Infertilidade conjugal é uma doença do sistema reprodutivo que acomete cerca de 20% dos casais e na qual o fator masculino responde por metade desses casos. A infertilidade masculina é um fenótipo complexo que abrange diferentes fatores. Os fatores genéticos envolvidos variam desde mutações pontuais, microdeleções no cromossomo Y, até alterações cromossômicas, como a Síndrome de Klinefelter. Mesmo após avaliação clínicolaboratorial detalhada, metade dos pacientes permanece sem a identificação de um fator causal, caracterizando a infertilidade idiopática. Nesse grupo, observamos com maior frequência os pacientes com falha espermatogênica primária, que clinicamente apresentam oligozoospermia grave ou azoospermia não obstrutiva (ANO) e, no qual, preponderam fatores genéticos ainda desconhecidos. Para auxiliar na compreensão de possíveis alterações genômicas, sejam as variantes de número de cópias (CNVs) ou as regiões de perda de heterozigosidade (LOHs), envolvidas com infertilidade masculina idiopática, 16 pacientes com ANO e 6 controles foram investigados pela técnica de hibridação genômica comparativa (aCGH) utilizando a plataforma 4x180 CGH+SNP Agilent® com análise dos dados pelo software Nexus 8.0. Não foram observadas diferenças significativas tanto no número, como no tamanho das alterações genômicas em ambos os grupos. Foram descritas 18 novas alterações genômicas com efeito sobre a produção espermática, distribuídas na forma de 12 ganhos, 3 perdas e 3 LOHs. Os ganhos mais significativos para o fenótipo azoospermia não obstrutiva foram descritos em 7q36.3, 17q21.33, Xq21.1 e Yp11.2. Nessas regiões, os genes com maior impacto sobre o fenótipo foram, respectivamente, SHH, COL1A1, COX7B e LINC00279. Ganhos envolvendo a sub-banda Yq11.223 e contendo cópias dos genes DAZ1 e DAZ4 foram considerados benignos. As três perdas detectadas em 2q31.1, 3p21.1-21.31 e 15q11.2, contendo, respectivamente, os genes DLX1, CACNA2D2 e representantes da família de receptores olfatórios foram consideradas relevantes. A análise das LOHs em fenótipos complexos é escassa e desafiadora. No presente trabalho, foram descritas 3 dessas alterações, localizadas em 1p31.1, 7q21.1 e 12q21.1-21.2 e compartilhadas por mais de um indivíduo infértil. A descrição dessas alterações genômicas contribui para a compreensão de mecanismos complexos e ainda pouco estudados, que resultam em azoospermia não obstrutiva decorrente da falha espermatogênica primária. / Infertility is a disease of the reproductive system that affects about 20% of all couples, with half of the cases being related to the male factor. Male infertility is a complex phenotype associated with an interaction of different factors. The genetic factors involved may range from point mutations, microdeletions on the Y chromosome to chromosomal changes such as Klinefelter syndrome. Even after detailed clinical-laboratory evaluation, the etiology may remain unknown in approximately half of the patients, and, in such cases, the infertility can be classified as idiopathic. This group of patients more frequently present with primary spermatogenic failure, with severe oligozoospermia or non-obstructive azoospermia (NOA). Nevertheless, the underlying genetic factors are still largely unknown. In order to better understand the potential genomic changes involved with idiopathic male infertility, sixteen patients with NOA and 6 controls were investigated in this study. Copy number variants (CNVs) and regions of loss of heterozygosity (LOHs) were assessed by array comparative genomic hybridization technique (aCGH), using the Agilent® 4x180 CGH + SNP platform. Data analyses was performed using Nexus 8.0 software. No significant differences between the groups were observed in relation to either the number or the size of the genomic changes. Eighteen new genomic alterations were described that were associated with sperm production (12 gains, 3 losses and 3 LOHs). The most important gains for the nonobstructive azoospermia phenotype were observed in 7q36.3, 17q21.33, Xq21.1 and Yp11.2. In these regions, the genes related to greatest impact on the phenotype were SHH, COL1A1, COX7B and LINC00279, respectively. Gains involving the Yq11.223 sub-band and containing copies of the DAZ1 and DAZ4 genes were considered benign. All 3 losses detected in 2q31.1, 3p21.1-21.31 and 15q11.2, containing, respectively, the DLX1, CACNA2D2 genes and representatives of the olfactory receptor family were considered relevant. Analysis of LOHs in complex phenotypes such as male infertility has been infrequently reported and is challenging. In the present study, three significants LOHs were found (1p31.1, 7q21.1 and 12q21.1-21.2) and were identified in more than one infertile individual. The description of these genomic alterations contributes to a better understanding of this complex and poorly explored mechanisms that results in non-obstructive azoospermia due to primary spermatogenic failure.

A Fibrose Cística (FC) é a doença autossômica recessiva mais freqüente em caucasianos e está associada, em seu amplo espectro de apresentação clínica, a mais de 1500 mutações no gene CFTR (Cystic Fibrosis Transmembrane conductance regulator). O papel de CFTR é especialmente relevante no desenvolvimento da porção reprodutiva dos ductos mesonéfricos. Em 98% dos pacientes masculinos portadores da FC, mutações em CFTR são responsáveis pela ausência bilateral congênita dos ductos deferentes (CBAVD), associada à anomalias variáveis das vesículas seminais, ductos ejaculatórios e da porção distal dos epidídimos. A ausência uni ou bilateral congênita dos ductos deferentes (CAVD), na ausência de outros sinais clínicos de FC, é conhecida causa de infertilidade masculina, presente em 1%-2% de todos os homens inférteis, e em cerca de 10% dos azoospérmicos. A reprodução assistida utilizando a injeção intracitoplasmática de espermatozóides (ICSI) obtidos preferencialmente por aspiração microcirurgica de espermatozóides do epidídimo (MESA) permite a paternidade biológica a esses pacientes. Em função da alta morbi-mortalidade da FC e da alta freqüência de portadores assintomáticos, estimada em 1:25, é recomendável que seja realizado teste para identificação de mutações em CFTR em todos os pacientes com CAVD antes de serem submetidos à ICSI. Em populações de etnia homogênea, a mutação F508 é identificada em 90% dos pacientes com FC e em 70% a 85% dos pacientes com CAVD. No Brasil, onde diferenças étnicas refletem a heterogeneidade genética, a freqüência da mutação F508 varia entre 23% e 50% em paciente com FC indicando que outras mutações devam estar envolvidas. Este dado levou ao estudo completo do gene CFTR de 20 homens inférteis com CAVD visando a identificação das mutações mais prevalentes em nossa população. Foram identificadas mutações em 17 pacientes (85%): três DF508 representando 15% (3/20), uma G542X, uma 875+1G>A e 4 mutações ainda não descritas na literatura, a S753R, G149W identificada em dois irmãos, V580F e a 712-1G>T. A variação no trato polipirimidínico em IVS8 (alelo 5T), seja como segunda mutação ou presente em homozigose, está diretamente relacionada com a CAVD, com freqüências em população caucasiana masculina infértil variando entre 21% e 30%. No presente estudo, 15 (15/20=75%) pacientes apresentaram o alelo a variante alélica 5T sendo que em 8 pacientes essa variante alélica foi identificada em heterozigose composta com outra mutação. Anomalias renais foram identificadas em 6 pacientes, todos com CBAVD. O presente estudo pode correlacionar o fenótipo da CAVD a alterações no genótipo de CFTR em 100% dos pacientes investigados / Cystic Fibrosis (CF) is the most common autosomal recessive disorder in caucasians and is associated, in an wide variety of different clinical manifestatons. More than 1500 mutations in the CFTR gene (Cystic Fibrosis regulator Transmembrane conductance) have been described and an even growing number of mutations are being currently studied worldwide. The role of CFTR gene is especially important in reproductive tissues of the mesonephric tract sensitive to the expression of the CFTR gene. The great majority of infertile males with CF (98%) have clinical manifestations and mutations in CFTR are responsible for the congenital bilateral absence of the vas deferens (CBAVD), associated to the abnormalities of the seminal vesicles, ejaculatory ducts and/or the distal portion of the epididymis. The congenital absence, uni or bilateral, of the vas deferens (CAVD), in the absence of other clinical signals of CF is a known cause of male infertility present in 1%-2% of all men investigated and in about 10% of men with obstructive azoospermia. Serious considerations should be drawn about the lack of proper diagnosis of infetile males with CFTR that seek reproductive clinics for assisted reproductive techniques (ARTs), as well as the lack of proper consideratins of the existance of this disease as a potential cause of male infetility among male are takers, like urologistas, andrologistas and gynecologists that rush for the misuse of ARTs. The introduction of Intracytoplasmic Sperm Injection (ICSI), has given new reproductive potetntial for these couples, but again as in the majority of cases it is obstructive azoospermia, couples should be advised about proper microsurgical sperm retrieval, preferentialy microsurgical epydidymal sperm aspiration (MESA). As a consequence of the potential high mortality rate of the CF descendents and the high frequency of carriers, estimated in 1:25,it is highly recommended that tests for correct identification of mutations in CFTR gene are carried out for all patients with CAVD before considered being submitted to ICSI. In populations of homogeneous ethnic origin, the mutation F508 is identified in 90% of the patients with CF and between 70% and 85% of the patients with CAVD. In Brazil, where ethnic differences reflect the genetic heterogeneity, the frequency of the mutation in F508 varies between 23% and 50%, indicating that other mutations must have a role. Our data looked carefully in the CFTR gene of 20 infertile men with CAVD aiming at the identification of the most prevalent mutations in our population. Mutations had been identified in 17 patients (85%): three DF508 representing 15% (3/20), one G542X, one 875+1G>A and 4 mutations not yet described in literature, S753R, G149W identified in two brothers, V580F and 712-1G>T. In the literature the allelic variant in IVS8 (allele 5T), either as a second mutation or in homozygosis, is directly related with the CAVD, with reported frequencies in the infertile caucasian male population varying between 21% and 30%. In the present study, 15 (15/20=75%) patients presented the CFTR mutation in the IVS8/5T: eight of them in heterozygosis composed with another mutation. Regarding genitourinary tract malformations, kidney anomalies were identified in 6 patients, all with CBAVD. In the present study we could correlationate the phenotype of the CAVD with the genotype alterations of CFTR gene in 100% of the investigated patients

INTRODUÇÃO: O cromossomo Y contém genes primordiais para o desenvolvimento testicular e a espermatogênese. Sua conformação repetitiva predispõe à ocorrência de deleções e duplicações, que tem impacto clínico. As microdeleções nas regiões AZF afetam loci responsáveis pela espermatogênese e são uma das causas mais frequentes de azoospermia e oligozoospermia. Seu diagnóstico tem valor preditivo para o sucesso na recuperação cirúrgica de espermatozoides testiculares em homens azoospérmicos. A técnica utilizada para detectá-las é a reação de polimerização em cadeia (PCR), porém os protocolos divergem entre si, havendo muita variabilidade na incidência destas deleções. À vista disso, sugerimos utilizar a técnica de Amplificação de Múltiplas Sondas dependente de Ligação (MLPA). OBJETIVO: Comparar os resultados obtidos com as técnicas de PCR e MLPA na detecção de microdeleções do cromossomo Y em homens inférteis. RESULTADOS: Analisamos 43 pacientes inférteis (azoospérmicos e oligozoospérmicos) e 40 homens férteis (controle) pelas técnicas de PCR e MLPA. Encontramos 7 deleções por PCR (16,2%) e 9 por MLPA (21%), além de 5 duplicações e um mosaico. DISCUSSÂO: Além das deleções, as duplicações também podem gerar instabilidades nos genes do cromossomo, podendo levar a infertilidade. CONCLUSÕES: Os resultados obtidos por ambas as técnicas revela que a MLPA é uma técnica mais sensível que a técnica de PCR para detectar microdeleções do cromossomo Y / INTRODUCTION: The Y chromosome contains several genes responsible for testicular development and spermatogenesis. Its repetitive conformation predisposes this chromosome to deletions and duplications that have clinical impact. Microdeletions in the AZF regions affect loci responsible for spermatogenesis and are one of the most frequent causes of azoospermia and oligozoospermia. This diagnosis may have a predictive value for success in the surgical recovery of testicular spermatozoa in azoospermic men. The gold standard method for this detection is polymerase chain reaction (PCR), but protocols diverge among them, generating a great variability in the incidence of these deletions. PURPOSE: We evaluated another molecular diagnostic method, Multiplex Ligand Probe Dependent Amplification (MLPA), which generates more genomic data (such as duplications and rearrangements) in a single reaction, leading to a better understanding of these patients phenotype. OBJECTIVE: To compare the results obtained with PCR and MLPA techniques in the detection of Y chromosome microdeletions in infertile men. RESULTS: We analyzed 43 infertile patients (azoospermic and oligozoospermic) and 40 fertile men (control) by PCR and MLPA techniques. We found 7 deletions by PCR (16.2%) and 9 by MLPA (21%), in addition to 5 duplications and one mosaic. DISCUSSION: Besides deletions, duplications can also generate instability in the chromosome genes, which may lead to infertility, and it is important being capable to diagnose these alterations with a faster and more effectively method. CONCLUSIONS: The results obtained by both techniques reveal that MLPA is more sensitive than PCR to detect microdeletions of the Y chromosome

INTRODUÇÃO: O cromossomo Y contém genes primordiais para o desenvolvimento testicular e a espermatogênese. Sua conformação repetitiva predispõe à ocorrência de deleções e duplicações, que tem impacto clínico. As microdeleções nas regiões AZF afetam loci responsáveis pela espermatogênese e são uma das causas mais frequentes de azoospermia e oligozoospermia. Seu diagnóstico tem valor preditivo para o sucesso na recuperação cirúrgica de espermatozoides testiculares em homens azoospérmicos. A técnica utilizada para detectá-las é a reação de polimerização em cadeia (PCR), porém os protocolos divergem entre si, havendo muita variabilidade na incidência destas deleções. À vista disso, sugerimos utilizar a técnica de Amplificação de Múltiplas Sondas dependente de Ligação (MLPA). OBJETIVO: Comparar os resultados obtidos com as técnicas de PCR e MLPA na detecção de microdeleções do cromossomo Y em homens inférteis. RESULTADOS: Analisamos 43 pacientes inférteis (azoospérmicos e oligozoospérmicos) e 40 homens férteis (controle) pelas técnicas de PCR e MLPA. Encontramos 7 deleções por PCR (16,2%) e 9 por MLPA (21%), além de 5 duplicações e um mosaico. DISCUSSÂO: Além das deleções, as duplicações também podem gerar instabilidades nos genes do cromossomo, podendo levar a infertilidade. CONCLUSÕES: Os resultados obtidos por ambas as técnicas revela que a MLPA é uma técnica mais sensível que a técnica de PCR para detectar microdeleções do cromossomo Y / INTRODUCTION: The Y chromosome contains several genes responsible for testicular development and spermatogenesis. Its repetitive conformation predisposes this chromosome to deletions and duplications that have clinical impact. Microdeletions in the AZF regions affect loci responsible for spermatogenesis and are one of the most frequent causes of azoospermia and oligozoospermia. This diagnosis may have a predictive value for success in the surgical recovery of testicular spermatozoa in azoospermic men. The gold standard method for this detection is polymerase chain reaction (PCR), but protocols diverge among them, generating a great variability in the incidence of these deletions. PURPOSE: We evaluated another molecular diagnostic method, Multiplex Ligand Probe Dependent Amplification (MLPA), which generates more genomic data (such as duplications and rearrangements) in a single reaction, leading to a better understanding of these patients phenotype. OBJECTIVE: To compare the results obtained with PCR and MLPA techniques in the detection of Y chromosome microdeletions in infertile men. RESULTS: We analyzed 43 infertile patients (azoospermic and oligozoospermic) and 40 fertile men (control) by PCR and MLPA techniques. We found 7 deletions by PCR (16.2%) and 9 by MLPA (21%), in addition to 5 duplications and one mosaic. DISCUSSION: Besides deletions, duplications can also generate instability in the chromosome genes, which may lead to infertility, and it is important being capable to diagnose these alterations with a faster and more effectively method. CONCLUSIONS: The results obtained by both techniques reveal that MLPA is more sensitive than PCR to detect microdeletions of the Y chromosome

A Fibrose Cística (FC) é a doença autossômica recessiva mais freqüente em caucasianos e está associada, em seu amplo espectro de apresentação clínica, a mais de 1500 mutações no gene CFTR (Cystic Fibrosis Transmembrane conductance regulator). O papel de CFTR é especialmente relevante no desenvolvimento da porção reprodutiva dos ductos mesonéfricos. Em 98% dos pacientes masculinos portadores da FC, mutações em CFTR são responsáveis pela ausência bilateral congênita dos ductos deferentes (CBAVD), associada à anomalias variáveis das vesículas seminais, ductos ejaculatórios e da porção distal dos epidídimos. A ausência uni ou bilateral congênita dos ductos deferentes (CAVD), na ausência de outros sinais clínicos de FC, é conhecida causa de infertilidade masculina, presente em 1%-2% de todos os homens inférteis, e em cerca de 10% dos azoospérmicos. A reprodução assistida utilizando a injeção intracitoplasmática de espermatozóides (ICSI) obtidos preferencialmente por aspiração microcirurgica de espermatozóides do epidídimo (MESA) permite a paternidade biológica a esses pacientes. Em função da alta morbi-mortalidade da FC e da alta freqüência de portadores assintomáticos, estimada em 1:25, é recomendável que seja realizado teste para identificação de mutações em CFTR em todos os pacientes com CAVD antes de serem submetidos à ICSI. Em populações de etnia homogênea, a mutação F508 é identificada em 90% dos pacientes com FC e em 70% a 85% dos pacientes com CAVD. No Brasil, onde diferenças étnicas refletem a heterogeneidade genética, a freqüência da mutação F508 varia entre 23% e 50% em paciente com FC indicando que outras mutações devam estar envolvidas. Este dado levou ao estudo completo do gene CFTR de 20 homens inférteis com CAVD visando a identificação das mutações mais prevalentes em nossa população. Foram identificadas mutações em 17 pacientes (85%): três DF508 representando 15% (3/20), uma G542X, uma 875+1G>A e 4 mutações ainda não descritas na literatura, a S753R, G149W identificada em dois irmãos, V580F e a 712-1G>T. A variação no trato polipirimidínico em IVS8 (alelo 5T), seja como segunda mutação ou presente em homozigose, está diretamente relacionada com a CAVD, com freqüências em população caucasiana masculina infértil variando entre 21% e 30%. No presente estudo, 15 (15/20=75%) pacientes apresentaram o alelo a variante alélica 5T sendo que em 8 pacientes essa variante alélica foi identificada em heterozigose composta com outra mutação. Anomalias renais foram identificadas em 6 pacientes, todos com CBAVD. O presente estudo pode correlacionar o fenótipo da CAVD a alterações no genótipo de CFTR em 100% dos pacientes investigados / Cystic Fibrosis (CF) is the most common autosomal recessive disorder in caucasians and is associated, in an wide variety of different clinical manifestatons. More than 1500 mutations in the CFTR gene (Cystic Fibrosis regulator Transmembrane conductance) have been described and an even growing number of mutations are being currently studied worldwide. The role of CFTR gene is especially important in reproductive tissues of the mesonephric tract sensitive to the expression of the CFTR gene. The great majority of infertile males with CF (98%) have clinical manifestations and mutations in CFTR are responsible for the congenital bilateral absence of the vas deferens (CBAVD), associated to the abnormalities of the seminal vesicles, ejaculatory ducts and/or the distal portion of the epididymis. The congenital absence, uni or bilateral, of the vas deferens (CAVD), in the absence of other clinical signals of CF is a known cause of male infertility present in 1%-2% of all men investigated and in about 10% of men with obstructive azoospermia. Serious considerations should be drawn about the lack of proper diagnosis of infetile males with CFTR that seek reproductive clinics for assisted reproductive techniques (ARTs), as well as the lack of proper consideratins of the existance of this disease as a potential cause of male infetility among male are takers, like urologistas, andrologistas and gynecologists that rush for the misuse of ARTs. The introduction of Intracytoplasmic Sperm Injection (ICSI), has given new reproductive potetntial for these couples, but again as in the majority of cases it is obstructive azoospermia, couples should be advised about proper microsurgical sperm retrieval, preferentialy microsurgical epydidymal sperm aspiration (MESA). As a consequence of the potential high mortality rate of the CF descendents and the high frequency of carriers, estimated in 1:25,it is highly recommended that tests for correct identification of mutations in CFTR gene are carried out for all patients with CAVD before considered being submitted to ICSI. In populations of homogeneous ethnic origin, the mutation F508 is identified in 90% of the patients with CF and between 70% and 85% of the patients with CAVD. In Brazil, where ethnic differences reflect the genetic heterogeneity, the frequency of the mutation in F508 varies between 23% and 50%, indicating that other mutations must have a role. Our data looked carefully in the CFTR gene of 20 infertile men with CAVD aiming at the identification of the most prevalent mutations in our population. Mutations had been identified in 17 patients (85%): three DF508 representing 15% (3/20), one G542X, one 875+1G>A and 4 mutations not yet described in literature, S753R, G149W identified in two brothers, V580F and 712-1G>T. In the literature the allelic variant in IVS8 (allele 5T), either as a second mutation or in homozygosis, is directly related with the CAVD, with reported frequencies in the infertile caucasian male population varying between 21% and 30%. In the present study, 15 (15/20=75%) patients presented the CFTR mutation in the IVS8/5T: eight of them in heterozygosis composed with another mutation. Regarding genitourinary tract malformations, kidney anomalies were identified in 6 patients, all with CBAVD. In the present study we could correlationate the phenotype of the CAVD with the genotype alterations of CFTR gene in 100% of the investigated patients

L’infertilité est considérée actuellement par l’organisation mondiale de la santé (OMS) comme une préoccupation majeure de santé affectant plus de 50 millions de couples dans le monde. Dans les pays occidentaux, la majorité des couples infertiles ont recours aux techniques d’assistance médicale à la procréation (AMP) pour obtenir une grossesse. Malgré le succès de ces techniques, près de la moitié des couples qui ont recours à l’AMP sortent du parcours de soin sans enfant. Une partie de ces échecs est expliquée par l’altération de la gamétogenèse. Chez l’homme, la spermatogenèse fait interagir des centaines de gènes spécifiquement exprimés dans le testicule. L’abondance de ces gènes suggère que les troubles de la spermatogenèse présentent une forte composante génétique. Récemment, les avancées techniques ont favorisé l’identification de gènes responsables de ces anomalies mais la grande majorité des cas d’infertilité masculine reste classée comme idiopathique. L’objectif de la thèse est d’identifier de nouvelles causes génétiques responsables d’infertilité masculine et d’élucider les mécanismes physiopathologiques associés à ces anomalies.Au cours de ma thèse j’ai participé avec l’équipe GETI (génétique, épigénétique et thérapies de l’infertilité) à l’exploration génétique et moléculaire de deux phénotypes distincts d’anomalies spermatiques liés à des défauts post-méiotiques de la spermatogenèse : une forme rare d’azoospermie non obstructive (ANO) et le phénotype d’anomalies morphologiques multiples du flagelle spermatique (AMMF). Enfin j’ai joué un rôle important dans la création et l’analyse de modèles murins pour caractériser la pathogénie de ces anomalies.L’analyse génétique de deux frères infertiles nés de parents consanguins et présentant une ANO idiopathique associée à un arrêt post-méiotique de la spermatogenèse nous a permis d’identifier un variant homozygote délétère dans le gène SPINK2 qui code pour un inhibiteur de sérine-protéases. L’étude des souris KO pour ce gène nous a permis d’observer que les souris mâles adultes sont infertiles et miment parfaitement les phénotypes spermatique et testiculaire observés chez nos patients. Nous avons montré que la protéine codée par ce gène est exprimée dans l’acrosome à partir du stade de spermatide ronde. En l’absence de Spink2, l’activité protéolytique non-neutralisée des protéases cibles qui transitent par le Golgi cause sa fragmentation et bloque la spermiogénèse au stade de spermatide ronde. Nous avons également pu observer que les spermatozoïdes provenant de patients et de souris hétérozygotes présentent un taux élevé d’anomalies morphologiques et une baisse de la mobilité progressive conduisant à une hypofertilité à expressivité variable. Ces résultats montrent pour la première fois que l’oligo-tératozoospermie et l’azoospermie peuvent constituer un continuum pathologique dû à une même pathogénie.Nous avons également réalisé le séquençage exomique complet d’une cohorte de 78 individus AMMF non apparentés et avons identifié chez 49 sujets des mutations bi-alléliques délétères dans 11 gènes candidats dont DNAH1, CFAP43, CFAP44, WDR66 et FSIP2, soit un rendement diagnostique de 63%. Ces résultats confirment l’hétérogénéité génétique du phénotype MMAF et l’efficacité diagnostique du séquençage haut débit dans son exploration. Nous avons également validé l’implication de certains gènes candidats (n=4) dans ce phénotype chez le modèle murin knock-out créé par la nouvelle technologie d’édition du génome, CRISPR/Cas9.Dans son ensemble, ce travail montre l’intérêt et l’efficacité de la combinaison du séquençage exomique et de la technique de CRISPR/Cas9 pour étudier les troubles de la spermatogenèse et l’infertilité masculine. / Infertility is currently considered by the World Health Organization (WHO) as a major health concern affecting more than 50 million couples worldwide. In western countries, the majority of infertile couples seek assisted reproductive technologies (ART) to achieve a pregnancy. Despite the success of these techniques, almost half of these couples fail to obtain a child. Part of these failures are explained by the alteration of gametogenesis. In humans, spermatogenesis involves hundreds of genes specifically expressed in the testis. The abundance of these genes suggests that spermatogenic defects are associated with a strong genetic component. Recently, technical advances have led to the identification of numerous causative genes, but the vast majority of male infertility cases remain idiopathic. The aim of the present thesis is to identify new genetic causes responsible for male infertility and to elucidate the physiopathological mechanisms associated with these anomalies.During my thesis, I participated with the team GETI (genetics, epigenetics and therapies of infertility) in the genetic exploration of two phenotypes of male infertility related to post-meiotic defects of spermatogenesis: a rare form of non-obstructive azoospermia and the phenotype of multiple morphological abnormalities of the sperm flagella (MMAF). I have also played a key role in creation and analysis of transgenic mice to better characterize the pathogeny of the identified genetic causes in Human.Genetic analyses performed on two infertile brothers born form consanguineous parents and presenting an-idiopathic non-obstructive azoospermia associated with a post-meiotic arrest of spermatogenesis allowed us to identify a homozygous variant in the SPINK2 gene that encodes a serine-protease inhibitor. Phenotypic analysis of Spink2-/- adult male mice showed that they are infertile and perfectly mimic the sperm and testicular phenotypes observed in our patients. We showed that Spink2 protein is expressed from the round spermatid stage and localized in the acrosome, a lysosomal-like vesicle rich in proteases that play a key role during fertilization. When Spink2 is absent, the deregulated proteolytic activity of the targeted proteases such as acrosin leads to the fragmentation of the Golgi apparatus and arrest of spermiogenesis at the round spermatid stage. We also showed that sperm from heterozygous human and mice present a high level of morphological abnormalities and a decrease of progressive motility leading to a variable subfertility. These results showed for the first time that oligo-teratozoospermia and azoospermia could present a pathological continuum due to the same pathogeny.We also performed exome sequencing in a cohort of 78 non related MMAF subjects and identified in 49 cases deleterious bi-allelic mutations in a total of 11 candidate genes including DNAH1, CFAP43, CFAP44, WDR66 and FSIP2 giving a genetic diagnosis yield of 63%. These results confirm the genetic heterogeneity of MMAF and the efficiency of high throughput sequencing in genetic exploration of this phenotype. We also demonstrated the pathogenic implication of certain candidate genes (n=4) using knock-out mice created by the new technology of genome editing, CRISPR/Cas9.Overall, this work demonstrates the interest and effectiveness of combining exome sequencing and CRISPR/Cas9 system to study spermatogenesis disorders and male infertility.

Spermatogenesis is a complex and coordinated process of formation of sperms from the precursor spermatogonia, occurring inside the unique environment existing in the seminiferous epithelium. This process of development, characterized by concomitant changes in the cellular morphology, metabolism and differential gene expression, can be divided into 3 distinct phases: i) proliferation of the spermatogonia through mitosis; ii) meiosis or reduction division, which commences with transformation of the type B spermatogonia into primary spermatocytes and their subsequent entry into the meiotic prophase I. These primary spermatocytes, divide to form secondary spermatocytes, and then divide again to form haploid round spermatids; (iii) spermiogenesis or differentiation and maturation of the round spermatids without further division to form the unique spermatozoa (Kerr and De Kretser, 2006, Clermont, 1966, Heller and Clermont, 1964). This complex process of division and differentiation is regulated at three distinct levels: i) The extrinsic level where the gonadotropins and testosterone regulate gene expression in the germ cells sustaining their survival and differentiation (French, 2012); ii) The interactive regulation that involves interactions between the somatic cells such as the Sertoli cells and the germ cells; iii) The intrinsic gene expression associated with each step of development of the germ cells (Eddy, 2002) wherein each stage of differentiation is accompanied by precise stage-specific differential gene expression. (Kleene, 1996, Kierszenbaum et al., 2003, Sassone-Corsi, 2002, Kleene, 2001, Sassone-Corsi, 1997). Any alterations in this gene expression pattern leads to disruption and/or arrest of spermatogenesis at various stages, causing male infertility (Zorrilla and Yatsenko, 2013, Krausz et al., 2015). Many studies have been focused on investigating the underlying molecular mechanisms governing the process of germ cell development such as self-renewal, meiotic recombination and differentiation (Hecht, 1998, Grootegoed et al., 2000, Robles et al., 2017). Analysis of differential gene expression in isolated and purified populations of different germ cells have been very useful in the understanding of the genetic regulation of human spermatogenesis by providing information about the cell type-specific gene expression and regulation. (Meistrich et al., 1973, Bellvé, 1993, Meistrich et al., 1981, Chalmel et al., 2007). However, these methods are limited by the large amounts of tissue required, which is difficult to obtain in the case of humans (Schultz et al., 2003). Large-scale gene expression studies and the “omics revolution” have also helped in identifying some of the regulators of spermatogenesis (Carrell et al., 2016). In spite of advances in the current understanding of the regulation of spermatogenesis, the exact molecular mechanisms of how the genetic and epigenetic alterations affect human spermatogenesis are still unclear (Neto et al., 2016). The present study is an attempt to investigate the human testicular gene expression pattern in the germ cells of patients with various types of azoospermia, and correlate the same to infertility. Comparative analysis of the testicular transcriptomes of infertile individuals (with arrested spermatogenesis) with the control, fertile individuals (with normal spermatogenesis) would allow identification of the cell type-specific altered genes. Analysis of these genes would provide an insight into the genetic regulation of the progress of spermatogenesis as well as allow identification of the crucial genes responsible for the arrest. The first step in this study was to ascertain the exact status of spermatogenesis in patients’ testes. Forty-four azoospermic patients were classified clinically into two major groups – obstructive (OA) and non-obstructive (NOA) azoospermia and further classified using flow cytometric analysis of the germ cells. The patients with OA exhibited presence of the diploid, tetraploid and haploid cells indicating complete spermatogenesis (Group I: DTH). The patients with NOA showed incomplete spermatogenesis with arrest at either the meiotic stage showing the presence of diploid and tetraploid cells, but not the haploid cells (Group II: DT), or at the pre-meiotic stage with only diploid cells (Group III: D). This was further verified by RT-PCR analyses for specific markers for different testicular cells. The Group I patients showed expression of markers specific for the Leydig cell (LHCGR, HSD3B2 and HSD17B3), the Sertoli cell (FSHR, KITL), spermatogonia (KIT), tetraploid cells (CCNA1, LDHC) and haploid cells (PRM1). The Group II patients showed expression of CCNA1 and LDHC, but not of PRM1. The Group III patients did not express any of the haploid or tetraploid specific markers. The germ cell pattern was further confirmed by histology where a clear difference was seen across the groups in accordance with their flow cytometric profiles. Subsequent to grouping of the patient samples based on their testicular germ-cell pattern, microarray analysis was carried out with representative samples from each group leading to identification of diploid-/tetraploid-/haploid-specific (D/T/H) genes. The enrichment, probable pathways and network interactions of these identified genes were analyzed and found to be in agreement with the classification made in this study. Further, based on their network interactions, the genes that were under multiple modes of regulation and the transcription factors that regulated multiple pathways were selected for further analysis. In absence of an in-vitro system to study germ cell differentiation, the importance of the selected genes in the progression of human spermatogenesis was analyzed from the data extrapolated from information available in the literature about expression of each gene in the human testes (wherever available), known function of the genes in various somatic cells, function in developing and adult testes of model organisms and the data from the knockout or transgenic animals where disruption of the gene/s resulted in an arrest or disruption of spermatogenesis. Expression of all the putative crucial genes was analyzed in all the patients including the control patients at the transcript level and three selected genes (one from each group- D, T and H) were further validated at the protein level using immunohistochemistry. All the genes showed a similar pattern of amplification in the different groups of patients to the pattern observed from the microarray. The diploid-specific genes (selected based on the available literature) were mainly the inhibitors or regulators of the cell cycle (CDKN1A, GADD45A, FOXM1) (Xiong et al., 1991, Jin et al., 2002, Laoukili et al., 2005) and regulators of cellular proliferation (KLFs, FOS, SRF, ATFs, SMADs) (Garrett-Sinha et al., 1996, Persengiev and Green, 2003, Angel and Karin, 1991, Ten Dijke et al., 2002). Six diploid-specific genes that were potential regulators of spermatogenesis were identified to be probable causes for the arrest of spermatogenesis at the pre-meiotic stage. CDKN1A showed elevated expression at the transcript level which suggested that DNA-damage induced proliferation check (mediated through CDKN1A) in the diploid cells probably prevented these cells from entering meiosis. This was further verified at the protein level by immuno-staining for CDKN1A. Further, GADD45A, KLF4, FOS, MCL1 and SERPINE1 were identified as genes crucial for transition from the diploid to the tetraploid stage and their aberrant expression correlated to the arrest of spermatogenesis in the Group D patients. Six tetraploid-specific genes and four haploid-specific genes were identified to be potential regulators of the tetraploid-haploid transition and responsible for the meiotic arrest. Over expression of the pro-inflammatory genes such as CCL3, IL1B and IL8 (Guazzone et al., 2009) was seen in the testis of the arrested patients which suggested that there was a potential alteration of the normal testicular micro-environment. Expression of EGR2 (a spermatogonial-maintenance gene controlling mitosis (Joseph et al., 1988)) was seen in the nucleus of spermatocytes in group DT patients which indicated its role in the meiotic arrest. To understand the role of the haploid-specific genes in the context of spermatocyte differentiation, only those genes whose expression are reported in the spermatocytes and persist till the spermatid stage were selected. Lack of expression of CST8 was identified to be potentially responsible for loss of germ cell integrity, and the loss of GGN expression in the Group DT patients seemed to be a significant contributor to the genotoxic stress in these patients. In the arrested patients RFX2 (reported to be master regulator of spermiogenesis (Wu et al., 2016)) was seen to be down regulated at the transcript level which indicated its role in the control of meiosis. This was further confirmed by IHC, where expression of RFX2 was seen to be present in the tetraploid cells of the Group DTH patients while no expression was seen in the tetraploid cells of Group DT patients. Thus, this study identified a role for RFX2 in the regulation of meiosis in humans, similar to the findings reported in rats (Horvath et al., 2009). The study also identified autophagy as a mechanism for the clearance of the arrested cells in NOA patients. IHC data using αLC3B showed that autophagy was up regulated in the arrested patients as compared to the Group DTH patients suggesting its role in cell survival and recycling of nutrients. Further, in-situ TUNEL labeling of tissue sections from the different groups (DTH, DT and D) revealed that there were no difference in the status of apoptosis in the azoospermic patients. The latter observation further corroborated with the elevated expressions of CDKN1A, GADD45A, MCL1, TNFAIP3 (reported to ensure cell survival by negatively modulating apoptosis) as seen in the NOA patients. In conclusion, this study identifies several genes that control the progression of spermatogenesis, including the genes whose alterations contribute towards an arrest in spermatogenesis, especially in azoospermia. These identified genes may be used as novel markers in the diagnosis of male infertility. The study opens up the possibility of using the identified genes as future therapeutic targets using small molecular regulators for treatment of infertility as well as targets for male contraception. The study also identifies a novel role for autophagy in patients with NOA which opens up new avenues for further investigation. Thus, this study is the beginning of understanding the complex events that regulate spermatogenesis.