WO2024165507A1 - Use of hif-1α stabilizing agents for the treatment of type i interferonopathies - Google Patents

Abstract

Type I interferonopathies represent a subgroup of autoinflammatory diseases caused by mutations in genes associated with proteasome degradation or cytoplasmic RNA- and DNA- sensing pathways. Among them, Aicardi-Goutières syndrome (AGS) ischaracterised by both neurological and extra-neurological involvement with onset in childhood. Chronic inflammation in response to uncontrolled type I IFN production is, among other things, associated with IP-10 secretion. The inventors analysed, at the single-cell transcriptomic levels, peripheral blood samples from patients bearing mutations in three AGS-causing genes, i.e., SAMHD1, RNASEH2B or ADAR1 genes. Using machine-learning approaches and differential gene expression performed on these single-cell data, they identified a drastic loss of transcription factor hypoxia induced factor 1 α (HIF-1α) expression associated with features of a metabolic switch and mitochondrial stress in monocytes/dendritic cells of patients. Chemical stabilization of HIF-1α, with a synthetic drug (DMOG) in an in vitro model of AGS, allowed the inventors to reverse the energy metabolic switch, attenuate mitochondrial stress and markedly reduce IP-10 production. The inventors therefore propose that inappropriate energy metabolic switch contributes to exacerbated chronic inflammation in AGS, and that targeting this pathway might represent a promising therapeutic approach.

Description

USE OF HIF-1A STABILIZING AGENTS FOR THE TREATMENT OF TYPE I

INTERFERONOPATHIES

FIELD OF THE INVENTION:

The present invention is in the field of medicine, in particular immunology.

BACKGROUND OF THE INVENTION:

The term “interferonopathy” first appeared in 2003, when some authors identified phenotypic overlaps between Aicardi-Goutieres syndrome (AGS) encephalopathy, viral congenital infections, and some autoimmune diseases such as systemic lupus erythematosus (SLE), postulating a common pathological feature as an upregulation of interferon (IFN) a activity. AGS is a monogenic disorder with onset frequently observed in the first months of life. Mutations causing AGS have been identified in genes encoding for the exonuclease TREX1 (AGS1) the components B, C and A of the RNase H2 complex (RNASEH2B (AGS2), RNASEH2C (AGS3), RNASEH2A (AGS4)) ², the deoxynucleotide triphosphate triphosphohydrolase SAMHD1 (AGS5) ³, the double-stranded RNA-specific adenosine deaminase (AD ARI (AGS6)) ⁴, the cytosolic dsRNA sensor IFIH1 (AGS7) ⁵ and components of the replication-dependent histone pre-mRNA-processing complex LSM11 (AGS8) and RNU7-1 (AGS9) ⁶. The encoded proteins are involved in the processing (AGS1-6) or sensing (AGS7) of cellular nucleic acid, and disruption of their normal function can induce an aberrant antiviral transcriptomic response ⁷. These pathways can be separated into two arms: on the one hand, detection of dsDNA and RNA/DNA hybrid leading to the activation of the stimulator of interferon genes (STING) pathway; and on the other, dsRNA sensing by the mitochondrial antiviral-signalling protein (MAVS). Both STING and MAVS activation converge on the translocation and the activation by phosphorylation of the transcription factor IRF3, leading to the production of type I IFN and the strengthening of cell defence mechanisms against viruses ⁸. Mutations described in AGS lead to constitutive triggering of the nucleic acid sensing pathways and the production of type I IFN, even in absence of viral infection ⁶’⁹. Thus, AGS is a chronic, inflammatory disorder that can mimics in

viral infection but differs, notably, in the type and nature of cytokines produced. In AGS, the inflammatory response is orchestrated by type I IFN signalling but also involves inflammatory signals through IFN induced protein 10 (IP- 10), CCL2 and vascular endothelial growth factor (VEGF), with no obvious upregulation of IL-6 or IL-8 ¹⁰ that is observed typically in response to viral infections. The major clinical features of AGS relate to the brain, most particularly intracranial calcification, white matter disease and cerebral atrophy. Extra-neurological features can also occur, including vasculitic chilblain-like lesions, and features consistent with systemic lupus erythematosus (SLE) in some cases ⁹. While type I IFN levels can be difficult to measure in patients its activity can be monitored by following the expression of IFN stimulated genes (ISG) such as IP-10, which concentration levels are consistently elevated both in serum and the cerebrospinal fluid of patients ¹⁰. Importantly, the type I IFN response and IP- 10 have been shown to induce direct neurotoxicity in human ¹¹. Of note, type I IFN targeted therapies have proven of limited impact on the main neurological features of AGS ¹², arguing for a need to better understand cellular pathways involved in AGS.

HIF-la, encoded by HIF1A, is a sensor of energy metabolism stress and, in response, its transcriptional activity allows the adaptation of energy production pathways. In the context of energy metabolism stress, two events can lead to HIF-la stabilisation: a lack of oxygen (hypoxia) ¹³, or the elevation of reactive oxygen species (ROS) ^{14 15}. Both hypoxia and ROS accumulation serve as danger signals for mitochondrial energy production through oxidative phosphorylation. Consequently, part of the HIF-la transcriptional program aims to inhibit mitochondrial respiration, thereby limiting oxygen consumption and ROS accumulation ¹⁶. To compensate for the lack of energy production in the cell, the aerobic glycolysis pathway is then induced by HIF-la transcriptional activity ¹⁶. Thus, HIF-la allows cells to adapt their energy metabolism when required, a process shown to be necessary in various homeostatic^17-19, and also in certain pathological²⁰, contexts. However, the involvement of HIF-la in the regulation of energy metabolism in type I IFN-related diseases such as AGS remains to be explored.

SUMMARY OF THE INVENTION:

The present invention is defined by the claims. In particular, the present invention relates to the use of HIF-la stabilizing agents for the treatment of type I interferonopathies.

DETAILED DESCRIPTION OF THE INVENTION:

Aicardi-Goutieres syndrome (AGS) is a rare genetic type I interferon (IFN)-mediated disease characterised by both neurological and extra-neurological involvement with onset in childhood. Chronic inflammation in response to uncontrolled type I IFN production is, among other things, associated with IP-10 secretion. The inventors analysed, at the single-cell transcriptomic levels, peripheral blood samples from patients bearing mutations in three AGS-causing genes, i.e., SAMHD1, RNASEH2B or AD ARI genes. Using machine-learning approaches and differential gene expression performed on these single-cell data, we identified a drastic loss of transcription factor hypoxia induced factor 1 a (HIF-la) expression associated with features of a metabolic switch and mitochondrial stress in monocytes/dendritic cells of patients. Chemical stabilization of HIF-la, with a synthetic drug (DMOG) in an in vitro model of AGS, allowed the inventors to reverse the energy metabolic switch, attenuate mitochondrial stress and markedly reduce IP- 10 production. The inventors therefore propose that inappropriate energy metabolic switch contributes to exacerbated chronic inflammation in AGS, and that targeting this pathway might represent a promising therapeutic approach.

Thus the first object of the present invention relates to a method of treating a type I interferonopathy in patient in need thereof comprising administering to the subject a HIF-la stabilizing agent.

As used herein, the term “type I interferon” or “type I IFN” has its general meaning in the art and refers to members of the type I interferon family of molecules. Examples of type I interferons are interferon alpha 1, 2a, 2b, 4, 5, 6, 7, 8, 10, 13, 14, 16, 17, 21, interferon beta and interferon omega.

As used herein, the term “type I interferonopathy” has its general meaning in the art and refers to a subgroup of autoinflammatory diseases caused by mutations in genes associated with proteasome degradation or cytoplasmic RNA- and DNA-sensing pathways. The term “interferonopathy” first appeared in 2003, when some authors identified phenotypic overlaps between Aicardi-Goutieres syndrome (AGS) encephalopathy, viral congenital infections, and some autoimmune diseases such as systemic lupus erythematosus (SLE), postulating a common pathological feature as an upregulation of interferon (IFN) a activity (Crow, Yanick J. "Type I interferonopathies: a novel set of inborn errors of immunity. "Annals of the New York Academy of Sciences 1238.1 (2011): 91-98). In particular, the term includes Aicardi-Goutieres syndrome (AGS), STING-associated vasculopathy with onset in infancy, and chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature syndrome. Patients with type I interferonopathy shared several clinical characteristics, including bilateral calcifications of the basal ganglia, chilblain-like rashes, and liver dysfunction. Each subtype includes diseasespecific severe complications, such as early-onset encephalopathy associated with AGS and pulmonary hypertension observed in patients diagnosed with STING-associated vasculopathy with onset in infancy.

As used herein, the term "patient" or "patient in need thereof", is intended for a human or non-human mammal. Typically the patient is affected or likely to suffer from a type I interferonopathy. In some embodiments, the patient is a human infant. In some embodiments, the patient is a human child. In some embodiments, the patient is a human adult.

As used herein, the term "treatment" or "treat" refer to both prophylactic or preventive treatment as well as curative or disease modifying treatment, including treatment of patient at risk of contracting the disease or suspected to have contracted the disease as well as patients who are ill or have been diagnosed as suffering from a disease or medical condition, and includes suppression of clinical relapse. The treatment may be administered to a patient having a medical disorder or who ultimately may acquire the disorder, in order to prevent, cure, delay the onset of, reduce the severity of, or ameliorate one or more symptoms of a disorder or recurring disorder, or in order to prolong the survival of a patient beyond that expected in the absence of such treatment. By "therapeutic regimen" is meant the pattern of treatment of an illness, e.g., the pattern of dosing used during therapy. A therapeutic regimen may include an induction regimen and a maintenance regimen. The phrase "induction regimen" or "induction period" refers to a therapeutic regimen (or the portion of a therapeutic regimen) that is used for the initial treatment of a disease. The general goal of an induction regimen is to provide a high level of drug to a patient during the initial period of a treatment regimen. An induction regimen may employ (in part or in whole) a "loading regimen", which may include administering a greater dose of the drug than a physician would employ during a maintenance regimen, administering a drug more frequently than a physician would administer the drug during a maintenance regimen, or both. The phrase "maintenance regimen" or "maintenance period" refers to a therapeutic regimen (or the portion of a therapeutic regimen) that is used for the maintenance of a patient during treatment of an illness, e.g., to keep the patient in remission for long periods of time (months or years). A maintenance regimen may employ continuous therapy (e.g., administering a drug at regular intervals, e.g., weekly, monthly, yearly, etc.) or intermittent therapy (e.g., interrupted treatment, intermittent treatment, treatment at relapse, or treatment upon achievement of a particular predetermined criteria [e.g., disease manifestation, etc.]). As used herein, the term “HIF-la” has its general meaning in the art and refers to a subunit of a heterodimeric transcription factor hypoxia-inducible factor 1 (HIF-1) that is encoded by the HIF1A gene. In the context of energy metabolism stress, two events can lead to HIF-la stabilisation: a lack of oxygen (hypoxia) ¹³, or the elevation of reactive oxygen species (ROS) ^{14 15}. Both hypoxia and ROS accumulation serve as danger signals for mitochondrial energy production through oxidative phosphorylation. Consequently, part of the HIF-la transcriptional program aims to inhibit mitochondrial respiration, thereby limiting oxygen consumption and ROS accumulation ¹⁶. To compensate for the lack of energy production in the cell, the aerobic glycolysis pathway is then induced by HIF-la transcriptional activity ¹⁶. Thus, HIF-la allows cells to adapt their energy metabolism when required, a process shown to be necessary in various homeostatic^17-19, and also in certain pathological²⁰, contexts.

As used herein, the term “HIF-la stabilizing agent” has its general meaning in the art and refers to any compound that inhibits or reduces the degradation of HIF-la. The term also includes any compound that increases the accumulation of, or stability of, HIF-la or increases the expression of HIF-la.

Examples of HIF-la stabilizing agents include but are not limited to cofactor-based inhibitors such as 2-oxoglutarate analogues, ascorbic acid analogues and iron chelators such as desferri oxamine (DFO), the hypoxia mimetic cobalt chloride (C0CI2), and mimosine, 3- Hydroxy-4-oxo-l(4H)-pyridinealanine, or other factors that may mimic hypoxia. Also of interest are hydroxylase inhibitors, including deferiprone, 2,2'-dipyridyl, ciclopirox, dimethyloxallyl glycine (DMOG), L-Mimosine (Mim) and 3-Hydroxy-l,2-dimethyl-4(lH)- Pyridone (OH-pyridone). Other EUF hydroxylase inhibitors are described herein, including but not limited to, oxoglutarates, heterocyclic carboxamides, phenanthrolines, hydroxamates, and heterocyclic carbonyl glycines (including, but not limited to, pyridine carboxamides, quinoline carboxamides, isoquinoline carboxamides, cinnoline carboxamides, beta-carboline carboxamides, including substituted quinoline-2-carboxamides and esters thereof; substituted isoquinoline-3-carboxamides and N-substituted arylsulfonylamino hydroxamic acids (see, e.g., PCT Application No. WO 05/007192, WO 03/049686 and WO 03/053997), and the like. Compounds reported to stabilize HIF-la also include [(3-hydroxy-6-isopropoxy-quinoline-2- carbonyl)-amino]-acetic acid, [3-hydroxy-pyridine-2-carbonyl)-amino]-acetic acid, [N-((l- chloro-4-hydroxy-isoquinoline-3-carbonyl)-amino)-acetic acid, [(7-bromo-4-hydroxy- isoquinoline-3-carbonyl)-amino]-acetic acid, [(7-chloro-3-hydroxy-quinoline-2-carbonyl)- amino]-acetic acid, [(l-bromo-4-hydroxy-7-kifluoromethyl-isoquinoline-3-carbonyl)-amino]- acetic acid, [(l-Bromo-4-hydroxy-7-phenoxy-isoquinoline-3-carbonyl)-amino]-ace-tic acid, [(l-Chloro-4-hydroxy-7-phenoxy-isoquinoline-3-carbonyl)-amino]-acetic acid, [(l-Chloro-4- hydroxy-7-methoxy-isoquinoline-3-carbonyl)-amino]-acetic acid, [(l-chloro-4-hydroxy- isoquinoline-3-carbonyl)-amino]-acetic acid, [(4-Hydroxy-7-phenoxy-isoquinoline-3- carbonyl)-amino]-acetic acid, [(4-Hydroxy-7-phenyl sulfanyl isoquinoline-3-carbonyl)-amino]- acetic acid, [(4-Hydroxy-6-phenylsulfanyl-isoquinoline-3-carbonyl)-amino]-acetic acid, 4- Oxo-l,4-dihydro-[l, 10]phenanthroline-3-carboxylic acid, 4-hydroxy-5-methoxy- [l,10]phenanthroline-3 -carboxylic acid ethyl ester, [(7-benzyloxy-l-chloro-4-hydroxy- isoquinoline-3-carbonyl)-amino]-acetic acid methyl ester, and 3-{[4-(3,3-Dibenzyl-ureido)- benzenesulfonyl]-[2-(4-methoxy-phenyl)-ethyl]-amino}-N-hydroxy -propionamide.

In some embodiments, the HIF-la stabilizing agent is a prolyl hydroxylase inhibitor. As used herein, the term “HIF prolyl hydroxylase” or “HIF PH” refer to any enzyme capable of hydroxylating a proline residue in the HIF protein. As used herein, the term “prolyl hydroxylase inhibitor” or “PH” refers to any compound that reduces or otherwise modulates the activity of HIF prolyl hydroxylase. Prolyl hydroxylase inhibitors are described in U.S. Pat. No. 7,811,595, which is incorporated herein by reference in its entirety. The synthesis of such prolyl hydroxylase inhibitors is described in U.S. Patent Publication No. 2012/0309977, which is incorporated herein by reference in its entirety. One such compound is {[5-(3-chlorophenyl)- 3-hydroxypyridine-2-carbonyl]amino}acetic acid (“Compound (I)”) and methods of making the compound were disclosed in U.S. Pat. No. 7,811,595, filed Jun. 26, 2007 (See inter alia Schemes I and II and accompanying synthetic procedures in columns 15-17, and 25) and U.S. Publication 2012-0309977 (U.S. patent application Ser. No. 13/488,554), filed in Jun. 5, 2012 (See inter alia paragraphs [0254]-[267]), the entireties of each of which are incorporated by reference herein. In some embodiments, the prolyl hydroxylase inhibitor is AKB-4924 (Okumura CY, Hollands A, Tran DN, Olson J, Dahesh S, von Kockritz-Blickwede M, Thienphrapa ffl, Corle C, Jeung SN, Kotsakis A, Shalwitz RA, Johnson RS, Nizet V. A new pharmacological agent (AKB-4924) stabilizes hypoxia inducible factor-1 (HIF-1) and increases skin innate defenses against bacterial infection. J Mol Med (Berl). 2012 Sep;90(9): 1079-89. doi: 10.1007/s00109-012-0882-3). In some embodiments, the prolyl hydroxylase inhibitor is selected from the group consisting of enarodustat, vadadustat, roxadustat, and N-oxalyl glycine. In some embodiments, the prolyl hydroxylase inhibitor is FG- 4497 (F ' orristai, Catherine E., et al. "FG-4497, a Pharmacological Stabilizer of HIF-la Protein, Synergistically Enhances Hematopoietic Stem Cells (HSC) Mobilization in Response to G-CSF and Plerixafor. "Blood 120.21 (2012): 216).

In some embodiments, the type I interferonopathy is resistant to a JAK inhibitor.

As used herein, the term “resistant” means a decreased response or lack of response to a standard dose of the therapeutic agent, relative to the subject's response to the standard dose of the therapeutic agent in the past, or relative to the expected response of a similar subject with a similar disorder to the standard dose of the therapeutic agent. Thus, in some embodiments, a subject may be resistant to therapeutic agent although the subject has not previously been given the therapeutic agent, or the subject may develop resistance to the therapeutic agent after having responded to the agent on one or more previous occasions.

As used herein the term “JAK” has its general meaning in the art and refers to the family of Janus kinases (JAKs) which are cytoplasmic tyrosine kinases that transduce cytokine signaling from membrane receptors to STAT transcription factors. Four JAK family members are described, JAK1, JAK2, JAK3 and TYK2 and the term JAK may refer to all the JAK family members collectively or one or more of the JAK family members as the context indicates.

As used herein the term “JAK inhibitor” is intended to mean compounds inhibit the activity or expression of at least JAK2. JAK inhibitors down-regulate the quantity or activity of JAK molecules.

JAK inhibitors are well known in the art and typically include AG490, AUB-6-96, AZ960, AZD1480, baricitinib (LY3009104, INCB28050), BMS-911543, CEP-701 , CMP6, CP352,664, CP690,550, CYT-387, INCB20, Jak2-IA, lestaurtinib (CEP-701), LS104, LY2784544, NS018, pacritinib (SB1518), Pyridone 6, ruxolitinib (INCBO 18424), SB1518, TG101209, TG101348 (SAR302503), TG101348, tofacitinib (CP-690,550), WHI-PI 54, WP1066, XL019, and XLOI 9.

As used herein, the expression "therapeutically effective amount" means a sufficient amount of the compound to provide a therapeutic effect (e.g. reducing the inflammation). It will be understood, however, that the total daily usage of the compounds and compositions of the present invention will be decided by the attending physician within the scope of sound medical judgment. The specific therapeutically effective dose level for any particular subject will depend upon a variety of factors including the disorder being treated and the severity of the disorder; activity of the specific compound employed; the specific composition employed, the age, body weight, general health, sex and diet of the subject; the time of administration, route of administration, and rate of excretion of the specific compound employed; the duration of the treatment; drugs used in combination or coincidental with the specific inhibitor employed; and like factors well known in the medical arts. For example, it is well within the skill of the art to start doses of the compound at levels lower than those required to achieve the desired therapeutic effect and to gradually increase the dosage until the desired effect is achieved. However, the daily dosage of the products may be varied over a wide range from 0.01 to 1,000 mg per adult per day. Typically, the compositions contain 0.01, 0.05, 0.1, 0.5, 1.0, 2.5, 5.0, 10.0, 15.0, 25.0, 50.0, 100, 250 and 500 mg of the active ingredient for the symptomatic adjustment of the dosage to the subject to be treated. A medicament typically contains from about 0.01 mg to about 500 mg of the active ingredient, preferably from 1 mg to about 100 mg of the active ingredient. An effective amount of the drug is ordinarily supplied at a dosage level from 0.0002 mg/kg to about 20 mg/kg of body weight per day, especially from about 0.001 mg/kg to 7 mg/kg of body weight per day.

In some embodiments, the HIF-la stabilizing agent is administered in combination with at least one additional drug suitable for the treatment of type I interferonopathy. In some embodiments, the HIF-la stabilizing agent is administered in combination with at least one JAK inhibitor (see supra). In some embodiments, the HIF-la stabilizing agent is administered in combination with at least one reverse transcriptase inhibitor (RTI) selected from the group consisting of Abacavir, Delavirdine, Didanosine, Efavirenz, Emtricitabine, Etravirine, Lamivudine, Nevirapine, Rilpivirine, Stavudine, Tenofovir, Zalcitabine and Zidovudine. In some embodiments, the HIF- la stabilizing agent is administered in combination with a RTI tri -therapy that comprises zidovudine, abacavir, and lamivudine.

Typically, the HIF-la stabilizing agent is combined with pharmaceutically acceptable excipients, and optionally sustained-release matrices, such as biodegradable polymers, to form pharmaceutical compositions. The term "Pharmaceutically" or "pharmaceutically acceptable" refers to molecular entities and compositions that do not produce an adverse, allergic or other untoward reaction when administered to a mammal, especially a human, as appropriate. A pharmaceutically acceptable carrier or excipient refers to a non-toxic solid, semisolid or liquid filler, diluent, encapsulating material or formulation auxiliary of any type.

|^|;ome embodiments, the method of the present invention comprises the step consisting of i) determining the level and/or activity of HIF-la in a sample obtained from the patient, ii) comparing the level determined at step i) with a predetermined reference value and iii) administering the HIF-la stabilizing agent when the level determined at step i) is lower than the predetermined reference value.

In some embodiments, the sample is a sample of monocytes. As used herein, the term “monocyte” has its general meaning in the art and refers to a large mononuclear phagocyte of the peripheral blood. The main phenotypic markers of human monocyte cells include CDl lb, CDl lc, CD33 and CD115. Methods for isolating monocytes are well known in the art and typically include on cell sorting methods such as fluorescence activated cell sorting (FACS) or magnetic activated cell sorting (MACS). For instance non-monocytes cells may be magnetically labeled with a cocktail of monoclonal antibodies chosen antibodies directed against CD3, CD7, CD19, CD56, CD123 and CD235a. Kits for isolation of monocytes are commercially available from Miltenyi Biotec (Auburn, CA, USA), Stem Cells Technologies (Vancouver, Canada) or Dynal Bioech (Oslo, Norway).

In some embodiments, the sample also comprises dendritic cells. As used herein, the term "Dendritic cells" or “DCs” refer to any member of a diverse population of morphologically similar cell types found in lymphoid or non-lymphoid tissues. These cells are characterized by their distinctive morphology, high levels of surface MHC-class II expression (Steinman, et al., Ann. Rev. Immunol. 9:271 (1997); incorporated herein by reference for its description of such cells).

The measurement of the expression and/or activity of HIF-la in the sample is typically carried- out using standard protocols known in the art.

Methods for determining the expression level of a gene product such as nucleic acid (e.g. RNA) are also well known in the art. Conventional methods typically involve polymerase chain reaction (PCR). For instance, U.S. Pat. Nos. 4,683,202, 4,683,195, 4,800,159, and 4,965,188 disclose conventional PCR techniques. PCR typically employs two oligonucleotide primers that bind to a selected target nucleic acid sequence. Primers useful in the present invention include oligonucleotides capable of acting as a point of initiation of nucleic acid synthesis within the target nucleic acid sequence. A primer can be purified from a restriction digest by conventional methods, or it can be produced synthetically. PCR involves use of a thermostable polymerase. Typically, the polymerase is a Taq polymerase (i.e. Thermus aquaticus polymerase). Quantitative PCR is typically carried out in a thermal cycler with the capacity to illuminate each sample with a beam of light of a specified wavelength and detect the fluorescence emitted by the excited fluorophore. The thermal cycler is also able to rapidly heat and chill samples, thereby taking advantage of the physicochemical properties of the nucleic acids and thermal polymerase. In order to detect and measure the amount of amplicon (i.e. amplified target nucleic acid sequence) in the sample, a measurable signal has to be generated, which is proportional to the amount of amplified product. All current detection systems use fluorescent technologies. Some of them are non-specific techniques, and consequently only allow the detection of one target at a time. Alternatively, specific detection chemistries can distinguish between nonspecific amplification and target amplification. These specific techniques can be used to multiplex the assay, i.e. detecting several different targets in the same assay. For example, SYBR® Green I probes, High Resolution Melting probes, TaqMan® probes, LNA® probes and Molecular Beacon probes can be suitable. TaqMan® probes are the most widely used type of probes. They were developed by Roche (Basel, Switzerland) and ABI (Foster City, USA) from an assay that originally used a radio-labelled probe (Holland et al. 1991), which consisted of a single-stranded probe sequence that was complementary to one of the strands of the amplicon. A fluorophore is attached to the 5’ end of the probe and a quencher to the 3’ end. The fluorophore is excited by the machine and passes its energy, via FRET (Fluorescence Resonance Energy Transfer) to the quencher. Traditionally, the FRET pair has been conjugated to FAM as the fluorophore and TAMRA as the quencher. In a well-designed probe, FAM does not fluoresce as it passes its energy onto TAMRA. As TAMRA fluorescence is detected at a different wavelength to FAM, the background level of FAM is low. The probe binds to the amplicon during each annealing step of the PCR. When the Taq polymerase extends from the primer which is bound to the amplicon, it displaces the 5’ end of the probe, which is then degraded by the 5 ’-3’ exonuclease activity of the Taq polymerase. Cleavage continues until the remaining probe melts off the amplicon. This process releases the fluorophore and quencher into solution, spatially separating them (compared to when they were held together by the probe). This leads to an irreversible increase in fluorescence from the FAM and a decrease in the TAMRA. In some embodiments, the level and/or activity of HIF-la is determined by RNA sequencing, and more particularly by single cell RNA sequencing. Typically, a signature of gene expression can be suitable for determining the level and/or activity of HIF-la. For instance, a set of genes known to be HIF-la targets from the NCI-Natures database to generate a signature of HIF-la activity can be used (Nesterova, A.P., Klimov, E.A., Zharkova, M., Sozin, S., Sobolev, V., Shkrob, M., and Yuryev, A. eds. (2020). Disease pathways: an atlas of human disease signaling pathways (Elsevier) . Typically signature of genes associated with the activity of HIF-la is depicted in Table 1. In some embodiments, a score that is a composite of the levels of the different genes includes in the signature is calculated and compared to the predetermined reference value for determining whether the patient is to be administered with the HIF-la stabilizing agent.

As used, the term "RNA sequencing” refers to sequencing performed on RNA (or cDNA) instead of DNA, where typically, the primary goal is to measure expression levels, detect fusion transcripts, alternative splicing, and other genomic alterations that can be better assessed from RNA. RNA sequencing typically includes whole transcriptome sequencing or targeted exome sequencing. In some embodiments, targeted exome sequencing may be preferred. As used herein, the term “whole transcriptome sequencing” refers to the use of high throughput sequencing technologies to sequence the entire transcriptome in order to get information about a sample's RNA content. As used herein, the term “targeted exome sequencing” refers to the use of high throughput sequencing technologies to sequence some specific targeted sequencing. RNA sequencing can be done with a variety of platforms for example, the Genome Analyzer (Illumina, Inc., San Diego, Calif.) and the SOLiD™ Sequencing System (Life Technologies, Carlsbad, Calif.), However, any platform useful for whole transcriptome sequencing may be used. Typically, the RNA is extracted, and ribosomal RNA may be deleted as described in U.S. Pub, No. 2011/0111409. cDNA sequencing libraries may be prepared that are directional and single or paired-end using commercially available kits such as the ScriptSeq™ M mRNA-Seq Library Preparation Kit (Epicenter Biotechnologies, Madison, Wis.). The libraries may also be barcoded for multiplex sequencing using commercially available barcode primers such as the RNA sequencing Barcode Primers from Epicenter Biotechnologies (Madison, Wis.). PCR is then carried out to generate the second strand of cDNA to incorporate the barcodes and to amplify the libraries. After the libraries are quantified, the sequencing libraries may be sequenced. Nucleic acid sequencing technologies are suitable methods for expression analysis. The principle underlying these methods is that the number of times a (DNA sequence is detected in a sample is directly related to the relative RNA levels corresponding to that sequence. These methods are sometimes referred to by the term Digital Gene Expression (DOE) to reflect the discrete numeric property of the resulting data. Early methods applying this principle were Serial Analysis of Gene Expression (SAGE) and Massively Parallel Signature Sequencing (MPSS). See, e.g., S. Brenner, et al., Nature Biotechnology 18(6):630-634 (2000). Typically RNA sequencing uses Next Generation Sequencing or NGS. As used herein, the term "Next Generation Sequencing" (NGS) refers to a relatively new sequencing technique as compared to the traditional Sanger sequencing technique. For review, see Shendure et al., Nature Biotech., 26(10): 1135-45 (2008), which is hereby incorporated by reference into this disclosure. For purpose of this disclosure, NGS may include cyclic array sequencing, microelectrophoretic sequencing, sequencing by hybridization, among others. By way of example, in a typical NGS using cyclic-array methods, genomic DNA or cDNA library is first prepared, and common adaptors may then be ligated to the fragmented genomic DNA or cDNA. Different protocols may be used to generate jumping libraries of mate-paired tags with controllable distance distribution. An array of millions of spatially immobilized PCR colonies or "polonies" is generated with each polonies consisting of many copies of a single shotgun library fragment. Because the polonies are tethered to a planar array, a single microliter- scale reagent volume can be applied to manipulate the array features in parallel, for example, for primer hybridization or for enzymatic extension reactions. Imaging-based detection of fluorescent labels incorporated with each extension may be used to acquire sequencing data on all features in parallel. Successive iterations of enzymatic interrogation and imaging may also be used to build up a contiguous sequencing read for each array feature. As used herein, the term “single-cell RNA sequencing” has its general meaning in the art and examines the sequence information from individual cells with optimized next-generation sequencing technologies. The method provides a higher resolution of cellular differences and a better understanding of the function of an individual cell in the context of its microenvironment (Eberwine, James, et al. "The promise of single-cell sequencing. "Nature methods 11.1 (2014): 25-27).

In some embodiments, the level of HIF-la is determined at the protein level by any immunoassays well known in the art (e.g. Westernblot).

In some embodiments, the predetermined reference value is a threshold value or a cut-off value. Typically, a "threshold value" or "cut-off value" can be determined experimentally, empirically, or theoretically. A threshold value can also be arbitrarily selected based upon the existing experimental and/or clinical conditions, as would be recognized by a person of ordinary skilled in the art. For example, retrospective measurement in properly banked historical subject samples may be used in establishing the predetermined reference value. The threshold value has to be determined in order to obtain the optimal sensitivity and specificity according to the function of the test and the benefit/risk balance (clinical consequences of false positive and false negative). Typically, the optimal sensitivity and specificity (and so the threshold value) can be determined using a Receiver Operating Characteristic (ROC) curve based on experimental data. For example, after determining the quantification of the selected population in a group of reference, one can use algorithmic analysis for the statistic treatment of the expression levels determined in samples to be tested, and thus obtain a classification standard having significance for sample classification. The full name of ROC curve is receiver operator characteristic curve, which is also known as receiver operation characteristic curve. It is mainly used for clinical biochemical diagnostic tests. ROC curve is a comprehensive indicator that reflects the continuous variables of true positive rate (sensitivity) and false positive rate (1- specificity). It reveals the relationship between sensitivity and specificity with the image composition method. A series of different cut-off values (thresholds or critical values, boundary values between normal and abnormal results of diagnostic test) are set as continuous variables to calculate a series of sensitivity and specificity values. Then sensitivity is used as the vertical coordinate and specificity is used as the horizontal coordinate to draw a curve. The higher the area under the curve (AUC), the higher the accuracy of diagnosis. On the ROC curve, the point closest to the far upper left of the coordinate diagram is a critical point having both high sensitivity and high specificity values. The AUC value of the ROC curve is between 1.0 and 0.5. When AUC>0.5, the diagnostic result gets better and better as AUC approaches 1. When AUC is between 0.5 and 0.7, the accuracy is low. When AUC is between 0.7 and 0.9, the accuracy is moderate. When AUC is higher than 0.9, the accuracy is high. This algorithmic method is preferably done with a computer. Existing software or systems in the art may be used for the drawing of the ROC curve, such as: MedCalc 9.2.0.1 medical statistical software, SPSS 9.0, ROCPOWER.SAS, DESIGNROC.FOR, MULTIREADER POWER. SAS, CREATE- ROC.SAS, GB STAT VIO.O (Dynamic Microsystems, Inc. Silver Spring, Md., USA), etc.

In some embodiments, the method of the present invention comprises the steps consisting of i) determining the level of oxidative phosphorylation in sample obtained from the patient, ii) comparing the level determined at step i) with a predetermined reference value and iii) administering the HIF-la stabilizing agent when the level determined at step i) is higher than the predetermined reference value.

In some embodiments, the sample is a sample of monocytes.

The assessment of level of oxidative phosphorylation in the population of monocytes is typically carried-out using standard protocols known in the art. For instance, the level is determined by RNA sequencing, and more particularly by single cell RNA sequencing. In some embodiments, a signature of genes associated with oxidative phosphorylation can be used for detecting an increase in oxidative phosphorylation. Said signatures are well known in the art and typically include those described in Nesterova, A.P., Klimov, E.A., Zharkova, M., Sozin, S., Sobolev, V., Shkrob, M., and Yuryev, A. eds. (2020). Disease pathways: an atlas of human disease signaling pathways (Elsevier) and Liberzon, A., Birger, C, Thorvaldsdottir, H., Ghandi, M., Mesirov, J.P., and Tamayo, P. (2015). The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst 1, 417-425. 10.1016/j.cels.2015.12.004. Typically a signature of genes associated with oxidative phosphorylation is depicted in Table 1. Alternatively, the metabolic profile of the cells can also be determined by using any cellbased assay such as described in the EXAMPLE. For instance, the Agilent Seahorse XF ATP Real-Time rate assay measures and quantifies the rate of ATP production from glycolytic and mitochondrial system simultaneously using label-free technology in live cells. Alternatively, the Scenith® method that is a flow cytometry based method to functionally profile energy metabolism with single cell resolution may be used (Arguello, Rafael J., et al. "SCENITH: a flow cytometry-based method to functionally profde energy metabolism with single-cell resolution. " Cell metabolism 32.6 (2020): 1063-1075).

In some embodiments, the method further comprises the step of a) determining the level of aerobic glycolysis in the sample, b) comparing the level determined at step i) with a predetermined reference value and iii) administering the HIF-la stabilizing agent when the level determined at step i) is lower than the predetermined reference value.

In some embodiments, the patient is administered with the HIF-la stabilizing agent when an increase in the level of oxidative phosphorylation and a decrease in the level of aerobic glycolysis are detected in the sample. Typically a signature of genes associated with aerobic glycolysis is depicted in Table 1.

The invention will be further illustrated by the following figures and examples. However, these examples and figures should not be interpreted in any way as limiting the scope of the present invention.

FIGURES:

Figure 1: Stabilization of HIF-la, at the protein level, revert energy metabolism switch, release mitochondrial stress and inhibit IP-10 production.

At day 4 after transduction, MDDC transduced with shRNA targeting SAMHD1 were treated with the HIF-lu stabilizing drug dimethyloxalyl glycine (DMOG) for indicated time and concentration. DMOG is an antagonist of the cofactor necessary for HIF prolylhydroxylase that is necessary for HIF-lu degradation by the proteasome.

(A) MDDC at day4 after transduction were treated with 500pM of DMOG for 5h before being lysed for protein extraction. HIF-lu level of protein expression observed by western blot revelation is represented and compared to u-tubulin (left panel). Quantification of HIF-lu protein level is normalized to u-tubulin protein level (right panel).

(B) MDDC at day4 after transduction were treated with increasing concentration of DMOG (250pM and 500pM) for 48h before undergoing the Seahorse XF Analyzer time course to determine oxygen consumption rate (OCR) (top panels) and extracellular acidification rate (ECAR) (Bottom panels). A Seahorse XF Analyzer time course is represented for 1 donor for both assays showing mean measurement with standard error of triplicates at each time point (Left panels). ATP synthase activity, Max respiratory capacity, actual glycolysis and max glycolytic capacity were quantified for 4 unrelated healthy donors assessed by Seahorse XF Analyzer experiment untreated or with 500pM of DMOG (right panels).

(C) RT-qPCR of CLPP and HsP60 on RNA extracted from MDDC at day 5 after transduction (24h of DMOG treatment) for 4 unrelated healthy donors. Data were normalized to the level of expression of the HPRT1 gene measured in each sample.

(D) ISG score based on mRNA quantification by RT-qPCR of ISG15, IFI44L, IFIT1, IFI27, RSAD2 and SIGLEC1 genes on MDDC at day 5 post transduction (24 hours of DMOG) from 4 unrelated healthy donors. Score refers as mean of relative mRNA level normalized to HPRT1 of all 6 genes. (E) Quantification of the percentage of ISG15 positive cells in flow cytometry among MDDC at day6 post transduction (48hours of DMOG) on 6 different donors.

(F) Cytokine measurement by legendplex in supernatant of MDDC at day 6 post transduction (48 hours of DMOG) for 6 unrelated healthy donors. Data are represented as fold change to scr condition.

(B and C) Mean and paired T test statistics are represented as: ns: not significant, ***:/?<0.001, **:/?<0.01, * p<0.05

(D,E and F) Mean and one way anova with Sidak multiple correction statistical test are represented as: ns: not significant, ***: ><0.001, **:/?<0.01, *p<0.05

EXAMPLE:

Methods

Experimental model and subject details

Patients

Patients were aged between 3 and 16 years old with the exception of P8 being 36 years old and were included based on molecular diagnosis of mutations in SAMHDL RNASEH2B or ADAR1 genes for AGS or COP A gene for patients with COPA disease. Pl, P2, P3 and P4 studied in this work have been previously reported by Rice at al. respectively labelled POOH, P0003, P0002 and P0004 ²¹. PBMC were frozen the day of sampling and thaw once for scRNAseq library generation. Pl was undergoing RTI tri -therapy (zidovudine, abacavir, lamivudine) for 5 months. PBMC from P2 were extracted before and after 1 month of RTI tri-therapy (zidovudine, abacavir, lamivudine). Other patients and healthy donors were not under therapy. The study was approved by the Comite de protection des personnes Iles de France II and the French advisory committee on data processing in medical research (ID-RCB: 2014-A01017-40). Consent of parents and/or patients, depending on age, was obtained for conducting the experiment.

Cells

To generate monocytes derived dendritic cells (MDDCs) form human primary cells, we acquired buffy coats from healthy human donors (Etablissement Frangais de Sang). We performed Ficoll extraction of peripheral blood mononuclear cells (PBMC) and used antihuman CD 14 magnetic beads (Miltenyi) to purify monocytes. Magnetic sorting was performed on AutoMACS pro separator device (Miltenyi Biotech). Isolated monocytes were cultured in RPMI (Gibco), 10% FBS (heat inactivated, Sigma), lOmM Hepes, 55pM P-mercaptoethanol, 6mM L-glutamine, 50pg/ml Gentamicin and lOOU/ml Penicilin/Streptomycin (Gibco) in the presence of lOng/mL of human recombinant GM-CSF and 50ng/mL of human recombinant IL- 4 (Miltenyi) to induce MDDC differentiation ⁵⁸. Multiple lot of FBS were tested and selected according minimal induction of CD86 expression on MDDC. 5xl0⁶ cells/mL of CD14⁺ cells were plated in 10cm culture plates (Sigma) and cultured for 4 days before functional assays to allow complete MDDC differentiation which was assessed by flow cytometry following CD 14 loss of expression and DC-SIGN upregulation. Fresh media with cytokines (20% of total volume) was added at day 1 and day3. At day 4 of cultures, cells were washed with fresh medium, counted and plated on 96 wells culture plates (Falcon) for stimulation and functional assays at the indicated times.

PBMC collected from patients were isolated by Ficall-Paque density gradient (Lymphoprep, Proteogenix) from blood samples using standard procedures. PBMC were frozen in liquid nitrogen in DMSO 10% FBS 90%. scRNAseq experiment were performed immediately after thawing.

Method details

Plasmids and viral particle production shSAMHDl coding plasmids were purchased from SIGMA company and are detailed in key resources table. Vpx coding plasmid was used to allow monocytes transduction as previously described ⁵⁹. pLKO.1-GFP was used to assess transduction efficiency while pCMV- AR8.91 and pCMV-VSVg were used for packaging and pseudotyping of viral particles as previously described ⁶⁰. Plasmid amplification was achieved by transformation of Stblt3 bacteria (ThermoFisher) and ampicillin selection. All plasmids were then purified using Purelink Hipure plasmid midiprep kit (ThermoFisher). To produce VLP we transfected plasmids into HEK293FT cells as previously described ⁶¹. HEK293FT cells (Invitrogen) were cultured in DMEM (Gibco), 10% fetal bovine serum (FBS, heat inactivated) (Sigma), 0.1 mM MEM non- essential Amino Acids, 6mM L-glutamine, ImM MEM sodium pyruvate, and antibiotics (Gibco). For generation of VLP containing plasmid of interest (scramble, GFP or shRNA targeting SAMHD1), the day before transfection, cells were seeded on 10cm culture plates to reach 70% confluence for transfection. To produce transfection reagent, 9,6pg of plasmid of interest was prepared together with 6pg A8.91 plasmid and 2,4pg pCMV-VSV-g plasmid in DMEM complemented with 1/25 TransIT-292 transfection reagent (Mirus-Euromedex). The transfection reagent was then added drop by drop to HEK292FT. For generation of VLP containing Vpx (VLP/Vpx), HEK293FT were transfected by calcium-phosphate. The day before transfection, cells were seeded on 15cm culture plates to reach 70% confluence for transfection. Transfection reagent was composed of 50pg pSIV3+ (Vpx coding plasmid⁶²) with lOpg pCMV-VSV-g plasmid in FEO + 0,25M CaC12 buffered by HBS according ProFection Mammalian Transfection protocol (Promega). Transfection reagent for VLP/Vpx was added drop by drop to HEK292FT. After one day, media of all transfected HEK293T cells (both for Vpx or plasmid of interest) was replaced with fresh medium. At day2, supernatant was harvested and filtered twice using 0.45pm sterile filters (Sigma). VLP were immediately used for monocytes transduction.

Lentiviral transduction

Monocytes were transduced immediately after magnetic sorting. Medium was complemented with 8pg/ml polybrene (Sigma) to facilitate transduction, and cells were plated in 10cm culture plates. 5xl0⁶ cells/mL of CD14⁺ in 5ml of complemented medium were transduced with 2,5mL of VLP/Vpx and 4mL of VLP containing plasmid coding shSAMHDl, scramble or GFP as described previously ⁵⁹. Scramble transduced MDDC present unaffected transcriptomic profile and served as a point of reference for analysis. GFP coding plasmid allowed an assessment of transduction efficiency by flow cytometry. MDDC transduction efficiency > 90% at day6 was considered satisfactory.

Flow cytometry analysis

For flow cytometry analysis, MDDCs were washed with PBS and first stained with LIVE/DEAD Fixable dye (TermoFisher) in PBS for 15min at 4°C in the dark. After washing in PBS, cells were stained with antibodies for extracellular labelling for 25min at 4°C in PBS 1% Bovine Serum Albumin (BSA, Roche) and ImM EDTA (Life technology). For intracellular staining (ISG15), cells were fixed with cytofix/cytoperm (BD) for 15min at room temperature then stained with antibodies in permeabilization buffer for 45min at 4°C. After a wash, cells were resuspended in PBS and processed on BD fortessa flow cytometer. Analyses were performed of Flow Jo software.

Seahorse XF Analyser experiment The metabolic assays on MDDC cell culture were performed as previously described ⁶³. Briefly, 100,000 cells were harvested at day 5 after transduction and plated in culture medium in 96- well Seahorse plates (Agilent). At day 6, cells were balanced for 1 h in unbuffered XF assay media (Agilent) supplemented for OCR analysis with either 2 mM Glutamine, 10 mM Glucose and 1 mM Sodium Pyruvate or just 2 mM Glutamine for ECAR measurement. For OCR measurements, compounds were injected during the assay at the following final concentrations: Oligomycin (ATP synthase inhibitor, 1 pM), FCCP (uncoupling agent measuring the maximal respiration capacity; 1 pM), Rotenone and Antimycin A (ETC inhibitors; 1 pM). For ECAR measurements, Glucose (10 mM), Oligomycin (1 pM), and 2-Deoxyglucose (2-DG, glycolytic inhibitor; 500 mM) were injected. For each cell line 4 to 6 technical replicates were evaluated. All OCR and ECAR measurements were normalized to the protein concentration dosed at the end of every experiment.

Quantitative RT PCR

RNA was extracted from 100 000 cells following High Pure RNA isolation kit (Roche) instructions. RNA amplification was performed using Taqman primer probes and PCR master mix (TermoFisher) according to the manufacturer’s instructions or, for CLPP and Hsp60 transcripts, the fast superior cDNA synthesis with SuperScript IV Vilo Reverse Transcriptase (Invitrogen) ⁶⁴. Samples were analyzed on Viia7 Real-Time PCR system (ThermoFisher) and final data were plotted as expression relative to HPRT.

Western Blot

1 million cells were lysed in lOOpL of lysis buffer, composed of protease inhibitor, phosphatase inhibitor and NP40 (Sigma) in MiliQ Water, for 30min on ice. Concentration of proteins were determined by Pierce BCA assay (ThermoFisher). After centrifugation clearance, 30pg of protein lysates were boiled at 95°C for 5min in DTT and loading buffer (Invitrogen). Lysates were deposit on Bolt SDS page gel (Invitrogen) and transferred on nitrocellulose membrane (ThermoFisher) using Trans Blot Turbo transfert (Biorad). Membranes were saturated with TBS (Euromedex) 0,05% Tween20 (SIGMA) 5% BSA (Euromedex). Membranes were stained with indicated antibodies in TBS 0,05% Tween20 and ECL signal was recorded on Chemidoc (Biorad). Images were analysed and band were quantified on Imaged software.

Single-cell RNA sequencing Frozen PBMC of AGS patients or healthy donors (CTRL) were processed by the LabTech single-cell@Imagine facility for cell encapsulation and NGS (Next generation Sequencing). A first experiment was performed on PBMC from healthy donors Cl and C2 and on PBMC from patients P2 before treatment and Pl. A second experiment was performed on PBMC from healthy donors C3 and another vial from Cl (Cl bis) and on PBMC from another vial of P2 before treatment (P2_before_treatment_bis) and PBMC from P2 after treatment. The scRNA- seq libraries were generated using Chromium Single Cell 3' Library & Gel Bead Kit v.2 (lOx Genomics) according to the manufacturer’s protocol. Briefly, cells were counted, diluted at 1000 cells/pL in PBS+0,04% and 20 000 cells were loaded in the lOx Chromium Controller to generate single-cell gel-beads in emulsion. After reverse transcription, gel-beads in emulsion were disrupted. Barcoded complementary DNA was isolated and amplified by PCR. Following fragmentation, end repair and A-tailing, sample indexes were added during index PCR. The purified libraries were sequenced on a Novaseq (Illumina) with 26 cycles of read 1, 8 cycles of i7 index and 98 cycles of read 2. Both experiments were integrated together on integration#!, thus including Cl, Cl bis, C2, C3, Pl, P2_before_treatment, P2_before_treatment_bis and P2_after_treatment.

Later, a new experiment was performed on PBMC from two healthy donors C4 and C5 as well as AGS patients with RNASEH2B mutation (P3 and P4), AD ARI mutations (P5 and P6) and COPA patients (P7 and P8). This experiment account for integration#2. The scRNA-seq libraries were generated using Chromium Single Cell 3' Library & Gel Bead Kit v.3 (lOx Genomics) according to the manufacturer’s protocol. The purified libraries were sequenced on a Novaseq (Illumina) with 28 cycles of read 1, 8 cycles of i7 index and 91 cycles of read 2. During analysis, Pl, P2_before_treatment, P2_before_treatment_bis and P2_after_treatment were often grouped together and labelled as “AGS5” group. Similarly, P3 and P4 were grouped as “AGS2”, P5 and P6 grouped as “AGS6” and P7 and P8 grouped as “COPA”. The integrations!# and #2 could not be merged due to technical advances (kit lOx for library generation version 2 versus kit version 3) made between experimental setup that do not allow direct comparison. Essentially, integration#2 allow significantly more genes to be detected. To assess RTI treatment effect on transcriptomic profile of P2, P2_before_treatment_bis and P2_after_treatment were extracted from integration#! and respectively labelled AGS(5) and AGS(5)T. These data were integrated together with data from healthy donors of the same experimental procedure Cl bis and C3 to generate integration#3 (not shown). Bioinformatic analysis

Sequenced reads from libraries generated were demultiplexed and aligned to the human reference genome (hg38), by Institut Imagine bioinformatic facility, using CellRanger Pipelin V6.0. R version 4.1.2 was used for data processing. Quality control, data integration and downstream analyses were produced using Seurat v4 ⁶⁵. Apoptotic cells and empty sequencing droplets were removed by filtering cells with low features (nfeatures<250 for integration#! and nfeatures<500 for integration#2) or a mitochondrial content <20%. Data were normalized using sctransform. For all three integrations generated, the Seurat clustering resolution 1.6 was selected for cluster identification. Labelling of cell types was defined based on expression of curated list of marker genes previously defined ⁶⁶. Pathway enrichment were performed by applying the indicated list of DEG on EnrichR software ⁶⁷. Enrichment of pathways in the list of DEG were ranked based on combined score. Combined score is a combination of adjusted p-value and z-score. P-value is computed using a standard statistical method used by most enrichment analysis tools: Fisher's exact test or the hypergeometric test. This is a binomial proportion test that assumes a binomial distribution and independence for probability of any gene belonging to any set. z-score is computed using a modification to Fisher's exact test in which we compute a z-score for deviation from an expected rank. EnrichR has lookup table of expected ranks and variances for each term in the library. These expected values are precomputed using Fisher's exact test for many random input gene sets for each term in the gene set library. Enrichr uses this lookup table to calculate the mean rank and standard deviation from this expected rank as the z-score.

Gene regulatory network (GRN) inference analysis

To infer GRN, we apply pySCENIC (Single-Cell rEgulatory Network Inference and Clustering) ²⁵ to a compendium of single-cell data from 8 AGS, 2 COPA patients and 6 healthy controls. Firstly, the gene modules that are co-expressed with transcription factors are identified using GRNboost²⁶. Secondly, for each co-expressed module, the predicted transcription factor binding motifs and candidate transcription factors for a gene list are identified using cis- regulatory motif analyses using RcisTarget. To build the final regulons, we merge the predicted target genes of each TF-module that show enrichment of any motif of the given TF. Finally, regulator-target relationships are extracted and emitted as a set of network edges. Following the above pipeline, we obtain GRN for each sample. To further investigate the TF activity in AGS and control samples collectively, the GRNs of similar samples were integrated using similarity network fusion (SNF)⁶⁸ algorithm. The network-fusion step of SNF uses a non-linear method based on message passing theory⁶⁹ that iteratively updates every network, making it more similar to the others with every iteration. After a few iterations, SNF converges to a single network. The final network is analyzed and the top 50 TFs in AGS and control GRN were obtained by measuring out-degrees. To understand loss or gain of TF activity in AGS, we then performed differential out-degree analysis (DOA)⁷⁰. DOA scores for the top 50 TFs were obtained by taking the ratio of the number of targets in AGS and control GRN. TFs are then ranked based on absolute value of DOA score.

Cytokine measurement

Supernatant of the MDDC were obtained after centrifugation to remove debris at day6 after transduction. Cytokines were measured using LEGENDplex Human Anti-Virus Response Panel (Biolegend) following the manufacturer’s protocol. Supernatant were not diluted during quantification. Fluorescence was measured on a BD fortessa cytometer.

Quantification and statistical analysis

Statistical analyses were performed on Prism 8.0 (Graphpad) to calculate two-way ANOVA with Sidak’s multiple comparisons test or Wilcoxon rank test as indicated. ***:/?<0.001, **:/?<0.01, *:/?<0.05. Experiment with lower than 2 replicates were not statistically tested.

Results:

Marked changes in the transcriptomic profile of monocytes/cDCs in AGS

We extracted peripheral blood mononuclear cells (PBMC) from 6 unrelated patients with AGS. Two patients harbour biallelic mutations in SAMHD1 gene (AGS5). scRNAseq libraries for these two AGS5 patients (AGS5: Pl and P2) were generated and integrated with PBMC from 3 healthy individuals (CTRL: Cl, C2 and C3) to generate integration#! (data not shown). We then generated scRNAseq libraries from 2 patients with biallelic mutations in RNASEH2B gene (AGS2: P3 and P4) and from 2 patients with mutations in ADAR1 gene (AGS6: P5 and P6). Further, we added data from 2 symptomatic patients bearing a dominant-negative mutation in COPA gene (COPA: P7 and P8). COPA pathology is mediated through chronic STING activation even in absence of nucleic acid sensing ²². Thus, COPA syndrome is also a type I IFN driven disease, although, in contrast to AGS patients, COPA manifests as chronic lung inflammation in the absence of overt neurological involvement. None of these patients were under immunosuppressive therapy or RTI at the time of sampling. Finally, we used PBMC from 2 healthy individuals (CTRL: C4 and C5) to complete generation of integration#2 (data not shown). As libraries generated in integrations #1 and #2 were performed with different chemical versions of 10X genomics technology, we analysed these samples separately so as to minimize any batch effect. A first UMAP was generated for integration#! (data not shown), including patients harbouring SAMHD1 mutations and healthy controls, and a UMAP for integration#2 (data not shown) with healthy controls and patients bearing mutations in RNASEH2B, AD ARI and COP A genes (data not shown).

We first assessed cell type proportions in all samples. We detected a decreased proportion of effector memory T cells, NK cells, MAIT and y6 T cells in PBMC of AGS5 patients (data not shown). In parallel, B cell proportion was increased. We obtained comparable results for AGS2, AGS6 and COPA patients, except for NK CD56int, MAIT and CD8 effector memory cells, with proportions similar to control for AGS6 patients (data not shown).

As both AGS and COPA are type I IFN-related diseases, we then assessed IFN-stimulated gene (ISG) expression at the single-cell transcriptomic level. Specifically, we used a previously described signature of 272 ISG ²³ and observed, as expected, an elevated type I IFN signature in AGS5 (data not shown), AGS2, AGS6 and COPA (data not shown) patients compared to respective controls. Interestingly, the strongest IFN response was observed in monocytes and classical dendritic cells (cDC). A six-ISG “IFN signature” has proven useful for diagnostic purposes in type I interferonopathies ²⁴. As expected, the expression of these six ISG was raised in all patients tested, although to a lesser extend in COPA patients compared to AGS2 and AGS6 patients (data not shown). The six ISG signature also remain higher in monocytes/cDC than in other cell types (data not shown). Moreover, when we extracted all differentially expressed genes (DEG), it appeared that monocytes and eDC demonstrate at least 40% more genes differentially regulated between AGS patients and CTRL than other cells types in integration#! and 15,7% more in integration#2 (data not shown). This higher number of DEG in monocytes/cDC is not due to higher cell proportion in this group (data not shown). Thus, this could suggest that greater transcriptomic changes occur in monocytes and eDC of AGS patients, even if we should stress that monocytes and DC express more genes than other cell types analyzed (data not shown).

Gene regulatory network analysis reveals loss of HIF-la transcription factor activity in PBMC of AGS patients In order to gain an unsupervised understanding of factors driving the observed transcriptomic differences in AGS, we performed a gene regulatory network (GRN) analysis ²⁵ applying pySCENIC (Single-Cell rEgulatory Network Inference and Clustering)²⁵ on all cells from our scRNAseq data sets. Briefly, the target genes that are co-expressed along with transcription factors (TF) are identified as modules using GRNboost²⁶. Then, for each co-expressed module to a given TF, the transcription factor binding motifs are identified using RcisTarget. To build the final regulons, we selected the predicted target genes of each TF -module that shows enrichment of any binding motif of the given TF. Predictions of TF importance in the transcriptomic profile of patients and controls are defined on the number of TF targets predicted by the GRN (termed out-degrees). According to this parameter, TF can be ranked in each group. Doing so, we extracted TF expected to be important in the transcriptomic state of PBMC in the CTRL group of integration#! (data not shown) and integration#2 (data not shown). We found recurrent TF among those predicted to be the most important in CTRL groups, including CEBP, SIS 1, MAFB and HIF1A. We produced the same GRN for all AGS patients of integration#! (data not shown) and integration#2 (data not shown), to find that CEBP, SPI1 and MAFB are conserved among the top TF in patients but not HIF1A. We then focused our analysis on predictions of TF regulation between each other. We identified regulatory events between HIF1A and some of the most important TF in CTRL groups (CEBPB, CEBPD and F0SL2) (data not shown), highlighting its importance in shaping the transcriptional profile of CTRL PBMC. To better understand variations in the number of targets predicted for each TF between CTRL and AGS patients, we performed a differential out-degree analysis, by comparing the number of targets associated with each TF in AGS versus CTRL. Most of the TFs with increased out-degree between CTRL and AGS patients relate to inflammatory mechanisms and a type I IFN response such as ERF6, IRF7 or STAT1 (data not shown). In contrast, HIF-la stands out as the transcription factor with the most important loss of targets in AGS compared to the CTRL group in integration#! (data not shown) and integration#2 (data not shown). Indeed, the number of predicted targets of HIF1A in AGS reached 0 in both integrations (data not shown). Overall, using an unsupervised approach with network inference analyses, we determined that in AGS2, AGS5 and AGS6 patients’ cells, beside an already described strong response to type I IFN and inflammation, a marked loss of HIF-la activity was predicted to explain most of the downregulation observed in AGS patients at the transcriptomic level.

Defective HIF-la transcriptional program in monocytes/cDC of AGS patients To validate and investigate the importance of the loss of HIF- la-mediated transcriptional activity in AGS patients highlighted by our unsupervised approach, we decided to focus first on monocytes/cDC, where the highest response to type I IFN and number of dysregulated genes were detected (data not shown). HIF1A, the gene coding FUF-la, appears among the most differentially downregulated genes in monocytes/cDC of both AGS5 (data not shown) and AGS2/AGS6 (data not shown) patients compared to CTRL. Indeed, HIF1A expression is decreased at the mRNA level in monocytes/cDC of AGS patients (data not shown), and similarly in all subpopulations of monocytic cell type (data not shown). Not only was HIF1A expression lower in each cell, but the percentage of cells expressing detectable levels of HIF1A mRNA was also reduced in AGS patients’ cells, to a greater extent in monocytes/cDC than in any other cell type (data not shown). COPA patients presented an intermediate level of HIF1A expression between CTRL and AGS2/AGS6 patients (data not shown). Yeh et al. first suggested that HIF-la protein levels can be decreased by ISG15-induced proteasome degradation in response to type I IFN ²⁷. Thus, we plotted HIF1A expression against type I IFN signature in monocytes/cDC. Cells expressing high levels of type I IFN signature expressed low levels of HIF1A (data not shown). However, the correlation is not strict, as monocytes/cDC with a low IFN signature demonstrated variable levels of HIF1A (data not shown). These data suggest that enhanced type I IFN signalling in AGS and COPA patients’ cells, might repress HIF1A expression at the mRNA level. To verify that lower HIF1A expression correlates with decreased HIF-la transcriptomic activity, we applied a set of genes known to be HIF-la targets from the NCLNatures database ²⁸ to generate a signature of HIF- la activity in our datasets. Monocytes/cDC of AGS patients demonstrated a lower signature score of HIF-la targets compared to CTRL (data not shown). Again, cells from COPA patients reached an intermediate level between CTRL and AGS2/AGS6 patients. Among all HIF-la targets, vascular endothelial growth factor A (VEGFA) is one of the most sensitive to HIF-la regulation ²⁹. Following HIF1A expression, VEGFA was mainly expressed in monocytes/cDC of CTRL, and completely lost in patient cells (data not shown). We confirm here a clear loss of HIF1A expression and transcriptional activity in AGS patients. Regarding COPA syndrome, patients present an intermediate phenotype between CTRL and AGS, corresponding to an intermediate type I IFN response.

Energy metabolism switch with mitochondrial stress in monocytes/cDC of AGS patients To further explore the transcriptomic dysregulations observed in our scRNA-seq datasets, we assessed differentially expressed genes (DEG) between AGS patients and CTRL cells among monocytes/cDC to reveal cellular pathways modified by the disease. We applied separately upregulated and downregulated DEG lists to several independent databases for pathway enrichment using EnrichR software. As expected, in both integrations, the top upregulated genes belonged to pathways directly related to type I IFN signalling (data not shown). However, we also observed an upregulation of a metabolic pathways, in particular oxidative phosphorylation, that has not been reported before in this context. Importantly, this effect was associated with a downregulation of genes associated with aerobic glycolysis (also known as the Warburg effect) and HIF-la signalling. We observed the same pathways when analysing DEG between AGS patients and CTRL among all PBMC (data not shown). Oxidative phosphorylation represents the most efficient way to produce energy from glucose, and involves the respiratory chain in mitochondria. In contrast, aerobic glycolysis is less efficient in energy production, but more responsive, and does not require mitochondria oxygen consumption ³⁰. HIF-la stands at the fork between oxidative phosphorylation and aerobic glycolysis, favouring glucose catabolism through the later and inhibiting oxidative phosphorylation. We further assessed energy metabolism in AGS and COPA patient cells by applying publicly available signatures for aerobic glycolysis and oxidative phosphorylation pathways ^2X'^{3 1}. Accordingly, we observed that AGS patients demonstrated increased expression of genes involved in oxidative phosphorylation, together with a decreased in the expression of genes involved in aerobic glycolysis (data not shown). Interestingly, COPA patients presented, again, intermediate signatures expression between CTRL and AGS2/AGS6 patients for both metabolic pathways (data not shown), which followed HIF1A expression and activity. When compared to other cell types, monocytes/cDC had the most noticeable increase in oxidative phosphorylation signature associated with a strong decrease in genes related to aerobic glycolysis in patients (data not shown). Elevated expression of genes involved in oxidative phosphorylation led us to assess features of mitochondrial stress in transcriptomic data from patients. We applied a mitochondrial stress signature (data not showns ³²“³⁷) to our dataset and observed a high expression of mitochondrial stress genes in all PBMC of AGS patients, in particular in monocytes/cDC (data not shown). Again, this mitochondrial stress signature was intermediate in COPA patient’s cells (data not shown). Thus, the major transcriptomic changes observed in monocytes/cDC from patients with AGS reflect an energy metabolism switch towards oxidative phosphorylation associated with mitochondrial stress in AGS patients. COPA patients present an intermediate phenotype, correlated with an intermediate response to type I IFN and HIF1A expression.

Detection by scRNAseq of a monocyte/cDC cluster of cells specific to the response to the treatment with RTI

As scRNAseq allowed us to establish important transcriptomic changes in AGS patients particularly in monocytes/cDC, we aimed at following the transcriptomic profile of patients before and after treatment with reverse transcriptase inhibitor (RTI) triple therapy ²¹. We therefore focused on P2 harbouring biallelic SAMHD1 mutations from whom we were able to generate scRNAseq data before (P2_before_T), and 1 month after the initiation of treatment (P2_after_T). Both samples were integrated with data from controls Cl and C3 to generate integration#3 (data not shown). First, we assessed cell type proportions in each sample and observed a decrease in the proportion of memory T cells and NK cells and an increase in the proportion of B cells in patient#2 compared to CTRL group (data not shown). Then, we assessed the type I IFN response and confirmed that monocytes/cDC presented the greater score of type I IFN signalling (data not shown). Focusing on monocytes/cDC, cells from CTRL presented a lower score to type I IFN response, while patient before treatment (P2_before_T) highly responded to type I IFN. After 1 month of treatment (P2_after_T), most cells remained high for type I IFN signature, although some cells demonstrated a level of type I IFN signature as low as in CTRL (data not shown). Interestingly, these cells demonstrating a low type I IFN response clustered together in a unique cluster labelled cluster 12 (data not shown). This cluster was seen exclusively in the treated patient (data not shown), and consists of a mixture of eDC, classical monocytes and non-classical monocytes based on marker expression of, respectively, FCER1A, S100A8 and FCGR3A (data not shown). We hypothesized that cluster 12 might correspond to treatment-responsive cells of the monocytic lineage. We then further characterized cells belonging to this cluster, and detected higher levels of HIF1A expression compared to other patient cells, reaching CTRL levels of expression (data not shown). Concordantly, cells with restored HIF1A expression demonstrated very low levels of type I IFN signature (data not shown). The activity of HIF-la was also restored in cells of cluster 12, based on the expression of HIF-la targets (data not shown). By performing DEG analysis between cells from cluster 12 and other monocytes/DCs of P2 after treatment, we confirmed a downregulation of genes belonging to type I IFN response and to oxidative phosphorylation (data not shown). Indeed, cells from cluster 12 presented a lower level of oxidative phosphorylation signature associated with a lower mitochondrial stress score, compared to monocytes/cDC from P2, to reach a profile similar to CTRL group (data not shown). These results suggest that treatment impacted a relatively small number of cells in which it reduced type I IFN response in monocytes/cDC, alongside a restored HIF-la activity, reduced oxidative phosphorylation and decreased mitochondrial stress. Our results further suggest a link between the type I IFN response and HIF-la regulation of energy metabolism in AGS.

SAMHD1 knock-down in MDDC replicates some molecular features of AGS in vitro

To validate the observations drawn from scRNAseq analysis of AGS patients’ material, we first established an in vitro cellular model, based on primary monocytes from healthy blood donors differentiated into dendritic cells (MDDC, Monocyte Derived Dendritic Cell), by targeting SAMHD1 expression at the mRNA level with short hairpin RNA (shRNA). We obtained 2 shRNA (shSAMHDl_3 and shSAMHDl), both effectively inhibiting SAMHD1 at the mRNA level (data not shown), with shSAMHDl also presenting a noticeable downregulation at the protein level (data not shown). Upon knock-down (KD) of SAMHD1, MDDC produced detectable amounts of IFN-P protein and high amounts of IP-10 but not IL-ip or TNFa (data not shown), consistent with what has been reported in AGS patients. ISG15 expression, a hallmark of a type I IFN mediated response, was tracked by flow cytometry staining (data not shown). ISG15 protein expression was also significantly upregulated in shSAMHDl transduced MDDC (data not shown), and reaching a maximal response at day 6 after transduction (data not shown). A type I IFN response was also observed at the mRNA level by measuring the representative six-ISG signature described above (comprising: ISG15, IFI44L, IFIT1, IFI27, RSAD2 and SIGLECF) (data not shown). We then aimed to confirm our observations related to HIF1A dysregulation in AGS, and detected lower amounts of HIF-la protein in SAMHDl deficient MDDC than in control MDDC (data not shown). To address potential energy metabolism imbalance resulting from observed HIF-la dysregulation, we measured key indicators of cellular metabolism using Seahorse XF Analyser (Agilent). Briefly, at day 4 after transduction with shRNA, extracellular acidification rate (ECAR), which is a readout of the lactate secretion by cells, were monitored to measure aerobic glycolysis activity. In parallel, oxygen consumption rate (OCR) is measured and reflects oxidative phosphorylation activity through oxygen consumption by the complex 5 of the mitochondrial electron transport chain. Several drugs were added during the time course and allowed us to measure ATP synthase activity, maximal respiratory capacity, actual glycolysis and maximal glycolytic capacity (data not shown). These experiments revealed that the glycolytic capacity of shSAMHDl -transduced MDDC was reduced (data not shown). Reduced glycolytic capacity was associated with qRT-PCR results showing increased expression of the mitochondrial stress associated genes CLPP and Hsp60 in SAMHD1 deficient MDDC patients (data not shown). All these results were strengthened with the observation of the same trend with the second shRNA shSAMHDl_3, but to a lesser extent (data not shown). Finally, we used a cocktail of type I IFN blocking antibodies and observed an abolished ISG15 expression and IP-10 production, confirming the strict type I IFN dependency of these cellular responses (data not shown) in our in vitro system. Here, by generating an in vitro cellular model of AGS5 in human primary MDDC, we reproduced induced type I IFN production and response to type I IFN as well as high IP-10 secretion, alongside reduction of HIF-la expression. This model validates at the functional level what was observed at the scRNAseq level in monocytes/cDC of AGS patients, i.e., a metabolic imbalance in SAMHD1 deficient cells, with increased oxidative phosphorylation and mitochondrial stress.

Stabilization of HIF-la expression level reverts the energy metabolism switch, relieves mitochondrial stress and inhibits IP-10 production

As HIF-la activity is known to control energy metabolism, we investigated whether we could correct the metabolic switch observed in our in vitro model of AGS5 MDDCs by using the prolyl hydroxylase inhibitor dimethyloxalylglycine (DMOG), known to prevent HIF-la degradation. 500 pM of DMOG effectively increased HIF-la protein levels as early as 5h post incubation, although it was less efficient in shSAMHDl KD cells due to HIF-la decreased level in these cells (Figure 1A). To monitor HIF-la function, the energy metabolism state of SAMHDl deficient MDDC was assessed by Seahorse XF Analyser (Agilent), in the presence or absence of 2 different doses of DMOG (250 pM and 500 pM) for 48 hours. We noted that HIF-la stabilisation reduced, in a dose dependent manner, the oxidative phosphorylation capacity of cells, while increasing aerobic glycolysis activity, in a dose dependent manner (Figure IB). In doing so, glucose consumption was redirected from mitochondrial respiration toward aerobic glycolysis. Consistently, following DMOG treatment, CLPP and Hsp60 gene expression was reduced in SAMHDl deficient MDDC, indicative of reduced mitochondrial stress (Figure 1C). We then assessed the impact of DMOG on the type I IFN production and response in this model. HIF-la stabilisation partially reduced mRNA levels of some ISG (Figure ID), but did not affect ISG15 protein levels (Figure IE), nor IFN- P production (Figure IF). However, SAMHDl deficient MDDC produced between 2 to 4 times less IP- 10 after DMOG treatment, depending on the donor (Figure IF). Similar results, following DMOG treatment, were reproduced with shSAMHDl_3 (data not shown), albeit to a lesser extent. In conclusion, we are reporting that HIF-la stabilisation by DMOG reverts the energy metabolism switch observed in SAMHD1 deficient MDDC, relieves mitochondrial stress and markedly inhibits the production of the neurotoxic cytokine IP-10.

Discussion:

Characterization of AGS pathogenesis at the molecular level

Based on our findings from scRNAseq analysis of PBMC from AGS patients, combined with in vitro validations, we propose an additional model of AGS pathogenesis, in accordance with existing literature. First, in healthy monocytes/cDC, oxidative phosphorylation is known to be the favoured pathway to generate energy. Although it represents the most efficient way to process glucose, this metabolic pathway has the downside of being a major source of ROS, a known inducer of oxidative stress. HIF-la, a sensor of oxidative stress, is stabilized at the protein level upon ROS accumulation, and regulates ROS production by favouring glucose processing through aerobic glycolysis, bypassing mitochondrial processing through oxidative phosphorylation. This mechanism is expected to be even more important in cells with high ROS content such as myeloid cells. Once ROS levels decrease, glucose catabolism through oxidative phosphorylation recommences in the presence of reduced HIF-la levels, implying a cyclic regulation of energy metabolism based on HIF-la stabilization. In AGS, our results suggest that constant type I IFN production induces molecular processes that inhibit HIF-la stabilisation through both reduced mRNA expression and proteasome-induced degradation. Cells thus remain in an enhanced oxidative phosphorylation state and are no longer able to appropriately regulate their energy metabolism. In the absence of HIF-la stabilisation ROS accumulates, increasing oxidative stress and inducing several events such as NF-kB activation and mitochondrial stress. While NF-kB activation is known to aggravate an IP-10 response, chronic mitochondrial stress affects mitochondrial membrane permeability, favouring mtNA release in the cytoplasm. As cytoplasmic nucleic sensing pathways can detect mtNA, we propose that this feeds a vicious cycle of type I IFN production and chronic IP-10 production in AGS. In an in vitro cellular model of AGS, HIF-la stabilisation by the synthetic chemical compound DMOG, allowed a switch of energy metabolism to anaerobic glycolysis, even in presence of type I IFN. In this way, mitochondrial stress is relieved and IP-10 production is reduced. The different results of our study presented hereby in the model are now discussed below.

Type I IFN regulation of HIF1A

Based on the use of a machine-learning algorithm, we established the gene regulatory network (GRN) of PBMC from AGS patients and compared them with the GRN of healthy donors. The transcription factor HIF-la has been predicted to have the most marked loss of activity in all AGS patients examined. This result potentially indicates that the complete loss of HIF-la activity in AGS is among the most dysregulated pathways once the pathology and type I IFN chronic production is established. Only one study reported a potential link between type I IFN and HIF-la, showing ISG15 capacity to decrease HIF-la protein levels through proteasome induced degradation ²⁷. Here, we report an inhibition of HIF1A at the mRNA level in PBMC of patients, alongside a decrease at the protein level in our in vitro cellular model with SAMHD1 inhibition by shRNA. These results could indicate that a chronic type I IFN exposure leads to HIF1A inhibition at the mRNA level. While ISG15 likely has a role, we do not exclude potential role of other ISG in the regulation of HIF1A.

Role of HIF-la in monocytes/cDC

HIF1A downregulation was associated with a decreased expression of its target genes, especially in monocytes/cDC of AGS patients. HIF1A is a sensor of metabolic stress in cells, and is stabilized when ROS reach a dangerous levels or if oxygen is lacking ¹⁵. Several sources of ROS can be listed, with the main one involving electron leakage from the mitochondria electron transport chain ³⁸. HIF-la stabilisation inhibits mitochondrial activity in order to supress ROS generation and oxygen consumption by the electron transport chain complexes ¹⁶. To compensate for the lack of energy production following mitochondria inhibition, conversion of glucose into lactate (aerobic glycolysis) is favoured. Several enzymes of the aerobic glycolysis are upregulated by HIF-la, including the lactate dehydrogenase that transforms the glycolysis end product pyruvate into lactate ¹⁶. The switch of energy production from mitochondria activity toward aerobic glycolysis is called the Warburg effect, first described in cancer cells ²⁰. Later studies have established the importance of this energy metabolism switch in critical cellular processes such as immune cell proliferation ¹⁹. We hypothesize that HIF1A is more highly expressed in monocytes/cDC, as these cells rely more heavily on the Warburg effect than other cell types. Indeed, upregulation of aerobic glycolysis has been shown to be necessary for the proper activation and maturation of dendritic cells ¹⁷ and macrophages ¹⁸. On top of that, monocytes rely on ROS generation for pathogen clearance ³⁹, and dendritic cells antigen presentation capacity also involve ROS ⁴⁰, implying a greater susceptibility to ROS induced stress and, consequently, HIF-la stabilisation. Our results place HIF1A at the centre of the transcriptomic profile regulation of monocytes/cDC, concordant with current literature establishing the importance of energy metabolism regulation in those cells.

Uncontrolled oxidative phosporylation induces mitochondrial stress

We observed increased expression of genes involved in oxidative phosphorylation in AGS patients’ cells associated with decreased signatures of aerobic glycolysis, especially in monocytes/cDC. Moreover, cells from AGS patients present increased expression of mitochondrial stress genes. Altogether, these data suggest that lack of HIF1A expression confers a defect in cell capacity to regulate oxidative phosphorylation by switching to aerobic glycolysis. We hypothesize that overactivation of oxidative phosphorylation, both in intensity and in time, favours mitochondrial stress. Two independent teams recently performed genome wide association studies (GWAS) on large cohorts to identify single nucleotide polymorphisms (SNP) associated with disruption in mtDNA copy number ⁴¹’⁴². Both studies revealed that SNP in SAMHD1 correlate with an increase in mtDNA copy number, strengthening our hypothesis of high mitochondrial stress in AGS (at least in SAMHD1 mutated patients) as disruption of mtDNA copy number is indicative of oxidative stress in the mitochondria ^43,44. One good candidate for oxidative phosphorylation induction of mitochondrial stress is ROS generation by the electron transport chain complexes. ROS accumulation has been shown to be an inducer of mitochondrial stress which, if uncontrolled, leads to a loss of mitochondria integrity ⁴⁵. Thus, we speculate that lack of HIF1A expression in AGS would result in defective sensing of ROS accumulation from oxidative phosphorylation, in turn inducing mitochondrial stress. Accordingly, when HIF-la was stabilized in our in vitro cellular model of SAMHD1 deficiency, oxidative phosphorylation was reduced, and mitochondrial stress relieved. Of possible relevance, we note that AGS has been suggested to involve a disturbance of mitochondria activity in some patients ⁴⁶'⁴⁷. Notably, a recent report revealed alteration of mitochondrial integrity associated with uncontrolled oxidative phosphorylation, increased ROS activity and mitochondrial stress in AGS patients mutated \nRNASEH2A or RNASEH2B ⁴⁸. We propose here, that HIF-la loss of activity could be a driver of mitochondrial dysfunction in AGS patients. It has been shown that chronic mitochondrial stress drives a loss of mitochondrial membrane integrity ⁴⁵. Consequently, mtDNA and mtRNA can be released into the cytoplasm and then sensed by nucleic acid receptors, resulting in type I IFN production ⁴⁵’⁴⁹’⁵⁰. Thus, it is tempting to speculate that type I IFN inhibition of HIF1A expression could further fuel nucleic acid sensing in AGS through uncontrolled mitochondrial activity followed by mtDNA/RNA release in the cytosol, resulting in exacerbated inflammation.

Uncontrolled production of IP-10 in AGS

We also observed high IP-10 production in MDDC KD for SAMHD1. IP-10 is one of the most highly expressed inflammatory cytokines detected in plasma and CSF of AGS patients ¹⁰. Its proinflammatory role has been attributed to an induction of chemotaxis of inflammatory cells bearing the CXCR3 receptor, which includes monocytes, NK cells, activated T cells and neutrophils ⁵¹. Of note, IP-10 can also negatively regulate angiogenesis ⁵² which might relate to the high frequency of vasculitic chilblain-like lesion seen in patients with AGS. Moreover, IP- 10 possess neurotoxicity capacity ⁿ, which might be correlated with deleterious neurological manifestations in AGS. IP-10 production is notably induced following type I IFN signalling, but cellular sources of IP-10 are uncertain in AGS. Endothelial and epithelial cells are potent producers, but some studies suggest that immune cells represent important IP-10 producers in human neurological disease ⁵³. We show that, among PBMC, cells with the highest score for type I IFN response are monocytes/cDC. Moreover, in vitro, MDDC deficient for SAMHD1 produced high levels of IP-10, arguing for an important role for monocytes/cDC in the production of IP-10 in AGS patients. Nonetheless, this IP-10 production was reduced following HIF-la stabilisation by DMOG, indicative of a suppressive effect of EUF-la on IP-10 production. It has been shown that regulation of energy metabolism, in particular by HIF-la, is crucial for DC maturation and other cytokines production ¹⁷’⁵⁴. Thus, we hypothesize here that high levels of oxidative phosphorylation are required for IP-10 production in MDDC and that HIF-la stabilisation limits IP-10 production by reverting energy metabolism towards aerobic glycolysis. We show here that HIF-la stabilisation inhibits IP-10 production without affecting type I IFN production, suggesting that high IP-10 secretion in AGS patient is favoured by HIF1A reduced expression in monocytes/cDC.

Results in COPA patients reinforce the link between type I IFN and HIF-la regulation of energy metabolism We compared samples from AGS patients to both healthy donors and to COPA patients in integration#2. Like AGS, COPA is a monogenic disease associated with enhanced type I IFN signalling. The disease results from dominant-negative mutations in COPA, a protein involved in the vesicular transport from the Golgi apparatus to the endoplasmic reticulum ²². In COPA syndrome, chronic IFN signalling results from a failure of retrograde transport of STING from the endoplasmic reticulum to the Golgi ²². While the pathogenetic mechanism in AGS and COPA overlaps, clinical manifestation are quite distinct, COPA syndrome is characterized by lung inflammatory involvement but can also involve joint and kidney inflammation ²². In our scRNAseq data, transcriptomic profiling of COPA patients followed the trend of AGS patients (AGS2 and AGS6) with high type I IFN response, decreased HIF1A expression, an energy metabolism switch towards oxidative phosphorylation and induced mitochondrial stress. However, for all signatures tested, cells from COPA patients manifest an intermediate response, between AGS and CTRL. Such a direct correlation between type I IFN response levels, HIF- la inhibition, oxidative phosphorylation activation and mitochondrial stress strengthens our hypothesis placing type I IFN as a direct driver of HIF-la inhibition and, consequently, metabolic dysfunction. On this basis, we suggest that energy metabolism, and in particular HIF- la regulation, should be tested in other type I IFN mediated diseases.

Cell type proportion in AGS

Using scRNAseq data from AGS patients' cells, we identified a loss of effector memory T cells proportion in patients. This result is coherent with reports in a mouse model indicating that type I IFN exposure can induce apoptosis in the memory compartment ⁵⁵ and also with observation in patients with type I IFN mediated diseases like SAVI ⁵⁶. However, we also identified a loss in MAIT, T y6 and NK cells, not yet reported to our knowledge in AGS nor in other type I IFN associated disorders. Interestingly, in the two AGS6 patients with mutations nADARl, normal levels of CD8 memory cells, NK and MAIT were observed, highlighting potential differences between AGS -associated genotypes. Indeed, while most AGS patients share common features like type I IFN signalling and neurological involvement, each genotypes can present accessory manifestations including vasculitis (chilblain lesions), glaucoma, cardiomyopathy or lung involvement ⁹. How mutations in the same pathway can have such different outcomes remains poorly understood. Concerning AGS6 (ADAR1 mutation), one could hypothesize that it activates type I IFN production through MAVS sensor, while other AGS associated mutations studied here activate STING sensor. Links between molecular pathways, different cellular proportions and particular clinical features among different genotypes will be interesting to further decipher in future studies on AGS.

Treatment-specific cluster of cells detected by scRNAseq

Using scRNAseq data we were able to detect cells responsive to RTI treatment in one patient with AGS. Responsive cells grouped in a unique cluster composed of subpopulations of both monocytes and cDC. These cells share a reduced type I IFN response compared to other monocytes/cDC correlated with restored HIF1A expression and reduced mitochondrial stress. A phase I clinical trial with RTI therapy over 12 months showed a reduced IFN-a levels in serum and CSF, and a lower type I IFN score in the whole blood of patients with AGS 1-5 mutations ²¹. This indicates that circulating cells in treated patients are likely to be exposed to lower levels of type I IFN allowing some cells to escape ISG induction. Although, this result needs to be reproduced in other patients, it highlights the potential importance of performing single-cell analyses in finely monitoring the effect of a treatment at the molecular level, and its potential value as a new tool to be used in personalized medicine. This encourages future research studies to assess individual follow up of transcriptional profile by scRNAseq during therapy, particularly in diseases with heterogeneous treatment responses.

TABLES:

Table 1: List of genes in “HIF1A activity”, “Oxidative phosphorylation” and “aerobic glycolysis” signatures.

REFERENCES:

Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure.

1. Rice, G., Newman, W.G., Dean, J., Patrick, T., Parmar, R., Flintoff, K., Robins, P., Harvey, S., Hollis, T., O’Hara, A., et al. (2007). Heterozygous Mutations in TREX1 Cause Familial Chilblain Lupus and Dominant Aicardi-Goutieres Syndrome. The American Journal of Human Genetics 80, 811-815. 10.1086/513443.

2. Crow, Y.J., Leitch, A., Hayward, B.E., Garner, A., Parmar, R., Griffith, E., Ali, M., Semple, C., Aicardi, J., Babul-Hirji, R., et al. (2006). Mutations in genes encoding ribonuclease H2 subunits cause Aicardi-Goutieres syndrome and mimic congenital viral brain infection. Nat Genet 38, 910-916. 10.1038/ngl842. 3. Rice, G.I., Bond, J., Asipu, A., Brunette, R.L., Manfield, I.W., Carr, I.M., Fuller, J.C.,

Jackson, R.M., Lamb, T., Briggs, T.A., et al. (2009). Mutations involved in Aicardi-Goutieres syndrome implicate SAMHD1 as regulator of the innate immune response. Nat Genet 41, 829- 832. 10.1038/ng.373.

4. Rice, G.I., Kasher, P.R., Forte, G.M.A., Mannion, N.M., Greenwood, S.M., Szynkiewicz, M., Dickerson, J.E., Bhaskar, S.S., Zampini, M., Briggs, T.A., et al. (2012). Mutations in ADAR1 cause Aicardi-Goutieres syndrome associated with a type I interferon signature. Nat Genet 44, 1243-1248. 10.1038/ng.2414.

5. Rice, G.I., Del Toro Duany, Y., Jenkinson, E.M., Forte, G.M., Anderson, B.H., Ariaudo, G., Bader-Meunier, B., Baildam, E.M., Battini, R., Beresford, M.W., et al. (2014). Gain-of- function mutations in IFIH1 cause a spectrum of human disease phenotypes associated with upregulated type I interferon signaling. Nat Genet 46, 503-509. 10.1038/ng.2933.

6. Uggenti, C., Lepelley, A., Depp, M., Badrock, A.P., Rodero, M.P., El-Daher, M.-T., Rice, G.I., Dhir, S., Wheeler, A.P., Dhir, A., et al. (2020). cGAS-mediated induction of type I interferon due to inborn errors of histone pre-mRNA processing. Nat Genet 52, 1364-1372. 10.1038/s41588-020-00737-3.

7. Schlee, M., and Hartmann, G. (2016). Discriminating self from non-self in nucleic acid sensing. Nat Rev Immunol 16, 566-580. 10.1038/nri.2016.78.

8. Tanaka, Y., and Chen, Z.J. (2012). STING Specifies IRF3 Phosphorylation by TBK1 in the Cytosolic DNA Signaling Pathway. Sci. Signal. 5. 10.1126/sci signal.2002521.

9. Crow, Y.J., and Manel, N. (2015). Aicardi-Goutieres syndrome and the type I interferonopathies. Nat Rev Immunol 15, 429-440. 10.1038/nri3850.

10. Van Heteren, J.T., Rozenberg, F., Aronica, E., Troost, D., Lebon, P., and Kuijpers, T.W. (2008). Astrocytes produce interferon-alpha and CXCL10, but not IL-6 or CXCL8, in aicardi- Goutieres syndrome. Glia 56, 568-578. 10.1002/glia.20639.

11. van Marie, G., Henry, S., Todoruk, T., Sullivan, A., Silva, C., Rourke, S.B., Holden, J., McArthur, J.C., Gill, M.J., and Power, C. (2004). Human immunodeficiency virus type 1 Nef protein mediates neural cell death: a neurotoxic role for IP-10. 17.

12. JAK Inhibition in the Aicardi-Goutieres Syndrome (2020). N Engl J Med 383, 2190- 2193. 10.1056/NEJMc2031081.

13. Epstein, A.C., Gleadle, J.M., McNeill, L.A., Hewitson, K.S., O’Rourke, J., Mole, D.R., Mukherji, M., Metzen, E., Wilson, M.I., Dhanda, A., et al. (2001). C. elegans EGL-9 and mammalian homologs define a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell 107, 43-54. 10.1016/s0092-8674(01)00507-4.

14. Liu, Q., Berchner-Pfannschmidt, U., Moller, U., Brecht, M., Wotzlaw, C., Acker, H., Jungermann, K., and Kietzmann, T. (2004). A Fenton reaction at the endoplasmic reticulum is involved in the redox control of hypoxia-inducible gene expression. Proc Natl Acad Sci U S A 101, 4302-4307. 10.1073/pnas.0400265101.

15. Taylor, C.T., and Scholz, C.C. (2022). The effect of HIF on metabolism and immunity. Nat Rev Nephrol 18, 573-587. 10.1038/s41581-022-00587-8.

16. Papandreou, I., Cairns, R.A., Fontana, L., Lim, A.L., and Denko, N.C. (2006). HIF-1 mediates adaptation to hypoxia by actively downregulating mitochondrial oxygen consumption. CELL METABOLISM, 11.

17. Jantsch, J., Chakravortty, D., Turza, N., Prechtel, A.T., Buchholz, B., Gerlach, R.G., Volke, M., Glasner, J., Warnecke, C., Wiesener, M.S., et al. (2008). Hypoxia and Hypoxia- Inducible Factor-la Modulate Lipopolysaccharide-Induced Dendritic Cell Activation and Function. J Immunol 180, 4697-4705. 10.4049/jimmunol.180.7.4697.

18. Peyssonnaux, C., Datta, V., Cramer, T., Doedens, A., Theodorakis, E.A., Gallo, R.L., Hurtado-Ziola, N., Nizet, V., and Johnson, R.S. (2005). EUF-lalpha expression regulates the bactericidal capacity of phagocytes. J Clin Invest 115, 1806-1815. 10.1172/JCI23865.

19. Wang, T., Marquardt, C., and Foker, J. (1976). Aerobic glycolysis during lymphocyte proliferation. Nature 261, 702-705. 10.1038/261702a0.

20. Warburg, O. (1956). On the Origin of Cancer Cells. 123, 7.

21. Rice, G.I., Meyzer, C., Bouazza, N., Hully, M., Boddaert, N., Semeraro, M., Zeef, L.A.H., Rozenberg, F., Bondet, V., Duffy, D., et al. (2018). Reverse-Transcriptase Inhibitors in the Aicardi-Goutieres Syndrome. N Engl J Med 379, 2275-2277. 10.1056/NEJMcl810983.

22. Lepelley, A., Martin-Niclos, M.J., Le Bihan, M., Marsh, J.A., Uggenti, C., Rice, G.I., Bondet, V., Duffy, D., Hertzog, J., Rehwinkel, J., et al. (2020). Mutations in COPA lead to abnormal trafficking of STING to the Golgi and interferon signaling. Journal of Experimental Medicine 217, e20200600. 10.1084/jem.20200600.

23. Rosenberg, B.R., Freije, C.A., Imanaka, N., Chen, S.T., Eitson, J.L., Caron, R., Uhl, S.A., Zeremski, M., Talal, A., Jacobson, I.M., et al. Genetic Variation at IFNL4 Influences Extrahepatic Interferon-Stimulated Gene Expression in Chronic HCV Patients. 6.

24. Hadjadj, J., Yatim, N., Barnabei, L., Corneau, A., Boussier, J., Smith, N., Pere, H., Charbit, B., Bondet, V., Chenevier-Gobeaux, C., et al. (2020). Impaired type I interferon activity and inflammatory responses in severe COVID-19 patients. 8.

25. Van de Sande, B., Flerin, C., Davie, K., De Waegeneer, M., Hulselmans, G., Aibar, S., Seurinck, R., Saelens, W., Cannoodt, R., Rouchon, Q., et al. (2020). A scalable SCENIC workflow for single-cell gene regulatory network analysis. Nat Protoc 15, 2247-2276. 10.1038/s41596-020-0336-2. 26. Moerman, T., Aibar Santos, S., Bravo Gonzalez-Blas, C., Simm, J., Moreau, Y., Aerts, J., and Aerts, S. (2019). GRNBoost2 and Arboreto: efficient and scalable inference of gene regulatory networks. Bioinformatics 35, 2159-2161. 10.1093/bioinformatics/bty916.

27. Yeh, Y.-H., Yang, Y.-C., Hsieh, M.-Y., Yeh, Y.-C., and Li, T.-K. (2013). A Negative Feedback of the HIF-la Pathway via Interferon-Stimulated Gene 15 and ISGylation. Clinical Cancer Research 19, 5927-5939. 10.1158/1078-0432.CCR-13-0018.

28. Nesterova, A.P., Klimov, E.A., Zharkova, M., Sozin, S., Sobolev, V., Shkrob, M., and Yuryev, A. eds. (2020). Disease pathways: an atlas of human disease signaling pathways (Elsevier).

29. Forsythe, J.A., Jiang, B.H., Iyer, N.V., Agani, F., Leung, S.W., Koos, R.D., and Semenza, G.L. (1996). Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. MOL. CELL. BIOL. 16, 10.

30. Vander Heiden, M.G., Cantley, L.C., and Thompson, C.B. (2009). Understanding the Warburg Effect: The Metabolic Requirements of Cell Proliferation. Science 324, 1029-1033. 10.1126/science.1160809.

31. Liberzon, A., Birger, C., Thorvaldsdottir, H., Ghandi, M., Mesirov, J.P., and Tamayo, P. (2015). The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst 1, 417-425. 10.1016/j.cels.2015.12.004.

32. Zhao, Q., Wang, J., Levichkin, I.V., Stasinopoulos, S., Ryan, M.T., and Hoogenraad, N.J. (2002). A mitochondrial specific stress response in mammalian cells. EMBO J 21, 4411— 4419. 10.1093/emboj/cdf445.

33. Martinus, R.D., Garth, G.P., Webster, T.L., Cartwright, P., Naylor, D.J., Hoj, P.B., and Hoogenraad, N.J. (1996). Selective Induction of Mitochondrial Chaperones in Response to Loss of the Mitochondrial Genome. European Journal of Biochemistry 240, 98-103. 10.1111/j .1432- 1033.1996.0098h.x.

34. Benedetti, C., Haynes, C.M., Yang, Y., Harding, H.P., and Ron, D. (2006). Ubiquitin- Like Protein 5 Positively Regulates Chaperone Gene Expression in the Mitochondrial Unfolded Protein Response. Genetics 174, 229-239. 10.1534/genetics.106.061580.

35. Holzerova, E., Danhauser, K., Haack, T.B., Kremer, L.S., Melcher, M., Ingold, I., Kobayashi, S., Terrile, C., Wolf, P., Schaper, J., et al. (2016). Human thioredoxin 2 deficiency impairs mitochondrial redox homeostasis and causes early-onset neurodegeneration. Brain 139, 346-354. 10.1093/brain/awv350. 36. Brambley, C.A., Marsee, J.D., Halper, N., and Miller, J.M. (2019). Characterization of Mitochondrial YME1L Protease Oxidative Stress-Induced Conformational State. Journal of Molecular Biology 431, 1250-1266. 10.1016/j.jmb.2019.01.039.

37. Rainbolt, T.K., Atanassova, N., Genereux, J.C., and Wiseman, R.L. (2013). Stress- regulated translational attenuation adapts mitochondrial protein import through Timl7A degradation. Cell Metab 18, 908-919. 10.1016/j.cmet.2013.11.006.

38. Treberg, J.R., Quinlan, C.L., and Brand, M.D. (2011). Evidence for Two Sites of Superoxide Production by Mitochondrial NADH-Ubiquinone Oxidoreductase (Complex I). Journal of Biological Chemistry 286, 27103-27110. 10.1074/jbc.Ml 11.252502.

39. Wikraiphat, C., Pudla, M., Baral, P., Kitthawee, S., and Utaisincharoen, P. (2014). Activation of NADPH oxidase is essential, but not sufficient, in controlling intracellular multiplication of Burkholderia pseudomallei in primary human monocytes. Pathogens Disease 71, 69-72. 10.1111/2049-632X.12122.

40. Lahiri, A., Lahiri, A., Das, P., Vani, J., Shaila, M.S., and Chakravortty, D. (2010). TLR

9 Activation in Dendritic Cells Enhances Salmonella Killing and Antigen Presentation via Involvement of the Reactive Oxygen Species. PLoS ONE 5, el3772.

10.1371/journal. pone.0013772.

41. Pillalamarri, V., Shi, W., Say, C., Yang, S., Lane, J., Guallar, E., Pankratz, N., and Arking, D.E. (2023). Whole-exome sequencing in 415,422 individuals identifies rare variants associated with mitochondrial DNA copy number. Human Genetics and Genomics Advances 4, 100147. 10.1016/j.xhgg.2022.100147.

42. Chong, M., Mohammadi-Shemirani, P., Perrot, N., Nelson, W., Morton, R., Narula, S., Lali, R., Khan, I., Khan, M., Judge, C., et al. (2022). GWAS and ExWAS of blood mitochondrial DNA copy number identifies 71 loci and highlights a potential causal role in dementia. eLife 11, e70382. 10.7554/eLife.70382.

43. Liu, C.-S., Tsai, C.-S., Kuo, C.-L., Chen, H.-W., Lii, C.-K., Ma, Y.-S., and Wei, Y.-H. (2003). Oxidative Stress-related Alteration of the Copy Number of Mitochondrial DNA in Human Leukocytes. Free Radical Research 37, 1307-1317. 10.1080/10715760310001621342.

44. Qiu, C., Hevner, K., Abetew, D., Sedensky, M., Morgan, P., Enquobahrie, D., and Williams, M. (2013). Mitochondrial DNA Copy Number and Oxidative DNA Damage in Placental Tissues from Gestational Diabetes and Control Pregnancies: A Pilot Study. Clin. Lab. 59. 10.7754/Clin.Lab.2012.120227.

45. Gao, Y., Xu, W., Dou, X., Wang, H, Zhang, X., Yang, S., Liao, H., Hu, X., and Wang, H. (2019). Mitochondrial DNA Leakage Caused by Streptococcus pneumoniae Hydrogen Peroxide Promotes Type I IFN Expression in Lung Cells. Front. Microbiol. 10, 630. 10.3389/fmicb.2019.00630.

46. Barnerias, C., Giurgea, I., Hertz-Pannier, L., Bahi-Buisson, N., Boddaert, N., Rustin, P., Rotig, A., Desguerre, I., Munnich, A., and de Lonlay, P. (2006). Respiratory chain deficiency in a female with Aicardi-Goutieres syndrome. Dev Med Child Neurol 48, 227-230. 10.1017/S001216220600048X.

47. Tumiene, B., Voisin, N., Preiksaitiene, E., Petroska, D., Grikiniene, J., Samaitiene, R., Utkus, A., Reymond, A., and Kucinskas, V. (2017). Inflammatory myopathy in a patient with Aicardi-Goutieres syndrome. Eur J Med Genet 60, 154-158. 10.1016/j.ejmg.2016.12.004.

48. Dragoni, F., Garau, J., Sproviero, D., Orcesi, S., Varesio, C., De Siervi, S., Gagliardi,

S., Cereda, C., and Pansarasa, O. (2022). Characterization of Mitochondrial Alterations in Aicardi-Goutieres Patients Mutated in RNASEH2A and RNASEH2B Genes. IJMS 23, 14482. 10.3390/ijms232214482.

49. Dhir, A., Dhir, S., Borowski, L.S., Jimenez, L., Teitell, M., Rotig, A., Crow, Y.J., Rice, G.I., Duffy, D., Tamby, C., et al. (2018). Mitochondrial double-stranded RNA triggers antiviral signalling in humans. Nature 560, 238-242. 10.1038/s41586-018-0363-0.

50. Lepelley, A., Wai, T., and Crow, Y.J. (2021). Mitochondrial Nucleic Acid as a Driver of Pathogenic Type I Interferon Induction in Mendelian Disease. Front. Immunol. 12, 729763. 10.3389/fimmu.2021.729763.

51. Zeng, X., Moore, T.A., Newstead, M.W., Deng, J.C., Lukacs, N.W., and Standiford,

T.J. (2005). IP-10 mediates selective mononuclear cell accumulation and activation in response to intrapulmonary transgenic expression and during adenovirus-induced pulmonary inflammation. J Interferon Cytokine Res 25, 103-112. 10.1089/jir.2005.25.103.

52. Yang, J., and Richmond, A. (2004). The angiostatic activity of interferon-inducible protein- 10/CXCL 10 in human melanoma depends on binding to CXCR3 but not to glycosaminoglycan. Molecular Therapy 9, 846-855. 10.1016/j.ymthe.2004.01.010.

53. Buttmann, M., Goebeler, M., Toksoy, A., Schmid, S., Graf, W., Berberich-Siebelt, F., and Rieckmann, P. (2005). Subcutaneous Interferon-P injections in patients with multiple sclerosis initiate inflammatory skin reactions by local chemokine induction. Journal of Neuroimmunology 168, 175-182. 10.1016/j.jneuroim.2005.07.011.

54. Fliesser, M., Morton, C.O., Bonin, M., Ebel, F., Hiinniger, K., Kurzai, O., Einsele, H., and Loftier, J. (2015). Hypoxia-inducible factor la modulates metabolic activity and cytokine release in anti-Aspergillus fumigatus immune responses initiated by human dendritic cells. Int J Med Microbiol 305, 865-873. 10.1016/j.ijmm.2015.08.036. 55. McNally, J.M., Zarozinski, C.C., Lin, M.-Y., Brehm, M.A., Chen, H.D., and Welsh,

R.M. (2001). Attrition of Bystander CD8 T Cells during Virus-Induced T-Cell and Interferon Responses. J Virol 75, 5965-5976. 10.1128/JVI.75.13.5965-5976.2001.

56. Fremond, M.-L., Hadchouel, A., Berteloot, L., Melki, I., Bresson, V., Barnabei, L., Jeremiah, N., Belot, A., Bondet, V., Brocq, O., et al. (2021). Overview of STING-Associated Vasculopathy with Onset in Infancy (SAVI) Among 21 Patients. The Journal of Allergy and Clinical Immunology: In Practice 9, 803-818. el l. 10.1016/j .j aip.2020.11.007.

57. Brigandi, R.A., Johnson, B., Oei, C., Westerman, M., Olbina, G., de Zoysa, J., Roger,

S.D., Sahay, M., Cross, N., McMahon, L., et al. (2016). A Novel Hypoxia-Inducible

Factor-Prolyl Hydroxylase Inhibitor (GSK1278863) for Anemia in CKD: A 28-Day, Phase 2A Randomized Trial. American Journal of Kidney Diseases 67, 861-871.

10.1053/j .ajkd.2015.11.021.

58. Sallusto, F., and Lanzavecchia, A. (1994). Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor alpha. J Exp Med 179, 1109-1118. 10.1084/jem.179.4.1109.

59. Mangeot, P.-E., Duperrier, K., Negre, D., Boson, B., Rigal, D., Cosset, F.-L., and Darlix, J.-L. (2002). High levels of transduction of human dendritic cells with optimized SIV vectors. Mol Ther 5, 283-290. 10.1006/mthe.2002.0541.

60. Manel, N., Hogstad, B., Wang, Y., Levy, D.E., Unutmaz, D., and Littman, D.R. (2010). A cryptic sensor for HIV-1 activates antiviral innate immunity in dendritic cells. Nature 467, 214-217. 10.1038/nature09337.

61. Johnson, J.S., Lucas, S.Y., Amon, L.M., Skelton, S., Nazitto, R., Carbonetti, S., Sather, D.N., Littman, D.R., and Aderem, A. (2018). Reshaping of the Dendritic Cell Chromatin Landscape and Interferon Pathways during HIV Infection. Cell Host Microbe 23, 366-381.e9. 10.1016/j.chom.2018.01.012.

62. Mangeot, P.-E., Negre, D., Dubois, B., Winter, A.J., Leissner, P., Mehtali, M., Kaiserlian, D., Cosset, F.-L., and Darlix, J.-L. (2000). Development of Minimal Lentivirus Vectors Derived from Simian Immunodeficiency Virus (SIVmac251) and Their Use for Gene Transfer into Human Dendritic Cells. J Virol 74, 8307-8315. 10.1128/JVI.74.18.8307- 8315.2000.

63. White, S.M., Avantaggiati, M.L., Nemazanyy, I., Di Poto, C., Yang, Y., Pende, M., Gibney, G.T., Ressom, H.W., Field, J., Atkins, M.B., et al. (2019). YAP/TAZ Inhibition Induces Metabolic and Signaling Rewiring Resulting in Targetable Vulnerabilities in NF2- Deficient Tumor Cells. Developmental Cell 49, 425-443. e9. 10.1016/j.devcel.2019.04.014.

64. Fali, T., Fabre-Mersseman, V., Yamamoto, T., Bayard, C., Papagno, L., Fastenackels, S., Zoorab, R., Koup, R.A., Boddaert, J., Sauce, D., et al. (2018). Elderly human hematopoietic progenitor cells express cellular senescence markers and are more susceptible to pyroptosis. JCI Insight 3, e95319. 10.1172/j ci.insight.95319.

65. Hao, Y., Hao, S., Andersen-Nissen, E., Mauck, W.M., Zheng, S., Butler, A., Lee, M.J., Wilk, A.J., Darby, C., Zager, M., et al. (2021). Integrated analysis of multimodal single-cell data. Cell 184, 3573-3587.e29. 10.1016/j.cell.2021.04.048.

66. Monaco, G., Lee, B., Xu, W., Mustafah, S., Hwang, Y.Y., Carre, C., Burdin, N., Visan, L., Ceccarelli, M., Poidinger, M., et al. (2019). RNA-Seq Signatures Normalized by mRNA Abundance Allow Absolute Deconvolution of Human Immune Cell Types. Cell Rep 26, 1627- 164O.e7. 10.1016/j.celrep.2019.01.041.

67. Chen, E.Y., Tan, C.M., Kou, Y., Duan, Q., Wang, Z., Meirelles, G.V., Clark, N.R., and Ma’ayan, A. (2013). Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics 14, 128. 10.1186/1471-2105-14-128.

68. Wang, B., Mezlini, A.M., Demir, F., Fiume, M., Tu, Z., Brudno, M., Haibe-Kains, B., and Goldenberg, A. (2014). Similarity network fusion for aggregating data types on a genomic scale. Nat Methods 11, 333-337. 10.1038/nmeth.2810.

69. Glass, K., Huttenhower, C., Quackenbush, J., and Yuan, G.-C. (2013). Passing Messages between Biological Networks to Refine Predicted Interactions. PLoS ONE 8, e64832. 10.1371/journal. pone.0064832.

70. Lichtblau, Y., Zimmermann, K., Haldemann, B., Lenze, D., Hummel, M., and Leser, U. (2016). Comparative assessment of differential network analysis methods. Brief Bioinform, bbw061. 10.1093/bib/bbw061.

71. Edgar, R., Domrachev, M., and Lash, A.E. (2002). Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res 30, 207-210. 10.1093/nar/30.1.207.

Claims

CLAIMS:

1. A method of treating a type I interferonopathy in patient in need thereof comprising administering to the subject a HIF-la stabilizing agent.

2. The method of claim 1 wherein the type I interferonopathy is Aicardi-Goutieres syndrome (AGS).

3. The method according to claim 1 or 2 wherein the HIF-la stabilizing agent is a prolyl hydroxylase inhibitor.

4. The method according to claim 1 or 2 wherein the HIF-la stabilizing agent is dimethyloxalylglycine (DMOG).

5. The method according to any one of claims 1 to 4 wherein the type I interferonopathy is resistant to a JAK inhibitor.

6. The method according to any one of claims 1 to 5 wherein the HIF-la stabilizing agent is administered in combination with at least one JAK inhibitor.

7. The method according to any one of claims 1 to 5 wherein the HIF-la stabilizing agent is administered in combination with at least one reverse transcriptase inhibitor (RTI) selected from the group consisting of Abacavir, Delavirdine, Didanosine, Efavirenz, Emtricitabine, Etravirine, Lamivudine, Nevirapine, Rilpivirine, Stavudine, Tenofovir, Zalcitabine and Zidovudine.

8. The method of claim 7 wherein the HIF-la stabilizing agent is administered in combination with a RTI tri-therapy that comprises zidovudine, abacavir, and lamivudine.

9. The method according to any one of claims 1 to 8 that omprises the step consisting of i) determining the level and/or activity of HIF-la in a sample obtained from the patient, ii) comparing the level determined at step i) with a predetermined reference value and iii) administering the HIF-la stabilizing agent when the level determined at step i) is lower than the predetermined reference value.

10. The method of claim 9 wherein the sample is a sample of monocytes.

11. The method of claim 10 wherein the sample also comprises dendritic cells.

12. The method of claim 9 wherein the level and/or activity of HIF-la is determined by RNA sequencing, and more particularly by single cell RNA sequencing.

13. The method according to any one of claims 1 to 8 that comprises the steps consisting of i) determining the level of oxidative phosphorylation in sample obtained from the patient, ii) comparing the level determined at step i) with a predetermined reference value and iii) administering the HIF-la stabilizing agent when the level determined at step i) is higher than the predetermined reference value.

14. The method of claim 13 that further comprises the step of a) determining the level of aerobic glycolysis in the sample, b) comparing the level determined at step i) with a predetermined reference value and iii) administering the HIF-la stabilizing agent when the level determined at step i) is lower than the predetermined reference value.

15. The method of claims 13 and 14 wherein the patient is administered with the HIF-la stabilizing agent when an increase in the level of oxidative phosphorylation and a decrease in the level of aerobic glycolysis are detected in the sample.