CN104845992A - Biomarkers for treating prostatic cancer, therapeutic target and application thereof - Google Patents

Abstract

The invention relates to biomarkers for treating prostatic cancer, which comprise fusion gene, long non-coding RNA, gene mutation and an alternative splicing variant. The invention also provides an application of the biomarkers in therapeutic target as a reagent for diagnosing prostatic cancer or a medicine for treating prostatic cancer.

Description

The biological markers, therapy target and uses thereof of prostate cancer

Related application

The application to be the applying date be September 16 in 2011 day, are called the divisional application of the Chinese patent application 201180073445.7 of " biological markers, therapy target and uses thereof of prostate cancer ".

Technical field

The present invention relates to cancer field, particularly prostate cancer.Meanwhile, the present invention relates to and use sequencing technologies of future generation, to find for diagnosing, the drug target of the biological markers of prognosis and therapeutic response prediction and effective treatment prostate cancer, especially for the biological markers of prostate cancer.In the present invention, employ RNA-Seq technology especially, namely transcript profile sequencing technologies analyzes the transcript profile of prostate cancer tissue and Carcinoma side normal tissue, discloses the transcripting spectrum that Chinese's prostate cancer is complete.

Background technology

In developed country, prostate cancer is still the highest tumour of sickness rate, ranked second position in male cancer associated death simultaneously.The sickness rate of whole world prostate cancer is in continuous rising, but in country variant and race, its sickness rate is widely different.What sickness rate was the highest is western countries, as the U.S.; What sickness rate was minimum is East Asian countries, and as China, this species diversity may partly be caused by not agnate gene difference.In addition, prostate cancer is a kind of different substantiality disease.Each tumour tumour evolve and biological behaviour (as Tumor dormancy, local growth, at a distance diffusion, to the reaction for the treatment of and recurrence etc.) upper widely different.Therefore, histopathology Classification and stage and the patient that Gleason scoring is identical, treatment plan is identical, its Clinical Outcome and tumour progression history may be completely different.Its tumour of some patients is in dormant state, is confined to prostate gland, can be more than Ten Year Survival, and other patients after diagnosis 2-3 die from the distant metastasis of tumour.All evidences show, the heterogeneity of prostate cancer clinical behavior be in tumour progression process by it molecular mechanism difference cause.

Between more than ten years in the past, DNA and RNA chip technology is widely used on analysis biological mechanism.It helps us to have new understanding to the pathogenesis of prostate cancer, for we find for diagnosing, prognosis and therapeutic response prediction biological markers provide the foundation.Although so far, detecting for the prognosis of prostate cancer genome of OncotypeDx and MammoPrint of similar mammary cancer is few, and the prostate cancer molecules that some are found changes and is applied to clinical practice.(the Taylor BS such as Taylor, et al. (2010) Integrative genomic profiling of human prostate cancer.Cancer Cell 18 (1): 11-22.) by finding the comprehensive gene group analysis of prostate cancer, the change of some gene copy number may distinguish evolving tumor and dormant trait tumour, and this discovery is significant.But we still also improve tumour progression and the predictive ability for the treatment of final result to detect prostate cancer more accurately in the urgent need to new biological markers.

Although it is pointed out that the research based on gene chip has made major contribution to us to the understanding that development occurs human tumor, this technology has significant limitation, as do not detected change and the base mutation of genome structure.

Summary of the invention

In the past few years, the develop rapidly of sequencing technologies of future generation (Next Generation Sequencing, NGS) overcomes above-mentioned deficiency.NGS makes us can with unprecedented high resolving power and the whole Oncogenome of high throughput analysis and transcript profile.The data of NGS from multiple angle analysis genome, as sudden change, can be transcribed, structure variation and transcribe rear adjustment (as methylating).In addition, updating of NGS technology makes scientist can check order to the genome of main tumor type.

At present, nearly all research for prostate cancer genome and the change of transcript profile level is all carry out in white man, and the research of yellow is few.In this research, we use RNA-Seq technology, and namely transcript profile sequencing technologies analyzes the transcript profile of 14 pairs of prostate cancer tissues and Carcinoma side normal tissue.All transcription product types are analyzed by we, disclose the transcripting spectrum that Chinese's prostate cancer is complete.We have found a lot of isomer and comprise: the reservation of exon skipping, intron, 5 ' and 3 ' end alternative splicing, gene fusion, point mutation, long-chain non-coding RNA, these all may work in the generation of prostate cancer and development.Our research illustrates the complicated collection of illustrative plates of prostate cancer genome change, confirms the heterogeneity of prostate cancer, advances the understanding of our centering Chinese Prostate Cancer.

1. the discovery of prostate cancer New Fusion gene and checking

(1). carry out RNA-Seq (i.e. transcript profile sequencing technologies) in Shanghai Changhai Hospital 14 pairs of prostate cancers and cancer beside organism, find USP9Y-TTTY15, CTAGE5-KHDRBS3, RAD50-PDLIM4, SDK1-AMACR totally 4 documents do not report high frequency fusion gene and other dozens of fusion gene, see such as following table 1.

Table 1. prostate cancer New Fusion gene

(2). we verify these fusion genes in 54 pairs of prostate cancers and cancer beside organism.We devise the specific PCR primer of gene fusion.After PCR and agar electrophoresis, (QiagenQIAquick Gel Extraction kit) parallel Sanger order-checking is reclaimed in all RT-PCR amplified fragments rubber tapping.We find 4 New Fusion genes specifically expressing, the frequency higher (the results are shown in Figure 2-4) in cancerous tissue verified.Be not in the news before these fusion genes, but its in the higher prompting of this research medium frequency, it plays an important role in the generation of Chinese's prostate cancer, these are expected to be elucidated in follow-up research.

(3). potential applicability in clinical practice: express in cancerous tissue, the fusion gene of not expressing by cancer and in healthy tissues, it is the prostate cancer marker of high degree of specificity, detected by real time PCR in blood, urine, situation is there is in prostate biopsy tissue and postoperative tissue by FISH detection fusion gene, for the early diagnosis of prostate cancer patient, molecule parting and judgement patient prognosis, fusion gene can be used as the target spot of targeted therapy simultaneously.

2. the long-chain non-coding RNA of finding differences property expression

The transcripting spectrum of long-chain non-coding RNA in prostate cancer.Increasing evidence shows to work during long-chain non-coding RNA is in cytobiology is many, points out it in the nosetiology of disease, comprises in tumour mechanism and working.Up to the present, research before does not all set foot in the overall transcriptional level of long-chain non-coding RNA in tumour and changes.Therefore, the entirety that first we analyze long-chain non-coding RNA in prostate cancer tissue and pairing Carcinoma side normal tissue thereof transcribes spectrum, finds on average have 1599 known long-chain non-coding RNAs to express in each sample.Next, we compare the expression level of long-chain non-coding RNA at prostate cancer tissue and pairing Carcinoma side normal tissue, (multiple changes >=2 to find on average there is the variant expression between the two of 406 long-chain non-coding RNAs, false positive rate, False positive Rate, FDR<=0.001), wherein 137 long-chain non-coding RNAs all present consistent rise or downward in the prostate cancer of 50%.

Because most of long-chain non-coding RNA is found relevant with transcriptional regulatory, we have studied the change of long-chain non-coding RNA expression amount to the impact of prostate cancer genetic expression.We analyze the dependency of each long-chain non-coding RNA and all gene expression amounts.Use that absolute correlation coefficient is greater than 0.85, False discovery rate is less than 0.01 for dividing value, we find the gene with long-chain non-coding RNA height correlation.Absorbing, have hundreds of gene significant correlations in 23 long-chain non-coding RNAs and full-length genome, and other most of genes only with several gene-correlation, or not relevant.This prompting long-chain non-coding RNA may have the function beyond transcriptional regulatory, such as in the adjustment of post-transcriptional level.Surprisingly, except two long-chain non-coding RNAs, nearly all long-chain non-coding RNA and genetic expression are proportionate, and point out these long-chain non-coding RNAs may promote the expression of gene.

In order to study the relation of long-chain non-coding RNA and prostate cancer, we have selected 4 long-chain non-coding RNAs, and (two known: DD3 and MALAT1; Two new discovery: FR257520 and FR348383), and in two groups of prostate gland samples, their expression amount is detected with qRT-PCR.First group is 40 pairs of prostate cancer tissues and pairing Carcinoma side normal tissue thereof, and second group is 15 normal human prostate tissues and 15 prostate cancer tissues.QRT-PCR and RNA-seq result has very strong dependency.Consistent with RNA-Seq result, PCA3, MALAT1 and FR348383 process LAN in most of prostate cancer sample, and FR257520 expression amount reduces.The result of PCA3 process LAN with think that the research that it may become new diagnosis marker is similar before, but we Late Cambrian MALAT1, FR257520 and FR348383 express in prostate cancer and normal prostatic has notable difference.

Potential applicability in clinical practice: detect long-chain non-coding RNA by real time PCR and there is situation in blood, urine, for early diagnosis, the molecule parting of prostate cancer patient, can be used as the target spot of targeted therapy simultaneously, judge patient's prognosis.Our result of study shows that 137 long-chain non-coding RNAs can as biomarker, specifically see table 2.

A table 2.137 long-chain non-coding RNA

3, the detection of single nucleotide polymorphism and point mutation

We use SOAPsnp (Li RQ, Li YR, Fang XD, Yang HM, Wang J, et al. (2009) SNPdetection for massively parallel whole-genome resequencing.Genome Research 19:1124-1132.) detect single nucleotide polymorphism.Sanger sequence verification is suddenlyd change.We reduce by following steps the false positive rate that single nucleotide polymorphism detects, comprise delete consistence quality lower than 20 SNP, be positioned at the SNP that SNP within donor splicing site 5bp and reading support are no more than 2.In order to find new SNP, we carry out screening (YH, 1000genomes, Yoruba, Korean, Watson and NCBIdbSNP) at reported six large snp databases further.

Prostate cancer mutation spectrum.We on average find 1725 point mutation in prostate cancer tissue.But, only have sub-fraction (average 1.5%) to be positioned at the coding region of gene.What is interesting is, some point mutation are positioned at long-chain non-coding RNA.Most sudden changes (91.7%) are the sudden changes of T:A to C:G.Reasonably explain it is that time this point mutation occurs in rna editing, rna editing, by adenosine is changed into inosine, is pronounced guanosine-when the latter translates, thus causes the change of specific RNA Nucleotide for one to this discovery.

309 point mutation are found altogether in the coding region of 290 genes.Wherein 115 be silent mutation, 181 missense mutation, 13 be nonsense mutation.These sudden changes all do not find in more than one tumor tissues, point out and do not have hot spot mutation in these prostate cancer samples.But we find that there is 3 samples the sudden change being positioned at UTP14C gene different positions, there are two samples to have to be positioned at the sudden change of 4 gene (CBARA1, FRG1, NAMPT and ZNF195) different positionss.We confirm 30 sudden changes with Genomic PCR, RT-PCR and Sanger order-checking.Wherein 27 confirm in genomic level, and 29 confirm in cDNA level.

The gene that we also find 183 to have a sudden change, but great majority are all low frequency sudden changes.138 genetic results that this and Taylor etc. (Taylor BS, et al. (2010) Integrative genomic profiling of human prostate cancer.Cancer Cell18 (1): 11-22.) are reported are consistent.Carry out suddenling change at 30 genes and verify that discovery RNA-Seq finds that the accuracy of sudden change is respectively 96.7% (cDNA level) and 90% (genomic level).1 sample has KLK3 transgenation.It is shocking, all samples all do not have P53 and PTEN to suddenly change, and these two genes are genes the highest with the prostate cancer degree of correlation in COSMIC database.Although be not in the news in prostate cancer before the gene of great majority sudden change, wherein 118 were found in other tumour, pointed out the sudden change of these genes also may cause prostate cancer.

Potential applicability in clinical practice: send order-checking detection SNP and point mutation to there is situation extract DNA from prostate biopsy tissue or Post operation tissue after after performing PCR, for prostate cancer patient's molecule parting and pharmacological agent target, judge patient's prognosis.194 sudden changes of 183 genes provided by the invention are see table 3, and wherein preferred 30 transgenations are as shown in table 8

The transgenation of table 3. prostatic cancer specific

4. the detection of alternative splicing

Alternative splicing (alternative splicing, AS) is the universal phenomenon in eukaryotic cell, and it can make genetic transcription go out different mRNA products, and then may translate different isomer proteins.

(1). we use SpliceMap to find shearing site, then use different methods to detect dissimilar alternative splicing and comprise exon skipping, intron reservation and selectivity 5 ' and 3 ' shearing site.First we find alternative splicings all in 28 sample transcript profile.Then we find and exist only in cancerous tissue sample and its alternative splicing of not having of pairing cancer beside organism.We have found thousands of alternative splicing, read sequence sift out the reliably otherness shearing of one group of height by nonredundancy.The intron finding that there is KLK3 (being also PSA) gene in the prostate cancer sample exceeding half retains, and this may produce a kind of new protein sequence.The transcription product of alternative splicing and albumen all may as the neontology markers of prostate cancer diagnosis.The exon skipping of AMACR gene is found that there is in a part of prostate cancer sample.These two kinds of alternative splicing modes all use RT-PCR to obtain checking in order-checking group.We verify with RT-PCR simultaneously in other 40 pairs of samples, find have PSA intron to retain in most cancerous tissue sample, and almost do not have in cancer beside organism.PSA is the diagnostic biological markers of several routines few in number.But the examination means accuracy at present based on PSA is limited.Our newfound PSA intron retains the Sensitivity and Specificity that may contribute to improving PSA.In 40 cancerous tissue samples only 9 have AMACR gene extron to jump.

(2). potential applicability in clinical practice: in blood, urine by real time PCR or ELISA detect alternative splicing there is situation, for early diagnosis, the molecule parting of prostate cancer patient, can be used as the target spot of targeted therapy simultaneously, judge patient's prognosis.

Table 4. alternative splicing body, comprise the variation of 3' shearing site, 5' shearing site makes a variation, and exon skipping and intron retain four kinds of modes.

3' shearing site makes a variation

5' shearing site makes a variation

Exon skipping

Intron retains

Change to understand above-mentioned molecular genetics in prostate cancer, we compare to the signal path that gene fusion, point mutation, differential expression, tumour-specific otherness shear the dysregulation that relevant tumour and Taylor describe.According to documents and materials, we are defined as activated gene the gene of process LAN in tumour and known oncogene, and the gene of down-regulated expression in tumour and known cancer suppressor gene are defined as inactivated gene.We calculate each activated gene, the frequency of inactivated gene in 14 samples.If tumor specimen has one or more gene to have point mutation, gene fusion, differential expression or TS alternative splicing in signal path, we just think that tumour there occurs change at this signal path.We find that there is 3 very common signal paths (AR, Ras-PI3K-AKT and RB) and there occurs change in prostate cancers.

The same with other a lot of tumour, prostate cancer is a kind of heredopathia, is that the accumulation changed by series of genes causes.Therefore, more detailed gene expression characteristics analysis is understood these diseases better by contributing to and is promoted the targeted therapy of the individuation that research and development are new.In addition, not agnate particularly between white man and yellow prostate-cancer incidence and clinical prognosis widely different.But although the prostate cancer gene profile of white man is studied very deep, the correlative study in yellow is few.In this research, we carry out RNA-Seq by 14 pairs of cancerous tissues and pairing Carcinoma side normal tissue and have studied above-mentioned two problems.This is also the many aspects simultaneously disclosing prostate cancer transcript profile first simultaneously, comprises expression and the somatic mutation of gene fusion, alternative splicing, virus transcription fragment and long-chain non-coding RNA.By the research to above-mentioned aspect, we find that different prostate cancer patient's transcript profile has very large heterogeneity.Find, to Chinese's prostate cancer, relevant signal path occurs to the comprehensive analysis of these different gene alterations and white man similar.These pathogenesis being found to be Study of China human prostata cancer provide new possibility, provide the possible mode for the treatment of prostate cancer simultaneously.

Accompanying drawing explanation

Fig. 1. system tumor transcriptome analysis schema.

Fig. 2. fusion gene schematic diagram.Wherein Fig. 2 c is CTAGE5-khdrbs3 fusion gene schematic diagram, and the 23rd exon and khdrbs3 the 8th exon of ctage5 merge; Fig. 2 d is Tmprss2-erg fusion gene schematic diagram, and Tmprss2 the 1st exon and ERG the 4th exon merge; Fig. 2 e is the occurrence frequency of 5 fusion genes.

Fig. 3. fusion gene schematic diagram.Wherein Fig. 3 a is that USP9Y-TTTY15 merges schematic diagram, and the 3rd exon of USP9Y and the 4th exon of TTTY15 merge; Fig. 3 b is the RT-PCR result of USP9Y-TTTY15.

Fig. 4. fusion gene schematic diagram.Wherein Fig. 4 a RAD50-PDLIM4 fusion gene RT-PCR and sanger sequencing result; Fig. 4 b is SDK1-AMACR fusion gene RT-PCR and sanger sequencing result.

Fig. 5. the noncoding differential expression of long-chain.Wherein Fig. 5 c is the differential expression of long-chain non-coding RNA DD3MALAT1FR0257520FR0348383 in 40 pairs of cancers and cancer beside organism; Fig. 5 d is long-chain non-coding RNA: DD3, MALAT1, FR0257520 and FR0348383 differential expression in prostate cancer and Benign Prostatic Hyperplasia Tissues:.

Embodiment

Below in conjunction with embodiment, embodiment of the present invention are described in detail, but it will be understood to those of skill in the art that the following example only for illustration of the present invention, and should not be considered as limiting scope of the present invention.

Unless otherwise defined, otherwise Science and Technology term used herein has the implication that those skilled in the art understand usually.For a better understanding of the present invention, the definition of following term is specifically provided.

Find fusion gene, long-chain non-coding RNA, sudden change, the common step of alternative splicing: checked by pathologist after collecting prostate cancer patient's sample-> cancerous tissue and cancer beside organism's row frozen section ensure the quality of products-> prepares the long-chain non-coding RNA that sequencing result is found differential expression by cDNA library->RNA-Seq-> after genome and transcript profile location-> are by gene and the stdn of long-chain non-coding RNA expression level, the sudden change of alternative splicing and tumour-specific, fusion gene.

One aspect of the present invention provides the biological markers for prostate cancer, comprises one or more in the long-chain non-coding RNA shown in fusion gene as shown in table 1, table 2, the transgenation shown in table 3, the alternative splicing shown in table 4.

Biological markers of the present invention, it can be used as the early diagnosis marker of prostate cancer, pharmacological agent Effective judgement mark or patient's prognostic marker further.

In the specific embodiment of the present invention, in described biological markers, described fusion gene comprise in 83 fusion genes of table 6 one or more, preferably comprise in 35 fusion genes in table 6 shown in underscore one or more.

In the specific embodiment of the present invention, in described biological markers, described fusion gene comprise in USP9Y-TTTY15, CTAGE5-KHDRBS3, RAD50-PDLIM4, SDK1-AMACR one or more, the primer preferably described in fusion gene USP9Y-TTTY15, CTAGE5-KHDRBS3, RAD50-PDLIM4, SDK1-AMACR table 5 increases.

In the specific embodiment of the present invention, in described biological markers, described long-chain non-coding RNA comprise in DD3, MALAT1, FR0257520, FR0348383 one or more, preferably described long-chain non-coding RNA: the primer described in DD3, MALAT1, FR0257520, FR0348383 table 7 increases.

In the specific embodiment of the present invention, in described biological markers, described transgenation comprise in 30 transgenations as shown in table 8 one or more, the primer described in 30 transgenation tables 9 shown in preferred earth's surface 8 increases.

In the specific embodiment of the present invention, in described biological markers, described alternative splicing comprises PSA or AMACR, and preferably alternative splicing PSA or the primer described in AMACR table 10 increase.

The opposing party of the present invention provides the purposes of described biological markers in the target spot as the reagent of diagnosing prostate cancer or the medicine for the treatment of prostate cancer, in particular as the purposes of the early diagnosis marker of prostate cancer, pharmacological agent Effective judgement mark or patient's prognostic marker.

The present invention further provides on the other hand for the primer of the described biological markers that increases or the probe of described biological markers for the preparation of being the purposes in the reagent of diagnosing prostate cancer.Wherein, described primer can be used for biological markers described in specific amplification, and described probe specificity is combined with described biological markers, thus indicates the existence of described biological markers.

In the specific embodiment of the present invention, be provided for the primer of the biological markers described in increasing, wherein said primer preferably includes the primer described in table 5, and it is for fusion gene USP9Y-TTTY15, CTAGE5-KHDRBS3, RAD50-PDLIM4, SDK1-AMACR; Primer shown in table 7, it is for the long-chain non-coding RNA that increases: DD3, MALAT1, FR0257520, FR0348383; Primer shown in table 9, it is for the transgenation of 30 shown in table 8 of increasing; Primer shown in table 10, it is for alternative splicing PSA or AMACR that increase.

In the specific embodiment of the present invention, provide the purposes of primer in the reagent preparing diagnosing prostate cancer described in table 5.

In the specific embodiment of the present invention, provide the purposes of the primer shown in table 7 in the reagent preparing diagnosing prostate cancer.

In the specific embodiment of the present invention, provide the purposes of the primer shown in table 9 in the reagent preparing diagnosing prostate cancer.

In the specific embodiment of the present invention, provide the purposes of the primer shown in table 10 in the reagent preparing diagnosing prostate cancer.

Embodiment

Embodiment 1. differential genes expression analysis

1. collect prostate cancer patient's sample

Patient and sample.

14 pairs of prostate cancer tissues for RNA-Seq and Carcinoma side normal tissue take from Shanghai Changhai Hospital.54 pairs for gene fusion checking samples: 23 to from Shanghai Changhai Hospital, 17 to from Jiangsu provincial hospital, 14 to coming affiliated hospital of this Zhongshan University the 3rd.One group 40 Shanghai Changhai Hospital is taken to the prostate cancer verified for alternative splicing, long-chain non-coding RNA and cancer beside organism.Another group be used for long-chain non-coding RNA checking 15 tumor samples and 15 BPH (benign prostatic hyperplasia) samples take from Jiangsu provincial hospital and Shanghai Changhai Hospital respectively.The code of RNA-Seq and its follow-up test obtain the approval of 3 Hospital Ethical Committees.All patients fill in Written informed consent, authorize us to use their sample.

2. ensured the quality of products by pathologist's inspection after cancerous tissue and cancer beside organism's row frozen section

Pathologic finding

Checked by the pathologist of this research after cancerous tissue and Carcinoma side normal tissue frozen section carry out HE dyeing (hematoxylin-eosin staining) organizes cancerous tissue density more than 80% selected by ensureing, does not have cancerous tissue in Carcinoma side normal tissue simultaneously.All pathology samples are checked by another pathologist.If there is the situation that conclusion is inconsistent, two pathologists inquire into determine conclusion jointly.

3. prepare cDNA library and RNA-Seq

Oligomerization deoxythymidine magnetic bead is used for from total serum IgE, be separated poly A mRNA.With fragmentation buffer by purified mRNA fragmentation.Using these short-movie sections as template, synthesize first paragraph cDNA chain with random hexamers.Second segment cDNA chain damping fluid, dNTPs, RNase H and DNA polymerase I synthesize.Short double stranded cDNA fragment QIAQuick PCRextraction kit (vendor) purifying and by EB buffer solution elution to repair end and to add " A ".Then, short-movie section is connected on Illumina sequencing adaptors.The DNA of object clip size is tapped rubber purifying for pcr amplification.Use Illumina HiSeq ^tMcheck order in 2000 pairs of amplification libraries.

CDNA library builds the mRNA-Seq 8-Sample Prep Kit (article No. is: RS-100-0801) using Illumina company to provide to carry out, and its concrete operations flow process is: oligomerization deoxythymidine magnetic bead is used for from total serum IgE, be separated poly AmRNA.With fragmentation buffer by purified mRNA fragmentation.Using these short-movie sections as template, synthesize first paragraph cDNA chain with random hexamers.Second segment cDNA chain damping fluid, dNTPs, RNase H and DNA polymerase I synthesize.Short double stranded cDNA fragment QIAQuick PCR extraction kit (Qiagen) purifying and by EB buffer solution elution to repair end and to add " A ".Then, short-movie section is connected on Illumina sequencing adaptors.The DNA of object clip size is tapped rubber purifying for pcr amplification.After carrying out quality examination by using Agilent 2100Bioanalyzer biological analyser and Stepone plus quantitative real time PCR Instrument to cDNA library, (criterion of acceptability is: pcr amplification product size is 322 ± 20bp, wherein inserting short clip size is 200 ± 20bp, library volumetric molar concentration is not less than 1.3nM), use Illumina HiSeq ^tMcheck order in 2000 pairs of amplification libraries.

4. data analysis

Original reading screening

The image that sequenator generates is carried out base calling process by supporting sequenator control software design.Original series saves as fastq form.Dirty reading is deleted before analytical data.We delete dirty reading by three standards:

1) dirty reading is deleted;

2) reading of " N " base more than 2% is deleted;

3) the inferior quality reading having more than 50% QA≤15 base is deleted.

All following analyses are all based on the reading after arrangement.

Reading is located on human genome and transcript profile.

The reference sequences of the genome that we use and transcript profile downloads (hg18version) from UCSC website.We use SOAP2 (Short Oligonucleotide Analysis Package (SOAP) aligner (SOAP2); Li R, Yu C, Li Y, Lam TW, Yiu SM, et al. (2009) SOAP2:an improved ultrafast tool for short read alignment.Bioinformatics 25:1966-1967) method by arrange after reading contrast with genome and transcript profile respectively.The number that do not mate of each reading can not more than 3.

The stdn of gene and long-chain non-coding RNA expression level.

The reading of specific gene can be positioned to for calculation expression level.Gene expression dose be come from the every kilobase length of certain gene in every 1,000,000 sections of reading read hop count.Formula is as follows:

$\mathrm{RPKM}=\frac{{10}^{6}C}{\mathrm{NL}/{10}^3}$

C is the copy number of selected gene reading; N is the copy number of all reading genes; L is the total length of selected gene extron.For the gene had more than an alternative transcription product, the longest transcription product is for calculating RPKM.RPKM method can eliminate the impact that different genes length and sequence difference calculate genetic expression.Therefore, the differential expression that can be directly used in gene between comparative sample of RPKM.

We calculate non-coding RNA expression level by same procedure.

5. difference expression gene analysis

With reference to " significance of digital gene express spectra " (such as Audic S & Claverie JM (1997) The significanceof digital gene expression profiles.Genome Res 7 (10): 986-995), we have found the gene of differential expression in 14 pairs of prostate cancer tissues and pairing Carcinoma side normal tissue with False discovery rate <=0.001 and multiple change >=2 as standard.Each sample generates the Nucleotide of the order-checking of average 66,432,064 reading and 5.98Gb size.By SOAP2 technology, we navigate to human genome (UCSC hg18version) the reading of 84.4%.By the transcript profile sequence of contrast cancerous tissue and pairing Carcinoma side normal tissue, we have found the gene of some gene fusion, the long-chain non-coding RNA of differential expression, alternative splicing and differential expression in each of front in row gland cancer sample.In addition, we find that average each cancerous tissue sample has 1725 point mutation.These results disclose in prostate cancer and there is very large heterogeneity, and synchronous signal path and molecular mechanism work in the generation of prostate cancer.

The discovery of embodiment 2. prostate cancer New Fusion gene and checking

Found with reference to during genome comparison by short rna reading at us, some sequences will be divided into two sections of ability and genome matches.This kind of reading need meet the following conditions:

A) comparatively short fragment size is not shorter than 8bp;

B) notice no matter intron is where (from 5 ' to 3 ', normal chain or minus strand)

To the contraposition analysis of two sections, we allow to be no more than one not mating and without room contraposition.

RT-PCR and sequence verification gene fusion.We verify the gene fusion that RNA-Seq obtains at transcriptional level.We devise the specific PCR primer of gene fusion.After PCR and agar electrophoresis, (Qiagen QIAquick Gel Extraction kit) parallel Sanger order-checking is reclaimed in all RT-PCR amplified fragments rubber tapping.We demonstrate 5 fusion genes in this way, are TMPRSS2-ERG, USP9Y-TTTY15, SDK1-AMACR, CTAGE5-KHDRBS3, RAD50-PDLIM4 respectively, and other 4 fusion genes wherein except TMPRSS2-ERG are that the present inventor is newfound.

4 newfound fusion genes are:

>39a fwd chrY 155 39b fwd chrY

USP9Y-TTTY15

GATAACTACATAAAGAGACAAAAAAAAGAAAAAAGAGCAAAGATCTGTGCTGTGTCAAGTATGACAGCCATCACTCATGGCTCTCCAGTAGGAGGGAACGACAGCCAGGGCCAGGTTCTTGATGGCCAGTCTCAGCATCTCTTCCAACAGAACCAGgaatcaaacttgacgtatggagccaagaaagcccttggaaaaactggcctcatattttgtgtacacagtccctgtacagggtttctgacctgtg

>31a fwd chr7 121 31b rev chr5

SDK1-AMACR

ACCTTCCTGGTGCCCCATCCAACCTGGTCATTTCCAACATCAGCCCTCGCTCCGCCACCCTTCAGTTCCGGCCAGGCTATGACGGGAAAACGTCCATCTCCAGGTGGATTGTTGAGGGGCAGgtgtcatggagaaactccagctgggcccagagattctgcagcgggaaaatccaaggcttatttatgccaggctgagtggatttggccagtcaggaagcttctgccggttagctggccacgatatcaactatttggctttgtcag

>2a site:235 ID:4253 fwd_chr14<＝>fwd_chr8 ID:10656

CTAGE5-KHDRBS3

AATTTAAATGTGCCTGATTCATCTCTCCCTGCTGAAAATGAAGCCACTGGCCCTGGCTTTGTTCCTCCACCTCTTGCTCCAATCAGAGGTCCATTGTTTCCAGTGGATGCAAGAGGCCCATTCTTGAGAAGAGGACCTCCTTTCCCCCCACCTCCTCCAGGAGCCATGTTTGGAGCTTCTCGAGATTATTTTCCACCAGGGGATTTCCCAGGTCCACCACCTGCTCCATTTGCAAtggtgctgattactatgattacggacatggactcagtgaggagacttatgattcctacg

>44a fwd chr5 113 44b fwd chr5 10111(RAD50) 8572(PDLIM4)

CAAAAAGAAACTGAACTTAATAAAGTAATAGCTCAACTAAGTGAATGCGAGAAACACAAAGAAAAGATAAATGAAGATATGAGACTCATGAGACAAGATATTGATACACAGAAGgtccatgctggcagcaaggctgcattggctgccctgtgcccaggagacctgatccaggccatcaatggtgagagcacagagctcatgacacacctggaggcacagaaccgcatcaagggctgccacgatcacctcacactgtctgtgagcag

Wherein capitalization represents the sequence of first gene, and lowercase represents the sequence of second gene.

For the amplimer of these 5 fusion genes as following table 5.

The amplimer of table 5.5 fusion gene

PCR condition is: 95 DEG C 10 seconds; 60 DEG C 30 seconds; 72 DEG C 90 seconds; 38-43 circulation.

Use PCR purification kit PCR Cleanup Kit 50-prep (AXYGEN, Cat No.AP-PCR-50, LotNo.KB10101204-G) PCR primer purifying is carried out, 2% agarose gel electrophoresis is carried out to PCR primer, glue is used to reclaim test kit DNA Gel Extraction Kit 50-prep (AXYGEN, Cat No.AP-GX-50, LotNo.KE10101204-G) carry out glue recovery.

There is the electrophoresis picture of fusion gene, respectively attend and see Fig. 2 d (TMPRSS2-ERG and CTAGE5-KHDRBS3), Fig. 3 a and b (USP9Y-TTTY15) and Fig. 4 a (RAD50-PDLIM4), Fig. 4 b (SDK1-AMACR).

The gene fusion of screening high frequency.After demonstrating gene fusion with RT-PCR, we demonstrate each (above 4) fusion gene in other 54 pairs of samples.First the RNA of all samples is extracted and reverse transcription is cDNA.RT-PCR primer is identical with above-mentioned checking primer.The cDNA of order-checking sample is as positive control.

Prostate cancer gene fusion collection of illustrative plates.Transcript profile order-checking is used to detect the gene fusion phenomenon in prostate cancer the earliest.Use pairing end reading, we have found altogether 84 gene fusion.Except well-known TMPRSS2-ERG gene fusion, we have found 83 new gene fusion, and these were not all in the news in before for the research of white man.35 new being detected in prostate cancer tissue with 1 gene fusion known in the past and not seeing in pairing Carcinoma side normal tissue (fusion gene see underscore part), fusion gene is had to express (see black matrix thickened portion) at Carcinoma side normal tissue in addition, concrete biological significance is temporarily failed to understand, also has following 4 fusion genes all to have in cancer and cancer side.

The gene fusion only expressed in cancer is defined as tumor-specific genes and merges.The gene fusion number of each cancerous tissue sample is not respectively 1 to 6 not etc.Described 83 new gene merge as shown in table 6, line out below wherein 35 new gene fusion

Table 6.83 new gene merges

Modal gene fusion is TMPRSS2-ERG and USP9Y-TTTY15.The two all sees 3 samples in 14 order-checking prostate cancer tissue samples.By RNA-Seq, we detect that another modal fusion gene is the USP9Y-TTTY15 be positioned on Y chromosome.USP9Y encodes one and is similar to the albumen of ubiquitin-specific protease, and TTTY15 is a non-coding RNA.USP9Y genetically deficient or sudden change relevant with male sterility.But research before does not all disclose above-mentioned two kinds of genes and occurs relevant with tumour.In RNA-Seq result, it is identical with TMPRSS2-ERG that 3 exons of USP9Y gene and 3 exons of TTTY15 gene merge the USP9Y-TTTY15 frequency (3/14=21.4%) formed.But RT-PCR finds that in 54 prostate cancer tissues, 19 have USP9Y-TTTY15.Be not in the news before this fusion gene, but its in the higher prompting of this research medium frequency, it plays an important role in the generation of Chinese's prostate cancer, these are expected to be elucidated in follow-up research.What is interesting is, find that the transcription product of this fusion gene seems do not have open reading frame with open reading frame (ORF) forecasting tool Six-Frame Translation, point out it may be non-coding RNA.We also find that this fusion may cause the disappearance of USP9Y function and a new noncoding fusion gene transcription product.This fusion gene higher frequency of occurrences in order-checking sample and checking sample points out it to play an important role in prostate cancer.

In these 54 pairs of prostate cancer samples, we also demonstrate other 3 (CTAGE5-KHDRBS3, SDK1-AMACR and RAD50-PDLIM4) gene fusion, and their frequency is 37%, 20%, 33.3% respectively.

The discovery of embodiment 3. prostate cancer long-chain non-coding RNA and checking

(1). download ncRNA database from http://www.ncrna.org/frnadb/download, then delete fragment and be less than ncRNA, zRNA of 200nt and non-human RNA and obtain 2981 long-chain non-coding RNAs.Next we calculate the expression level of long-chain non-coding RNA with this database.The standard of the long-chain non-coding RNA differential expression of pairing cancer and the other sample of cancer is: False discovery rate <=0.001, multiple change >=2.The consistent long-chain non-coding RNA raised or lower in more than 50% sample is selected to exercise supervision cluster analysis (using cluster 3.0 pairs of genes and long-chain non-coding RNA express spectra to carry out hierarchical cluster analysis).The correlation analysis of further row long-chain non-coding RNA and gene.We select the consistent long-chain non-coding RNA that raises or lower in more than 50% prostate cancer sample and analyze the dependency of they and all genes found in prostate cancer tissue.The expression level (RPKM) of long-chain non-coding RNA and gene is used as to calculate coefficient R.

(2) .qRT-PCR verifies that (we use Power SYBR Green Mastermix reagent to be qRT-PCR at Applied Biosystems Step One Plus to long-chain non-coding RNA.GAPDH primer is used as internal reference.Described above one group of 40 pairs of prostate cancer and cancer beside organism take from Shanghai Changhai Hospital, and another group is used for 15 tumor samples and 15 BPH samples take from Jiangsu provincial hospital and Shanghai Changhai Hospital respectively, verifies for long-chain non-coding RNA.Use two-step approach pcr amplification standard program: Stage1: denaturation (Reps:1; 95 DEG C 30 seconds); Stage2:PCR reacts (Reps:40; 95 DEG C 5 seconds; 60 DEG C 34 seconds); Dissociation Stage (dissociating the stage).

Devise primer for 4 long-chain non-coding RNAs following table 7:

The primer of a table 7.4 long-chain non-coding RNA

All experiments all use two or three holes to carry out parallel repetition and test, and result draws (Fig. 5) to change relative to the average fold of GAPDH.We find that there is 137 long-chain non-coding RNAs and present consistent rise or downward in the prostate cancers of 50%.The dependency that we analyze each long-chain non-coding RNA and all gene expression amounts finds that there is hundreds of gene significant correlations in 23 long-chain non-coding RNAs and full-length genome, and other most of genes only with several gene-correlation, or not relevant.

Interpretation of result part

We find in checking in 40 pairs of prostate cancers and cancer beside organism, in 15 normal human prostate tissues and 15 prostate cancer tissues, PCA3 (being also called DD3), MALAT1 and FR0348383 process LAN in most of prostate cancer sample, and FR0257520 expression amount reduces (Fig. 5).The result of PCA3 process LAN with think that the research that it may become new diagnosis marker is similar before, but the frequency of our Late Cambrian MALAT1 process LAN is very high in prostate cancer.

The invention provides 137 long-chain non-coding RNAs to can be used for diagnosis, judge patient's prognosis and drug reaction, and the target spot for the treatment of, see table 2.

The discovery of embodiment 4. single nucleotide polymorphism and point mutation and checking

(1). we use SOAPsnp to detect single nucleotide polymorphism.This software by sequencing sequence and known array being contrasted, the consensus sequence of the individuality newly checked order is assembled into genome with repetition sequence measurement.By consensus sequence and reference sequences being compared, single nucleotide polymorphism can be found.

(2). candidate's base pair variation that we filter out with RT-PCR associating Sanger sequence verification RNA-Seq.PCR condition is: 95 DEG C 10 seconds; 60 DEG C 30 seconds; 72 DEG C 90 seconds; 38-43 circulation.Sample is from Shanghai Changhai Hospital 14 pairs of prostate cancers and cancer beside organism.30 the encoding histone sudden changes of our Stochastic choice are verified.Wherein 27 exist only in cancerous tissue (all having in cDNA and DNA), and do not see Carcinoma side normal tissue (in cDNA and DNA equal nothing).2 variations are rarely seen with cancerous tissue cDNA, and do not see healthy tissues cDNA.1 variation does not all have in cancerous tissue and Carcinoma side normal tissue.

30 sudden changes of table 8. empirical tests, wherein the rightest one to arrange the template being be that CDNA and DNA, S represent successfully respectively, and F represents unsuccessfully.

Table 9.30 primer that sudden change uses

(3). all samples all do not have P53 and PTEN to suddenly change, and these two genes are genes the highest with the prostate cancer degree of correlation in COSMIC database.Although be not in the news in prostate cancer before the gene of great majority sudden change, wherein 118 were found in other tumour, pointed out the sudden change of these genes also may cause prostate cancer.

The invention provides 183 sudden changes, these sudden changes can be used as diagnosis marker, Index for diagnosis, curative effect of medication judgement and therapy target, specifically see table 3.

The discovery of embodiment 5. alternative splicing and checking

We mainly comprise two steps for the method detecting alternative splicing:

1) we use SOAPsplice 1.1 that reading is navigated to people's reference sequences, then shearing site is found according to the comparing result of tie point reading (reading that the independent segments two or more with reference sequences is corresponding is separated by intron between these two fragments).We use the default parameters of SOAPsplice as far as possible, and the reading for complete comparison allows 3 mispairing, and each fragment of the reading for segmentation comparison only allows 1 mispairing.

2) according to alternative splicing mechanism, we use shearing site and comparing result to detect four kinds of basic alternative splicings, comprise exon skipping, selectivity 5 ' shearing site, selectivity 3 ' shearing site and intron and retain.

After finding out four kinds of alternative splicings, we select and are present in cancerous tissue and the alternative splicing not being present in Carcinoma side normal tissue.To each cancerous tissue sample, we calculate the mean depth of the intron remained in the tie point number of readings per taken of support 3 kinds of alternative splicings (exon skipping, selectivity 5 ' shearing site and selectivity 3 ' shearing site) corresponding connection site and intron reservation event respectively.Because often kind of alternative splicing enormous amount, we obtain the alternative splicing of high confidence level by getting 0.99 percentile, and are schemed to disclose some common patterns by picture circos.For 1T, it has 2047 selectivity 3 ' shearing sites.Support that the tie point reading of selectivity 3 ' shearing site is not from 1 to 609 etc., its 0.99 percentile is 69.Therefore, we retain the selectivity 3 ' shearing site of tie point reading >=69.In addition, we also delete the alternative splicing also had in Carcinoma side normal tissue.Finally, the cancer specific selectivity that we obtain one group of high confidence corresponding with each sample is sheared.RT-PCR verifies alternative splicing.We extract total serum IgE from freezing cancerous tissue and cancer beside organism, and then getting 5 μ gRNA reverse transcriptions is cDNA (Qiagen QuantiTect Reverse Transcription kit).We verify alternative splicing with RT-PCR in 40 pairs of cancerous tissues and Carcinoma side normal tissue.

PCR condition is: 95 DEG C 10 seconds; 60 DEG C 30 seconds; 72 DEG C 90 seconds; 33-36 circulation.Wherein two gene primers are as follows especially:

The amplimer of table 10.PSA and AMACR alternative splicing

Invention provides the alternative splicing of tumour-specific as shown in table 4, and these alternative splicings as the diagnosis marker of blood, urine and tissue, can also can be used as the mark of judging prognosis, result for the treatment of, can also as the target spot of oncotherapy.

The intron finding that there is KLK3 (being also PSA) gene in the prostate cancer sample exceeding half retains, and finds that there is the exon skipping of AMACR gene in a part of prostate cancer sample.These two kinds of alternative splicing modes all use RT-PCR to obtain checking in order-checking group.We verify with RT-PCR simultaneously in 40 pairs of samples (40 samples from Changhai hospital), find have PSA intron to retain in most cancerous tissue sample, and almost do not have in cancer beside organism.In 40 cancerous tissue samples only 9 have AMACR gene extron to jump.

Although the specific embodiment of the present invention has obtained detailed description, it will be understood to those of skill in the art that.According to disclosed all instructions, can carry out various amendment and replacement to those details, these change all within protection scope of the present invention.Four corner of the present invention is provided by claims and any equivalent thereof.

Claims (10)

1., for the biological markers of prostate cancer, it comprises fusion gene SDK1-AMACR as shown in table 1.

2. biological markers according to claim 1, it comprises one or more in the long-chain non-coding RNA shown in fusion gene as shown in table 1, table 2, the transgenation shown in table 3, the alternative splicing shown in table 4 further,

Optionally, comprise in the fusion gene of 83 shown in table 6 further one or more, preferably further comprise in 35 fusion genes in table 6 shown in underscore one or more.

3. the biological markers described in claim 1 or 2, it can be used as the early diagnosis marker of prostate cancer, pharmacological agent Effective judgement mark or patient's prognostic marker.

4. the biological markers described in claim 1 or 2, wherein said fusion gene comprise in USP9Y-TTTY15, CTAGE5-KHDRBS3, RAD50-PDLIM4, SDK1-AMACR one or more, the primer preferably described in fusion gene USP9Y-TTTY15, CTAGE5-KHDRBS3, RAD50-PDLIM4, SDK1-AMACR table 5 increases.

5. the biological markers described in claim 1 or 2, wherein said long-chain non-coding RNA comprise in DD3, MALAT1, FR0257520, FR0348383 one or more, preferably described long-chain non-coding RNA: the primer described in DD3, MALAT1, FR0257520, FR0348383 table 7 increases.

6. the biological markers described in claim 1 or 2, wherein said transgenation comprise in 30 transgenations as shown in table 8 one or more, the primer described in 30 transgenation tables 9 shown in preferred earth's surface 8 increases.

7. the biological markers described in claim 1 or 2, wherein said alternative splicing comprises PSA or AMACR, and preferably alternative splicing PSA or the primer described in AMACR table 10 increase.

8. the purposes of the biological markers according to any one of claim 1-7 in the target spot as the reagent of diagnosing prostate cancer or the medicine for the treatment of prostate cancer, in particular as the purposes of the early diagnosis marker of prostate cancer, pharmacological agent Effective judgement mark or patient's prognostic marker.

9. for the primer of the biological markers according to any one of claim 1-7 that increases or the probe of described biological markers for the preparation of being the purposes in the reagent of diagnosing prostate cancer.

10. purposes according to claim 9, wherein said primer comprises the primer described in table 5, and it is for fusion gene USP9Y-TTTY15, CTAGE5-KHDRBS3, RAD50-PDLIM4, SDK1-AMACR; Primer shown in table 7, it is for the long-chain non-coding RNA that increases: DD3, MALAT1, FR0257520, FR0348383; Primer shown in table 9, it is for the transgenation of 30 shown in table 8 of increasing; Primer shown in table 10, it is for alternative splicing PSA or AMACR that increase.