The mammalian-specific Tex19.1 gene plays an essential role in spermatogenesis and placenta supported development_Tarabay_2013

ORIGINAL ARTICLE Reproductive biology

The mammalian-specific Tex19.1gene plays an essential role inspermatogenesis and placenta-supported developmentYara Tarabay1,†, Emmanuelle Kieffer1,2,†, Marius Teletin1,2,Catherine Celebi1,2, Aafke Van Montfoort3, Natasha Zamudio4,Mayada Achour1, Rosy El Ramy1, Emese Gazdag1,5, Philippe Tropel1,Manuel Mark1,2, Deborah Bourc’his4, and Stephane Viville1,6,*1Institut de Genetique et de Biologie Moleculaire et Cellulaire (IGBMC), Institut National de Sante et de Recherche Medicale (INSERM) U964/Centre National de Recherche Scientifique (CNRS) UMR 1704/Universite de Strasbourg, 67404 Illkirch, France 2Service de Biologie de laReproduction, Centre Hospitalier Universitaire, 67000 Strasbourg, France 3Department of Obstetrics and Gynaecology, GROW – School forOncology and Developmental Biology, Maastricht University Medical Centre, Maastricht, The Netherlands 4Unite de genetique et biologie dudeveloppement, UMR 3215-Inserm U934, Institut Curie, 26, rue d’Ulm, 75005 Paris, France 5Department of Molecular Biology, Faculty ofScience, Nijmegen Centre of Molecular Life Sciences, Radboud University Nijmegen, Nijmegen, The Netherlands 6Centre HospitalierUniversitaire, F-67000 Strasbourg, France

*Correspondence address. E-mail: [email protected]

Submitted on November 8, 2012; resubmitted on March 22, 2013; accepted on March 27, 2013

study question: What is the consequence of Tex19.1 gene deletion in mice?

summary answer: The Tex19.1 gene is important in spermatogenesis and placenta-supported development.

what is known already: Tex19.1 is expressed in embryonic stem (ES) cells, primordial germ cells (PGCs), placenta and adult gonads.Its invalidation in mice leads to a variable impairment in spermatogenesis and reduction of perinatal survival.

study design, size, duration: We generated knock-out mice and ES cells and compared them with wild-type counterparts. Thephenotype of the Tex19.1 knock-out mouse line was investigated during embryogenesis, fetal development and placentation as well as duringadulthood.

participants/materials, setting, methods: We used a mouse model system to generate a mutant mouse line in which theTex19.1 gene was deleted in the germline. We performed an extensive analysis of Tex19.1-deficient ES cells and assessed their in vivo differen-tiation potential by generating chimeric mice after injection of the ES cells into wild-type blastocysts. For mutant animals, a morphological char-acterization was performed for testes and ovaries and placenta. Finally, we characterized semen parameters of mutant animals and performedreal-time RT–PCR for expression levels of retrotransposons in mutant testes and ES cells.

main results and the role of chance: While Tex19.1 is not essential in ES cells, our study points out that it is important forspermatogenesis and for placenta-supported development. Furthermore, we observed an overexpression of the class II LTR-retrotransposonMMERVK10C in Tex19.1-deficient ES cells and testes.

limitations, reasons for caution: The Tex19.1 knock-out phenotype is variable with testis morphology ranging from severelyaltered (in sterile males) to almost indistinguishable compared with the control counterparts (in fertile males). This variability in the testis pheno-type subsequently hampered the molecular analysis of mutant testes. Furthermore, these results were obtained in the mouse, which has a secondisoform (i.e. Tex19.2), while other mammals possess only one Tex19 (e.g. in humans).

wider implications of the findings: The fact that one gene has a role in both placentation and spermatogenesis might open newways of studying human pathologies that might link male fertility impairment and placenta-related pregnancy disorders.

† Equal first authors.

& The Author 2013. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.For Permissions, please email: [email protected]

Human Reproduction, Vol.0, No.0 pp. 1–14, 2013

doi:10.1093/humrep/det129

Hum. Reprod. Advance Access published May 14, 2013 by guest on M

ay 15, 2013http://hum

rep.oxfordjournals.org/D

ownloaded from

http://humrep.oxfordjournals.org/

study funding/competing interest(s): This work was supported by the Centre National de la Recherche Scientifique(CNRS), the Institut National de la Sante et de la Recherche Medicale (INSERM) (Grant Avenir), the Ministere de l’Education Nationale, del’Enseignement Superieur et de la Recherche, the Universite de Strasbourg, the Association Francaise contre les Myopathies (AFM) and the Fon-dation pour la Recherche Medicale (FRM) and Hopitaux Universitaires de Strasbourg.The authors have nothing to disclose.

Key words: spermatogenesis / meiosis / stem cells / Tex19 / transposons

IntroductionIn mammals, a tiny cell population, the primordial germ cells (PGCs), isplaying a crucial role by ensuring the production of haploid gametesthat will generate a new individual after fertilization. During their specifi-cation (Lawson and Hage, 1994; Sasaki and Matsui, 2008) and further de-velopment, these cells undergo a repression of the somatic programmeavoiding their differentiation towards somatic lineages (Ohinata et al.,2005; Surani and Hajkova, 2010) and undergo a two-step reprogram-ming process leading to the acquisition of an epigenetic status compatiblewith future embryonic development (Hajkova et al., 2008; Sasaki andMatsui, 2008; Hayashi and Surani, 2009; Surani and Hajkova, 2010). Atthe end of this process, PGCs come out with an epigenome that ishighly similar to the one observed in pluripotent embryonic stem cells(ESCs) (Surani and Hajkova, 2010). The second reprogramming stepnotably leads to the acquisition of genomic imprinting, through the de-position of sex-specific DNA methylation patterns, which will lead tothe differential expression of paternally and maternally inherited allelesof a handset of genes after fertilization (Barlow, 2011; Kaneda, 2011).Other important targets of the germline-specific programming ofDNA methylation patterns are transposable elements (TEs). Themouse genome hosts both long terminal repeats (LTRs) derived ele-ments, such as intracisternal A particles (IAPs) and mouse endogenousretroviruses (MERVs), as well as non-LTR elements, like long inter-spersed nucleotide elements (LINEs) and short interspersed nucleotideelements (SINEs). De novo DNA methylation of these elements has beenrecently linked to a pathway involving germ line-specific small RNAs,called piwi-interacting RNAs (piRNAs) (Aravin et al., 2007; Ollingeret al., 2010; Zamudio and Bourc’his, 2010; Pillai and Chuma, 2012).Mutant mouse models for the genes composing the Piwi pathway invari-ably show a massive transposon reactivation in the male germline, with aprecocious interruption of the spermatogenetic process and a completesterility phenotype (Pillai and Chuma, 2012). The control of transposonactivity, therefore, appears as a major determinant of fertility and repro-ductive success particularly in males (Bourc’his and Bestor, 2004; Ollin-ger et al., 2010; Pillai and Chuma, 2012).

It was recently shown that ESCs may be derived from early PGCs thusestablishing a link between pluripotency and PGC ontogeny (Chu et al.,2011). Considering the therapeutic potential of pluripotent stem cells(Wu and Hochedlinger, 2011) and the fact that fertility is a majorhealth concern (World Health Organization, 1999), it is fundamentallyimportant to understand the molecular mechanisms that can potentiallybe involved in both of them. Tex19.1 (testis expressed gene 19) is one ofthe rare genes specifically expressed by all types of pluripotent stem cellsand in spermatogonia. Its study could be important in revealing newaspects of pluripotency and/or fertility.

Tex19 was initially cloned as a spermatogonia specifically expressedgene (Wang et al., 2001). Our initial characterization showed that

Tex19 is restricted to mammals, is present as a unique gene in humansand has been duplicated in mouse and rat giving rise to the paralogsTex19.1 and Tex19.2. By multiple sequence alignment of Tex19 proteins,two highly conserved domains named MCP and VPTEL domains werecharacterized. However, none of them shares homologies with knownproteins, therefore preventing functional prediction (Kuntz et al.,2008). Tex19.1 is expressed throughout the pluripotent cycle in vivofrom the preimplantation embryo to the gonads at the embryonic andadult stages and to spermatocytes up to the pachytene stage (Celebiet al., 2012). Tex19.1 is also expressed in in vitro pluripotent stem cellsderived from the ICM (ESCs), the epiblast (embryonic carcinomascells) or the PGCs (EGC). Its expression is lost when these cells differen-tiate upon either retinoic acid treatment or embryoid body formation(Kuntz et al., 2008).

Analyses of Tex19.1 knock-out (KO) mice have highlighted two mainphenotypes. Tex19.1-deficient males are infertile presenting an interrup-tion of spermatogenesis at meiosis between pachytene and metaphase I(Ollinger et al., 2008). This phenotype is reminiscent of a failure tocontrol transposon expression in the male germline. Indeed,Tex19.12/2 spermatogenesis impairment occurs in the context of aspecific up-regulation of MMERVK10C retrotransposons (Ollingeret al., 2008). In addition, around half of the expected homozygousanimals are missing from the litters for currently unknown reasons (Ollin-geret al., 2008). It wasalso suggested thatTex19.1KO phenocopies Ubr2knock-out (Yang et al., 2010).

We present here a thorough analysis of the Tex19.1 mutant pheno-type. We confirm that spermatogenesis is altered in mutant males andfurthermore show that females are fertile. We determine that the lethal-ity onset soon after birth is due to a placental defect and describe a severehypotrophy of newborn homozygous mutant animals with no sex differ-ence. Furthermore, we show that MMERVK10C retrotransposon ex-pression is altered upon Tex19.1 deficiency not only in testes, butalso in ESCs, together with LINE and IAP families of retrotransposons.Altogether, our results suggest a role of Tex19.1 in two essential func-tions of the mammalian life cycle, i.e. placenta-supported in uterogrowth and male fertility.

Materials and Methods

Antibody productionTo generate the anti-Tex19 monoclonal antibody, the entire protein wasproduced and injected into 8-week-old female BALB/c mice intraperitone-ally with 200 mg of poly (I/C) as adjuvant. Three injections were performedat 2 week intervals. Four days prior to hybridoma fusion, mice with positivelyreacting sera were reinjected. Spleen cells were fused with Sp2/0.Agl4myeloma cells as described by de StGroth and Scheidegger (de StGrothand Scheidegger, 1980). Hybridoma culture supernatants were tested onDay 10 by ELISA for cross-reaction with Tex19 peptides. Positive

2 Tarabay et al.

by guest on May 15, 2013


Dow

nloaded from


supernatants were then tested by immunofluorescence and western blot onTex19 transfected COS-1 cells. Specific cultures were cloned twice on softagar. Specific hybridomas were established and ascites fluid was preparedby injection of 2 × 106 hybridoma cells into pristane-primed BALB/c mice.

The specificity of the produced antibody was tested by western blotting onwild-type and Tex19.1 knock-out testes and ESCs. This antibody(7Tex-1F11) detects a 42 kDa band in the cytoplasmic fraction from WTESCs and WT adult testes and this band was absent in Tex19 knock-outtestes and ESCs (Fig. 1D).

All animal experimental procedures were performed according to theEuropean authority guidelines.

RT-qPCRRNAwas prepared using the RNeasy mini or micro kit (Qiagen) following themanufacturer’s instructions. After DNase I digestion (Roche), 1 mg RNAwasreverse-transcribed by random priming using Superscript II (Invitrogen). Theresulting cDNA was diluted in a final volume of 80 ml, and 0.25 ml cDNA wasused for each qPCR reaction, which were performed using SYBRw greenJumpStartTM Taq ReadyMixTM (Sigma) and LightCycler 480 (Roche). The ef-ficiency and specificity of each primer pair was checked using a cDNA stand-ard curve. All samples were normalized to b-actin or Rrm2 expression.Oligonucleotide sequences and PCR conditions are listed in Supplementarydata, Table SI. The amplicon size was measured to validate the specificity of allprimers (data not shown).

ESC cultureCK35 ESC line (kindly provided by Chantal Kress, Institut Pasteur, Paris) wascultured on a feeder layer in ESC-FCS, a medium containing DMEM (Gibco)15% fetal calf serum (FCS), 2 mM glutamine (Gibco), 100 mM non-essentialamino acids, 100 mM b-mercaptoethanol, antibiotics and 1000 U/ml LIF(Esgro, Millipore).

iPSC derivationMouse-induced pluripotent stem cells (iPSCs) were generated as previouslydescribed (Madan et al., 2009). Apparition of iPSC clones was monitored byidentifying their characteristic morphology and individual clones wereexpanded. RNA was extracted from three individual clones and theTex19.1 expression level was compared with the one of the original fibro-blasts and with that of wild-type (WT) ESCs.

Tex19.1 knock-out mice generationConditional knock-out (KO) mice were generated by the MCI (Mouse Clin-ical Institute, Strasbourg) transgenesis facility. Using homologous recombin-ation (HR) in ESCs, the whole Tex19.1 gene was replaced by the floxed genefollowed by the Neo-frt cassette in ESCs (Fig. 1A). HR events in C57Bl/6ESCs were first screened by PCR and confirmed by Southern blot. Hetero-zygous Tex19.1+/2 ESCs were microinjected into host blastocysts. Chi-meric animals were crossed with Flp transgenic mice and backcrossed withC57Bl/6 females in order to eliminate the Neo cassette and the Flp trans-gene, respectively. The floxed gene was eliminated by crossing the heterozy-gous mice with pCMV-Cre mice. Heterozygous animals were backcrossedwith C57Bl/6 animals to retrieve the Cre transgene. The present analyseshave been performed with animals having between 93.73 and 96.86%;C57/Bl6 background. Tex19.1 genotyping was carried out using a duplexPCR. Oligonucleotide sequences and PCR conditions are listed in Supple-mentary data, Table SI.

The absence of Tex19.1 mRNA and protein was tested by RT–PCR usingoligonucleotides described in Supplementary data, Table SI and by westernblotting. Cytoplasm and nuclear protein extractions were prepared andtreated as described (Achour et al., 2008, Wang et al., 2012). The mouse

monoclonal antibodies (clone 7Tex-1F11) raised against Tex19.1, cloneTUB-2A2 raised against b-tubulin and clone UHRF-1C10 raised againstUHRF1 were engineered at the IGBMC Monoclonal Antibodies Facility.Briefly, 150 mg protein of extracts were separated on SDS–PAGE 10%.Blots were probed with anti-Tex19.1, anti-b-tubulin or anti-UHRF1 (dilution1:1000). Blots for Tex19.1 were incubated with protein A conjugated-HRP(Abcam, Cambridge, UK) diluted to 1:10 000. Blots for UHRF1 andb-tubulin were incubated with secondary peroxidase-conjugated antibodies(Jackson Immunoresearch, West Grove, PA, USA) diluted to 1:10 000.Signals were detected by chemiluminescence using the ECL detectionsystem (Amersham Biosciences Europe GmbH, Saclay, France).

PhenotypingFor the phenotyping, heterozygous Tex19.1+/2 mice were bred. 0.5 dpcwas considered the day of plug (day post-coitum). Embryos or post-natalanimals, and when applicable, placentas were collected, weighed and mea-sured at 10.5 dpc, 13.5 dpc, 17.5 dpc, 19.5 dpc, 0.5 dpp, 5 dpp and thengenotyped. Placentas and gonads were used for further histological orRT-qPCR analyses.

Sperm analysesEpididymes of nine adult males (five knock-outs and four wild-type) were dis-sected in 1 ml of EmbryoMaxw Human Tubal Fluid (HTF) medium (Milli-pore), pre-equilibrated at 378C, 5% CO2, and incubated for 30 min to letspermatozoa leave the epididymis. Sperm cells were counted and mobilitywas measured with an IVOS Computer Assisted Sperm Analyzer (HamiltonThorne, USA) after a 1/20 dilution. For each animal, a smear was realizedwith the initial solution and the 1/20 diluted one, air dried, fixed with 70%ethanol for 5–10 min, air dried and counterstained according to HarrisSchorr staining. Sperm morphology was based on head shape analysisaccording to Burruel et al. (1996). Sperm head morphology analysis was per-formed on 100 spermatozoa when possible (all Tex19.1+/+ and oneTex19.12/2 animal). Otherwise, 16 and 50 sperm cells have beencounted for two other Tex19.12/2 animals.

Histology and TUNEL assaysPlacentas and ovaries were collected and fixed in 4% (wt/vol) buffered for-malin for 24 h, whereas testes were collected and fixed in Bouin’s fluid for48 h and then embedded in paraffin. For histological analyses, 5 mm-thicksections were stained with haematoxylin/eosin. All slides were examinedusing a DMLA microscope (Leica) with 10×, 20×, 40× and 100× objec-tives with apertures of 0.3, 0.5, 0.7 and 1.3, respectively and with DMLAM420 macroscope (Leica) with Leica Apozoom. Images were taken with adigital camera (CoolSnap; Photometrix) using the CoolSnap v.1.2 softwareand then processed with Photoshop CS2 v.9.0.2 (Adobe). For detection ofapoptotic cells, TUNEL assays were performed as described (Ghyselincket al., 2006).

Tex19.12/2 ESC derivationTo obtain Tex19.12/2 ESC lines, embryos from heterozygous crosseswere collected at 3.5 dpc, by flushing the uterus with ESC-KSR medium con-taining KO-DMEM, 15% Knock-out Serum (KSR, Gibco), 2 mM glutamine(Gibco), 100 mM non-essential amino acids, 100 mM b-mercaptoethanol,antibiotics and 1000 U/ml LIF (Esgro, Millipore). The protocol wasadapted from Bryja et al. (2006). Genotyping was performed after feederremoval using platings to avoid WT allele contamination, and was confirmedby a RT–PCR to detect Tex19.1 transcripts and by western blot. These EScell lines present a C57Bl/6 background varying from 93.8 to 95.4%.

Tex19.1 in development and spermatogenesis 3



Dow

nloaded from

http://humrep.oxfordjournals.org/lookup/suppl/doi:10.1093/humrep/det129/-/DC1






Functional tests for Tex19.1 ESC linesClonality and proliferationFor clonality and proliferation tests, feeders were removed by adsorption. Toassess the clonality, 400 cells were plated on a 60 mm gelatin-coated dish in amodified ESC-KSR medium containing 10% KSR and 10% FCS to help ESCattachment and growth without feeders. Medium was refreshed every 2days. After 6 days, the dishes were rinsed with PBS and fixed with ice-coldmethanol for 2 min. Dishes were dried and washed with Tris HCl 100 mM,

pH 9.5, NaCl 100 mM and MgCl2 10 mM, and stained for alkaline phosphat-ase activity using the BCIP/NBT liquid substrate system (SIGMA B1911), for20 min in the dark at room temperature (RT). Dishes were finally washedwith distilled water and dried.

The proliferation was measured by plating 75 000 cells/well in 6-wellplates, and counting negative Trypan blue cells for every 24 h. ThreeTex19.1+/+ or Tex19.12/2 cell lines were used. For each cell line,the measurement was done in duplicate. The experiment was repeatedthree times.

Figure 1 Genetic targeting of Tex19.1 gene in mice. (A) The Tex19.1 locus is shown before and after homologous recombination (HR). The neomycine(Neo) gene was removed by crossing the mice with a Flp-recombinase strain. The Knock-out allele was obtained by crossing Floxed mice with pCMV-Cremice; (B) example of PCR genotyping; one sample of each genotype (Tex19.1+/+, +/2 and 2/2) is shown and mutant (Mt) and Wild-type (WT) allelebands are indicated; (C) RT–PCR assessment of Tex19 expression in ESCs, 16 dpp testes and E18.5 placenta. The upper band indicates the Tex19.1 tran-script and the lower band indicates the Tex19.2 transcript. Gapdh expression is used as a control; (D) western blot for Tex19.1 in nuclear (N) and cyto-plasmic (C) fractions from ESCs (WT and Tex19.12/2) and adult testes (WT and Tex19.12/2). Subcellular fractionation was monitored using UHRF1(Ubiquitin-like containing PHD and RING Finger domain 1) and b-tubulin as nuclear and cytoplasmic markers, respectively. Tex19.1 is detected in thecytoplasm.

4 Tarabay et al.



Dow

nloaded from


Blastocysts injection: chimera analysisWT Balb/c blastocysts were microinjected with Tex19.1+/+ orTex19.12/2 ESCs at the microinjection platform of the MCI and transferredin pseudogestant females. Their chimerism rate was evaluated by coat colourobservation. High-rate chimeric males were bred with Balb/c females toassess the germline transmission ability of the ESC lines.

Chimeric animals were analysed using flow cytometry and PCR techni-ques. Flow cytometric analysis of lymphocyte populations was performedas described by Madan et al. (2009).

For WT chimeras, the Klrk1 locus on chromosome 6, carrying a restrictionfragment length polymorphism and containing a XbaI site, was amplified byPCR. DNA was then digested with XbaI (20 000 U/ml, BioLabs), purifiedBovine Serum Albumin (BSA) 10× and NEBuffer 2 10× (Biolabs) for 2 hat 378C. For KO chimeras, Tex19.1 WT and KO alleles were specifically amp-lified in a duplex single reaction. Oligonucleotide sequences and PCR condi-tions are listed in Supplementary data, Table SI.

Results

Loss of Tex19.1 leads to growth defectand early post-natal lethalityTo precisely decipher Tex19.1 function in vivo as well as in in vitro derivedESCs, we developed a strategy that removes the complete Tex19.1gene. Mice carrying loxP-flanked Tex19.1 alleles were crossed withmice bearing the CMV-Cre transgene to generate constitutive Tex19.1mutants (Fig. 1A). Cre-mediated LoxP site recombination was moni-tored by genotyping-PCR (Fig. 1B). To confirm that our KO strategygave rise to a null allele, we performed RT–PCR on Tex19.12/2

ESCs, 16 dpp testis and 18.5 dpc placenta. No mRNA could bedetected, showing that Tex19.1 is not expressed, in relevant tissues ofhomozygous mutant animals (Fig. 1C). The absence of the Tex19.1protein was also checked by western blot using nuclear or cytoplasmicextracts of WT, KO ESCs or adult testis (Fig. 1D). It is noteworthythat in our last published work (Kuntz et al., 2008), the antibodywe used showed a nuclear localization, which was in contradiction toOllinger et al. findings (Ollinger et al., 2008). The new antibody used inthe present study is in accordance with other studies showing cytoplas-mic expression and points out the possible aspecificity of our previousantibody.

Interbreeding of Tex19.1+/2 animals resulted in a statistically signifi-cant deviation frequency of homozygous animals from the expectedMendelian 1:2:1 ratio (158 Tex19.1+/+, 29.9%; 299 Tex19.1+/2,56.6% and 71 Tex19.12/2, 13.5%) (P , 0.0001, Khi2 test) (Fig. 2A).We obtained an equal number of female and male Tex19.12/2

animals (44% and 56% of Tex19.12/2 pups, respectively, non-significant), suggesting no gender-specific lethality. Genotyping was ini-tially performed at 15 days post-partum (dpp). We first checked if lethal-ity could occur sometimes between birth and 15 dpp. We analysed 63living pups at 0.5 dpp and 45 at 5 dpp. Their genotyping revealed no sig-nificant distortion of the normal Mendelian ratio at birth or at 5 dpp, sug-gesting a lethality onset between 5 dpp and 15 dpp (Fig. 2A). ThreeTex19.12/2 animals at 0.5 dpp and 5 dpp, respectively, were founddead and one Tex19.1+/+ at 5 dpp. We noticed a significant 7% sizereduction and 19.5% weight loss in Tex19.12/2 compared withTex19.1+/+ animals at 0.5 dpp (Fig. 2B). No size or weight defectswere observed at the adult stage (data not shown), suggesting that the

surviving Tex19.12/2 animals are able to recover from their growthdefect.

We then wondered when this growth defect takesplace during in uterodevelopment. Having analysed 231 fetuses in total, 32 at 10.5 dpc, 101 at13.5 dpc, 65 at 17.5 dpc and 33 at19.5 dpc, wedid not observeany Men-delian ratio distortion (Fig. 2A). While size and weight of the fetuses wereidentical in all genotypes at 13.5 dpc, a weight reduction was observedfor Tex19.12/2 embryos at 17.5 dpc, and size and weight differenceswere more accentuated at 19.5 dpc (P , 0.02, Fig. 2B). Tex19.12/2

placentas also showed weight and size reduction, with statistically signifi-cant differences from 17.5 dpc onwards (P , 0.002, Fig. 2C). All placen-tal layers could be identified in Tex19.12/2 placentas at 13.5, 17.5 and19.5 dpc. However, starting from 17.5 dpc, Tex19.12/2 placentasshowed diminished thickness affecting all placental layers (Fig. 3). Add-itionally, at 17.5 dpc, Tex19.12/2 placentas displayed signs of necrosis(arrow in Fig. 3B and D) in the junctional zone, which was also detectablebut to a lesser extent at 19.5 dpc (data not shown). This phenotype wasconsistent in all Tex19.12/2 placentas analysed at 17.5 and 19.5 dpc (atleast five placentas per group).

Loss of Tex19.1 leads to a heterogeneousspermatogenic defect and testiculardegenerationNo obvious defect or difference could be noticed in the survivingTex19.12/2 male and female animals. Considering the germ cell ex-pression of Tex19.1, we assessed the fertility of Tex19.12/2 animals.From the nine males tested (mean cross duration of 7 weeks, rangingfrom 4 to 17 weeks), seven never gave rise to pups despite the presenceof plugs, while two were able to reproduce, with litter sizes rangingbetween two and nine pups. Morphological analysis revealed thatTex19.12/2 males have smaller testes (Fig. 4A) with a mean weightof 61 mg ranging from 30 to 105 mg, compared with a mean weight of102 mg ranging from 77 to 112 mg in Tex19.1+/+ males (Fig. 4B).

We analysed the sperm count of five Tex19.12/2 compared withfour Tex19.1+/+ males, and found features of almost complete azoo-spermia to normospermia. Three Tex19.12/2 males had ,1 millionof spermatozoa per millilitre and two had a count of 2.2 and 3.2 millionsof spermatozoa per millilitre, respectively, which is less than the numberobserved in Tex19.1+/+ males (Fig. 4D, P , 0.003). All tested maleshad a reduced sperm motility compared with Tex19.1+/+ animals(Fig. 4E). Three Tex19.12/2 animals had no motility at all, whereasthe other two had a reduced total motility of 40 and 50%, respectively,compared with a range of 81–95% for Tex19.1+/+ animals (Fig. 4E).In order to analyse sperm morphology, a spermocytogram was realizedon three out of five Tex192/2 samples and compared with fourWT animals. Normal and subnormal morphology was drasticallyreduced in Tex19.12/2 animals (from 20 to 44%) compared withTex19.1+/+ animals (86–98%; Fig. 4C). These results suggest an in-volvement of Tex19.1 in spermatogenesis and spermiogenesis.

Histological analysis of Tex19.12/2 testes showed variation in sem-iniferous epithelium degeneration among individuals, allowing a classifi-cation into three groups according to the severity of the phenotype.One-third of the mutant males presented no spermatozoa in thecaudal epididymis, a complete absence of post-meiotic germ cells in alltubules and the most advanced meiotic cells at pachytene stage(Fig. 4H–K). Another third displayed a less severe phenotype, with a




Dow

nloaded from



proportion of cells that were able to complete meiosis. In thesetestes, the seminiferous epithelium showed a reduced thickness anda diminished number of post-meiotic germ cells (Fig. 4G) comparedwith WT males (Fig. 4F). Moreover, the seminiferous epitheliumof Tex19.1–/– mutants showed scattered large vacuoles (VA;compare Fig. 4F and G), in the layer of meiotic cells and desquamationof round cells (black arrowhead in Fig. 4G). In this group, the caudalepididymis contained low spermatozoa stores (compare SZ in Fig. 4I

and J), but numerous necrotic round cells (R; Fig. 4J and K). In agree-ment with a spermatocyte maturation failure, the seminiferous tubulesexhibited large amounts of TUNEL-positive and pycnotic nuclei with apachytene-like morphology (Fig. 4M). Therefore, Tex19.1–/–pachytene-like spermatocytes undergo apoptosis. In the last third,no obvious defects were seen. In line with these findings, the caudalepididymis sperm store was indistinguishable from those of wild-typeanimals (data not shown).

Figure 2 Generation of Tex19 mutant mice. (A) Genotype distribution after heterozygous crossings and at different stages of development. Daggerindicates the number of dead pups. P indicates that genotype distribution is different from what is expected for a Mendelian ratio (Khi2 test). ns, non-significant (P . 0.05); (B) mean and SD for 13.5 dpc, 17.5 dpc, 19.5 dpc embryos, 0.5 dpp, 5 dpp pups size and weight for WT and Tex19.12/2 indi-viduals; (C) mean and SD for 13.5 dpc, 17.5 dpc and 19.5 dpc placenta size and weight. For B and C: Numbers of studied pups and placentas areshown below each histogram point. *P , 0.02; §P , 0.002.

6 Tarabay et al.



Dow

nloaded from


Altogether, these data are in accordance with the fact that somemutant males were fertile and are indicative of heterogeneity of testiculardegeneration with a complete spermatogenesis-arrest in pachytenestage as the most severe phenotype. This heterogeneity was not influ-enced by the age of the animals (data not shown).

Eleven Tex19.12/2 females were tested for their fertility. All of themgave rise to pups at a frequency comparable with control Tex19.1+/+females (5+/22.5 pups/litter for Tex192/2 compared with6+/22.8 pups/litter for their control littermates). Histological analysesdid not detect any obvious ovarian abnormalities in Tex19.12/2

females (Supplementary data, Fig. S1).

Tex19.1 null blastocysts give rise to pluripotentES cell linesThe specific Tex19.1 expression in pluripotent stem cells prompted usto analyse how the gene was regulated during somatic cell reprogram-ming, using the iPSC system. While Tex19.1 mRNA was absent inmouse embryonic fibroblasts, its expression was induced whenthose cells were reprogrammed into iPSCs (Fig. 5A), with a level ofexpression at the same range as in ESC lines. This result suggeststhat Tex19.1 transcription might correlate with pluripotency. To geta deeper understanding of the function of Tex19.1 in pluripotency,we established ES cell lines from Tex19.12/2 mice. Starting from

49 blastocysts, 46 (93.9%) of them attached and 22 (47.8%) gaverise to ES cell lines. Genotyping of these cell lines demonstrated Men-delian genotype distribution (Fig. 5B) with seven Tex19.1+/+ (32%),ten Tex19+/2 (45%) and five Tex19.12/2 (23%) ES cell lines, re-spectively. RT-qPCR (Fig. 5C) and western blotting (Fig. 1D) con-firmed the absence of Tex19.1 transcript and protein inTex19.12/2 cell lines. No difference could be noticed in ES cellmorphology (Fig. 5D) and ability to express Oct4, Nanog and Sox2among the different genotypes (data not shown). Tex19.2 was notexpressed in the Tex19.12/2 ES cell lines (data not shown) andcould, therefore, not be involved in compensatory mechanisms. Al-together, these results suggest that Tex19.1 does not seem to actas a major factor for the maintenance of ESC pluripotency.

We assessed the self-renewal ability by measuring ESC clonality andproliferation. For clonality assays, 400 cells were plated and culturedfor 6 days, and stained for tissue non-specific alkaline phosphatase(TNAP) activity. A significant decrease in the number of positiveclones was noticed. Indeed, on average 10% of the platedTex19.1+/+ and only 3% of the plated Tex19.12/2 ESCs gave riseto individual colonies, respectively (P ¼ 0.01; Fig. 6A). These resultssuggest that Tex19.1 plays a role in the aptitude of ESCs to form colonies.However, no significant growth rate, apoptosis or cell cycle differenceswere noticed between Tex19.12/2 and Tex19.1+/+ ESCs(Fig. 6B–D).

Figure3 Tex19.12/2 placental defects. Histological sections stained with haematoxylin and eosin of E17.5 (A–D) and E19.5 (E and F) placentas. Blackarrows in B and D point to necrosis area in the junctional zone; De, decidua; SZ, spongiotrophoblast; LZ, labyrinthe (Scale bars: 1 mm in A, B, E and F and100 mm in C and D).




Dow

nloaded from



Figure4 Tex19.12/2 testicular defects. (A) Testes from Tex19.12/2 mice are smaller than testes fromWT littermates (8-week-old). (B) Mean testisweights are reduced in adult Tex19.12/2 knock-out mice (Student’s t-test, P , 0.01). (C) Sperm morphology of Tex19.1+/+ (n ¼ 4) and Tex19.12/2

(n ¼ 3) adults. Mean + SD for morphology features. *P , 0.001; §P , 0.05; (D) Sperm concentration measurements for Tex19.1+/+ (n ¼ 4) andTex19.12/2 (n ¼ 5) animals, expressed in millions per millilitre. *P , 0.003; (E) motility, progressivity and rapidity measurements for Tex19.1+/+(n ¼ 4) and Tex19.12/2 (n ¼ 5) animals *P , 0.003. (F–K) Histological sections stained with haematoxylin and eosin through the testes (F–G) or epi-didymides (I–K) of 9-week-old mice. Black arrowhead in (G) points to round spermatids detaching from the seminiferous epithelium. (L and M) TUNELassays: the positive signal was converted into a red false colour and superimposed with the DAPI nuclear stain (blue false colour); R, round germ cells; RS,round spermatids; SZ, spermatozoa; VA, vacuoles (Scale bars: 5 mm in A, 100 mm in F–K and 50 mm in L and M).

8 Tarabay et al.



Dow

nloaded from


Ollinger et al. reported a 4-fold increase in the expression of the classIILTR-retrotransposon MMERVK10C in Tex19.12/2 16 dpp testes(Ollinger et al., 2008). When we analysed the level of MMERVK10C ex-pression by RT-qPCR in mRNA from Tex19.12/2 ESCs and testes incomparison with that of their WT counterparts, we found a 2-fold in-crease (Fig. 7). In addition, we tested the level of expression of otherTEs such as LINE-1, IAPs and MuERV1. We detected a significant4-fold over-expression of both LINE-1 and IAP elements inTex19.12/2 ESCs despite high variability. No significant change couldbe detected in Tex19.12/2 16 dpp testes (Fig. 7).

Tex19.12/2 ESCs contribute efficiently to allthe three germ layers, but not to the adultgerm lineTo assess in vivo the pluripotency of Tex19.12/2 ESCs, we tested theirperformance in chimeric animal colonization by injecting them into WTblastocyts. Two independent ES cell clones of either Tex19.1+/+ orTex19.12/2 background, with a normal male karyotype, were injectedinto Balb/c host blastocysts. Two cohorts of 81 blastocysts were injectedwith Tex19.1+/+ ESCs, resulting in 9 (11%) and 21 (26%) chimeric

Figure5 Generation of Tex19.1-deficient ES cell lines. (A) RelativeTex19.1expression by RT-qPCR in mouse iPSCs (iPS2, iPS7, iPS8 and iPS9) comparedwith two mouse ESC (mES BD10, mES CK35) lines and with the fibroblasts (fibro) that were used for reprogramming into iPSCs performed on duplicatesand normalized to beta-actin); (B) genotype distribution of ESCs obtained from cultured blastocysts of heterozygous crossings; (C) Tex19.1 expression byRT-qPCR in Tex19.1+/+, Tex19.1+/2 and Tex19.12/2 ES cell lines growing on a feeder layer. Feeders do not express Tex19.1 and are shown as acontrol; (D) Tex19.1+/+, Tex19.1+/2 and Tex19.12/2 ES cell lines were morphologically similar.




Dow

nloaded from


males for each cell line. For Tex19.12/2 ESCs, 77 and 87 blastocystswere injected and 16 (21%) and 7 (8%) chimeric males were obtained.These animals showed a high degree of chimerism, since 80 and 87%of the Tex19.1+/+ and Tex19.12/2 chimeric males presented acoat colour chimerism over 50%, respectively (Supplementary data,Table SII, Fig. 8A). To determine the contribution of the injected ESCs,we performed a cytofluorometry experiment to explore the major histo-compatibility class I (MHC-I) tissues constitution using antibodies recog-nizing either H2d or H2b corresponding to the MHC-I expressed byBalb/c or C57/Bl6, respectively, on spleen lymphocytes (Fig. 8B). In add-ition, a semi-quantitative PCR on genomic DNA using oligonucleotidesused for the genotyping was also performed on Tex19.12/2 ESC-derivedchimeras, to detect the mutant allele. For the Tex19.1+/+ ESC-derived chimeras, we performed a second semi-quantitative PCR todetect a RFLP in the Klrk1 locus between Balc/c and C57/Bl6 mice,the latter containing an XbaI restriction site. Due to the sensitivity ofthe PCR, this method is expected to detect even minor contributionsof ESCs in the chimeric organs (Fig. 8C and D). We found that bothTex19.12/2 and Tex19.1+/+ ESCs were able to contribute efficientlyto ectodermal, endodermal and mesodermal derived tissues.

To specifically assess germ line transmission, six Tex19.1+/+ and sixTex19.12/2 highly chimeric males (above 90%; Fig. 8A), produced

from two ESC lines of each genotype, were crossed with Balb/cfemales. All Tex19+/+ ESC-derived chimeric males were bred withWT females. All females gave birth to at least one or two litters ofagouti pups. Despite a prolonged breeding time with Balb/cfemales, Tex19.12/2 ESC-derived chimeric males were eithersterile (two males out of three for each Tex19.12/2 ESC line) orgave birth to litters of white coat colour pups only, originating fromWT Balb/c host blastocyst cells (Supplementary data, Table SIII).Thus, none of the Tex19.12/2 ESC-derived chimeric malesshowed germ line transmission.

Histological analysis of testes from Tex19.12/2 ESC-derived chi-meric mice showed signs of degeneration in most of seminiferoustubules (asterisks in Fig. 8F) compared with WT ESC-derived chimericmice (Fig. 8E), next to tubules of normal appearance (Fig. 8F), and thecorresponding caudal epididymis contained low spermatozoa counts(data not shown). In other chimeric males, the tubules contained veryfew germ cells with most of the tubules containing Sertoli cells only(Fig. 8G). This finding is likely to explain the infertility of Tex19.12/2

ESC-derived chimeric males.Taken together, these results suggest thatTex19.12/2 ESCs are able

to contribute widely to most (if not all) somatic tissues, but cannot con-tribute to formation of a mature germ line.

Figure 6 (A) Colony formation assay: Colonies were stained by their alkaline phosphatase enzyme activity. Mixed and undifferentiated colonies arecounted and shown for each genotype. The experiment was repeated three times in duplicates. (B) ES proliferation assay: ES cells were stained usingTrypan blue and counted over 4 days (D1-D4). The experiment was repeated three times in duplicates. (C) ES apoptosis detection assay: ES cellswere grown for 2 days and apoptosis was measured by Facs using an apoptosis inducer, using campthotecin as a control. (D) ES cell cycle assay: EScells were grown to confluence and different phases of the cell cycle are measured by FACS using PI staining.

10 Tarabay et al.



Dow

nloaded from





DiscussionTex19 proteins are linked to pluripotency and fertility by their expressionpatterns, but their molecular function is still unknown. To get deeperinsight into the biological relevance of Tex19.1, we carried out a knock-out (KO) strategy to delete the Tex19.1 gene in the mouse. We also con-firmed the phenotypic variation, the heterogeneous spermatogenicdefect and the over-expression of MMERVK10C retrotransposons in16 dpp testis initially associated with Tex19.1 deficiency (Ollinger et al.,2008).

As described by others (Ollinger et al., 2008; Yang et al., 2010), werecorded an early lethality of almost 50% of Tex19.12/2 individuals.However, in contrary to previous reports, we did not observe anygender bias in the lethality (Yang et al., 2010), which may be explainedby different genetic backgrounds. We could show that death is occurringin early neonatal stages, in association with a body size defect, whichcould be traced back to the second half of in utero development. Thisfetal onset of growth retardation is reminiscent of a placental default,for which we indeed provided evidence, showing significant size andweight reduction as well as a diminished thickness of all placental layersand necrosis in the regions of junctional zone. The early post-natal lethal-ity is, therefore, likely to result from a placenta defect during pregnancy,which affects growth during late gestation and compromises viabilitysoon after birth. This lethality could be due to a competition betweenWT and KO animals for maternal milk sources, the KO animals beingweaker and therefore less competent.

It is interesting to note that Tex19.1 promoter has a typical Zfx tran-scription factor binding site and that Zfx is expressed in the placenta.Zfx represents then a likely candidate for controlling Tex19.1 expressionin the placenta. Moreover, Zfx KO mice are smaller, less viable andpresent a germ cell depletion, as Tex19.12/2 mice do (Luoh et al.,1997).

Surviving Tex19.12/2 males and females are healthy, but males havereduced fertility with a variable penetrance. We, therefore, checked theexpression of Tex19.2 in Tex19.12/2 adult testes and noticed no com-pensation (data not shown). In contrast to previously published results(Ollinger et al., 2008), we did not notice any significant female fertility im-pairment compared with WT littermates, a discrepancy that could beagain explained by differences in genetic backgrounds or by subtledefects in female reproductive lifespan that we did not investigatefurther. Histological analyses showed a spermatogenetic arrest at thepachytene stage as the most severe phenotype in one-third of themutant animals and reduced spermatogenesis in another third. As previ-ously described (Ollinger et al., 2008), we also noticed, in the mostsevere cases, an absence of germ cells beyond pachytene stage as wellas variable levels of chromosomal pairing anomalies (data not shown).In addition, when spermatozoa were present, sperm parametersof Tex19.12/2 males were systematically perturbed, convergingtowards a severe form of oligoasthenoteratozoospermia. We postulatethat TEX19 mutations may represent potential causes of oligoastheno-teratozoospermia linked to human forms of infertility. Thus, it wouldalso be interesting to screen for mutation in TEX19 in a cohort of oli-goasthenoteratozoospermic patients.

To further investigate the link between Tex19 and pluripotency, weestablished ES cell lines from our KO mice. Tex19.1 deficiency did notaffect the efficiency of derivation of ESCs, their morphology and the

Figure 7 Retrotransposon expression profiles in (A) ESCs and(B) 16.5 dpp testes. Expression levels of MMERVK10C normalized toBeta-Actin housekeeping gene LINE-1, and IAPdelta1 and MuERVL ele-ments normalized to Rrm2 housekeeping gene. n values represent thenumber of biological replicates, SD, standard deviation. *P , 0.05.




Dow

nloaded from


expression level of pluripotency-related genes such as Oct4, Nanog orSox2. The only phenotype we could notice is a reduced ability ofTex19.12/2 ESCs to form colonies in vitro. In addition to its expressionin most pluripotent stem cells such as ESCs or EGC, we could show thatTex19.1 is induced following the reprogramming of somatic cells intoiPSCs. Tex19.1 expression seems to be strictly associated with the pluri-potent state, since it was not detected in multipotent stem cell popula-tions such as neural stem cells or mesenchymal stem cells (D’Amourand Gage, 2003; Galan-Caridad et al., 2007). It will be interesting tostudy the eventual role of Tex19 proteins in the establishment of iPSCs.

In vivo, Tex19.12/2 ESCs can contribute to all somatic tissues butnone of the tested Tex19.12/2 ES-derived chimeric males showed

germ cell transmission in association with a variable testicular phenotype.Accordingly, shRNA Tex19.1 knock-down experiments have highlightedthe need of TEX19.1 for a proper differentiation of ESCs into PGCs invitro (West et al., 2009). This also suggests a cell autonomous functionof Tex19.1 in spermatogenesis since Tex19.12/2 ESCs are notrescued in a WT environment.

The only insight into Tex19 function so far on a molecular level isprovided by its link with transposon control during spermatogenesis.As in Ollinger et al., we observed changes in the expression of theclass II LTR-retrotransposon MMERVK10C in 16 dpp mutant testes(Ollinger et al., 2008). Interestingly, we also reported an up-regulationof these elements in Tex192/2 ESCs, along with a more global

Figure 8 Analysis of mouse chimeras deriving from BALB/c blastocysts (MHC, H2d) injected with C57BL/6 Tex19.12/2 or WT ESCs (MHC, H2b).(A) Macroscopic appearance of chimeric animals; (B) flow cytometric analysis of spleen lymphocytes from chimeric mice derived from Tex19.12/2 andWT ESCs and appropriate haplotype-controls. B and T lymphocytes originating from Tex19.12/2 and WT ESCs are framed; (C) tissue contribution ana-lysis from Tex19.12/2 ES-derived chimeras by semi-quantitative PCR. Genomic DNA from indicated organs was amplified with primers that allow a dis-tinction between Tex19.1+/+ and 2/2 alleles corresponding to the lower and the upper bands, respectively; HZ indicated heterozygote genotype; (D)analysis of genomic DNA tissues from WT ES-derived chimeras by XbaI digestion on the Klrk1 locus. The contribution of the BALB/c strain to the chimericmice is shown by two bands at 100 and 600 bp, while the contribution of the C57BL/6 strain is pointed out by two bands at 300 bp and one band at 100 bp;(M : molecular size marker); (E–G) histological sections stained with haematoxylin and eosin through the testes of chimeric mice derived from WT (E) andTex19.12/2 ESCs (F and G) (Scale bar: 100 mm in E, F and G). Asterisks indicate degenerated seminiferous tubules.

12 Tarabay et al.



Dow

nloaded from


up-regulation of other transposable element families (LINE-1 and IAPs),while this was not observed in the mutant testis context. Such tissue-specificity of global transposon reactivation may originate from variablefunctional redundancy between Tex19.1 and Tex19.2: indeed, we re-cently showed that Tex19.1 and Tex19.2 are concomitantly expressedduring spermatogenesis (Celebi et al., 2012), while Tex19.1 is the onlyTex19 member to be expressed in ESCs. It would be interesting to seewhether the double inactivation of Tex19.1 and Tex19.2 generates amore severe phenotype in mutant males and if derepression of transpos-able elements is more pronounced in this context. Our study furtherhighlights that ESCs may provide a suitable cellular model to investigatethe functional link between Tex19.1 and transposon control, allowinglarge-scale experiments and functional assays that are not easily amen-able on germ cells.

With the exception of the placental phenotype, the Tex19.12/2

male traits are finally reminiscent of the ones observed in loss-of-functionof Piwi-related genes. Indeed, mutant males of the Piwi/piRNA pathwayshow drastic spermatogenesis defects with a clear reactivation of LINE1and/or IAP retrotransposons (Ollinger et al., 2010; Pillai and Chuma,2012). Despite our previous results (Kuntz et al., 2008), we confirmedthe cytoplasmic localization of Tex19.1 in mouse ES cells and testes asdescribed by others (Ollinger et al., 2008; Yang et al., 2010). Indeed,when tested on KO testis, it turned out that our initial Ab was non-specific. We, therefore, developed a new one. In regard to Tex19.1 cyto-plasmic localization (Ollinger et al., 2008; Yang et al., 2010), there is aneed to study its involvement and epistatic position within the Piwi/piRNA pathway by investigating its association with the chromatoidbody or other nuage-related cytoplasmic components as well as itsability to bind small RNAs. Finally, the fact that Tex19 deficiency leadsto transposon up-regulation in ES cells where the piRNA/PIWIpathway is not active leaves open the possibility that Tex19 functions in-dependently of this specialized pathway.

The link between Tex19 function in placental development and retro-transposon control in germ cells is not obvious at first sight. However, it isnoteworthy that in contrast to classical protagonists of the Piwi/piRNApathway, which are widely conserved, Tex19 genes have specificallyevolved in eutherians (Kuntz et al., 2008). This feature may suggestthat placental emergence provided the evolutionary force for the acqui-sition of Tex19 function, which could then have been co-opted for a germline- and pluripotency-related purpose. During the manuscript process-ing, the group of Dr I. Adams published a paper supporting our results onthe placenta (Reichmann et al., 2013)

Supplementary dataSupplementary data areavailable athttp://humrep.oxfordjournals.org/.

AcknowledgementsWe thank the IGBMC common facilities for technical support. This workwas supported by the Centre National de la Recherche Scientifique(CNRS), the Institut National de la Sante et de la Recherche Medicale(INSERM) (Grant Avenir), the Ministere de l’Education Nationale, del’Enseignement Superieur et de la Recherche, the Universite de Stras-bourg, the Association Francaise contre les Myopathies (AFM) and theFondation pour la Recherche Medicale (FRM) and Hopitaux Universi-taires de Strasbourg.

Authors’ rolesY.T., E.K., M.T., C.C., A.v.M., N.Z., M.A., R.E.R., E.G., P.T., M.M., D.B.,S.V.: Data analysis and interpretation, manuscript writing, final approvalof manuscript.

FundingThis work was supported by the Centre National de la Recherche Scien-tifique (CNRS), the Institut National de la Sante et de la Recherche Med-icale (INSERM) (Grant Avenir), the Ministere de l’Education Nationale,de l’Enseignement Superieur et de la Recherche, the Universite de Stras-bourg, the Association Francaise contre les Myopathies (AFM) and theFondation pour la Recherche Medicale (FRM) and Hopitaux Universi-taires de Strasbourg.

Conflict of interestNone declared.

ReferencesAchour M, Jacq X, Ronde P, Alhosin M, Charlot C, Chataigneau T,

Jeanblanc M, Macaluso M, Giordano A, Hughes AD et al. Theinteraction of the SRA domain of ICBP90 with a novel domain ofDNMT1 is involved in the regulation of VEGF gene expression.Oncogene 2008;27:2187–2197.

Aravin AA, Hannon GJ, Brennecke J. The Piwi-piRNA pathway provides anadaptive defense in the transposon arms race. Science 2007;318:761–764.

Barlow DP. Genomic imprinting: a mammalian epigenetic discovery model.Annu Rev Genet 2011;45:379–403.

Bourc’his D, Bestor TH. Meiotic catastrophe and retrotransposonreactivation in male germ cells lacking Dnmt3L. Nature 2004;431:96–99.

Bryja V, Bonilla S, Cajanek L, Parish CL, Schwartz CM, Luo Y, Rao MS,Arenas E. An efficient method for the derivation of mouse embryonicstem cells. Stem Cells 2006;24:844–849.

Burruel VR, Yanagimachi R, Whitten WK. Normal mice develop fromoocytes injected with spermatozoa with grossly misshapen heads. BiolReprod 1996;55:709–714.

Celebi C, Montfoort AV, Skory V, Kieffer E, Kuntz S, Mark M, Viville S. Tex 19paralogs exhibit a gonad and placenta-specific expression in the mouse.J Reprod Dev 2012;58:360–365.

Chu LF, Surani MA, Jaenisch R, Zwaka TP. Blimp1 expression predictsembryonic stem cell development in vitro. Curr Biol 2011;21:1759–1765.

D’Amour KA, Gage FH. Genetic and functional differences betweenmultipotent neural and pluripotent embryonic stem cells. Proc Natl AcadSci USA 2003;100 Suppl 1:11866–11872.

de StGroth SF, Scheidegger D. Production of monoclonal antibodies: strategyand tactics. J Immunol Methods 1980;35:1–21.

Galan-Caridad JM, Harel S, Arenzana TL, Hou ZE, Doetsch FK, Mirny LA,Reizis B. Zfx controls the self-renewal of embryonic and hematopoieticstem cells. Cell 2007;129:345–357.

Ghyselinck NB, Vernet N, Dennefeld C, Giese N, Nau H, Chambon P,Viville S, Mark M. Retinoids and spermatogenesis: lessons from mutantmice lacking the plasma retinol binding protein. Dev Dyn 2006;235:1608–1622.

Hajkova P, Ancelin K, Waldmann T, Lacoste N, Lange UC, Cesari F, Lee C,Almouzni G, Schneider R, Surani MA. Chromatin dynamics during




Dow

nloaded from








epigenetic reprogramming in the mouse germ line. Nature 2008;452:877–881.

Hayashi K, Surani MA. Self-renewing epiblast stem cells exhibit continualdelineation of germ cells with epigenetic reprogramming in vitro.Development 2009;136:3549–3556.

Kaneda M. Genomic imprinting in mammals-epigenetic parental memories.Differentiation 2011;82:51–56.

Kuntz S, Kieffer E, Bianchetti L, Lamoureux N, Fuhrmann G, Viville S. Tex19, amammalian-specific protein with a restricted expression in pluripotentstem cells and germ line. Stem Cells 2008;26:734–744.

Lawson KA, Hage WJ. Clonal analysis of the origin of primordial germ cells inthe mouse. Ciba Found Symp 1994;182:68–84.

Luoh SW, Bain PA, Polakiewicz RD, Goodheart ML, Gardner H, Jaenisch R,Page DC. Zfx mutation results in small animal size and reduced germ cellnumber in male and female mice. Development 1997;124:2275–2284.

Madan B, Madan V, Weber O, Tropel P, Blum C, Kieffer E, Viville S, Fehling HJ.The pluripotency-associated gene Dppa4 is dispensable for embryonicstem cell identity and germ cell development but essential forembryogenesis. Mol Cell Biol 2009;29:3186–3203.

Ohinata Y, Payer B, O’Carroll D, Ancelin K, Ono Y, Sano M, Barton SC,Obukhanych T, Nussenzweig M, Tarakhovsky A et al. Blimp1 is a criticaldeterminant of the germ cell lineage in mice. Nature 2005;436:207–213.

Ollinger R, Childs AJ, Burgess HM, Speed RM, Lundegaard PR, Reynolds N,Gray NK, Cooke HJ, Adams IR. Deletion of the pluripotency-associatedTex19.1 gene causes activation of endogenous retroviruses anddefective spermatogenesis in mice. PLoS Genet 2008;4:e1000199.

Ollinger R, Reichmann J, Adams IR. Meiosis and retrotransposon silencingduring germ cell development in mice. Differentiation 2010;79:147–158.

Pillai RS, Chuma S. piRNAs and their involvement in male germlinedevelopment in mice. Dev Growth Differ 2012;54:78–92.

Reichmann J, Reddington JP, Best D, Read D, Ollinger R, Meehan RR,Adams IR. The genome-defence gene Tex19.1 suppresses LINE-1retrotransposons in the placenta and prevents intra-uterine growthretardation in mice. Hum Mol Genet 2013;22:1791–1806.

Sasaki H, Matsui Y. Epigenetic events in mammalian germ-cell development:reprogramming and beyond. Nat Rev Genet 2008;9:129–140.

SuraniMA, Hajkova P. Epigenetic reprogramming of mouse germ cells towardtotipotency. Cold Spring Harb Symp Quant Biol 2010;75:211–218.

Wang PJ, McCarrey JR, Yang F, Page DC. An abundance of X-linked genesexpressed in spermatogonia. Nat Genet 2001;27:422–426.

Wang X, Chen WR, Xing D. A pathway from JNK through decreased ERKand Akt activities for FOXO3a nuclear translocation in response to UVirradiation. J Cell Physiol 2012;227:1168–1178.

West JA, Viswanathan SR, Yabuuchi A, Cunniff K, Takeuchi A, Park IH,Sero JE, Zhu H, Perez-Atayde A, Frazier AL et al. A role for Lin28 inprimordial germ-cell development and germ-cell malignancy. Nature2009;460:909–913.

World Health Organization. WHO laboratory manual for the examination ofhuman semen and sperm-cervical mucus interaction. Cambridge, UK:Cambridge University Press, 1999.

Wu SM, Hochedlinger K. Harnessing the potential of inducedpluripotent stem cells for regenerative medicine. Nat Cell Biol 2011;13:497–505.

Yang F, Cheng Y, An JY, Kwon YT, Eckardt S, Leu NA, McLaughlin KJ,Wang PJ. The ubiquitin ligase Ubr2, a recognition E3 component of theN-end rule pathway, stabilizes Tex19.1 during spermatogenesis. PLoSOne 2010;5:e14017.

Zamudio N, Bourc’his D. Transposable elements in the mammaliangermline: a comfortable niche or a deadly trap? Heredity 2010;105:92–104.

14 Tarabay et al.



Dow

nloaded from


Documents

The mammalian-specific Tex19.1 gene plays an essential role in spermatogenesis and placenta supported development_Tarabay_2013