Differential Geminin Requirement for Proliferation of Thymocytes and Mature T Cells

This information is current as Dimitris Karamitros, Panorea Kotantaki, Zoi Lygerou, of October 1, 2021. Henrique Veiga-Fernandes, Vassilis Pachnis, Dimitris Kioussis and Stavros Taraviras J Immunol 2010; 184:2432-2441; Prepublished online 27 January 2010; doi: 10.4049/jimmunol.0901983 http://www.jimmunol.org/content/184/5/2432 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2010/01/27/jimmunol.090198

Material 3.DC1 http://www.jimmunol.org/ References This article cites 54 articles, 25 of which you can access for free at: http://www.jimmunol.org/content/184/5/2432.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on October 1, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2010 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Differential Geminin Requirement for Proliferation of Thymocytes and Mature T Cells

Dimitris Karamitros,* Panorea Kotantaki,* Zoi Lygerou,† Henrique Veiga-Fernandes,‡,1 Vassilis Pachnis,x Dimitris Kioussis,‡ and Stavros Taraviras*

Stem/progenitor cells coordinate proliferation and differentiation, giving rise to appropriate cell numbers of functionally specialized cells during organogenesis. In different experimental systems, Geminin was shown to maintain progenitor cells and participate in fate determination decisions and organogenesis. Although the exact mechanisms are unclear, Geminin has been postulated to influence proliferation versus differentiation decisions. To gain insight into the in vivo role of Geminin in progenitor cell division and differentiation,wehavegeneratedmicethatspecificallylackGemininincellsoflymphoidlineagethroughCre-mediatedrecombination.Tcells lacking Geminin expression upregulate early activation markers efficiently upon TCR stimulation in vitro and are able to enter the S phase of cell cycle, but show a marked defect in completing the cycle, leading to a large proportion of T cells accumulating in S/G2/M Downloaded from phases. Accordingly, T cells deficient in Geminin show a reduced ability to repopulate lymphopenic hosts in vivo. Contrary to expect- ations, Geminin deficiency does not alter development and differentiation of T cells in vivo. Our data suggest that Geminin is required fortheproliferationeventstakingplaceeitherinvitrouponTCRreceptoractivationorduringhomeostaticexpansion,butappearstobe redundantfortheproliferationanddifferentiationofthemajorityofprogenitorTcellpopulations. TheJournalofImmunology,2010, 184: 2432–2441. http://www.jimmunol.org/ regulated balance between proliferation and differentia- a multisubunit prereplicative complex that is essential for S-phase tion is fundamental for development, organogenesis, and onset (1). Geminin binds to the DNA licensing factor Cdt1 and in- A tissue homeostasis. To achieve this, cells must integrate hibits its function, ensuring that genome replication takes place only diverse extrinsic signals, remodel chromatin, and alter their tran- once per cell cycle (1), thereby maintaining genomic integrity (2). In scriptional profiles. Changes in chromatin organization and tran- addition, it has been shown that Geminin is expressed in neural scription must be coordinated with replication to establish or progenitor cells and is downregulated in fate-restricted precursor maintain the transcriptional profile of progeny cells. cells (3), although it participates in the control of cell fate specifi- Geminin has been implicated in proliferation-differentiation cation and progenitor cell maintenance during nervous system de- by guest on October 1, 2021 decisions through balanced interactions with multiple binding velopment through modulation of proliferation, transcription, and partners. Geminin participates in cell cycle control as a negative modification of chromatin structure (4). Ectopic overexpression of regulator of licensing, a process that precedes S-phase initiation Geminin in Xenopus embryos and Drosophila induces neural spe- and consists of the formation on origins of DNA replication of cific genes and hypertrophy of neural tissue, whereas Geminin has been shown to regulate the initiation of Sox2 expression during the establishment of the neural plate in vertebrate embryos (5–7). *Department of Pharmacology and †Department of Biology, Medical School, Uni- During eye development in Medaka, Geminin acts antagonistically versity of Patras, Rio, Patras, Greece; and ‡Division of Molecular Immunology and xDivision of Molecular Neurobiology, Medical Research Council/National Institute to Six3 homeobox-containing transcription factor, regulating the for Medical Research, The Ridgeway, London, United Kingdom balance between proliferation and differentiation. Thus, Geminin 1Current address: Immunobiology Unit, Instituto de Medicina Molecular, Faculdade overexpression leads to the reduction in eye size, exhibiting reduced de Medicina de Lisboa, Lisboa, Portugal. numbers of proliferating cells and premature induction of neuronal Received for publication June 24, 2009. Accepted for publication December 23, marker expression (8). Furthermore, it inhibits Hox transcription 2009. factor function through direct and polycomb-mediated interactions, This work was supported by research funding from the Association for International determining axial patterning during chicken embryogenesis (9). Cancer Research, United Kingdom (to S.T.) and Leukemia Research Foundation (to D. Kioussis). D. Karamitros was supported by a European Molecular Biology Orga- Interestingly, Geminin directly binds to Brg1, the catalytic subunit nization short-term fellowship (ASTF 26.00-2007) and Lady Tata Memorial Trust of the SWI/SNF complex, and reduces the ability of helix-loop- studentship. P.K. was supported by a European Molecular Biology Organization helix–type transcription factors to activate downstream target genes, short-term fellowship (ASTF 60.00-04) and a Federation of European Biochemical Societies summer fellowship. suggesting a role of Geminin in the maintenance of neuronal pro- Address correspondence and reprint requests to Dr. Dimitris Kioussis, Division of genitor cells in an undifferentiated state (10). Recently, Geminin has Molecular Immunology, Medical Research Council/National Institute for Medical Re- been shown to be involved in sustaining hematopoietic stem cells, search, The Ridgeway, London NW7 1AA, United Kingdom or Dr. Stavros Taraviras, expanding its potential role in other systems (11). Consistent with an Department of Pharmacology, Medical School, University of Patras, 26504 Rio, Patras, Greece. E-mail addresses: [email protected] and [email protected] essential role in proliferation and differentiation, mice lacking The online version of this article contains supplemental material. Geminin expression are dying during the first days of embryogen- Abbreviations used in this paper: 7-AAD, 7-aminoactinomycin D; CFSE, carboxy- esis, revealing no formation of inner cell mass (12, 13). fluorescein diacetate succinimidyl ester; DN, double-negative; DP, double-positive; We have used T cell differentiation as a model to address the role of ES, embryonic stem; Foxp3, forkhead box p3; LCR, locus control region; LN, lymph Geminin in developmental processes that require a regulated balance node; PDBU, phorbol 12,13-dibutyrate; SP, single-positive; WT, wild-type. between proliferation and differentiation. During immune system de- Copyright Ó 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00 velopment and maturation, undifferentiated hematopoietic precursor www.jimmunol.org/cgi/doi/10.4049/jimmunol.0901983 The Journal of Immunology 2433 cells originating in the bone marrow enter the thymus and give rise to allele, and DNA was amplified under the following conditions: 94˚C for 45 s, lineage-committed cells destined to become T cells (14, 15). De- 62.7˚C for 45 s, and 72˚C for 1 min (primer location is indicated in the Gem veloping thymocytes are classified into four major populations based Dex3-4 allele, Fig. 1A). Verification of the pGKNeo/tk cassette removal was established via a PCR that uses primers that anchor the neo gene (59- on the expression of two coreceptor molecules, CD4 and CD8. Double- cctgcgtgcaatccatcttgttca-39) and exon 3 of the mouse geminin gene (59-gca- negative (DN) thymocytes, lacking both CD4 and CD8 expression, are gaaggctggatcatcttcagcg-39), and DNA was amplified for 30 cycles (95˚C for the earlier progenitors and can be further subdivided into four sub- 45 s, 65˚C for 45 s, and 72˚C for 120 s). CD2Cre and PC3Cre transgenes were populations, DN1–4, based on the expression of CD44 and CD25 also detected by PCR as previously described (24, 25). To analyze excision efficiency in GemDex3-4 mice, genomic DNAwas prepared from tail, thymus, markers (16), with a large proportion of cycling cells identified in DN2 spleen, or LN and analyzed by Southern blot using BamHI digestion and a 59 and DN4 T cell populations (17). Successful rearrangement of the probe (Fig. 1A, probe AflII-SacII, and 1D)orbyPCR(Fig.1B). Experiments TCR-b gene (TCRb) that takes place at the DN3 (18) stage triggers were performed with Fl/koCD2Cre, whereas Fl/wt served as controls. Similar subsequent proliferation (19, 20). DN cells that have successfully results were obtained using Fl/wtCD2Cre mice as controls (data not shown). passed b-selection point downregulate CD25 and initiate the expres- The Medical Research Council/National Institute for Medical Research Eth- ical committee approved all animal experiments. sion of CD4 and CD8 (double-positive [DP]) accompanied by rearrangement of the TCR-a gene (TCRa) (18, 21, 22). The expansion in Isolation of cells, in vitro activation and proliferation thymocyte numbers that occurs between the late DN stage and the DP Total splenocytes or LNs (5 3 105) were cultured in 96-well plates in stage is the one of the main events that determines the final number of triplicate in complete RPMI 1640 (Sigma-Aldrich, Dorset, U.K.) supple- thymocytes (19). It has been estimated that 9 to 10 divisions are re- mented with 10% FCS (Life Technologies, Paisley, U.K.), 100 mg/ml quired from the CD44+CD25+ (DN2) stage to the CD4+CD8+ (DP) streptomycin, 100 U/ml penicillin, 2 mM glutamine, and 50 mM 2-ME stage (23). DP thymocytes are screened by stromal cells for useful (Sigma-Aldrich). Cells were stimulated with plate-bound anti-CD3 mAb (145.2C11, eBioscience, San Diego, CA) and anti-CD28 (37.51, eBio- Downloaded from TCRs, a process that leads to elimination of harmful or useless spe- science), or 0.5 mg/ml phorbol 12,13-dibutyrate (PDBU) and 1 mg/ml cificities (negative selection) and retention of T cells with useful TCRs ionomycin or 5 mg/ml Con A (Sigma-Aldrich). T cells were pulsed with (positive selection). The positive selected thymocytes bear either CD4 1 mCi [3H]thymidine/well (ICN Pharmaceuticals, Asse-Relegem, Bel- or CD8 coreceptors only (single-positive [SP]), and they are exported gium) 48 h poststimulation and harvested after 16 h. to the periphery to participate in the immune functions in the body (14). Adoptive transfer experiments To investigate the in vivo function of Geminin in the devel-

∼ http://www.jimmunol.org/ opment and homeostasis of the lymphoid system, we generated T cells from LNs of 6-wk-old mice were purified to 90% homogeneity by magnetic cell sorting using a mix of biotin-conjugated Abs (anti-mCD19 mice that allow the conditional inactivation of the mouse Geminin MB19-1, anti-mNK DX5, and anti-mGR.1 RB6-8C5, all from eBioscience) gene through Cre-mediated recombination. A transgenic line that and streptavidin Dynabeads (Invitrogen, Paisley, U.K.). Cells were labeled drives Cre expression under the control of the promoter and locus with 5 mM carboxyfluorescein diacetate succinimidyl ester (CFSE) (Molec- control region (LCR) of the human CD2 gene has been used to ular Probes, Paisley, U.K.) and counted usingtrypanbluetoassessviability.A total of 1.5 3 106 T cells derived from the Fl/KOCD2Cre or control animals allow the lymphoid specific inactivation of Geminin (24). We 2 2 were transferred into Rag2KOgc / recipients that combine the NK de- show that, unexpectedly, Geminin deficiency has minor effects in ficiency with T and B cell block due to deletions in common g-chain (gc; thymocyte development and differentiation, whereas mature pe- a shared component of the receptors for IL-2, -4, -7, -9, -15, and -21) and the ripheral T cells lacking Geminin exhibit severe proliferation de- Rag2 gene (27). On days 3, 5, and 15 following transfer, recipients were by guest on October 1, 2021 fects upon in vitro activation and during homeostatic expansion sacrificed, and proliferation history, total cellularity, and absolute numbers of CD4 and CD8 cells were assessed. upon adoptive transfer to lymphopenic hosts. Cell cycle analysis Materials and Methods Cell cycle analysis was carried out using an FITC BrdU flow kit (BD Targeting vector, generation of mice, and genotyping Pharmingen, Heidelberg, Germany) according to the manufacturer’s instructions. Activated T cells were stained for BrdU and DNA content using AGeminin genomic clone was isolated from the RPCI21-471M19 PAC library 7-aminoactinomycin D (7-AAD). BrdU pulse and chase experiment cells derived from 129S6/SvEvTac mouse strain using Geminin cDNA as a probe were subjected to double immunostaining for Brdu and Ki67 (anti-Ki67 (3). The 59 and 39 regions of homology in the targeting vector consisted of PE; BD Pharmingen). Naive T cells were activated using antiCD3/CD28 a 3.5-kb SacII-Xho1 fragment including exons 1 and 2 and a 6.0-kb BstX1- Abs for 38 h and subsequently pulsed with BrdU for 2 h. BrdU was re- Asp718 fragment spanning exons 5–7, respectively (regions of homology are moved and cells were fixed at 2, 8, and 16 h (time is calculated from the indicated in Fig. 1A as interrupted lines). The Floxed-pGKNeo/tk cassette addition of BrdU). T cells were stained with BrdU, Ki67, and 7-AAD. (courtesy of Dr. T. Mandamadiotis, Victorian College of Pharmacy, Monash Cells that have exited the cell cycle were calculated as the fraction of BrdU University, Victoria, Australia) was introduced into the XhoI restriction site, expressing cells that are not expressing Ki67. whereas a third loxP site was introduced into the BstXI site. This resulted in a 1.7-kb region encompassing exons 3 and 4 being flanked by loxP sites. The CFSE labeling linearized vector was electroporated into embryonic stem (ES) cells derived Splenic or LN-derived T cells were resuspended at a density of 1.5 3 107 from transgenic mice carrying the protamine Cre (25). Postselection with cells/ml in PBS (Life Technologies). A 10 mM stock solution of CFSE G418, 480 ES cell clones were isolated, expanded, and frozen according to (Molecular Probes) in DMSO was added to a final concentration of 1 mM standard methods. To screen for homologous recombinants, ES cell DNAwas and was incubated at 37˚C for 10 min. Subsequently, cells were washed digested with BamHI, Asp718, and combined BamHI, Asp718, and hybrid- twice in air-buffered IMDM supplemented with 1%FCS and then re- ized to sequences located 59 (Fig. 1A, probe AflII-SacII, black colored box) suspended in RPMI 1640 (Sigma-Aldrich). We have used CD4 PE (RM4-5, and 39 external (Fig. 1A, probe BglII-BamHI, gray checkered box) to the re- eBioscience), CD8a PerCP (53-6.7, BD Pharmingen), and CD25 allophy- combination sites, respectively. Twelve correctly targeted clones were iden- cocyanin (PC61.5, eBioscience). The dilution of CFSE used was proper to tified by Southern blot analysis and five of them with a euploid karyotype were minimize bleed-through to the FL2 channel, and any remaining bleed- used for injection into C57BL/6 blastocysts. Successful germline transmission through was properly compensated for. was achieved upon the injection of one of the homologous recombined ES clones. Mice were genotyped by PCR (26) or Southern blot of DNA samples FACS analysis (flow cytometry, cell sorting, and Abs) prepared from tail, thymus, spleen, and lymph node (LN). Samples were PCR amplified for 34 cycles (94˚C for 45 s, 65˚C for 30 s, and 72˚C for 1 min) using Single-cell suspensions were prepared from thymocytes and splenocytes, and PCR primers 2003 (59-tttggacgcatgggacagacc-39)and2004(59-gtccaggctg- cells were stained with saturating concentrations of FITC-, PE-, PerCP-, or cagtgtctcac-39) to detect the floxed and wild-type (WT) alleles (generating allophycocyanin-conjugated Abs in PBS with 0,1% BSA and 0.05% NaN3 at products of 597 bp and 549 bp, respectively) (primer location is indicated in the 4˚C for 30 min, then washed with PBS and subsequently analyzed using BD Geminin floxed allele, Fig. 1A). PCR primers (59-ggaatatttagagttctgaaatga- Biosciences FACSCalibur and FlowJo 8.7.1 software (Tree Star, Ashland, gatg-39)and(59-ccaactcagtcactgtcctgtt-39) were used to detect the deleted OR). For cell sorting, thymocytes were stained at a concentration of 1 3 108 2434 CONDITIONAL INACTIVATION OF GEMININ Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 1. Targeting the mouse Geminin locus. A, The mouse Geminin locus (Gem WT allele), targeted locus (Gem floxed allele), and the deleted Geminin allele lacking exons 3 and 4 (GemDex3-4 allele). Boxes indicate the mouse Geminin gene exons and triangles the LoxP sites. SacII, XhoI, BstX1, and Asp718 restriction sites define the 59 and 39 arms of homology, respectively (shown in WT allele). 59 external (probe AflII-SacII) and 39 external (probe BglII-BamHI) are shown in WT allele. A, B, S, X, and Bs indicate sites that are recognized from Asp718, BamHI, SacII, XhoI, and BstXI restriction enzyme, respectively. Relevant restriction sites are shown. Regions of homology are indicated as interrupted lines. B, DNA isolated from tail, thymus, spleen, and LNs were used to genotype mice using PCR. The Gem WT allele (WT), Gem floxed allele, and GemDex3-4 allele (knockout) generate PCR- amplified fragments of 549, 597, and 240 bp, respectively. Primers used are indicated as arrows in A. C, The efficiency of Geminin deletion in different T cell populations was established by PCR. DNA isolated from tail and DN, DP, CD4, and CD8 FACS-sorted thymocytes was analyzed using primers detecting the WT, floxed, and deleted allele. D, Genomic DNA was isolated from tail, thymus, spleen, and lymph nodes, digested with BamHI, and analyzed by Southern blotting using the 59 external probe (probe AflII-SacII). An 11.4-kb band corresponds to the targeted locus (Gem floxed allele) and 9.8 kb to the deleted allele (GemDex3-4 allele). E, Whole-cell protein lysates from thymus, spleen, and T cells activated by anti-CD3/CD28 (10 mg/ml and 2 mg/ml) or PDBU/ionomycin (0.5 mg/ml and 1 mg/ml) were subjected to Western blot analysis using specific Abs against Geminin, tubulin, and actin.

cells/ml in air-buffered IMDM supplemented with 5% FCS and sorted using Geminin and Cdt1 (28), rabbit polyclonal Abs against Cdc2 pTyr15 (9111, a MoFlo sorter (DakoCytomation, Carpinteria, CA). The following Abs were Cell Signaling, Danvers, MA), Rb pSer807/811 (9308, Cell Signaling), cyclin used: TCR-b–FITC (H57-597, eBioscience), CD4-FITC (H129.19, Santa E (M-20, Santa Cruz Biotechnology), cyclin B1 (H-433, Santa Cruz Bio- Cruz Biotechnology, Santa Cruz, CA), CD44 FITC (Caltaq Laboratories, technology), mouse mAbs against cyclin-D3 (DCS22, Cell Signaling, Buckingham, U.K.), CD4 PE (RM4-5, eBioscience), CD62L PE (MEL-14, Danvers, MA), p27 (57/kip1/p27 BD Transduction Laboratories), cyclin A eBioscience), CD69 PE (H1.2F3, eBioscience), CD4-PerCP (RM4-5, (CY-A1, Sigma, Dorset, U.K.), tubulin (B512, Sigma-Aldrich), and goat eBioscience), CD8a PerCP (53-6.7, BD Pharmingen), CD4-allophycocya- polyclonal Ab against actin (clone I-19, Santa Cruz Biotechnology). nin (GK1.5, eBioscience), CD8a-allophycocyanin (53-6.7, BD Pharmin- gen), and CD25 allophycocyanin (PC61.5, eBioscience). Annexin V staining Western blotting analysis Activated T cells were stained for Annexin V according to the manufacturer’s instructions (BD Pharmingen). Whole-cell lysates from thymocytes, splenocytes, and activated T cells or Statistical analysis purifiednaiveTcellswerepreparedinSDS-PAGEloadingbuffer,subjectedto electrophoresis in 6%, 10%, 12%, or 15% acrylamide gels and transferred to Average values 6 SEM are given, and p values were calculated by em- polyvinylidene difluoride membranes (Millipore, Watford, U.K.). Immu- ploying the two-sample Student t tests using Microsoft Excel software nodetection was performed using affinity-purified polyclonal Abs against (Microsoft, Redmond, WA). The Journal of Immunology 2435

Results previous findings showing that Geminin is essential for early mouse Generation of mice lacking Geminin in the lymphoid lineage embryogenesis (12, 13) and verify effective targeting of the locus. Geminin is expressed in the developing thymus during mouse We have employed the Cre-LoxP system to generate mice per- embryogenesis and its mRNA is present in all progenitor T cell mitting the conditional inactivation of the mouse Geminin gene. populations and peripheral CD4 and CD8 cells (Supplemental Mouse genomic DNA harboring mouse Geminin exons 1–7 was Fig. 1 and data not shown). To investigate the in vivo role of used in the targeting construct (Fig. 1A). LoxP sites were inserted in the introns flanking exons 3 and 4 and a neomycin-thymidine kinase Geminin during lymphocyte progenitor cell development and homeostasis, Gemininflox/flox or Gemininflox/2 mice were crossed cassette flanked by two LoxP sites in intron 2. Cre-mediated re- +/2 combination would lead to deletion of exons 3 and 4. If exons 2 and 5 with Geminin mice carrying a transgene that drives expression were to be spliced and a stable mRNA be produced, a frame-shift of Cre recombinase under the control of the human CD2 promoter mutation is generated, ensuring no functional protein is produced and LCR (29). This transgene has been previously shown to be following Cre recombination. The targeting construct was electro- expressed in B and T cell lineages. A large proportion of DN1 porated into ES cells derived from transgenic mice expressing Cre cells and the vast majority of DN3 and DN4 are expressing Cre flox/2 recombinase under the control of the protamine gene promoter, recombinase (24). For simplicity, we refer to the Geminin flox/+ which directs Cre expression in spermatocytes (25). Neomycin- CD2Cre mice as Fl/koCD2Cre and refer to their Geminin resistant ES cell clones were isolated and correct integration events CD2Cre littermates as Fl/wtCD2Cre. DNA isolated from tail, confirmed using 59 and 39 probes external to the homology regions thymus, spleen, and LNs of Fl/ko, Fl/koCD2Cre, and Fl/wt mice of the targeting construct (Fig. 1A and data not shown). Chimeric were analyzed by PCR and Southern blotting to assess the effi- mice were generated and germ line transmission of the targeted ciency of the recombination (Fig. 1). Using specific primers that Downloaded from allele was achieved. The neomycin-thymidine kinase-resistant detect the three different Geminin alleles (floxed, knockout, WT), cassette was removed through intercrossing mice carrying the tar- we show that in Fl/koCD2Cre mice, the Geminin floxed allele is geted allele and the protamine Cre with WT mice and selecting absent from DNA derived from thymus, whereas it is significantly progeny carrying the Geminin floxed allele without the neomycin- reduced in DNA derived from spleen and LNs, where a residual thymidine kinase cassette. This breeding scheme also produced band is still detectable (Fig. 1B). Developing thymocytes can be subdivided into several subpopulations: ∼5% express neither CD4 mice in which the additional deletion of exons 3 and 4 had taken http://www.jimmunol.org/ place in the germ line, thus establishing a line carrying one WT and nor CD8 (DN cells); ∼80% express both CD4 and CD8 (DP cells); one Geminin null allele (Geminin+/2). Subsequent breeding with ∼10% express only CD4 (CD4 SP) and ∼5% express only CD8 WT mice segregated out the protamine Cre transgene. Geminin+/2 (CD8 SP). PCR analysis of DNA samples derived from all four and Gemininflox/+ were viable, fertile, and apparently normal. In- stages of T cells in thymus shows a complete absence of the floxed tercrosses of Geminin+/2 and Gemininflox/2 mice did not produce allele in the presence of Cre recombinase (Fig. 1C). Furthermore, live Geminin2/2 offspring, whereas analysis for the presence of Southern blot analysis of DNA samples derived from tail, thymus, embryos at embryonic day 7.5 did not reveal any live Geminin2/2 spleen, and LNs using a 59 external probe has confirmed the embryo at that age (data not shown). These data are consistent with efficient deletion in tissues where Cre recombinase is expressed by guest on October 1, 2021

FIGURE 2. Mice that lack Geminin expression show moderate reduction on T cell populations of thymus and spleen. A, Average absolute cell numbers of DP, CD4, CD8, DN1, DN2, DN3, and DN4 thymic subpopulations from Fl/wt (n = 12) and Fl/koCD2Cre (n = 17) mice are depicted in the bar graphs. Cell-surface markers are shown as coordinates. Representative FACS analysis of thymocytes is shown. B, Represen- tative FACS analysis of splenocytes using Abs recognizing CD4 and CD8 cells is shown. Number of CD4+ CD8+ and CD42CD82 populations is shown (Fl/wt, n = 12, and Fl/koCD2Cre, n = 17). Values indicate average number of cells (6 SE). pp , 0.05; ppp , 0.01. 2436 CONDITIONAL INACTIVATION OF GEMININ

FIGURE 3. Mice that lack Geminin expression show reduced numbers of naive, regulatory, and memory T cells. Representative FACS analysis of splenocytes using Abs recognizing naive, memory, and regulatory splenic T cells are shown. Average absolute cell 2 numbers for naive CD4CD44 CD62Lhigh (A) and Downloaded from CD8CD44-CD62Lhigh (B) T cells from Fl/wt (n =6) and Fl/koCD2Cre (n = 10) mice. Average absolute cell numbers of memory T cells (CD4+CD44high) from Fl/ wt (n = 8) and Fl/koCD2Cre (n = 8) mice (C) and regulatory T cells (CD4+CD25+FoxP3+) derived from Fl/wtCD2Cre (n = 6) and Fl/koCD2Cre (n = 3) mice were measured (D). Values indicate average number of http://www.jimmunol.org/ cells (6 SE). pp , 0.05; ppp , 0.01. by guest on October 1, 2021

(Fig. 1D). Moreover, total cellular extracts from thymus and protein present in spleen, possibly due to the presence of other spleen from Fl/koCD2Cre and control littermates show absence of hematopoietic cells in this organ in which Cre recombinase is not the Geminin protein in thymus and only a residual amount of expressed. Furthermore, although T cells derived from the spleen

FIGURE 4. Activated T cells that lack Geminin expression show reduced proliferation and are unable to undergo multiple divisions. A, T cells that lack Geminin expression show reduced [3H]thymidine incorporation. Proliferation of T cells in the presence of anti-CD3 (1 mg/ml) and anti-CD28 (2 mg/ml) mAbs, measured by [3H]thymidine. Values indicate average counts (6 SE) from a representative experiment of three. B, T cells from control (Fl/wt, Fl/wt CD2Cre) and Fl/koCD2Cre animals were stained with CFSE (1 mM) and cultured in vitro in the presence of plate-bound anti-CD3 (1 mg/ ml) plus anti-CD28 (2 mg/ml) for 96 h. CFSE dilution histograms gated on CD4+CD25high (left-hand column) and CD8+CD25high (right-hand column) populations are shown. pppp , 0.001. The Journal of Immunology 2437

FIGURE 5. T cells that lack Geminin expression show abnormal cell cycle proﬁle and reduced ability of re-entering into cell cycle. A, Cell cycle progression of asynchronous T cells from control and mutant animals stimulated in vitro with anti-CD3 (10 mg/ml) plus anti- CD28 (2 mg/ml) was examined. Cells were pulsed with BrdU (10 mM) for 1 h and then stained with anti-BrdU FITC and 7-AAD followed by FACS analysis. Flow cytometric analysis of BrdU and 7-AAD at 48 h is Downloaded from shown. The percentage of cells in each phase of the cell cycle is shown. B, Puriﬁed naive T cells from Fl/ koCD2Cre and Fl/wtCD2Cre control animals were stimulated with plate-bound anti-CD3 (10 mg/ml) and anti-CD28 (2 mg/ml) for 38 h. Cells were subsequently pulsed with BrdU (20 mM) for 2 h. BrdU was then re-

moved, and cells were ﬁxed and stained with anti-BrdU http://www.jimmunol.org/ FITC and 7-AAD at the indicated time points (2 h, 8 h, and 16 h following BrdU addition). Histogram distribution of 7-AAD staining in the BrdU+ fraction of cells is shown. C, The percentage of BrdU+ cells that are not expressing Ki67 (Ki672) is shown at the indicated time points (8 h and 16 h after BrdU addition). Control mice and Fl/koCD2Cre mice are indicated as black and white columns, respectively. by guest on October 1, 2021

of control mice show high expression of Geminin following do not show statistically significant differences. Further examina- in vitro activation with CD3/CD28 or PDBU/ionomycin, T cells tion of DN subsets shows that DN1 and DN4 cell populations ex- derived from Fl/koCD2Cre mice show no detectable Geminin hibit a statistically significant reduction of 24% (0.721 3 106, expression upon activation (Fig. 1E). We therefore conclude that n = 17, versus 0.951 3 106, n = 12; t test, p = 0.04) and 19% (2.75 3 expression of Cre recombinase under the control of human CD2 106, n = 17, versus 3.4 3 106, n = 12; t test, p = 0.03), respectively regulatory elements leads to Geminin deficiency in progenitor (Fig. 2A). Our results indicate that T cell development in the thymus T cell populations and their progeny. is not drastically affected in the absence of Geminin because all T cell subtypes are generated, showing only a moderate, statisti- Absence of Geminin does not alter the T cell developmental cally significant reduction in DN1, DN4, and DP cell populations. program, but reduces T cell populations of thymus and spleen FACS analysis of T cell populations in the periphery showed that Fl/koCD2Cre mice lacking Geminin expression in the T cell the CD4 and CD8 compartments of the Fl/koCD2Cre, when com- compartment develop and grow apparently normally. The total pared with control mice, are reduced by ∼35% (0.89 3 107, n = 17, number of thymocytes is reduced by 21% in Fl/koCD2Cre compared versus 1.36 3 107, n = 12; t test, p = 0.004) and 33% (4.73 3 106, n = with Fl/wt mice (2.55 3 108, n = 17, versus 3.24 3 108, n = 12; t test, 17, versus 7.01 3 106, n = 12; t test, p = 0.002), respectively (Fig. 2B). p = 0.009). Quantification of individual thymocyte subsets by FACS The heterogeneous circulating T cell populations are charac- using the CD4 and CD8 Abs showed that in mice deficient for terized by differences in life span and turnover rate (30). Looking Geminin, the DP population that lacks Geminin expression shows at specific T cell populations, we examined the effects of the a 22% decrease (1.84 3 108, n = 17, versus 2.36 3 108, n = 12; t absence of Geminin in naive, memory, and regulatory T cells. Fl/ test, p = 0.009) (Fig. 2A), whereas DN, CD4 SP, and CD8 SP cells koCD2Cre naive CD4 or CD8 T cells not expressing CD44 and 2438 CONDITIONAL INACTIVATION OF GEMININ

CD25 expression is upregulated as efficiently as in WT T cells in response to various stimuli in vitro (data not shown). We conclude that absence of Geminin leads to a reduction of at least three specific T cell subpopulations in spleen. Proliferation defects of T cells that lack Geminin As Geminin is involved in cell cycle control, we tested the ability of Geminin negative lymphocytes to proliferate. For this purpose, we measured the incorporation of [3H]thymidine into splenic T cells in response to mitogenic stimulation. T cells from Geminin-deficient mice show .60% reduction in thymidine uptake poststimulation for 64 h with anti-CD3 (1 mg/ml) plus anti-CD28 (2 mg/ml) Abs (Fig. 4A). To examine the possibility that TCR signaling might be impaired in the Geminin-deficient T cells, we examined changes in expression of cell surface molecules in Geminin-deficient cells. Following stimulation of T cells with anti-CD3 and anti-CD28 or with PDBU and ionomycin or Con A, the expression of the IL-2R (CD25), the upregulation of the early activation marker CD69, and CD44 levels were measured at different time points by FACS analysis. The induction of these cell surface markers did not show Downloaded from any statistically significant difference when T cells from control and Fl/koCD2Cre animals were compared (Supplemental Fig. 2 and data not shown). These results suggest that in Geminin- deficient T cells, TCR activation signals are efficiently transduced, resulting in appropriate upregulation of immediate response genes. To distinguish whether the lower incorporation of [3H]thymidine http://www.jimmunol.org/ was due to impairment of cell division and/or enhanced cell death, we measured cell death by measuring Annexin V binding and FIGURE 6. T cells that lack Geminin expression fail to undergo efficient caspase 3 activation. Fl/wt and Fl/koCD2Cre T cells showed similar homeostatic proliferation and to repopulate the spleen. T cells from control Annexin Vand caspase 3 staining after 24 or 48 h of TCR activation (Fl/ko, wt/koCD2Cre) and Fl/koCD2Cre mice were labeled with CFSE 2 2 (Supplemental Fig. 3 and data not shown), suggesting that Gem- (5 mΜ) and transferred into Rag2KOgc / mice. Splenocytes were recovered after 3 (first column) and 5 (second column) days, and CD8 T cells inin-deficient cells do not show enhanced cell death. To assess cell were analyzed by flow cytometry for CFSE dilution (n = 3). Two weeks division more precisely, LN T cells from Fl/koCD2Cre and Fl/wt were purified, labeled with CFSE and stimulated with plate-bound postinjections, CD4 and CD8 T cells that have repopulated spleens from by guest on October 1, 2021 control and mutant animals were measured (n = 4). Control mice and Fl/ Abs against CD3 and CD28. At 24 h postactivation, no obvious koCD2Cre mice are indicated as black and white columns, respectively. division was observed for either control or mutant T cells (data not Values indicate average number of cells (6 SE). ppp , 0.01. shown). At 96 h following stimulation, the majority of CD4-expressing cells derived from control animals had undergone up to four divisions, whereas only a small fraction of CD4+ cells lacking expressing high levels of CD62L, when compared with control Geminin had divided up to three times (Fig. 4B). Moreover, a sig- mice, show a reduction of 43% (2.83 3 106, n =10,versus nificant percentage of CD8+ T cells from WT mice performed four 5.03 3 106, n =6;t test, p = 0.002) and 32% (2.11 3 106, n = divisions, whereras CD8 T cells derived from Fl/koCD2Cre mice 10, versus 3.11 3 106, n =6;t test, p = 0.008), respectively (Fig. remained undivided (Fig. 4B). These data show that Geminin-de- 3A,3B). We have also examined the CD4+CD62L2 and CD8+ ficient T cells exhibit severe proliferation defects following TCR CD62L2 compartments. We find that CD4+CD62L2 cells are activation. reduced by 18% in Fl/koCD2Cre animals compared with con- To delineate defects in specific phases of the cell cycle for the trols (2.93 3 106, n = 10, versus 3.56 3 106, n =6;p = 0.38), Geminin-deficient cells, T cells from spleen were activated with whereas CD8+CD62L2 cells are reduced by 33% in Fl/koCD2- soluble anti-CD3 and anti-CD28 for 48 h and then pulse-labeled for Cre animals (1.26 3 106, n = 10, versus 1.88 3 106, n =6;p = 1 h with BrdU, fixed, permeabilized, and stained with 7-AAD and 0.04) (data not shown). High levels of CD44 expression are anti-BrdU FITC. Flow cytometry showed reduced BrdU in- characteristic of activated or memory type T cells. CD4+ corporation in Geminin-deficient cells (27% of Fl/koCD2Cre CD44high cells in Fl/koCD2Cre show a reduction of 52% when compared with 39% of Fl/wt T cells) and a greater percentage of the compared with the Fl/wt mice (1.42 3 106, n =8,versus2.963 Fl/ko cells remainining in G0/G1 phases (43.7% versus 39.6%). 106, n =8;t test, p = 0.009) (Fig. 3C). In normal mice, around Moreover, Geminin-deficient T cells showed an increased per- 10% of the CD4 T cells express the IL-2R a-chain (IL-2Ra; centage of cells at the G2/M phase (3.33% compared with 1.83% CD25). These CD4+CD25+ T cells contain a population of for control cells) (Fig. 5A). The presence of apoptotic or sub-G0 forkhead box p3+ (Foxp3+) cells known as regulatory T cells that cells was similar in T cell populations from both groups of mice, have the capacity to modulate immune responses and inhibit in agreement with our results for the Annexin V staining. More- development of autoimmune diseases (31, 32). FACS analysis over, we do not observe any statistically significant difference in showed that in the spleen of Fl/koCD2Cre mice, CD4+CD25+ cells with DNA content .4 N, suggesting that extended over- Foxp3+ T cells are reduced by ∼35% (0.58 3 106, n =3,versus replication does not take place in activated T cells derived from Fl/ 0.97 3 106, n =6;t test, p = 0.046) when compared with the koCD2Cre mice (data not shown). To follow a T cell population control mice (Fig. 3D). Notably, this does not reflect an inability during the cell cycle, naive T cells were stimulated using Abs of Fl/koCD2Cre T cells to express CD25 because in these cells, against CD3 and CD28 for 38 h and pulse-labeled with 20 mM The Journal of Immunology 2439

BrdU for 2 h. BrdU was then removed and cells were fixed at WT and defective for Geminin T cells to undergo homeostatic different time points (2 h, 8 h, 16 h). Immediately after BrdU proliferation. The CFSE dilution profile of CD8+ cells was mea- + removal, 7-AAD staining of BrdU+ T cells that derived from sured by FACS at 3 and 5 d posttransfer. Control CD8 T cells had control and mutant mice show similar numbers in different cell divided at least five times by 3 d, whereas the majority of T cells cycle phases (Fig. 5B, 2 h), suggesting that naive T cells that lack that lack Geminin expression had undergone a single division Geminin are entering into S phase as efficiently as the WT T cells. with only a small percentage of cells undergoing two or more. At + . At 8 h after BrdU addition, .49% of the BrdU+ T cells derived day 5, control CD8 T cells had undergone 11 divisions, with from Fl/koCD2Cre mice remain in S/G2/M phase compared with the vast majority of the Fl/koCD2Cre T cells having undergone less than four divisions (Fig. 6). Spleens were isolated from 23% of T cells derived from control animals (Fig. 5B, 8 h), sug- 2/2 gesting a delay of T cells that lack Geminin to progress to G1, Rag2KOgc host mice 2 wk posttransfer of T cells, and the whereas similar distribution was also observed at 16 h (Fig. 5B, CD4 and CD8 T cell populations that repopulated the recipient . mice were assessed by FACS. The numbers of CD4 and CD8 16 h). Furthermore, at the same time point, 18% of the BrdU- 2/2 labeled T cells derived from Fl/koCD2Cre mice do not express Ki67 T cells that were recovered from Rag2KOgc host spleens and 3 6 compared with 7.0% in the control T cells, indicating an increased derived from Fl/koCD2Cre are reduced by 51% (0.48 10 , n = 3 6 3 6 percentage of cells that withdraw from the cell cycle (Fig. 5C). Our 4, versus 0.99 10 , n =4;t test, p = 0.06) and 72% (0.43 10 , 3 6 data suggest that activated T cells lacking Geminin expression show n =4,versus1.61 10 , n =4;t test, p = 0.007), respectively, no defect in entry into the S phase, but they exhibit a partial im- when compared with the control mice (Fig. 6). These results pairment in exiting from the S/G2/M phase and entering into a sec- suggest that in lymphopenic environments, T cells that lack Geminin are not able to divide as efficiently as WT T cells, con- ond division, without a significant increase in ploidy or apoptosis. Downloaded from firming the significant defect in T cell proliferation kinetics seen Defective homeostatic mechanisms in Geminin-deficient T cells during homeostatic T cell proliferation. The overall size and composition of the peripheral T cell pool are controlled by homeostatic mechanisms that regulate cell survival Stimulated T cells that lack Geminin show increased Cdt1 and basal steady-state proliferation (33, 34), resulting in relatively expression levels and accumulate in G2 phase constant size. Thus, when the T cell pool is severely depleted, the Geminin exerts its role during the cell cycle by binding to and http://www.jimmunol.org/ remaining T cells undergo spontaneous proliferation (35, 36). To inhibiting Cdt1 (1), which is in addition negatively regulated by assess the ability of T cells that lack Geminin to proliferate under proteolysis through the Skp2/SCF and Cul4/DDB1 ubiquitin li- such conditions, we adoptively transferred into lymphopenic gases (37–39). To investigate whether thymocytes and stimulated Rag2KOgc2/2 recipients an equal number of CFSE-labeled T cells T cells may differ in their ability to regulate Cdt1 levels in the derived from Fl/koCD2Cre or control LNs. Tracking CFSE dilution absence of Geminin, protein extracts derived from WT and in adoptive transfer experiments allows us to study the ability of Geminin-deficient thymocytes and stimulated T cells were subjected to Western blotting analysis using Abs against Cdt1 (Fig. 7A). Cdt1 protein expression is increased in Geminin-deficient

peripheral T cells activated with anti-CD3/CD28 for 16 h when by guest on October 1, 2021 compared with stimulated T cells derived from control animals (Fig. 7A). However, Cdt1 protein expression in WT and mutant thymocytes is similar (Fig. 7A). These results suggest that differences in Cdt1 protein levels may, at least partially, account for the defects that we observed in activated T cells for Geminin. Our analysis shows that developing thymocytes and peripheral T cells exhibit a differential requirement for Geminin for efficient proliferation. Activated T cells that lack Geminin expression accumulate in S/G2/M phase of the cell cycle. To further investigate this cell cycle defect and address the underlying mechanism, we have examined the expression levels of molecules controlling the G1 to S and G2 to M transition. Naive T cells derived from control and Fl/koCD2Cre mice were stimulated with anti-CD3/CD28 Abs, and total protein extracts were isolated at 0, 16, and 44 h poststimulation and analyzed by Western blotting. At 44 h, expression of cyclin D3, cyclin E, phosphorylated Rb, and p27 expression appear similar, whereas expression of cyclin B1, cyclin A, and tyrosine 15 phosphorylated cdc2 are increased in activated T cells that lack Geminin expression (Fig. 7B). This suggests that stimulated T cells lacking Geminin accumulate in G2 phase before G2/M transition. FIGURE 7. Stimulated peripheral T cells show increased Cdt1 expression levels in the absence of Geminin and accumulation in G2. A, Protein Discussion lysates were prepared from thymocytes and naive peripheral T cells acti- Geminin has been implicated in the control of DNA replication and vated with anti-CD3/CD28 (10 mg/ml and 2 mg/ml) for 16 and 44 h. Western blotting analysis was performed using Abs against Cdt1 and actin. B, neural cell fate acquisition and is believed to work as a molecular Western blotting analysis was performed in protein extracts from un- link between proliferation and differentiation (4). Inactivation and stimulated naive peripheral T cells (0 h) and stimulated with anti-CD3/ overexpression experiments in cell lines and model organisms have CD28 for 16 and 44 h using Abs against cyclins D3, E, A, and B1, p27, Rb suggested an essential role for Geminin in establishing neural fate (pSer807/811), Cdc2 (pTyr15), and actin. determination and maintaining genomic integrity. The complete 2440 CONDITIONAL INACTIVATION OF GEMININ inactivation of the mouse Geminin gene leads to early lethality undergo a progressive commitment, maturation, and proliferation before the blastocyst stage and, therefore, cannot address the in the absence of Geminin. Increased protein expression levels of in vivo function of Geminin in regulating maintenance and dif- Cdt1 detected in stimulated T cells could partly account for the ferentiation of progenitor cells, organ development, and homeo- sensitivity of activated T cells to the lack of Geminin compared stasis (12, 13, and data in the current study). To study the in vivo with thymocytes. An alternative possibility could be that in vivo role of Geminin, we have specifically deleted Geminin from T cells proteolytic regulation of Cdt1 by two distinct E3 ubiquitin ligase using Cre recombinase under the regulation of the human CD2 systems (39) is able to restrict Cdt1 activity efficiently, and LCR and promoter sequences. Contrary to what has been suggested therefore Geminin in T cell development operates as a redundant for the role of Geminin in other experimental systems, conditional control (45–47). Furthermore, the relative contribution of multiple inactivation of Geminin in the lymphoid system appears not to be redundant pathways that regulate licensing in parallel to Geminin essential for the maintenance and differentiation processes of the (51) could also differ between cells at different stages of de- T cell lineage in the thymus, because all progenitor populations are velopment, thus rendering different cells differentially affected by present. DN1, DN4, and DP T cells show a small reduction in cell the absence of Geminin. number, suggesting that Geminin does not have a major role in In this report, we provide in vivo evidence that during T cell cellular divisions of T cell progenitors in the thymus. It is unlikely development, Geminin is not essential for progenitor T cell pro- that the lack of phenotype during T cell differentiation can be at- liferation, commitment, and specification. Our results suggest tributed to small amounts of protein remaining after Geminin de- a differentiation stage-specific role for Geminin similar to other letion, because previous studies in tumorigenic cell lines have central cell cycle regulators, such as cdk2 and cyclin E, that have shown that Geminin is an unstable protein; it is rapidly degraded been previously shown to have cell type-specific effects in vivo following ubiquitination from the anaphase promoting complex (52). Differences in Geminin requirements by developing thy- Downloaded from during mitosis and G1 (40), whereas even during S and G2 phases, mocytes could reflect the existence of alternative mechanisms that Geminin has a short t1/2 (41, 42). Therefore, our data indicate that operate redundantly or compensatory and control genomic sta- in vivo, Geminin may not be essential for the proliferation and bility, progenitor proliferation, and differentiation in these cells differentiation of T cell progenitor populations. Our data do not (47, 53, 54). The mice we report in this study, which specifically exclude that Geminin may be critical at an earlier stage of he- lack Geminin in T cells, may prove a valuable system to probe matopoietic system development, as previously suggested for these various pathways, but also assess the effects of Geminin http://www.jimmunol.org/ Geminin-Polycomb complex interactions (11). absence on the immune response and tumorigenesis. Moreover, In contrast to progenitor T cells in the thymus, we show that these mice might be a useful model to study immune system ac- peripheral T cells lacking Geminin expression present a dramatic tivation in the absence of efficient T cell expansion. defect in proliferation both in vitro, following TCR activation, and in vivo, following adoptive transfer into lymphopenic recipient Disclosures mice. Furthermore, we show that activated T cells lacking Geminin The authors have no financial conflicts of interest. are able to enter S phase; however, they show reduced ability to perform multiple rounds of division. T cells lacking Geminin show by guest on October 1, 2021 abnormal cell cycle profiles, with increased number of cells in G2/ References M following activation. Increased expression of cyclin B1 and the 1. Nishitani, H., and Z. Lygerou. 2004. DNA replication licensing. Front. Biosci. 9: 15 2115–2132. Y phosphorylated form of cdc2 suggests that activated T cells 2. Petropoulou, C., P. Kotantaki, D. Karamitros, and S. Taraviras. 2008. Cdt1 and are accumulated in G2 phase of the cell cycle in the absence of Geminin in cancer: markers or triggers of malignant transformation? Front. Geminin. G2 cell cycle arrest due to lack of Geminin has been Biosci. 13: 4485–4494. 3. Spella, M., O. Britz, P. Kotantaki, Z. Lygerou, H. Nishitani, R. G. Ramsay, linked with aberrant replication triggered by uncontrolled Cdt1 C. Flordellis, F. Guillemot, T. Mantamadiotis, and S. Taraviras. 2007. Licensing activity and induction of DNA damage response. This finding is regulators Geminin and Cdt1 identify progenitor cells of the mouse CNS in reminiscent of Geminin depletion experiments by antisense oli- a specific phase of the cell cycle. Neuroscience 147: 373–387. 4. Kroll, K. L. 2007. Geminin in embryonic development: coordinating transcrip- gonucleotides in Xenopus laevis oocytes, which also resulted in tion and the cell cycle during differentiation. Front. Biosci. 12: 1395–1409. G2 arrest (43). Defects during mitosis were also described fol- 5. Kroll, K. L., A. N. Salic, L. M. Evans, and M. W. Kirschner. 1998. Geminin, lowing RNA interference depletion of Geminin in U2OS, HCT116, a neuralizing molecule that demarcates the future neural plate at the onset of gastrulation. Development 125: 3247–3258. and TIG3 human cell lines, suggesting a role of Geminin for cen- 6. Papanayotou, C., A. Mey, A. M. Birot, Y. Saka, S. Boast, J. C. Smith, J. Samarut, trosome duplication and chromosome segregation (44). In Dro- and C. D. Stern. 2008. A mechanism regulating the onset of Sox2 expression in sophila (45), several human cell lines (46, 47), and early embryonic the embryonic neural plate. PLoS Biol. 6: e2. divisions in the mouse (12, 13), absence of Geminin led to over- 7. Quinn, L. M., A. Herr, T. J. McGarry, and H. Richardson. 2001. The Drosophila Geminin homolog: roles for Geminin in limiting DNA replication, in anaphase replication of the genome, in accordance with Geminin’s role as and in neurogenesis. Genes Dev. 15: 2741–2754. a negative regulator of the licensing factor Cdt1 (1), accompanied 8. Del Bene, F., K. Tessmar-Raible, and J. Wittbrodt. 2004. Direct interaction of in some systems by DNA damage and signs of apoptosis (46, 47). geminin and Six3 in eye development. Nature 427: 745–749. 9. Luo, L., X. Yang, Y. Takihara, H. Knoetgen, and M. Kessel. 2004. The cell-cycle We did not observe a detectable increase in ploidy, apoptosis in regulator geminin inhibits Hox function through direct and polycomb-mediated T cells lacking Geminin, though partial rereplication cannot be interactions. Nature 427: 749–753. excluded. Interestingly, a similar lack of obvious rereplication in 10. Seo, S., A. Herr, J. W. Lim, G. A. Richardson, H. Richardson, and K. L. Kroll. 2005. Geminin regulates neuronal differentiation by antagonizing Brg1 activity. the absence of Geminin was noted in HeLa cells (48, 49). Genes Dev. 19: 1723–1734. Activated T cells labeled for BrdU exhibited a decrease in Ki67 11. Ohtsubo, M., S. Yasunaga, Y. Ohno, M. Tsumura, S. Okada, N. Ishikawa, expression in the absence of Geminin, suggestive of cell cycle K. Shirao, A. Kikuchi, H. Nishitani, M. Kobayashi, and Y. Takihara. 2008. withdrawal. Our results suggest that peripheral T cells undergoing Polycomb-group complex 1 acts as an E3 ubiquitin ligase for Geminin to sustain hematopoietic stem cell activity. Proc. Natl. Acad. Sci. USA 105: 10396–10401. rapid divisions upon TCR activation or homeostatic proliferation 12. Gonzalez, M. A., K. E. Tachibana, D. J. Adams, L. van der Weyden, M. Hemberger, require Geminin for efficient proliferation and repeated divisions, N. Coleman, A. Bradley, and R. A. Laskey. 2006. Geminin is essential to prevent and its absence causes cell cycle delay and promotes cell cycle exit. endoreduplication and to form pluripotent cells during mammalian development. Genes Dev. 20: 1880–1884. In contrast, common lymphoid progenitors that migrate into the 13. Hara, K., K. I. Nakayama, and K. Nakayama. 2006. Geminin is essential for the thymus from the bone marrow (50) appear to a large extent able to development of preimplantation mouse embryos. Genes Cells 11: 1281–1293. The Journal of Immunology 2441

14. Aifantis, I., M. Mandal, K. Sawai, A. Ferrando, and T. Vilimas. 2006. Regulation 34. Jameson, S. C. 2002. Maintaining the norm: T-cell homeostasis. Nat. Rev. Im- of T-cell progenitor survival and cell-cycle entry by the pre-T-cell receptor. munol. 2: 547–556. Immunol. Rev. 209: 159–169. 35. Bell, E. B., S. M. Sparshott, M. T. Drayson, and W. L. Ford. 1987. The stable and 15. Mo¨roÿ, T., and H. Karsunky. 2000. Regulation of pre-T-cell development. Cell. permanent expansion of functional T lymphocytes in athymic nude rats after Mol. Life Sci. 57: 957–975. a single injection of mature T cells. J. Immunol. 139: 1379–1384. 16. Pear, W. S., L. Tu, and P. L. Stein. 2004. Lineage choices in the developing 36. Li, C. R., S. Santoso, and D. D. Lo. 2007. Quantitative analysis of T cell ho- thymus: choosing the T and NKT pathways. Curr. Opin. Immunol. 16: 167–173. meostatic proliferation. Cell. Immunol. 250: 40–54. 17. Tourigny, M. R., S. Mazel, D. B. Burtrum, and H. T. Petrie. 1997. T cell receptor 37. Li, X., Q. Zhao, R. Liao, P. Sun, and X. Wu 2003. The SCF(Skp2) ubiquitin (TCR)-beta gene recombination: dissociation from cell cycle regulation and de- ligase complex interacts with the human replication licensing factor Cdt1 and velopmental progression during T cell ontogeny. J. Exp. Med. 185: 1549–1556. regulates Cdt1 degradation. J. Biol. Chem. 278: 30854–30858. 18. Godfrey, D. I., J. Kennedy, T. Suda, and A. Zlotnik. 1993. A developmental 38. Takeda, DY, JD Parvin, and A Dutta. 2005. Degradation of Cdt1 during S phase pathway involving four phenotypically and functionally distinct subsets of CD3- is Skp2-independent and is required for efficient progression of mammalian cells CD4-CD8- triple-negative adult mouse thymocytes defined by CD44 and CD25 through S phase. J. Biol. Chem. 280: 23416–23423. expression. J. Immunol. 150: 4244–4252. 39. Nishitani, H., N. Sugimoto, V. Roukos, Y. Nakanishi, M. Saijo, C. Obuse, 19. Hoffman, E. S., L. Passoni, T. Crompton, T. M. Leu, D. G. Schatz, A. Koff, T. Tsurimoto, K. I. Nakayama, K. Nakayama, M. Fujita, et al. 2006. Two E3 M. J. Owen, and A. C. Hayday. 1996. Productive T-cell receptor beta-chain gene ubiquitin ligases, SCF-Skp2 and DDB1-Cul4, target human Cdt1 for proteolysis. rearrangement: coincident regulation of cell cycle and clonality during de- EMBO J. 25: 1126–1136. velopment in vivo. Genes Dev. 10: 948–962. 40. McGarry, T. J., and M. W. Kirschner. 1998. Geminin, an inhibitor of DNA 20. Dudley, E. C., H. T. Petrie, L. M. Shah, M. J. Owen, and A. C. Hayday. 1994. replication, is degraded during mitosis. Cell 93: 1043–1053. T cell receptor beta chain gene rearrangement and selection during thymocyte 41. Nishitani, H., S. Taraviras, Z. Lygerou, and T. Nishimoto. 2001. The human development in adult mice. Immunity 1: 83–93. licensing factor for DNA replication Cdt1 accumulates in G1 and is destabilized 21. von Boehmer, H., I. Aifantis, F. Gounari, O. Azogui, L. Haughn, I. Apostolou, after initiation of S-phase. J. Biol. Chem. 276: 44905–44911. E. Jaeckel, F. Grassi, and L. Klein. 2003. Thymic selection revisited: how es- 42. Kulartz, M., S. Kreitz, E. Hiller, E. C. Damoc, M. Przybylski, and R. Knippers. sential is it? Immunol. Rev. 191: 62–78. 2003. Expression and phosphorylation of the replication regulator protein 22. Godfrey, D. I., and A. Zlotnik. 1993. Control points in early T-cell development. geminin. Biochem. Biophys. Res. Commun. 305: 412–420. Immunol. Today 14: 547–553.

43. McGarry, T. J. 2002. Geminin deficiency causes a Chk1-dependent G2 arrest in Downloaded from 23. Peńit, C., B. Lucas, and F. Vasseur. 1995. Cell expansion and growth arrest Xenopus. Mol. Biol. Cell 13: 3662–3671. phases during the transition from precursor (CD4-8-) to immature (CD4+8+) 44. Tachibana, K. E., M. A. Gonzalez, G. Guarguaglini, E. A. Nigg, and R. A. Laskey. thymocytes in normal and genetically modified mice. J. Immunol. 154: 5103– 2005. Depletion of licensing inhibitor geminin causes centrosome overduplication 5113. and mitotic defects. EMBO Rep. 6: 1052–1057. 24. de Boer, J., A. Williams, G. Skavdis, N. Harker, M. Coles, M. Tolaini, T. Norton, 45. Mihaylov, I. S., T. Kondo, L. Jones, S. Ryzhikov, J. Tanaka, J. Zheng, L. A. Higa, K. Williams, K. Roderick, A. J. Potocnik, and D. Kioussis. 2003. Transgenic N. Minamino, L. Cooley, and H. Zhang. 2002. Control of DNA replication and mice with hematopoietic and lymphoid specific expression of Cre. Eur. J. Im- Mol. Cell. Biol. munol. 33: 314–325. chromosome ploidy by geminin and cyclin A. 22: 1868–1880. 46. Melixetian, M., A. Ballabeni, L. Masiero, P. Gasparini, R. Zamponi, J. Bartek,

25. O’Gorman, S., N. A. Dagenais, M. Qian, and Y. Marchuk. 1997. Protamine-Cre http://www.jimmunol.org/ recombinase transgenes efﬁciently recombine target sequences in the male germ J. Lukas, and K. Helin. 2004. Loss of Geminin induces rereplication in the line of mice, but not in embryonic stem cells. Proc. Natl. Acad. Sci. USA 94: presence of functional p53. J. Cell Biol. 165: 473–482. 14602–14607. 47. Zhu, W., Y. Chen, and A. Dutta. 2004. Rereplication by depletion of geminin is 26. Mantamadiotis, T.,S.Taraviras, F.Tronche,and G.Schu¨tz. 1998. PCR-based strategy seen regardless of p53 status and activates a G2/M checkpoint. Mol. Cell. Biol. for genotyping mice and ES cells harboring loxP sites. Biotechniques 25: 968–970, 24: 7140–7150. 972. 48. Machida, Y. J., and A. Dutta. 2007. The APC/C inhibitor, Emi1, is essential for 27. Colucci, F., C. Soudais, E. Rosmaraki, L. Vanes, V. L. Tybulewicz, and J. P. Di prevention of rereplication. Genes Dev. 21: 184–194. Santo. 1999. Dissecting NK cell development using a novel alymphoid mouse 49. Nishitani, H., Z. Lygerou, and T. Nishimoto. 2004. Proteolysis of DNA repli- model: investigating the role of the c-abl proto-oncogene in murine NK cell cation licensing factor Cdt1 in S-phase is performed independently of geminin differentiation. J. Immunol. 162: 2761–2765. through its N-terminal region. J. Biol. Chem. 279: 30807–30816. 28. Xouri, G., Z. Lygerou, H. Nishitani, V. Pachnis, P. Nurse, and S. Taraviras. 2004. 50. Shortman, K., M. Egerton, G. J. Spangrude, and R. Scollay. 1990. The genera-

Cdt1 and geminin are down-regulated upon cell cycle exit and are over- tion and fate of thymocytes. Semin. Immunol. 2: 3–12. by guest on October 1, 2021 expressed in cancer-derived cell lines. Eur. J. Biochem. 271: 3368–3378. 51. Sugimoto, N., K. Yoshida, Y. Tatsumi, T. Yugawa, M. Narisawa-Saito, S. Waga, 29. Greaves, D. R., F. D. Wilson, G. Lang, and D. Kioussis. 1989. Human CD2 39- T. Kiyono, and M. Fujita. 2009. Redundant and differential regulation of mul- flanking sequences confer high-level, T cell-specific, position-independent gene tiple licensing factors ensures prevention of re-replication in normal human cells. expression in transgenic mice. Cell 56: 979–986. J. Cell Sci. 122: 1184–1191. 30. Tough, D. F., and J. Sprent. 1994. Turnover of naive- and memory-phenotype 52. Berthet, C., and P. Kaldis. 2007. Cell-specific responses to loss of cyclin- T cells. J. Exp. Med. 179: 1127–1135. dependent kinases. Oncogene 26: 4469–4477. 31. Shevach, E. M. 2002. CD4+ CD25+ suppressor T cells: more questions than 53. Neganova, I., and M. Lako. 2008. G1 to S phase cell cycle transition in somatic answers. Nat. Rev. Immunol. 2: 389–400. and embryonic stem cells. J. Anat. 213: 30–44. 32. Read, S., and F. Powrie. 2001. CD4(+) regulatory T cells. Curr. Opin. Immunol. 54. Neganova, I., X. Zhang, S. Atkinson, and M. Lako. 2008. Expression and 13: 644–649. functional analysis of G1 to S regulatory components reveals an important role 33. Surh, C. D., and J. Sprent. 2000. Homeostatic T cell proliferation: how far can for CDK2 in cell cycle regulation in human embryonic stem cells. Oncogene 28: T cells be activated to self-ligands? J. Exp. Med. 192: F9–F14. 20–30.