loxP-kanMX- loxP, mat HIS3 reporter gene (growth assay without histidine). The following interaction pairs were tested: Smad3 / Smurfl (Hybrigenics' positive control, empty LexA bait vector / empty prey vector (negative control) , empty LexA bait vector / AD-APP (negative control) , LexA-Coxl/ empty prey vector (negative control), LexA-Coxl / AD-APP.
Interaction pairs were tested in triplicate, using a growth assay on +/-His and +/- 3- AT plates, and streaks from three independent clones from each diploid were picked for the growth assay. The DO-2 selective medium lacking tryptophan and leucine was used as a growth control and to verify the presence of the bait and prey plasmids. The DO-3 selective medium without tryptophan, leucine and histidine selects for the interaction between bait and prey.
3) Y2H library screen with COl bait Two yeast two-hybrid library screens were done to screen for protein interactions, with the region of COl containing V83 as "bait". The COl amino acid sequence was the same used in the ELISA assays, but codon usage was changed to comply with the standard (non-mitochondrial) genetic code and to optimize expression of the fusion proteins in yeast.
4) Y2H quantification of COl / UBQLN1 interaction
The coding sequence of the human Coxl wt and Cox-1 mut fragments (aa. 41-101 of SEQ ID NO: 3) were synthesized and cloned in frame with the Gal4 DNA binding domain (DBD) into plasmid pB66 (N-Gal4- Bait-C), derived from pAS2 18. The DBD constructs were checked by sequencing the entire insert. Hybrigenics reference for those baits are hgx3681vl_pB66 (for Coxl wt) and hgx3915_pB66 (for Cox-lmut). The prey fragments for the human SCN1A (aa. 1699-1820, GenBank accession number gi:260166632) and UBQLN1 (aa.108-388, GenBank accession number gi: 194328681) were extracted from the ULTImate Y2HTM screening of Coxlwt ((aa.41-101 of SEQ ID NO: 3) against the Human Adult Brain cDNA library). The prey fragments are cloned in frame with the Gal4 Activation Domain (AD) into plasmid pP6, derived from the original pGADGH 8. The AD construct was checked by sequencing. The prey Hybrigenics references are hgx3681vl_pB27_B7 (SCN1A) and hgx3681vl_pB27_A346 (UBQLN1). The pP7 prey plasmid used in the control assay was derived from the pP6 plasmid.
Bait and prey constructs were transformed in the yeast haploid cells, respectively
CG1945 (mata) and YHGX13 (Y187 ade2-101 ::loxP-kanMX-loxP, mat) strains. The diploid yeast cells were obtained using a mating protocol with both yeast strains 18 . These assays are based on the HIS3 reporter gene (growth assay without histidine).
The following interaction pairs were tested: 1) Smad3 / Smurfl (Hybrigenics' positive control, Colland et al, 2004), 2) empty Gal4 bait vector / empty prey vector
(negative control), 3) empty Gal4 bait vector / AD- SCN1A (negative control), 4) Gal4- Coxl wt / empty prey vector (negative control), 5) Gal4- Coxl wt / AD- SCN1 A
(interaction), 6) Gal4- Cox-1 mut / empty prey vector (negative control), 7) Gal4- Cox-1 mut / AD- SCN1A, 8) empty Gal4 bait vector / AD- UBQLN1 (negative control) , 9) Gal4- Coxl wt / AD- UBQLN1 (interaction), 10) Gal4- Cox-1 mut / AD- UBQLNl(interaction) . Interaction pairs were tested in triplicate in form of streaks as three independent clones from each diploid were picked for the growth assay. The DO-2 selective medium lacking tryptophan and leucine was used as a growth control and to verify the presence of the bait and prey plasmids. The DO-3 selective medium without tryptophan, leucine and histidine selects for the interaction between bait and prey.
Interaction of Αβ and COl peptides.
Sandwich ELISA assays were performed with biotinylated synthetic peptides to confirm the previously reported affinity of Αβ (1-42) peptide for an N-terminal region of COl, and to quantify the effect of the V83I amino acid replacement on this interaction. Purified APOE protein was used as a positive control for interaction with Αβ, and TBX3 protein (OraGene) and scramble Αβ peptide were used as negative controls for evaluating interaction with Αβ.
Sequencing identified additional missense variants inMT-COl.
Sanger sequencing of the N-terminal region οΐΜΤ-COl identified 21 missense variants, in addition to the three variants, that had been previously detected in the
POAAGG cohort 1 and found to have nominal association with POAG 14 (m.6150G>A (V83I), m.6253T>C (Ml 17T), m.6480G>A (V193I)). Seven of these twenty-four variants have been implicated in prostate cancer 59 , in addition to the three targeted variants (Table 6). SIFT analysis classified six of the missense variants (L20M, N46D, N46T, T146A, F148L, L215F) as deleterious mutations, however with a total of five observations in POAG cases and four in controls. Only two of the variants classified as deleterious by SIFT, L20M and T146A, were not observed in controls, but each was observed in only one case. A larger number of cases than controls were sequenced, so these results suggest that predicted deleterious rare variants, defined as SIFT score < 0.05) are unlikely to play a significant role in AA POAG. One variant, m.5913G>A (D4N), not predicted to be deleterious, had a relatively low SIFT score (0.11) and was observed in 5 cases and 0 controls; however the difference was not significant. D4N, in addition to being prostate- cancer associated, has also been reported in a Tunisian family with maternally inherited diabetes and deafness, but was not established as the cause of disease 74 .
The three variants that had been nominally associated with POAG, V83I, Ml 17T, and VI 931, were all classified as neutral by SIFT, albeit with scores towards the deleterious end of the spectrum, ranging from 0.09 (Ml 17T) to 0.23 (V193I). Because they are moderately well conserved, it is likely these sites are functionally constrained and could be adaptive mutations. Other missense variants, e.g. A3V and D4N, were observed more frequently in cases than controls, but the differences were not significant.
Table 6. Missense variants in COl identified in African American POAG cases and controls.
Figure imgf000051_0001
m.6678A>G** T259A neutral 0.25 1 1022 0 525
PC = prostate-cancer associated i MITOMAP
** = novel variant
Phylogenetic relationship ofMT-COl variants
The mtDNA of all living humans is descended from a "most recent common ancestor (MRCA), an African woman who lived approximately 200,000 years ago.
Common mtDNA variants are associated with one of more of the lineages (haplogroups), and African haplogroups are represented by the letter "L", with L0, LI, L2, L4, etc. representing the deepest roots and most ancient divergence times. FIG. 6 depicts the mtDNA family tree, and indicates the approximate locations of common synonymous and non-synonymous variants MT-COl variants in evolutionary time. The pair of missense mutations, V83I and Ml 17T, defines the Llc2 haplogroup, which arose more than 50,000 years ago. The third POAG-associated missense, V193I, occur on an offshoot of Llc2, Llc2blb. V83I also is found in a subset of POAAGG subjects belonging to the L2c haplogroup. All three variants may occur on non- African haplogroups, although none of these are common in the POAAGG cohort. The non-synonymous variant, m.6548C>T is uniquely associated with Lib haplogroups, which share LI ancestry, but lack the three missense variants that occur on Llc2 lineages.
Association of four MT-COl variants with POAG in AA males and females
Sanger sequencing data for four selected variants in MT-COl were re-analyzed after binning cases and controls by gender. These four variants were the three POAG- associated missense mutations, and a fourth synonymous variant, m.6548C>T (L215L), that defines Lib lineages (FIG. 6). As previously reported for the combined group (males plus females) the associations of the three missense variants, V83I, Ml 17T, and VI 931 with POAG were all nominally significant (P<0.04), with V83I having the most significant association (OR=1.8, p=0.01). When analyzed separately by gender, V83I was observed more frequently in female POAG cases than female controls, but this difference was not significant (OR 1.1, p=0.8) (Table 7). However, when male cases were compared to male controls, the difference was extremely significant (OR 6.5, p=0.0001); with 39 of the of the 42 V83I men having been diagnosed with POAG.
Association of Ml 17T with POAG was significant for both males and females.
Association of VI 931 with POAG was significant in the combined group, but not for males or females when analyzed separately. Association of the fourth variant, the Lib haplogroup-linked synonymous substitution m.6548C>T (L215L), with POAG was not significant in the combined group, males, or females (Table 7).
Table 7: Association οΐΜΤ-COl variants with POAG in AA males and females
Figure imgf000053_0001
Results for the combined group were reported previously , but have been adjusted to reflect subsequent changes in disease status, for example progression from case to control Characteristics of V831 POAG patients
We compared glaucoma-related traits, e.g. IOP, CCT, visual fields and family history of POAG for all patients having V83I to all V83 (wild type) patients. For a parallel comparison, we also binned patients by the synonymous variant m.6548C>T (L215L), which is associated with a related LI African haplogroup, Lib (FIG. 6).
The V83I patients also had worse visual field defects, differing significantly in pattern standard deviation (p=0.008) and mean deviation (p=0.02) (Table 8). When analyzed separately by gender, mean pattern deviation was higher in V83I males and in V83I females, but the difference was significant in males (p=0.047), but not females. The V83I patients had significantly lower IOP. The finding that V83I patients had worse visual function was surprising because IOP is a major risk factor, suggesting that V83I patients might be more vulnerable to elevated pressure. Mean central comeal thickness (CCT) was higher in the V83I group, but this difference was not significant. However the distribution of CCT across three bins was nominally significant (p=0.049), with the largest fraction of V83I patients having CCT greater than 540 mm. Thicker corneas may cause IOP to be overestimated by tonometry, so the finding that V83I patients had lower IOP does not appear to be due to thinner corneas. Thinner corneas are thought to be a risk factor for POAG, so the finding that the V83I comeas were thicker is also consistent with this group being potentially more vulnerable to disease. 73% of the V83I cases reported a positive family history of glaucoma, vs. 57% of V83 WT cases, and these proportions are also statistically significant (p=0.03). Maternal family history of glaucoma was also reported more often in V83I cases (33% vs. 20% for V83 WT), and this difference was also significant (p=0.03). Although females outnumber males cases in the cohort as a whole, by approximately 1.5 : 1, the fraction of females in the V83I group was nearly equal, that is the V83I patients were disproportionately male. Prostate cancer was noted more frequently in the V83I group, 10.2% vs. 5.6%, but this difference was not significant, and likely explained by the larger fraction of males.
The V83I group had significant differences in visual function, IOP, family history of POAG from V83 (WT) patients, however the L215L (m.6548C>T) group, representing haplogroup Lib and lacking all three missense variants, differed significantly from other patients only in that it was disproportionately female (p=0.049) and had a significantly lower incidence of prostate cancer (p=0.04), which is likely explained by the larger fraction of women (Table 8).
Table 8. Phenotypic characteristics of POAG patients having V83I (m.6150G>A) and patients having L215L (m.6548C>T).
Figure imgf000055_0001
(SD) (10.8)
Negat 424 13 403 34
Family
ive (42.6%) (27.1 %) (42.5%) (36.2%) history of 0.03 0.24
Positi 571 35 546 60
glaucoma
ve (57.4%) (72.9%) (57.5%) (63.8%)
Mother's Negat 818 33 771 80
side family ive (80.1 %) (67.3%) (79.2%) (83.3%)
0.03 0.33 history of Positi 203 16 203 16
glaucoma ve (19.9%) (32. 7%) (20.8%) (16.7%)
410 25 405 30
Male
(40.2%) (51.0%) (41.6%) (31.3%)
Gender 0.13 0.049
Fe611 24 569 66
male (59.8%) (49.0%) (58.4%) (68.8%)
Malignant 964 44 913 95
No
neoplasm / (94.4%) (89.8%) (93.7%) (99.0%)
Carcinoma 0.18 0.04
57 5 61
in situ of Yes 1 ( 1.0%)
(5.6%) (10.2%) (6.3%)
prostate
1019 48 971 96
Intellectual No
(99.8%) (98.0%) 0.02 (99.7%) (100.0%) 0.59 disabilities
Yes 2 (0.2%) 1 (2.0%) 3 (0.3%) 0 (0.0%)
Masked chart reviews ofhaplogroup Llc2 (V83I + M117T) vs. Lib patients
Because the V83I patients, mostly having the Ll c2 African haplogroup, had significantly worse visual function than V83 WT patients despite lower IOP, we followed up with a masked chart review, focused on Ll c2 patients (n=29). In order to control for LI mtDNA ancestry, patients from the Lib haplogroup (wild type for V83 and Ml 17) were used as the reference group (n=29). The two groups were also matched for age, gender, and family history of POAG (yes/no) in order to control for these potential confounders. The chart review was conducted with a trained glaucoma specialist. The Ll c2 group had worse disease, as assessed by ICD9 codes for glaucoma severity that had been previously entered in EPIC, with 52% of Ll c2 having "severe" disease as opposed to 9.5% of Lib (Table 9). Table 9. Glaucoma severity codes (ICD9) of matched Llc2 vs. Lib POAG patients.
Figure imgf000057_0001
The Llc2 group had significantly higher mean cup-to-disc ratio (CDR) (p=0.04), and worse visual function, as assessed by mean pattern standard deviation (p=0.009) and mean deviation (p=0.006) at visit closest to diagnosis (Table 10). IOP was lower in the Llc2 group despite their worse disease, although the difference did not quite reach statistical significance (p=0.07). CCT (not shown) did not differ significantly between the Llc2 and Lib groups.
Table 10. Phenotypic traits of matched Llc2 vs. Lib POAG patients.
Figure imgf000057_0002
Subcellular localization ofV83, Ml 17 and VI 93 on COl protein
Sequence analysis by UniProt predicted that the V83I replacement affects a residue located on the inner side of the inner mitochondrial membrane, which is exposed to the mitochondrial matrix, whereas Ml 17T and V193I are predicted to be located inside the mitochondrial inner membrane. The locations of these residues and domain organization of the COl protein are depicted in FIG. 10. The reported Αβ -interacting region of the COl protein, containing V83, spans an intermembrane, transmembrane and mitochondrial matrix region.
Yeast 2-hybrid (Y2H) test of COl / Αβ interaction, and cDNA library screen for COl interactors.
We were unable to demonstrate an interaction between COl and Αβ in the Y2H assay. The Y2H screen using the N-terminal COl fragment as "bait" retrieved products of the following genes: UBQLNlmd SCN1A.
Quantification of wild type and mutant (V83I) COl fragment with Ubqlnl fragment in yeast two-hybrid system.
The interaction between COl wt or mut and SCN1 A could not be confirmed. The
COl (WT) / Ubqll interaction was confirmed to be very strong; however the mutant COl
(V83I) / Ubqll interaction was equally strong, so V83I did not affect this interaction. Both interactions resisted up to 10 mM of 3 AT (data not shown).
Interaction of COl αηάΑβ peptides
The reported affinity of wild type COl (Val83) peptide for Αβι-42 2 was confirmed by ELISA, whereas mutant COl (V83I) was found to interact very weakly with Αβ (FIG.
11). At the highest peptide concentrations tested, V83I reduced the interaction by 92%, to a level that was comparable to the negative control with TBX3 protein. However wild type and mutant (V83I) COl peptides had similar lack of affinity for the scrambled Αβ negative control peptides, suggesting that the interaction of COl and Αβ is specific, and is disrupted by V83I.
Male-female dimorphism in POAG, mitochondrial variants and optic neuropathy
The association of gender with POAG is controversial, with different studies reporting higher prevalence in either gender or inconclusive results, however a recent meta analysis found men to be at higher risk of POAG than women. Estrogen appears to be protective for POAG and AD with a mechanism that may involve Αβ. A recent study of nuclear SNPs near 9p21 reported a significantly stronger (p=0.01) association with normal tension glaucoma in females (OR 1.5) than males (OR 1.35) 54 .
Inheritance of mitochondria is matrilineal, so male-specific phenotypes are not expected to have fitness consequences for mitochondria, and has been proposed to explain why LHON affects males much more severely than females, with much higher penetrance and an earlier age of onset in males 17 . LHON is caused by three mtDNA mutations, with penetrance higher in men than women. The association of m.6150G>A (V83I) with POAG showed an extreme male gender bias similar to LHON, with an odds ratio nearly 5.9-fold higher in men than in women (Table 6).
The V831 POAG phenotype
The V83I patients had worse visual function and degeneration of the optic nerve, despite significantly lower IOP, even after controlling for age, gender, family history of POAG, African-American ancestry and LI mtDNA haplogroup. This evidence is consistent with the proposal that this missense mutation may contribute to POAG pathogenesis, and these patients may more vulnerable to POAG in general, and at lower IOP than other African Americans. The potential involvement of AD genes APP, and UBQLNl, implies that AD / dementia or other intellectual disabilities could be more likely to be co-morbidities with POAG in these patients, although this can't be concluded from our results. The survey of ICD9 codes related to AD / dementia found the proportion of V83I POAG cases with intellectual disabilities was higher than non-V83I cases, however this was based on only one observation in V83I cases.
Interestingly, a study of the association of mtDNA variations with dementia risk and Αβ in elderly African Americans found that haplogroup LI participants were at elevated risk for dementia (OR 1.88, p=.004), lower plasma Αβι-42 levels (p=0.03), and greater risk for intellectual decline, relative to the most common African American macrohaplogroup, L378. It is tempting to speculate that V83I could predispose to both POAG and dementia in African Americans, however it is not known what fraction of the LI subjects in the dementia study had V83I. The majority of LI subjects in the POAAGG study belong to Lib mitochondrial haplogroups, lacking V83I, which were not associated with significant POAG risk in our study (FIG. 6, Tables 7 and 8). Additional work is needed to determine whether V83I may predispose to both POAG, dementia and intellectual disabilities, and whether these may share a common mechanism, perhaps involving Αβ-mediated mitochondrial dysfunction.
Potential functional effects ofMT-COl missense variants
The three missense mutations may occur either alone, for example V83I on an L2c mtDNA background, or in combination, for example V83I and M117T on Llc2-related haplogroups, which may also carry VI 931 (FIG. 6). All three of the POAG-associated missense mutations have been proposed as potentially pathogenic in studies of other diseases, and might relate directly to POAG pathogenesis, either alone or in combination. a) V83I (m.6150G>A)
V83I is the strongest candidate for being causal in addition to a marker for POAG susceptibility. The association of V83I with POAG in African American men was stronger than for Ml 17T or V193I. This is because the association of V83I with POAG stemmed not only from Llc2 subjects, who also have M117T, but was some L2c subjects who have V83I, but lack the other two missense mutations. V83I is also the only mutation within the Αβ binding region, and we showed that this replacement greatly abrogated this interaction in the ELISA assay, which suggests V83 may be critical for Αβ interaction.
Another prostate cancer-linked mutation, Met74Thr (m.6124T>C), is located only 9 residues from V83. This mutation was not observed in POAAGG patients, however it also within the Αβ-binding region, and was shown to be functional, causing resistance to statin induced apoptosis 72 , increase reactive oxygen production, and enhanced cellular proliferation 6 . Met74Thr cybrid cells had a significantly faster doubling time than wild type cells. So it is reasonable to speculate that V83I or other nearby mutations might also directly affect proliferation, albeit negatively, thereby promoting neurodegeneration. The retrieval of UBQLN1, an AD associated gene with neuroprotective function, by Y2H cDNA library screen is more evidence that this region may function, perhaps in autophagy. However, we found no evidence that V83I residue is involved in the interaction with UBQLN1.
b) M117T (m.62530T)
Because M117T was observed only on Llc2 haplogroups , which also carry V83I, the (weaker) association of Ml 17T with POAG implicates V83I. Cytochrome c oxidase is a "bigenomic" protein machine with components encoded by both the nuclear and mitochondrial genomes, and the mtDNA-encoded proteins, e.g. COl, are functionally constrained by the requirement to interact with the nuclear proteins. Ml 17T has been predicted to be pathogenic based on 3D structural modeling that suggests it participates in these protein-protein interactions, and might influence the stability of Complex IV 51 . c) V193I (m.6480G>A) The VI 931 variant was originally detected in a patient having cytochrome c oxidase deficiency, and was proposed to be pathogenic, based on its absence in 300 controls 9 . However VI 931 is now known to be a common polymorphism associated with the Llc2blb African haplogroup, which also carries V83I and M117T79 and with non-African haplogroups, but it is still possible that it may be deleterious. V193I has been proposed as a "helper" variant that acts in synergy with the primary LHON mutation m.11778G>A in a Chinese family 10 . V193I, like Ml 17T, has also been predicted to be pathogenic by 3D modeling and may affect interactions with Complex IV proteins encoded by nuclear genes 51 .
Potential for interaction among COl, Αβ and Ubiquilin-1
It was surprising that the Y2H library screen with the Αβ -interacting region of COl as "bait" retrieved the ubiquilin 1 gene {UBQLNl), implicated in AD, control of Αβ production, autophagy and the unfolded protein response (UPR) 86 . UBQLNl is well known as a neuroprotective gene associated with familial AD and also as "molecular chaperone" of Αβ, and regulator of the APP amyloid precursor protein gene. UBQLNl has not been linked to POAG, but was among a small number of anti-apoptotic genes found to be up-regulated by platelet derived growth-factor CC, which protects retinal ganglion cells from death, in optic nerve crush-injured mouse retinae76 . The ability of Αβ to interact 1) with itself, 2) with COl and 3) with UBQLNl suggests that all three interactions might occur simultaneously, with Αβ as either a soluble monomer or neurotoxic oligomer.
The POAG-associated V83I COl mutation diminished the interaction with Αβι.42 in the ELISA assay (FIG. 11). COl and Αβι-42 have been shown to coprecipitate from mitochondria of human neuroblastoma cells, so this interaction appears to occur in vivo 2 . These observations suggest that the interaction of soluble Αβ with COl might be normal and potentially beneficial. Αβ has the potential to self-interact, forming oligomers that may polymerize into the fibrillar plaques that are characteristic of AD. It is possible the aggregation of Αβ directly leads to mitochondrial dysfunction, including oxidative damage by reactive oxygen species, induction of apoptosis and ion channel formation20. Amyloid peptides Αβι-40 and a-synuclein are prone to oligomerize in lipid bilayers and cause abnormal conductance 61 .
V83I did not disrupt the strong interaction of COl with UBQLNl in the Y2H system (data not shown), but did disrupt the interaction of COl with Αβ (FIG. 11) in the ELISA assays. This suggests that V83I is not directly involved the UBQLNl interaction, and that COl might interact with UBQLNl and Αβ cooperatively, as opposed to competitively. We hypothesis that Αβ may bridge and facilitate this interaction, whereas V83I may prevent the binding of UBQLNl with monomeric Αβ bound. It is possible that oligomerization of Αβ might also be disruptive to the interaction of UBQLNl and COl, if this requires Αβ to be in soluble monomeric form.
Αβ and AD and glaucoma
Glaucoma and Alzheimer's Disease (AD) have several features in common, including increasing incidence with age, and loss of specific neuronal subpopulations. Down's syndrome patients have an extra copy and elevated expression of the APP gene, and frequently develop Alzheimer's at a young age, and glaucoma in their teens. This is consistent with overlapping molecular etiologies that involve amyloidβ (Αβ). Αβ is known to be present in inside mitochondria, and intramitochondrial Αβ may directly cause neurotoxicity and mitochondrial dysfunction, including impairment of OXPHOS and interaction with mitochondrial proteins 58 . Αβ is known to localize to the mitochondrial matrix and inner mitochondrial membrane so there is a potential for Αβ to interact directly with Complex IV. It was recently shown 69 that Αβ -induced mitochondrial function involves the forkhead box 03 a FOX03A gene, which is present in neuronal mitochondrial binds mtDNA and leads to decreased MT-COl expression, so the influence of Αβ on Complex IV function may be indirect. However the mechanism(s) of Αβ -induced neurotoxicity are not well understood.
Mechanisms for mitochondrial dysfunction and glaucoma
Glaucomatous changes might arise from multiple mechanisms of mitochondrial damage, some of which might involve Αβ. These factors might act alone, or in concert with elevated IOP. For example, elevated pressure causes release of cytochrome c and OPA1 release in RGC-5 retinal ganglion cells, resulting in apoptotic cell death 4 .
Mitochondrial damage from IOP was also demonstrated in glaucomatous DBA/2J mice, where elevated IOP resulted in mitochondrial fission, matrix swelling, cytochrome c release, and a moderate reduction of expression of MT-COl mRNA 44 . Oxidative stress caused by ozone exposure has been shown to increase Αβι-42 production and accumulation of Αβι-42 in mitochondria, with colocalization of OPA1, Αβ and COl . Another potential mechanism for mitochondrial damage in AD and POAG is an increased rate of somatic mutation, and this might involve Αβ and possibly estrogen. Increased expression of Αβ and abnormal APP metabolism is associated with RGC apoptosis in experimental glaucoma models, with colocalization of Αβ with apoptotic RGCs; however it is possible that the aggregation of Αβ starts out as protective response 29
The association of V83I, Ml 17T and V193I with POAG in African American men may be explained by unknown factors associated with the corresponding mtDNA ancestries (primarily Llc2 and L2c lineages). Alternatively, other mtDNA or nuclear variants or factors linked to these ancestries may explain the strong associations with POAG in AA men. Without wishing to be bound by theory, we hypothesize that the interaction between COl and UBQLNl, detected in the Y2H screens, occurs in vivo and is functionally relevant to glaucomatous neurodegeneration. Αβ interacts with a large number of proteins, and it's normal role is not understood. Accordingly the significance of the V83I mutation to the previously reported interaction between COl and Αβ, if any, is unknown.
In summary, these data supports that common germline mtDNA polymorphisms, such as the V83I mutation in COl, play a role in susceptibility to POAG, particularly in men, and impact disease severity. Because V83I is predominately associated with African mitochondrial ancestries (Llc2, L2c and others), this variant contributes to African Americans higher POAG risk relative to Americans with European ancestries. The m.6150G>A (V83I) variant in COl was present in 6.5% of male AA POAG patients, but only 1.1% of male AA controls, and this difference was extremely significant. V83I disrupts a reported protein-protein interaction with Αβ, which could be significant in light of proposals that the etiology of POAG may overlap with other neurodegenerations, particularly AD, with the APP gene common to both. We propose that a second gene, UBQLNl, interacts with COl, and that this gene, which is a potential therapeutic target for AD 75 , is relevant to POAG too.
Interventions to treat optic neuropathies by supporting mitochondrial function are under development 27 . Patients having V83I or other COl missense mutations may benefit from emerging neuroprotective therapies that directly support mitochondrial respiration by donating electrons to cytochrome c oxidase, e.g. exposure to near infrared light, or treatment with methylene blue, which preferentially enters neuronal mitochondria after systemic administration25 . Recently metformin and the antioxidant mitotempo have been shown to be protect human neural stems and cultured mouse neurons, respectively, against Αβ -induced mitochondrial dysfunction 12 7 . Alternatively, emerging gene-based therapies now offer the ability to correct deleterious germline alterations in mtDNA.
EXAMPLE 10: Additional Data
Subject recruitment and specimen collection
The POAAGG study population consists of self-identified Blacks (African Americans, African descent, or African Caribbean), aged 35 years or older. POAAGG subjects were recruited from the University of Pennsylvania from the Scheie Eye Institute and its satellite locations in Philadelphia. All subjects provided informed written consent, in accordance with the tenets of the Declaration of Helsinki, under an IRB-approved protocol. Following visual examination by a glaucoma specialist, subjects were classified as case, suspect, or control (POAAGG). Cases were defined by demonstrating
characteristic optic nerve defects as well as corresponding visual field loss, suspects were excluded from the current study.
Blood was collected by venipuncture in 10 ml purple top tubes with EDTA anticoagulant. These samples were frozen at -20 degrees prior to DNA isolation. For saliva collection, subjects were asked to refrain from drinking or eating prior to donating specimens. Two milliliters of saliva per subject were collected in Oragene DISCOVER (OGR-500) self-collection kits (DNA Genotek, Canada). The saliva specimens were mixed with stabilizing reagent within the collection tubes per manufacturer's instructions, and these were stored at room temperature until DNA extraction.
DNA extraction and quantitation
DNA was isolated from thawed blood samples using Gentra PureGene kits (Qiagen, Valencia, CA), and the optional RNase treatment step was included. DNA from saliva samples was extracted using the prepIT.L2P reagent (cat # PT-L2P-5, DNA
Genotek, Canada) and precipitated with ethanol according to manufacturer's instructions. The saliva DNA samples were RNAse treated by double digestion with RNaseA and RNase T and re-precipitated using ethanol according to manufacturer's instructions. The concentrations of DNA from blood and saliva samples were determined using the fluorescence-based Quant iT dsDNA Board-Range (BR) assay kit (cat # Q-33130, Life Technologies, CA). Fluorescence was measured with a Tecan Infinite M 200 Pro multimode microplate reader (Tecan, NC). Genotyping
A 25 μΐ aliquot of each sample was plated for array-based high throughput genotyping. The 3,684 DNA samples were genotyped using MEGA (Multi ethnic genotyping array) V2 (EX) consortium chip on the Infinium iSelect platform by Illumina FastTrack Services (Illumina, San Diego, CA). The MEGA chip has 1.8 million successfully genotyped SNPs and 1294 of those are Mitochondrial SNPs. The genotype calls were generated using the Genome Studio genotyping module (GT). Cluster optimization, reproducibility analysis for paired samples were performed as per standard practices at Illumina FastTrack services.
Phylogeny and Haplogroup assignment
Build 17 of the PhyloTree resource (http://www.phylotree.org/ref?) and the MITOMAP compendium (http://mitomap.org/MITOMAPref?) were used to associate mitochondrial haplogroups with the observed mtDNA variants. Haplogroup assignment of the individual subjects was performed using Haplogrep 2 prediction software
(http://haplogrep.uibk.ac.at/) based on genotypes from 916 mitochondrial positions. The genotype calls were formatted to VCF file and uploaded to Haplogrep standalone module. The haplogroups assigned were exported from the report generated by the software. Statistical analysis
The p-values for the odds ratios were calculated using chi-square tests using r software package and Graph pad prism V6. All the phenotype data was analyzed using linear regression, with generalized estimating equations (GEE) used to account for the correlation between the eyes of the same person. All the p-values for multiple comparisons (in both the chi-squares and linear regressions) were adjusted for with the Hochberg step-up method.
Results - Quality control analysis of the mitochondrial genotypes
The genotypes from the 1293 mitochondrial probes were extracted using Genome Studio software. Of the 1293 probes, 234 were added as a part of the custom content based on the preliminary pooled sequencing of 2000 subjects Mitochondria (MolVis). The frequency of each of the 1293 positions was calculated and classified to different bins based on the criteria below.
• Positions that have a call rate of 97% or below.
• Positions with a variant frequency of 1% or below (rare variants). • Positions with a nucleotide not matching the rCRS reference nucleotide (design failure)
• Duplicate probes for same SNP.
A total of 275 positions that have 1 % and above variance frequency (data not shown) are used for further disease association analysis.
Haplogroup assignment
The mtDNA haplogroup of each individual subject was classified by the Haplogrep 2 software using phytotreel7. To generate the vcf file 916 probes were used including duplicate probes and rare variants. Indels and design mismatch probes which are a total of 35 genotypes were excluded along with no variants probes which are 339 genotypes. Haplogrep2 is a public software that generates haplogroups automatically using a clustering algorithm based on phylotree classification. The haplogroup classification was primarily based on the Kulczynksi distance, a dissimilarity metric feature which implements 3 other algorithms based on Jaccard index, Hamming distance and kimura 2-parameter distance also runs in back ground to estimate the distance concordance. All the samples where at least one metric resulted in discordance with others are listed under a separate list; 47% of the POAAGG study samples were discordant. We have previously assigned haplogroups to 96 samples based on whole mitochondria sequencing using these samples we determined 100% concordance with the j accard index haplogroup assignment from Haploogroup2. All the discordant sample haplogroups were then reassigned based on the j accard classification. Table 11 shows the distribution of the macro haplogroups in the POAAGG population.
Table 11
Figure imgf000066_0001
Association of Mitochondrial Haplogroups with POAG The frequency of each individual haplogroup was calculated and association with the POAG risk our study cohort is analyzed, the association is also analyzed separated by gender. A total of 29 individual haplogroups which represent 57% of the study population have an individual frequency of 1% or above. The association of the 29 common haplogroups was calculated by using the frequency and calculating the odds ratio and the significance using chi square tests. The genders were analyzed separately as we have previously seen increased risk of POAG associated with men belonging to Llc2 groups. To analyze this difference in detail we have looked at all the 29 groups separately in male and females. The Llcl and Llc2bl significance is exclusively coming from males and not significant between the females. Table 12 shows the odds ratio and p values calculated by gender separated for all the 29 mtDNA common haplogroups in POAAGG. None of the significant associations survive the multiple corrections using the Hochberg step up method.
Table 12: The 29 haplogroups which are common in POAAGG cohort along with the rare haplogroups collapsed under the macro haplogroups. Significant groups denoted by **.
Figure imgf000067_0001
1.55) 2.06) 1.75)
L2al 1.15 (0.85, 0.36 1.18 (0.73, 0.5 1.10 (0.74, 0.64
1.56) 1.93) 1.63)
L2ala 0.95 (0.66, 0.81 0.79 (0.43, 0.43 1.05 (0.65, 0.84
1.39) 1.43) 1.70)
L2alal 1.20 (0.72, 0.49 2.94 (0.96, 0.047 0.87 (0.46, 0.67
2.02) 8.99) 1.64)
L2alc 1.16 (0.69, 0.57 0.83 (0.29, 0.73 1.36 (0.75, 0.31
1.94) 2.38) 2.45)
L2ale 1.40 (0.76, 0.28 1.25 (0.44, 0.68 1.48 (0.69, 0.31
2.59) 3.53) 3.18)
L2b 1.24 (0.74, 0.41 1.46 (0.61, 0.39 1.12 (0.59, 0.73
2.08) 3.51) 2.15)
L2bla 0.97 (0.64, 0.91 0.88 (0.45, 0.7 1.01 (0.58, 0.97
1.49) 1.71) 1.76)
L2c** 1.69 (1.14, 0.008 1.57 (0.83, 0.16 1.73 (1.05, 0.03
2.50) 2.97) 2.85)
L2c2 0.88 (0.49, 0.66 0.71 (0.24, 0.54 0.98 (0.49, 0.96
1.57) 2.12) 1.96)
L3b 0.96 (0.53, 0.88 1.31 (0.50, 0.58 0.76 (0.36, 0.48
1.71) 3.40) 1.63)
L3bla 0.87 (0.63, 0.39 0.79 (0.44, 0.42 0.93 (0.63, 0.7
1.20) 1.41) 1.36)
L3dr2'3'4'5'6 1.15 (0.63, 0.65 1.88 (0.58, 0.29 0.97 (0.48, 0.94
2.08) 6.13) 1.98)
L3dlb3 0.94 (0.51, 0.86 0.97 (0.32, 0.95 0.94 (0.44, 0.88
1.76) 2.90) 2.03)
L3d4 0.94 (0.50, 0.85 1.00 (0.30, 0.99 0.94 (0.44, 0.88
1.78) 3.28) 2.03)
L3el 0.84 (0.49, 0.52 0.70 (0.31, 0.38 0.92 (0.45, 0.81
1.43) 1.57) 1.85)
L3ele** 0.49 (0.30, 0.006 0.70 (0.31, 0.38 0.39 (0.19, 0.005
0.82) 1.57) 0.77) L3e2alb 1.10 (0.72, 0.67 1.77 (0.80, 0.16 0.89 (0.53, 0.67 1.68) 3.95) 1.51)
L3e2b 0.95 (0.71, 0.72 0.65 (0.39, 0.1 1.15 (0.81, 0.43
1.27) 1.08) 1.65)
L3e3b 1.16 (0.69, 0.57 1.08 (0.47, 0.86 1.18 (0.61, 0.62
1.94) 2.48) 2.29)
L3flbla 1.05 (0.64, 0.86 1.09 (0.52, 0.82 0.93 (0.47, 0.84
1.71) 2.26) 1.84)
L3flb4c 0.91 (0.60, 0.67 0.98 (0.51, 0.94 0.85 (0.49, 0.54
1.39) 1.88) 1.45)
H5a3a 0.85 (0.55, 0.44 1.17 (0.51, 0.71 0.77 (0.46, 0.32
1.30) 2.64) 1.29)
The distribution of these common haplogroups (FIG. 8) shows the distribution of the 29 haplotypes and 6 combined groups where the remaining rare haplotypes are grouped based on macrohaplogroup.
The phylogenetic relationship among the 29 common haplogroups is depicted in
FIG. 9, the associated risk based on the odds ratio is represented as protective, moderate and potential high risks groups. The gender distribution is with overlapping lines, one solid and one broken line.
Association by Mitochondria positions
275 positions are selected for POAG association analysis. The odds ratio and p values are calculated separated by gender as we have seen a gender effect with the haplogroup analysis above. Table 13 shows the top 20 mtDNA positions with the significant p value. The haplogroup assignment for the top positions show the same significant bins from the haplogroup association above.
Figure imgf000070_0003
Figure imgf000070_0002
Figure imgf000070_0001
TABLE 14
(Sequence Listing Free Text)
The following information is provided for sequences containing free text under identifier <223>.
Figure imgf000071_0001
All applications, including priority US Provisional patent application No.
62/330133 filed April 30, 2016, publications and documents recited in this specification are incorporated herein by reference. Numerous modifications and variations are included in the scope of the above-identified specification and are expected to be obvious to one of skill in the art. Such modifications and alterations to the compositions and processes are believed to be within the scope of the claims appended hereto.
REFERENCES
1. [Anonymous]. World Medical Association declaration of Helsinki.
Recommendations guiding physicians in biomedical research involving human subjects. JAMA. 1997;277:925-926.
2. Abu-Amero KK, et al. Mitochondrial genetic background in ghanaian patients with primary open-angle glaucoma. Mol Vis 2012; 18: 1955-9. 22876121
3. Abu-Amero KK, et al. Susceptibility to primary angle closure glaucoma in saudi arabia: The possible role of mitochondrial DNA ancestry informative haplogroups. Mol Vis 2011 ; 17:2171-6. 21850192
4. Abu-Amero KK, et al. The role of mitochondrial haplogroups in glaucoma: A study in an Arab population. Mol Vis 2008; 14:518-22. 18385785
5. Andrews R, et al. The role of mitochondrial haplogroups in primary open angle glaucoma. Br J Ophthalmol 2006; 90:488-90. 16547333
6. Arnold RS, et al. An inherited heteroplasmic mutation in mitochondrial gene COI in a patient with prostate cancer alters reactive oxygen, reactive nitrogen and proliferation. BioMed Res Int 2013; 2013:239257. 23509693
7. Astafurov K, et al. Oral microbiome link to neurodegeneration in glaucoma. PLoS One 2014; 9:el04416. 25180891
8. Barrel P, et al. Elimination of false positives that arise in using the two-hybrid system. BioTechniques. 1993;14:920-924.
9. Behar DM, et al. Genographic Consortium. The dawn of human matrilineal
diversity. Am J Hum Genet 2008; 82: 1130-40. 18439549
10. Cai W, et al. Mitochondrial variants may influence the phenotypic manifestation of Leber's hereditary optic neuropathy-associated ND4 Gl 1778A mutation. J Genet Genomics. 2008;35:649-655.
11. Charlson ES, et al. The primary open-angle african american glaucoma genetics study: Baseline demographics. Ophthalmology 2015; 122:711-20. PMID
25576993 Chiang MC, et al. Metformin activation of AMPK-dependent pathways is neuroprotective in human neural stem cells against Amyloid-beta-induced mitochondrial dysfunction. Exp Cell Res. 2016. Collins DW, et al. Mitochondrial sequence variation in african-american primary open-angle glaucoma patients. PLoS One 2013; 8:e76627. 24146900 Collins DW, et al. Association of primary open-angle glaucoma with
mitochondrial variants and haplogroups common in African Americans. Mol Vis. 2016;22:454-471. Coskun P, et al. A mitochondrial etiology of alzheimer and parkinson disease. Biochim Biophys Acta 2012; 1820:553-64. 21871538 Daudt DR, 3rd, et al. Methylene blue protects primary rat retinal ganglion cells from cellular senescence. Invest Ophthalmol Vis Sci. 2012;53:4657-4667. Frank SA, Hurst LD. Mitochondria and male disease. Nature. 1996;383:224. Fromont-Racine M, et al. Toward a functional analysis of the yeast genome through exhaustive two-hybrid screens. Nat Genet. 1997;16:277-282. Ghelli A, et al. The background of mitochondrial DNA haplogroup J increases the sensitivity of leber's hereditary optic neuropathy cells to 2,5-hexanedione toxicity. PLoS One 2009; 4:e7922. 19936068 Ghiso JA, et al. Alzheimer's disease and glaucoma: mechanistic similarities and differences. J Glaucoma. 2013;22 Suppl 5:S36-8. Gomez J, et al. Non optical semi-conductor next generation sequencing of the main cardiac QT-interval duration genes in pooled DNA samples. J Cardiovasc Transl Res 2014; 7: 133-7. 24190697 Gomez-Duran A, et al. Oxidative phosphorylation differences between mitochondrial DNA haplogroups modify the risk of leber's hereditary optic neuropathy. Biochim Biophys Acta 2012; 2012: 1216-22. 22561905 Gonder MK, et al. Whole-mtDNA genome sequence analysis of ancient african lineages. Mol Biol Evol 2007; 24:757-68. 17194802 Gong Z, et al. Humanin and age-related diseases: A new link? Front Endocrinol (Lausanne) 2014; 5:210. 25538685 Gonzalez-Lima F, Auchter A. Protection against neurodegeneration with low-dose methylene blue and near-infrared light. Front Cell Neurosci. 2015;9: 179. Gudiseva, HV et. al BMC Genomics 2016; PMID: 27052975 Gueven N, et al. Targeting mitochondrial function to treat optic neuropathy. Mitochondrion. 2016. Gupta A. Harnessing the microbiome in glaucoma and uveitis. Med Hypotheses 2015. 26238774 Gupta V, et al. One protein, multiple pathologies: multifaceted involvement of amyloid beta in neurodegenerative disorders of the brain and retina. Cell Mol Life Sci. 2016. Hao XD, et al. Mitochondrial DNA haplogroup Y is associated to leigh syndrome in Chinese population. Gene 2013; 512:460-3. 23111160 He Y, et al. Mitochondrial complex I defect induces ROS release and degeneration in trabecular meshwork cells of POAG patients: Protection by antioxidants. Invest Ophthalmol Vis Sci 2008; 49: 1447-58. 18385062 Hernandez-Zimbron LF, et al. Amyloid-beta peptide binds to cytochrome C oxidase subunit 1. PLoS One. 2012;7:e42344. Hernandez-Zimbron LF, Rivas-Arancibia S. Oxidative stress caused by ozone exposure induces beta-amyloid 1-42 overproduction and mitochondrial accumulation by activating the amyloidogenic pathway. Neuroscience.
2015;304:340-348. Horai R, et al. Microbiota-dependent activation of an autoreactive T cell receptor provokes autoimmunity in an immunologically privileged site. Immunity 2015; 43:343-53. 26287682 Howell N, et al. Low penetrance of the 14484 LHON mutation when it arises in a non-haplogroup J mtDNA background. Am J Med Genet A 2003 ; 119 A: 147-51. 12749053 Hsouna S, et al. Association study of mitochondrial DNA polymorphisms with type 2 diabetes in tunisian population. Mitochondrial DNA 2015; 26:367-72. 24102601 Hu H, Li M. Mitochondria-targeted antioxidant mitotempo protects mitochondrial function against amyloid beta toxicity in primary cultured mouse neurons.
Biochem Biophys Res Commun. 2016;478: 174-180. Iglesias Al, et al. Genes, pathways, and animal models in primary open-angle glaucoma. Eye (Lond). 2015;29: 1285-1298. Jaksch M, et al. A systematic mutation screen of 10 nuclear and 25 mitochondrial candidate genes in 21 patients with cytochrome c oxidase (COX) deficiency shows tRNA(Ser)(UCN) mutations in a subgroup with syndromal encephalopathy. J Med Genet. 1998;35:895-900. Janssen SF, et al. The vast complexity of primary open angle glaucoma: disease genes, risks, molecular mechanisms and pathobiology. Prog Retin Eye Res.
2013;37:31-67. Jeoung JW, et al. Mitochondrial DNA variant discovery in normal -tension glaucoma patients by next-generation sequencing. Invest Ophthalmol Vis Sci 2014; 55:986-92. 24448266 Johnson DC, et al. Mitochondrial DNA diversity in the african american population. Mitochondrial DNA 2015; 26:445-51. 24102597 Ju WK, et al. Elevated hydrostatic pressure triggers release of OPA1 and cytochrome C, and induces apoptotic cell death in differentiated RGC-5 cells. Mol Vis. 2009;15: 120-134. Ju WK, et al. Intraocular pressure elevation induces mitochondrial fission and triggers OPA1 release in glaucomatous optic nerve. Invest Ophthalmol Vis Sci. 2008;49:4903-4911. Kenney MC, et al. Molecular and bioenergetic differences between cells with african versus european inherited mitochondrial DNA haplogroups: Implications for population susceptibility to diseases. Biochim Biophys Acta 2014; 1842:208- 19. 24200652 Lascaratos G, et al. Mitochondrial dysfunction in glaucoma: Understanding genetic influences. Mitochondrion 2012; 12:202-12. 22138560
Lascaratos G, et al. Resistance to the most common optic neuropathy is associated with systemic mitochondrial efficiency. Neurobiol Dis 2015; 82:78-85. 26054436 Lee S, et al. Impaired complex-I-linked respiration and ATP synthesis in primary open-angle glaucoma patient lymphoblasts. Invest Ophthalmol Vis Sci 2012; 53:2431-7. 22427588 Lee S, et al. Mitochondrial dysfunction in glaucoma and emerging bioenergetic therapies. Exp Eye Res 2011; 93:204-12. 20691180 Li YJ, et al. FECD Genetics Consortium. Mitochondrial polymorphism A10398G and haplogroup I are associated with fuchs' endothelial corneal dystrophy. Invest Ophthalmol Vis Sci 2014; 55:4577-84. 24917144 Lloyd RE, McGeehan JE. Structural analysis of mitochondrial mutations reveals a role for bigenomic protein interactions in human disease. PLoS One.
2013;8:e69003. Lott MT, et al. mtDNA variation and analysis using MITOMAP and
MITOMASTER. Curr Protoc Bioinformatics 2013; 1 : 1-23. 25489354 Ma J, et al. mtDNA haplogroup and single nucleotide polymorphisms structure human microbiome communities. BMC Genomics 2014; 15:257. 24694284 Ng SK, et al. Genetic Association at the 9p21 Glaucoma Locus Contributes to Sex Bias in Normal-Tension Glaucoma. Invest Ophthalmol Vis Sci. 2016;57:3416- 3421. Osborne NN, et al. Targeting mitochondrial dysfunction as in aging and glaucoma. Drug Discov Today 2014; 19: 1613-22. 24880106 Otaegui D, et al. Mitochondrial polymporphisms in parkinson's disease. Neurosci Lett 2004; 370: 171-4. 15488317 Pacheu-Grau D, et al. Mitochondrial antibiograms in personalized medicine. Hum Mol Genet 2013; 22:1132-9. 23223015 Pagani L, Eckert A. Amyloid-Beta interaction with mitochondria. Int J Alzheimers Dis. 2011;2011 :925050. Petros JA, et al. mtDNA mutations increase tumorigenicity in prostate cancer. Proc Natl Acad Sci USA 2005; 102:719-24. 15647368 Pinazo-Duran MD, et al. Oxidative stress and mitochondrial failure in the pathogenesis of glaucoma neurodegeneration. Prog Brain Res 2015; 220: 127-53. 26497788 61. Quist A, et al. Amyloid ion channels: a common structural link for protein- misfolding disease. Proc Natl Acad Sci U S A. 2005;102: 10427-10432.
62. Ramos A, et al. Human mitochondrial DNA complete amplification and
sequencing: A new validated primer set that prevents nuclear DNA sequences of mitochondrial origin co-amplification. Electrophoresis 2009; 30: 1587-93.
19350543
63. Ray AM, et al. Sequence variation in the mitochondrial gene cytochrome c oxidase subunit I and prostate cancer in african american men. Prostate 2009; 69:956-60. 19267350
64. Rellstab C, et al. Validation of SNP allele frequencies determined by pooled next- generation sequencing in natural populations of a non-model plant species. PLoS One 2013; 8:e80422. 24244686
65. Robinson JT, et al. Integrative genomics viewer. Nat Biotechnol 2011; 29:24-6.
21221095
66. Rothberg JM, et al. An integrated semiconductor device enabling non-optical genome sequencing. Nature 2011; 475:348-52. 21776081
67. Schlebusch CM, et al. MtDNA control region variation affirms diversity and deep sub-structure in populations from southern africa. BMC Evol Biol 2013; 13:56. 23445172
68. Schlotterer C, et al. Sequencing pools of individuals - mining genome-wide
polymorphism data without big funding. Nat Rev Genet 2014; 15:749-63.
25246196
69. Shi C, et al. Mitochondrial FOX03a is involved in amyloid beta peptide-induced mitochondrial dysfunction. J Bioenerg Biomembr. 2016;48: 189-196.
70. Shu L, et al. Complete mitochondrial DNA sequence analysis in two southern Chinese pedigrees with leber hereditary optic neuropathy revealed secondary mutations along with the primary mutation. Int J Ophthalmol 2012; 5:28-31. 22553750
71. Sudoyo H, et al. Asian-specific mtDNA backgrounds associated with the primary G11778A mutation of leber's hereditary optic neuropathy. J Hum Genet 2002; 47:594-604. 12436196 72. Sun Q, et al. A mitochondrial DNA mutation influences the apoptotic effect of statins on prostate cancer. Prostate. 2015;75: 1916-1925
73. Sundaresan P, et al. Whole-mitochondrial genome sequencing in primary open- angle glaucoma using massively parallel sequencing identifies novel and known pathogenic variants. Genet Med 2015; 17:279-84. 25232845
74. Tabebi M, et al. A novel mutation MT-COIII m.9267G>C and MT-COI
m.5913G>A mutation in mitochondrial genes in a Tunisian family with maternally inherited diabetes and deafness (MIDD) associated with severe nephropathy. Biochem Biophys Res Commun. 2015;459:353-360.
75. Takalo M, et al. Targeting ubiquilin-1 in Alzheimer's disease. Expert Opin Ther Targets. 2013;17:795-810.
76. Tang Z, et al. Survival effect of PDGF-CC rescues neurons from apoptosis in both brain and retina by regulating GSK3beta phosphorylation. J Exp Med.
2010;207:867-880.
77. Tielsch JM, et al. A population-based evaluation of glaucoma screening: The baltimore eye survey. Am J Epidemiol 1991; 134: 1102-10. 1746520
78. Tranah GJ, et al. Mitochondrial DNA sequence associations with dementia and amyloid-beta in elderly African Americans. Neurobiol Aging. 2014;35:442.el- 442.e8.
79. van Oven M, Kayser M. Updated comprehensive phylogenetic tree of global human mitochondrial DNA variation. Hum Mutat 2009; 30:E386-94. 18853457
80. Vojtek AB, Hollenberg SM. Ras-Raf interaction: two-hybrid analysis. Methods Enzymol. 1995;255:331-342
81. Waisbourd M, et al. The Philadelphia Glaucoma Detection and Treatment Project:
Detection Rates and Initial Management. Ophthalmology. 2016;123: 1667-1674.
82. Watson B Jr, et al. Mitochondrial DNA mutations in black americans with
hypertension-associated end-stage renal disease. Am J Kidney Dis 2001; 38:529- 36. 11532685
83. Yamagishi Y, et al. Identification of essential amino acids in humanin, a
neuroprotective factor against alzheimer's disease-relevant insults. Peptides 2003; 24:585-95. 12860203 84. Yu-Wai-Man P, et al, editors. GeneReviews(R). Seattle (WA): University of Washington, Seattle; 1993
85. Zhang AM, et al. Mitochondrial DNA haplogroup background affects LHON, but not suspected LHON, in Chinese patients. PLoS One 2011; 6:e27750. 22110754
86. Lu A, et al. Effects of ubiquilin 1 on the unfolded protein response. J Mol
Neurosci. 2009;38: 19-30

Claims

CLAIMS:
1. A method for evaluating a subject's risk of developing glaucoma or having progressively severe glaucoma or diagnosing glaucoma in a subject comprising determining from a biological sample from a subject whether the patient has a nucleic acid variant in the mitochondrial DNA (mtDNA) defining a mitochondrial haplogroup selected from L2, L2c, L2b, L2alc, L2al'2'3'4 or Llc2, or Llc2blbl.
2. The method according to claims 1, comprising determining from said sample the existence of one or more mutational variants defining or representing that mitochondrial haplogroup.
3. The method according to claim 1, wherein the mitochondrial haplogroup is Llc2 or Llc2blbl.
4. The method according to any one of claims 1 to 3, wherein said variant is one or more missense variants.
5. The method according to claim 2, wherein said variant comprises a mutation within the MT-COl gene of the patient's mtDNA.
6. The method according to claim 2, wherein the variant is at one or more mtDNA positions 6150, 6253 or 6480.
7. The method according to claim 6, wherein the variant is a G to A variant at mtDNA position 6150.
8. The method according to claim 6, wherein the variant is a T to C variant at mtDNA position 6253.
9. The method according to claim 6, wherein the variant is a G to A variant at mtDNA position 6480.
10. The method according to claim 2, wherein the variant is identified in Table 1.
11. The method according to claim 2 wherein the variant alters an amino acid residue encoded by a mitochondrial gene.
12. The method according to claim 11, wherein the variant creates a polymorphism at amino acid 83 in the MT-COl protein, from Val to He (V83I) or a polymorphism at amino acid 117 in the MT-COl protein, from Met to Thr (Ml 17T) or a polymorphism at amino acid 193 in the MT-COl protein, from Val to He (VI 931) or a combination of two of more of these polymorphisms.
13. The method according to claim 11 or 12, wherein the altered amino acid residue alters the function or interaction of amyloid beta in the subject.
14. The method according to any of claims 1 to 13 further comprising evaluating said patient for clinical abnormalities in cup-to-disc ratio (CDR), visual field, or intraocular pressure (IOP).
15. The method according to any one of claims 1 to 14, further comprising obtaining a sample from a subject and assaying said sample to determine the haplotype of the subject.
16. The method according to claims 1 to 15, wherein the biological sample is saliva, urine or blood.
17. The method according to any one of claims 1 to 16, wherein the sample is subjected to an evaluation in an ELISAs, a platform multiplex ELISA, TAQMAN assay, ILLUMINA assay, mass spectrometry quantitative assays, PCR, RT-PCR, QPCR or next generation sequencing techniques.
18. The method according to any one of claims 1 to 17, wherein the sample is subjected to genotyping.
19. The method according to claim 1 wherein the sample is subjected to whole genome sequencing and computational analysis of the WGS data.
20. The method according to any of claims 1 to 19 further comprising treating the subject diagnosed as having an increased risk of glaucoma with therapeutic agents, therapeutic methods or dietary supplements to prevent or reduce advance to glaucoma or to preserve or enhance mitochondrial function.
PCT/US2017/030034 2016-04-30 2017-04-28 Methods for screening and diagnosing glaucoma WO2017189951A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201662330133P 2016-04-30 2016-04-30
US62/330,133 2016-04-30

Publications (1)

Publication Number Publication Date
WO2017189951A1 true WO2017189951A1 (en) 2017-11-02

Family

ID=60161167

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2017/030034 WO2017189951A1 (en) 2016-04-30 2017-04-28 Methods for screening and diagnosing glaucoma

Country Status (1)

Country Link
WO (1) WO2017189951A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112331267A (en) * 2020-09-25 2021-02-05 浙江大学 Acinetobacter database and construction method thereof based on mass spectrum
WO2022197765A1 (en) * 2021-03-16 2022-09-22 University Of North Texas Health Science Center At Fort Worth Macrohaplotypes for forensic dna mixture deconvolution

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
ABU-AMERO, KK ET AL.: "Eurasian and Sub-Saharan African Mitochondrial DNA Haplogroup Influences Pseudoexfoliation Glaucoma Development in Saudi patients", MOLECULAR VISION, vol. 17, 19 February 2011 (2011-02-19), pages 543 - 547, XP055439499 *
COLLINS, DW ET AL.: "Association of Primary Open-Angle Glaucoma with Mitochondrial Variants and Haplogroups Common in African American s", MOLECULAR VISION, vol. 22, 16 May 2016 (2016-05-16), pages 454 - 471, XP055439544 *
HORAN, MP ET AL.: "Quaternary Protein Modeling to Predict the Function of DNA Variation Found in Human Mitochondrial Cytochrome C Oxidase", JOURNAL OF HUMAN GENETICS, vol. 58, no. 3, 10 January 2013 (2013-01-10), pages 127 - 134, XP055439504 *
PINTO, M ET AL.: "Mitochondrial Genome Changes and Neurodegenerative Diseases", BIOCHIMICA ET BIOPHYSICA ACTA, vol. 1842, no. 8, 16 November 2013 (2013-11-16), pages 1198 - 1207, XP028873356 *
WANG, R ET AL.: "Common and Rare Genetic Risk Factors for Glaucoma", COLD SPRING HARBOR PERSPECTIVES IN MEDICINE, vol. 4, no. 12, 18 September 2014 (2014-09-18), pages 1 - 13, XP055439549 *
YANG, HJ ET AL.: "Vision from Next Generation Sequencing: Multi-Dimensional Genome-Wide Analysis for Producing Gene Regulatory Networks Underlying Retinal Development, Aging and Disease", PROGRESS IN RETINAL AND EYE RESEARCH, vol. 46, 7 February 2015 (2015-02-07), pages 1 - 30, XP055439541 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112331267A (en) * 2020-09-25 2021-02-05 浙江大学 Acinetobacter database and construction method thereof based on mass spectrum
WO2022197765A1 (en) * 2021-03-16 2022-09-22 University Of North Texas Health Science Center At Fort Worth Macrohaplotypes for forensic dna mixture deconvolution

Similar Documents

Publication Publication Date Title
Collins et al. Association of primary open-angle glaucoma with mitochondrial variants and haplogroups common in African Americans
Burdon et al. Genome-wide association study identifies susceptibility loci for open angle glaucoma at TMCO1 and CDKN2B-AS1
Martin-Merida et al. Toward the mutational landscape of autosomal dominant retinitis pigmentosa: a comprehensive analysis of 258 Spanish families
Krohn et al. Fine‐mapping of SNCA in rapid eye movement sleep behavior disorder and overt Synucleinopathies
Booth et al. Variants in CIB2 cause DFNB48 and not USH1J
Melia et al. Limb-girdle muscular dystrophy 1F is caused by a microdeletion in the transportin 3 gene
Cobos et al. Association between CFH, CFB, ARMS 2, SERPINF 1, VEGFR 1 and VEGF polymorphisms and anatomical and functional response to ranibizumab treatment in neovascular age‐related macular degeneration
Avila-Fernandez et al. Identification of an RP1 prevalent founder mutation and related phenotype in Spanish patients with early-onset autosomal recessive retinitis
Fernandez‐San Jose et al. Prevalence of Rhodopsin mutations in autosomal dominant Retinitis Pigmentosa in Spain: clinical and analytical review in 200 families
Blein et al. An original phylogenetic approach identified mitochondrial haplogroup T1a1 as inversely associated with breast cancer risk in BRCA2 mutation carriers
Lucas et al. Rare, potentially pathogenic variants in 21 keratoconus candidate genes are not enriched in cases in a large Australian cohort of European descent
Cooper et al. Progranulin levels in blood in Alzheimer's disease and mild cognitive impairment
Álvarez et al. LOXL1 gene variants and their association with pseudoexfoliation glaucoma (XFG) in Spanish patients
Scott et al. Expanding the phenotypic spectrum in RDH12-associated retinal disease
Zhong et al. Seven novel variants expand the spectrum of RPE65-related Leber congenital amaurosis in the Chinese population
Khan et al. Importance of glucokinase-258G/A polymorphism in Asian Indians with post-transplant and type 2 diabetes mellitus
Corton et al. Involvement of LCA5 in Leber congenital amaurosis and retinitis pigmentosa in the Spanish population
Selmer et al. Autosomal dominant pericentral retinal dystrophy caused by a novel missense mutation in the TOPORS gene
Wei et al. Next-generation sequencing identifies a novel compound heterozygous mutation in MYO7A in a Chinese patient with Usher Syndrome 1B
Bianco et al. Leber’s hereditary optic neuropathy (LHON) in an Apulian cohort of subjects
Tang et al. Analysis of SLC4A11, ZEB1, LOXHD1, COL8A2 and TCF4 gene sequences in a multi-generational family with late-onset Fuchs corneal dystrophy
Jurkute et al. Biallelic variants in coenzyme Q10 biosynthesis pathway genes cause a retinitis pigmentosa phenotype
Planté-Bordeneuve et al. The range of chronic demyelinating neuropathy of infancy: a clinico-pathological and genetic study of 15 unrelated cases
Sowińska‐Seidler et al. Heterozygous DLX5 nonsense mutation associated with isolated split‐hand/foot malformation with reduced penetrance and variable expressivity in two unrelated families
Li et al. Mitochondrial polymorphism A10398G and Haplogroup I are associated with Fuchs' endothelial corneal dystrophy

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17790497

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17790497

Country of ref document: EP

Kind code of ref document: A1