The 3 Require the Presence of the Hs4 Enhancer in Ig Synthesis and Class Switching Do

Ig Synthesis and Class Switching Do Not Require the Presence of the hs4 Enhancer in the 3′ IgH Regulatory Region

This information is current as Christelle Vincent-Fabert, Véronique Truffinet, Remi of October 1, 2021. Fiancette, Nadine Cogné, Michel Cogné and Yves Denizot J Immunol 2009; 182:6926-6932; ; doi: 10.4049/jimmunol.0900214 http://www.jimmunol.org/content/182/11/6926 Downloaded from

References This article cites 25 articles, 11 of which you can access for free at: http://www.jimmunol.org/content/182/11/6926.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on October 1, 2021

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 © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Ig Synthesis and Class Switching Do Not Require the Presence of the hs4 Enhancer in the 3؅ IgH Regulatory Region1

Christelle Vincent-Fabert,2 Ve´ronique Trufﬁnet,2 Remi Fiancette, Nadine Cogne´, Michel Cogne´, and Yves Denizot3

Several studies have reported that regulatory elements located 3؅ of the IgH locus (namely hs3a, hs1,2, hs3b, and hs4) might play a role during class switch recombination (CSR) and Ig synthesis. While individual deletion of hs3a or hs1,2 had no effect, pairwise deletion of hs3b (an inverted copy of hs3a) and hs4 markedly affected CSR and Ig expression. Among these two elements, hs4 was tentatively presented with the master role due to its unique status within the 3؅ regulatory region: distal position outside repeated regions, early activation in pre-B cells, strong activity throughout B cell ontogeny. To clarify its role, we generated mice with a clean deletion of the hs4 after replacement with a ﬂoxed neoR cassette. Surprisingly, and as for previous deletion of hs3a or hs1,2, deletion of hs4 did not affect either in vivo CSR or the secretion level of any Ig isotype. Downloaded from In vitro CSR and Ig secretion in response to LPS and cytokines was not affected either. The only noticeable effects of the hs4 deletion were a decrease in the number of B splenocytes and a decreased membrane IgM expression. In conclusion, while dispensable for CSR and Ig transcription in plasma cells, hs4 mostly appears to contribute to Ig transcription in resting B lymphocytes. The Journal of Immunology, 2009, 182: 6926–6932.

complex interplay of multiple regulatory elements is ments constitute a potent locus control region conferring position- responsible for the tissue-specific and stage-specific reg- independent and copy-dependent expression to transgenes (7). http://www.jimmunol.org/ A ulation of both transcription and rearrangement of the Studies with cell lines have suggested a role of the 3ЈRR on Ig IgH locus. Germline transcription of C␮, initiation of DJ and VDJ transcription. Thus, loss of the 3Ј IgH enhancers occurred con- rearrangement, expression of rearranged ␮ genes, and opening of comitantly with a dramatic decrease of Ig transcription in the the S␮ region to class switch recombination (CSR)4 mainly rely on LP1.2 IgA-secreting cell line (5, 8). Disruption of the 3ЈRR in the ␮ upstream regulatory elements such as VH promoters and the E 9921 cell line (which had an endogenous deletion of the E␮ ele- intronic enhancers (1, 2). In addition to upstream elements, a 3Ј ment) by inserting the neoR gene into hs1,2 abrogated IgH tran- regulatory region (3ЈRR) located downstream the locus has been scription (9). Transgenic mice bearing a bacterial artificial chro-

shown to include four lymphoid-specific transcriptional enhancers: mosome that included a rearranged VDJ gene and the 3ЈRR by guest on October 1, 2021 hs3a, hs1,2, hs3b, and hs4 (3). A region of open chromatin con- showed that the 3Ј end of the IgH locus enhanced germline tran- Ј taining numerous CTCF sites is found 3 of hs4 and upstream of scription and switch recombination to the four ␥ genes (10). Fi- Ј the non-B specific gene hole, thus marking the 3 terminus of the nally, studies with IgH minilocus have highlighted the role of the IgH locus (4). The hs4 distal element is active from the pre-B cell hs3b/hs4 pair in sustaining IgH gene transcription in 9921 cells stage and throughout B cell ontogeny (5). A larger region with a (11). To elucidate the function of 3Ј IgH enhancers, mice with global “palindromic” structure encompassing the hs1,2 central en- genomic mutations were generated. Class switching and Ig syn- hancer flanked by inverted repeats including hs3a and hs3b ele- thesis were normal in mice lacking hs3a or hs1,2 (12). In contrast, ments is active at late B cell stages (6). Altogether, the four ele- joint deletion of the last two 3Ј enhancer hs3b and hs4 severely impaired germline transcription and class switching to most isotypes in mice (13). Whether these effects were due to the deletion Centre National de la Recherche Scientifique Unite´Mixte de Recherche 6101, Uni- of hs3b alone, to hs4 alone, or to both regulatory elements re- versite´de Limoges, France mained an open question. Three potent arguments argued for a role Received for publication January 21, 2009. Accepted for publication March 30, 2009. mostly carried by hs4. First, knockout deletion of hs3a (which The costs of publication of this article were defrayed in part by the payment of page shares 98% of sequence similarity with hs3b) had no effect on class charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. switching and Ig synthesis (12). Second, hs3b is absent in other 1 This work was supported by grants from Ligue Contre le Cancer (Comite´De´par- species than the mouse, so that humans could be considered as temental de la Haute-Vienne et de la Corre`ze), Conseil Re´gional du Limousin, and carrying a natural knockout of hs3b (14, 15). Third, the hs4 en- “Lions Club de la Corre`ze, Zone 33 District 103 Sud”. hancer was associated with active chromatin (demethylated DNA Y.D. and M.C. wrote the paper and performed and designed research. C.V.-F., V.T., and acetylated histone H3 and H4) from pro-B cells to plasma R.F., and N.C. performed research. cells, while hs3b has become associated with activated chromatin 2 C.V.-F. and V.T. contributed equally to this work. markers later in B cell differentiation (B cell and plasma cell 3 Address correspondence and reprint requests to Dr. Yves Denizot, Centre National de la Recherche Scientifique Unite´ Mixte de Recherche 6101, Laboratoire stages) (4, 16). In the present study, we have tested the hypothesis d’Immunologie, 2, rue du Dr Marcland, 87025 Limoges Cedex, France. E-mail that hs4 is of importance for CSR and Ig production. For this address: [email protected] purpose we therefore undertook gene targeting experiments where 4 Abbreviations used in this paper: CSR, class switch recombination; ES, embryonic hs4 was either replaced with a thymidine kinase (Tk)-neoR cassette stem; LCR, locus control region; 3ЈRR, 3Ј regulatory region; wt, wild type. or deleted by the cre/loxP system. Replacement of hs4 by Neo or Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 the clean deletion of hs4 from the 3ЈRR gave the surprising result

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0900214 The Journal of Immunology 6927

that Ig synthesis and CSR were not affected as compared with 72 h of growth, cells were incubated for 15 min at 4°C with 7-aminoacti- wild-type (wt) animals. We conclude that the suggested role of hs4 nomycin D and FITC-labeled annexin V Abs (BD Biosciences) and ana- in driving CSR and/or Ig gene transcription can be largely sup- lyzed by flow cytometry. In a separate set of experiments, splenocytes were Ј treated 24 h with various concentration of etoposide before staining with planted in vivo by the activity of the remaining 3 IgH enhancers 7-aminoactinomycin D and annexin V Abs. or other elements within the IgH locus. Proliferation assay Materials and Methods ϫ 5 Vector construction, transfection, and embryonic stem (ES) cell Splenic B cells (2 10 cells/well) were cultured (in sextuplicates) in 96-well plates, either alone or in the presence of 20 ␮g/ml LPS from Sal- screening monella typhimurium,5␮g/ml anti-CD40, or 20 ␮g/ml anti-IgM for 72 h. The studies have been reviewed and approved by Centre National de la The number of viable cells in proliferation was assessed using the CellTiter Recherche Scientifique and the Universite´de Limoges review committee. 96 One Solution cell proliferation assay (Promega) according to the man- The hs4 targeting construct was similar to the one previously used for the ufacturer’s recommendations. joint deletion of hs3b and hs4 (13). The 3Ј arm was the same while the 5Ј arm was slightly longer to include hs3b and to generate an additional Spleen cell cultures for Ig production EcoR1 site facilitating the screening of recombinant clones. A neomycin ϫ 6 resistance gene (neoR) flanked by loxP sites was stuck in between the 5Ј Single-cell suspensions of spleen cells were cultured 5 days at 1 10 ␮ and 3Ј arm. At the 5Ј end, a phosphoglucokinase promoter-herpes simplex cells/ml in RPMI 1640 with 10% FCS and 40 g/ml LPS, with or without ␤ ␥ virus thymidine kinase gene (Tk) was included to permit negative selection addition of cytokines: 20 ng/ml IL-4, 2 ng/ml TGF- , or 2 ng/ml IFN- Ϫ against random integration. Cells of the ES cell line E14 were transfected (PeproTech). Supernatants were then harvested and stored at 20°C until with linearized vector by electroporation and selected using 300 ␮g/ml Ig evaluations. In parallel experiments, 4-day in vitro-stimulated splenocytes were washed in PBS and stained with various Abs: anti-B220-PC5 geneticin and 2 ␮g/ml gancyclovir. The ES cell line E14 was derived from Downloaded from the inbred mouse strain 129 (17). Southern blot analysis with two probes and anti-Ig subclasses conjugated with FITC. Cells were analyzed on a 5Ј and 3Ј of the construct-identified recombinants. An ES clone showing Coulter XL apparatus (Beckman Coulter). homologous recombination of hs4 was injected into C57BL/6 blastocysts, and the resulting chimeras were mated with C57BL/6 animals. Germline ELISA assays transmission in heterozygous and homozygous mutant mice was checked by Southern blot. In parallel, mutant mice were mated with cre transgenic Culture supernatants and sera from transgenic and wt mice were analyzed mice. The progeny was checked by PCR for the occurrence of a cre-me- for the presence of the various Ig classes (IgM, IgG1, IgG2b, IgG2b, IgG3, diated deletion. IgE, and IgA) by ELISA as previously described (13). http://www.jimmunol.org/

Southern blot analysis Immunization and Ab determinations Genomic DNA was extracted from ES cell clones and mouse progeny. For immunization experiments, batches of 8-wk-old mice were used (five DNA was digested with EcoRI loaded on 0.7% agarose gels and transferred ⌬/⌬ mice and five wt mice). The first immunization was performed with 50 on nylon sheets (Amersham Biosciences). Southern blot analysis with two ␮g of chicken OVA per animal in 50% CFA, and a second immunization probes (5Ј and 3Ј) of the construct identified recombinants (a 9.7-kb band was realized 13 days later with the same amount of Ag in 50% IFA. and 8.2-kb band for the neo replacement with 5Ј probe and 3Ј probe, re- Immunized mice were eye-bled at various intervals during the immu- spectively). An hs4 probe was also used to confirm hs4 deletion in trans- nization protocol, and sera were analyzed for the presence of OVA- genic animals.

specific IgM and IgG1 by ELISA. In a second set of experiments, im- by guest on October 1, 2021 ␮ PCR analysis of hs4 and cre-mediated deletion munization was performed with 50 g of DNP-Ficoll per animal in IFA followed by the same injection 13 days later. Animal sera were ana- PCR experiments for hs4 were conducted with specific sense 5Ј- lyzed for the presence of DNP-specific IgM and IgG3 by ELISA. CCAAAAATGGCCAGG CCTAGG-3Ј and anti-sense 5Ј-AGGTCTACA ELISA assays were performed as previously described (18). The pos- CAGGGGCTCTG-3Ј primers. DNA was denatured 180 s at 94°C, and then itive control consisted of a pool of sera from OVA- or DNP-Ficoll submitted to 30 cycles consisting of 94°C for 30 s, 61°C for 45 s, and 72°C immunized wt mice; the same control serum was used in all ELISA for 60 s. Amplification products were analyzed on a 1.2% agarose gel. assays. Dilution of the control serum allowed drawing a titration curve. Expected size of amplificated product was 602 bp. PCR experiments for A 1/100 to 1/2700 dilution of each serum was compared with the ti- cre-mediated deletion were conducted with specific sense 5Ј-GGAT tration curve obtained on the same multiwell plate, which allowed the GATCTGGACGAAGAGCAT-3Ј and anti-sense 5Ј-TCCCCTCAGAA quantification of OVA-specific Abs in arbitrary units. GAACTCGTCAA-3Ј primers located in the neo gene. DNA was denatured 180 s at 94°C, and then submitted to 30 cycles consisting of 94°C for 30 s, 55°C for 45 s, and 72°C for 60 s. Amplification products were analyzed on Results a 0.7% agarose gel. Expected size of amplificated product was 336 bp. Replacement and deletion of the hs4 enhancer within the 3Ј IgH locus control region (LCR) in ES cells and generation of mutant Blood sampling mice Blood samples were recovered from transgenic mice and wt controls with heparinized needles. Ten-week-old animals were used. Plasma samples A gene-targeting vector was constructed to replace the genomic R were recovered by centrifugation and stored at Ϫ20°C until used. fragment encompassing hs4 with a loxP/neo gene cassette (Fig. 1). Southern blotting and hybridization with two flanking probes Flow cytometry analysis located outside of the construct allowed the selection of 2 out of Single-cell suspensions from spleen and bone marrow were labeled with 2500 clones, which gave hybridizing fragments of the sizes ex- various Abs conjugated with Spectral Red (PC5), PE, and PE-Cy7 pected for the Nhs4 mutation (N), that is, replacement of hs4 with (PC7) (SouthernBiotech): anti-B220-PC5, anti-B220-PC7, anti-CD19- the neo gene (Fig. 2A). Mutant ES clones allowed the derivation of PE, anti-CD117-PE, anti CD25-PE, anti-IgM-PE, anti-CD138-PE, anti ⌬ ⌬ ⌬ CD4-PE, anti-CD8-PC5, and anti-CD11b-PE for bone marrow cells; two homozygous lines of Nhs4 (N/N) and hs4 ( / ) mutant an- anti-B220-PC5, anti-CD19-PE, anti-IgM-PE, anti-IgD-PE, anti-CD138- imals (⌬ refers to the cre deletion of the neo gene) (Fig. 1). Rep- PE, anti CD4-PE, anti-CD8-PC5, anti-CD23-PC7, anti-CD21-PE, and resentative Southern blots of wt/wt, N/wt, and N/N animals ob- anti-CD11b-PE for splenocytes. Control experiments included irrele- tained with 5Ј and 3Ј probes are reported in Fig. 2B.A vant isotype-matched Abs conjugated with PC5, PC7, or PE. Cells were analyzed on a Coulter XL apparatus (Beckman Coulter). representative Southern blot of wt/wt and N/N animals obtained with an hs4 probe is reported in Fig. 2B, right. A representative Apoptosis assay neo gene PCR experiment of wt/wt, N/wt, N/N, wt/⌬, and ⌬/⌬ Freshly isolated splenocytes (1 ϫ 106 cells/ml) were cultured in growth animals is reported in Fig. 2C, left. Finally, an hs4 PCR experiment medium in 24-well plates. Immediately after isolation and after 24, 48, and confirmed the absence of hs4 in N/N and ⌬/⌬ mice (Fig. 2C, right). 6928 Ig AND CSR IN hs4-DEFICIENT MICE

5’ probe 3’ probe Wt ~22 kb Wt ~22 kb NeoR ~8.2 kb NeoR~9.7 kb

eborp ’5 eborp ’3 eborp

Wt Wt/N N/N Wt Wt/N N/N

Wt ~22 kb Wt ~22 kb Hs4 probe R ~8.2 kb NeoR ~9.7 kb Neo Wt N/N

hs4

Neo PCR Hs4 PCR

9 19 C1 fi fi U t t/N /N t/ /fi UC t t/N /N t/ /fi p W W N W fi p W W N W fi

602 pb 336 pb

FIGURE 2. Southern blot and PCR of knockout ES cells or animals. A, Downloaded from Southern blot on ES cells. The 5Ј probe detected a genomic EcoR1 band of 9.7 kb. The 3Ј probe detected a genomic EcoR1 band of 8.2 kb. B, Southern blot on N/N animals. The 5Ј probe detected a unique genomic EcoR1 band of 9.7 kb. The 3Ј probe detected a unique genomic EcoR1 band of 8.2 kb. The hs4 probe did not detect the hs4 element in N/N mice. C, PCR experiments on N/N and ⌬/⌬ mice. No amplification of neo was found in ⌬/⌬ mice. No amplification of hs4 was found in N/N and ⌬/⌬ mice. http://www.jimmunol.org/ FIGURE 1. Targeting of the hs4 enhancer. Upper panel, Map of the mouse 3ЈRR and targeting construct. Open circles stand for transcriptional enhancers. Arrows indicate the location of sense and antisense primers for though not drastic, a reproducibly decreased surface IgM expres- ϩ hs4 PCR experiments. Locations of 5Ј,3Ј, and hs4 probes used for South- sion was found on B220 splenocytes from ⌬/⌬ mutants (mean ern blot experiments are indicated. Medium panel, Map of the mouse 3ЈRR fluorescence intensity of 3.8 Ϯ 0.3) by comparison to wt (mean in animals with neoR insertion (N/N mice). Arrows indicate the location of fluorescence intensity of 5.7 Ϯ 0.2) (Fig. 3D, left). The percentages sense and antisense primers for neo PCR experiments. Lower panel, Map of B220ϩIgMϩ and B220ϩIgDϩ cells were reduced in the spleen Ј of the mouse 3 RR in animals with hs4 deletion after cre-loxP recombi- of ⌬/⌬ mice as compared with wt animals (Table I). Surface IgD nation of the neo gene (⌬/⌬ mice). ϩ by guest on October 1, 2021 expression was not significantly different on B220 splenocytes from ⌬/⌬ mutants (mean fluorescence intensity of 6.6 Ϯ 0.6) as compared with wt mice (mean fluorescence intensity of 8.2 Ϯ 0.7). B cells and lymphoid tissue development in mice carrying a Finally, blood cell counts were similar in hs4-deficient mice and wt homozygous deletion of the hs4 element animals (Table I). Percentages of B220ϩ, B220ϩIgMϩ, and ϩ ϩ A decrease of cell counts was found in bone marrow of hs4-defi- B220 IgD cells were not affected (Table I). cient mice as compared with wt mice (Table I). However, percent- The spontaneous or etoposide-stimulated apoptotic process in B ages of B cells (B220ϩCD19ϩ), T cells (CD4ϩ and CD8ϩ), and cells of ⌬/⌬ mice and wt animals was similar, indicating that the monocytes/macrophages (CD11bϩ) in bone marrow were similar lower amounts of B cells in spleen of ⌬/⌬ mice was not related to (Fig. 3A). Percentages of pro-B cells (B220ϩCD117ϩ), pre-B cells a higher apoptotic rate (data not shown). The proliferation of (B220ϩCD25ϩ), immature B cells (B220ϩIgMϩ), and recirculat- splenic B cells of ⌬/⌬ mice in response to bacterial LPS (20 ␮g/ ing B cells (B220ϩIgDϩ) were also similar (Table I). Flow cy- ml) and anti-CD40 (5 ␮g/ml) was not altered as compared with tometry analysis revealed a reproducibly decreased surface IgM those of wt animals. In contrast, their proliferation was reduced by expression on bone marrow B220ϩ cells when ⌬/⌬ mutants were 20% in response to 20 ␮g/ml anti-IgM (Fig. 3D, right). compared with wt (Fig. 3B). Mean (ϮSEM) IgM fluorescence intensity of IgMϩ cells decreased ( p Ͻ 0.01, Mann-Whitney U test) Ig synthesis and CSR in mice carrying a homozygous deletion from 6.5 Ϯ 0.4 to 4.3 Ϯ 0.3 for eight wt and seven ⌬/⌬ animals. of the hs4 element The total cell number was similar in spleen of hs4-deficient mice In a first set of experiments we investigated Ig production in vivo. as compared with wt mice (Table I). In contrast to bone marrow, As shown in Fig. 4, ⌬/⌬ mice and their littermates had similar the percentage of B cells (B220ϩCD19ϩ) was significantly ( p ϭ levels of serum IgM, IgG1, IgG2a, IgG2b, IgG3, IgE, and IgA. To 0.02, Mann-Whitney U test) reduced in the spleen of ⌬/⌬ animals determine whether hs4 deletion affects CSR, Ig production was as compared with wt animals (Fig. 3C and Table I). Similar per- then assessed in vitro through stimulation of splenocytes with LPS centages of follicular B cells (B220ϩCD23ϩCD21Ϫ) were found and/or cytokines. Flow cytometric analysis allowed the counting of in spleen of ⌬/⌬ mice (90.0 Ϯ 1.0%, n ϭ 4) and wt animals cells and the study of surface expression of class-switched isotypes (91.1 Ϯ 1.5%, n ϭ 8). Identical percentages of marginal zone B on LPS only, LPS plus IL-4, LPS plus IFN-␥, or LPS plus TGF-␤ cells (B220ϩCD21ϩ) were found in spleen of ⌬/⌬ mice (9.9 Ϯ activated splenic cells. A shown in Fig. 5, hs4 deletion had no 1.0%, n ϭ 4) and wt animals (8.8 Ϯ 1.5%, n ϭ 8). The percentages evident effect on CSR. There was no change in the percentages of of B220ϩIgMϪ cells in spleen of wt and ⌬/⌬ mice were 12.7 Ϯ cells with surface expression of IgG3 and IgG2b among LPS-stim- 0.1% and 21.15 Ϯ 2.1%, respectively, with these percentage being ulated cells from ⌬/⌬ mice and wt controls. There was no change significantly different ( p ϭ 0.008, Mann-Whitney U test). Al- either in the percentages of cells with surface expression of IgG1 The Journal of Immunology 6929

Table I. B cell phenotyping results of ⌬/⌬ hs4-deﬁcient micea

wt ⌬/⌬ Mice Signiﬁcance

Bone marrow White cell counts (ϫ106) 10.3 Ϯ 1.2 (6) 6.6 Ϯ 0.9 (6) p ϭ 0.04 % B220ϩ cells 16.1 Ϯ 0.9 (8) 17.3 Ϯ 1.3 (8) NS % CD117ϩ cells among B220ϩ cells 1.1 Ϯ 0.1 (6) 1.3 Ϯ 0.1 (6) NS % CD25ϩ cells among B220ϩ cells 35.8 Ϯ 4.9 (8) 34.5 Ϯ 3.4 (8) NS % IgMϩ cells among B220ϩ cells 30.3 Ϯ 1.0 (8) 31.2 Ϯ 2.7 (8) NS % IgDϩ cells among B220ϩ cells 10.1 Ϯ 1.6 (6) 11.2 Ϯ 2.3 (6) NS Spleen White cell counts (ϫ106 per organ) 68.5 Ϯ 6.7 (6) 48.1 Ϯ 5.5 (6) NS % B220ϩ cells 31.9 Ϯ 3.5 (8) 22.2 Ϯ 3.5 (8) p ϭ 0.02 % IgMϩ cells among B220ϩ cells 87.2 Ϯ 1.1 (8) 78.8 Ϯ 2.4 (8) p ϭ 0.01 % IgDϩ cells among B220ϩ cells 80.2 Ϯ 1.0 (8) 72.3 Ϯ 2.8 (8) p ϭ 0.02 Blood White cell counts (ϫ106 per ml) 5.9 Ϯ 1.0 (6) 5.4 Ϯ 0.6 (6) NS % B220ϩ cells 23.9 Ϯ 4.3 (6) 22.1 Ϯ 2.0 (6) NS % IgMϩ cells among B220ϩ cells 87.0 Ϯ 1.6 (6) 89.6 Ϯ 1.7 (6) NS % IgDϩ cells among B220ϩ cells 89.4 Ϯ 1.0 (6) 88.2 Ϯ 1.1 (6) NS

a Results are reported as mean Ϯ SEM. The numbers of mice in each group are indicated in parentheses. Significance was assessed with the Mann-Whitney U test. Downloaded from among LPS plus IL-4-stimulated cells. There was finally no resting B cells, the mean fluorescence intensities obtained with change in surface expression of IgG2a among LPS plus IFN-␥- IgG1, IgG2a, IgG2b, IgG3, and IgA were similar in switched cells stimulated cells and of IgA among LPS plus TGF-␤-stimulated from ⌬/⌬ mice and wt animals (data not shown). Finally, we in- cells. Besides percentages of switched cells and contrary to IgM in vestigated whether hs4 deletion affected the in vitro Ig secretion in http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 3. B cells in ⌬/⌬ mice. A, Percentage (mean Ϯ SEM) of B cells, T cells, and monocytes/macrophages in the bone marrow of seven ⌬/⌬ and eight wt mice. B, B220ϩIgMϩ cells in the bone marrow of ⌬/⌬ and wt mice. One representative experiment out of seven is shown. C, Percentage (mean Ϯ SEM) of B cells, T cells, and monocytes/macrophages in the spleen -p Ͻ 0.05 as com ,ء .of seven ⌬/⌬ and eight wt mice pared with wt animals (Mann-Whitney U test). D, Left, IgMϩ cells in the spleen of ⌬/⌬ and wt mice. Cells were gated on B220ϩ cells. One representative experiment out of seven is shown. Right, B splenocytes from wt (white histograms) and hs4-deﬁcient mice (gray histograms) were grown for 3 days alone or with LPS (20 ␮g/ml), anti-CD40 (5 ␮g/ml), or anti-IgM (20 ␮g/ml). Mean Ϯ SEM of six experiments in sextuplicates. -p Ͻ 0.01 as compared with wt animals (Mann-Whit ,ء ney U test). 6930 Ig AND CSR IN hs4-DEFICIENT MICE Downloaded from http://www.jimmunol.org/

FIGURE 4. Serum Ig levels in ⌬/⌬ mice. ELISA analysis of IgM, IgG1, IgG2a, IgG2b, IgG3, IgE, and IgA in 8-wk-old mice sera. ⌬/⌬ mice are shown in the left part of graphs followed by wt littermates in the right. FIGURE 5. Surface Ig expression in activated splenic B cells. B cells 6

ϫ by guest on October 1, 2021 Mean Ϯ SEM of 10 animals in each group. No significant differences were were cultured for 4 days with LPS with or without cytokines at 1 10 found. cells/ml and stained with anti-B220-PC5 and anti-isotype FITC Abs. Rep- resentative results from ⌬/⌬ and wt experiments were shown. The percentages of class switching are the means Ϯ SEM of seven and eight independent experiments with ⌬/⌬ mice and wt animals, respectively. response to LPS and/or cytokines. As shown in Table II, B cells from ⌬/⌬ mice and wt mice secreted in vitro similar amounts of IgM, IgG2b, and IgG3 in response to LPS, of IgG1 and IgE in Homozygous hs4 knockout mice display efficient Ag-specific response to LPS plus IL-4, of IgG2a in response to LPS plus primary and secondary responses IFN-␥, and of IgA in response to LPS plus TGF-␤. hs4-Deficient mice showed germinal centers of normal size, normal serum Ig levels, and no class-switching defect. Additionally, B Analysis of N/N mutant animals cells from these mice did not display any general activation defect The phenotype of N/N mutant was similar to the one of ⌬/⌬ an- when stimulated with a polyclonal activator such as LPS. It was imals. The numbers of bone marrow B cells, T cells, and mono- questionable whether these cells were also able to produce high- cytes/macrophages were similar in N/N and wt animals (data not affinity Abs when challenged with specific Ags. hs4-Deficient shown). A decreased ( p Ͻ 0.01) number of B cells was found in mice were thus specifically immunized with OVA or DNP and N/N spleens (22.5 Ϯ 2.2%, six mice) as compared with wt (31.3 Ϯ assayed for the levels and kinetics of their humoral responses. IgM 3.4%, six animals). Flow cytometry analysis of freshly isolated splenic B cells reveals, although not drastic, a decreased ( p Ͻ 0.01) surface IgM expression when N/N mutants (mean fluores- Table II. Ig secretion in hs4-deleted micea cence intensity of 5.2 Ϯ 0.3, 6 animals) were compared with wt (mean fluorescence intensity of 7.0 Ϯ 0.4, 7 animals). Assessment Ig Isotype (␮g/ml) wt Mice ⌬/⌬ Mice of serum Ig levels revealed that N/N mice and their littermates had IgM 313.6 Ϯ 24.4 388.0 Ϯ 23.2 similar levels of serum IgM (299 Ϯ 56 vs 187 Ϯ 24 pg/ml, 10 mice IgG1 90.6 Ϯ 18.5 130.0 Ϯ 22.5 in each group), IgG1 (300 Ϯ 29 vs 327 Ϯ 43 ␮g/ml), IgG2a IgG2a 2.0 Ϯ 0.7 4.0 Ϯ 1.6 (406 Ϯ 109 vs 256 Ϯ 112 ␮g/ml), IgG2b (165 Ϯ 26 vs 214 Ϯ 29 IgG2b 14.2 Ϯ 3.2 18.5 Ϯ 4.1 ␮ Ϯ Ϯ ␮ Ϯ Ϯ IgG3 14.0 Ϯ 6.0 6.1 Ϯ 3.3 g/ml), IgG3 (275 101 vs 132 20 g/ml), IgE (2 1vs2 Ϯ Ϯ Ϯ Ϯ ␮ IgE 1.2 0.6 1.3 0.6 1 ng/ml), and IgA (97 10 vs 124 15 g/ml). Finally, replace- IgA 53.7 Ϯ 17.2 34.3 Ϯ 11.4 ment of hs4 with the neo gene did not affect CSR in B cells of N/N a ELISA analysis of Ig secretion by LPS (and/or appropriate cytokine)-stimulated mice in response to LPS and/or cytokines or the in vitro Ig syn- splenocytes. Mean Ϯ SEM values of nine independent experiments. For all isotypes, thesis (data not shown). p Ͼ 0.05 (Mann-Whitney U test). The Journal of Immunology 6931

the conclusion that hs4 is not required for CSR and Ig synthesis. Thus, hs4-deficient mice had serum levels of IgM, IgG1, IgG2a, IgG2b, IgG3, IgE, and IgA similar to wt littermates. In vitro class switching and membrane Ig expression in response to LPS and/or cytokines of B cells were similar in hs4-deficient mice and wt animals. Finally, the in vitro capacity to secrete various Ig isotypes was unaffected by the knockout. The fact that hs4 was not required for Ig secretion is in agreement with the recent study of Zhang et al. (19), who reported, using various bacterial artificial chromo- somes and the plasmocytoma cell line 9921, that transcription of a productively rearranged Ig VDJC␣ does not require the presence of hs4 in the 3ЈRR. Surprisingly, deletion of hs4 has no evident effect on CSR and Ig synthesis. These results markedly differ from those reported for the joint hs3b and hs4 deletion, where CSR and Ig production were severely impaired (13). A predominant role for hs3b in this later study is difficult to support since deletion of the hs3a element (which shares 98% of homology with hs3a) has no effect (12). As

previously suggested (12, 19), this finding might reflect extensive Downloaded from functional redundancy among these elements (in cooperation or not with other elements in the locus). Deletion of only one of the FIGURE 6. Humoral response of hs4-deficient mice immunized with four 3Ј IgH enhancers can be easily supplanted by the activity of OVA or DNP. ELISA analysis of OVA-specifif IgM or IgG1 secretion and the three remaining enhancers, with deletion of a pair of them DNP-specific IgM or IgG3 secretion in sera of 8-wk-old immunized mice. becoming more problematic for the cell. A, OVA-specifif IgM, (B) OVA-specifif IgG1, (C) DNP-specific IgM, and The phenotype observed in hs4-deficient animals was marked (D) DNP-specific IgG3. Ab levels are expressed in arbitrary units by com- http://www.jimmunol.org/ parison with a control serum. Time after immunization is indicated in days. by an apparent normal B cell differentiation in the bone marrow, Each point is the mean (ϮSEM) of serum determinations from five hs4- by a normal proliferation in response to LPS or anti-CD40, by a deficient mice (E) and five wt mice (F). slightly impaired proliferation in response to anti-IgM, by a slight decrease of the B cell number in spleen, and by a decreased expression of IgM on resting B lymphocytes. This latter phenomenon is of great interest since it was already found in the dual hs3b- class Abs (both against OVA and DNP) were raised with a normal hs4-deficient mice (13). This observation suggests that hs4 might kinetics and titers by comparison with those of wt mice (Fig. 6, A interact with E␮ in regulating ␮ expression in resting B cells. and C). Specific anti-OVA IgG1 (Fig. 6B) and specific anti-DNP by guest on October 1, 2021 IgG3 (Fig. 6D) were produced with a normal kinetic and raised Recently, the chromosome conformation capture technique has re- ␮ Ј levels at least equal to those in wt mice. Thus, B cell activation by vealed physical interactions between E and the 3 RR in B cells a specific Ag is not compromised in hs4 knockout mice. (20, 21). Several observations support hs4 being a candidate for such interaction: hs4 is the first acting 3Ј IgH enhancer during B Discussion cell development (5), hs3a and hs1,2 were deleted in the 70Z/3 pre-B cell line without affecting ␮ expression (22), and hs4 loca- The 3ЈRR is a complex structure containing four transcriptional Ј activators still in search of a clear function. The 5Ј part of this tion at the 3 end of the IgH regulatory region and outside the ␮ structure consists in a large region of 25 kb. Its global “palin- palindrome does allow such a long-range interaction with E (3). dromic” structure encompasses the hs1,2 central enhancer flanked During the past decade, the pgk-neo cassette has been inserted with inverted repeats, which (in mice) include the hs3a and hs3b at various sites within the CH locus (12, 23, 24). Taken together, elements. The hs4 element lies 2 kb after the end of the palin- these results indicated that such insertions inhibited CSR to up- dromic region, giving it a singular status as compared with hs3a, stream, but not to downstream, CH genes. This suggests the idea hs1,2, and hs3b. Besides its particular position, recent results also that the pgk-neo cassette insertion into the locus short-circuits the Ј demonstrated unique functional properties: hs4 is associated with ability of the 3 RR to facilitate CSR of dependent CH genes up- active chromatin not only in B cells and plasma cells, as for hs3a, stream of the insertion (25). Mice exhibiting a pgk-neo cassette hs1,2, and hs3b, but also in early pro-B and pre-B cells (4). The targeted 2 kb downstream of the hs4 element had normal serum Ig vast majority of studies investigating the role of the 3ЈRR in Ig levels and normal class switching after B cell activation (25). The Ј synthesis have been made with B cell lines or transgenic mice authors thus suggested that key elements of the 3 RR lie within the bearing a previously rearranged VDJ gene. Only a few studies 17-kb region between hs1,2 and 2 kb downstream of hs4. Our have investigated the effect of the direct disruption of 3Ј IgH en- results reveal that insertion of the neo gene at the position of hs4 hancers. Insertion of a promoter/neoR gene cassette into hs3a or had no effect on CSR and Ig synthesis. hs1,2 had severe effects on several isotypes (12). However, these In conclusion, hs4-deficient mice show that, despite several pre- effects merely resulted from the neo cassette insertion, since vious experimental studies, the proposed role of hs4 in driving IgH “clean” deletion of hs3a or hs1,2 with the cre-loxP system did not gene transcription and CSR can be largely overcome by the activ- alter expression of any H chain isotype (12). Beyond so-called ity of the remaining enhancers located within the locus. “neo effects”, only the “clean” deletion of a large fragment encompassing the last two 3Ј enhancer of hs3b and hs4 severely impaired germline transcription and class switching to most iso- Acknowledgments types in transgenic mice (13). In the present study, results obtained We thank S. Desforges and B. Remerand for help with animal manipula- with mice bearing a clean deletion of only the hs4 element led to tion. We also thank E. Pinaud for his helpful discussions. 6932 Ig AND CSR IN hs4-DEFICIENT MICE

Disclosures 14. Mills, F. C., N. Harindranath, M. Mitchell, and E. E. Max. 1997. Enhancer com- plexes located downstream of both human immunoglobulin C␣ genes. J. Exp. The authors have no financial conflicts of interest. Med. 186: 845–858. 15. Pinaud, E., C. Aupetit, C. Chauveau, and M. Cogne. 1997. Identification of a References homolog of the C␣3Ј/hs3 enhancer and of an allelic variant of the 3ЈIgH/hs1,2 ␣ 1. Ernst, P., and S. T. Smale. 1995. Combinatorial regulation of transcription, II: enhancer downstream of the human immunoglobulin 1 gene. Eur. J. Immunol. The immunoglobulin ␮ heavy chain gene. Immunity 2: 427–438. 27: 2981–2985. 2. Henderson, A., and K. Calame. 1998. Transcriptionnal regulation during B cell 16. Giambra, V., S. Volpi, A. V. Emelyanov, D. Pflugh, A. L. M. Bothwell, P. Norio, development. Annu. Rev. Immunol. 16: 163–200. Y. Fan, Z. Ju, A. I. Skoultchi, R. R. Hardy, et al. 2008. Pax5 and linker histone 3. Khamlichi, A. A., E. Pinaud, C. Decourt, C. Chauveau, and M. Cogne. 2000. The H1 coordinate DNA methylation and histone modifications in the 3Ј regulatory 3Ј IgH regulatory region: a complex structure in a search for a function. Adv. region of the immunoglobulin heavy chain locus. Mol. Cell Biol. 28: 6123–6133. Immunol. 75: 317–345. 17. Hooper, M., K. Hardy, A. Handyside, S. Hunter, and M. Monk. 1987. HPRT- 4. Garrett, F. E., A. V. Emelyanov, M. A. Sepulveda, P. Flanagan, S. Volpi, F. Li, deficient (Lesch-Nyhan) mouse embryos derived from germline colonization by D. Loukinov, L. A. Eckhardt, V. V. Lobanenkov, and B. K. Birshtein. 2005. cultured cells. Nature 326: 292–295. Ј Chromatin architecture near a potential 3 end of the IgH locus involves modular 18. Le Morvan, C., E. Pinaud, C. Decourt, A. Cuvillier, and M. Cogne´. 2003. The regulation of histone modifications during B-cell development and in vivo occu- immunoglobulin heavy-chain locus hs3b and hs4 3Ј enhancers are dispensable for pancy at CTCF sites. Mol. Cell. Biol. 25: 1511–1525. VDJ assembly and somatic hypermutation. Blood 102: 1421–1427. 5. Michaelson, J. S., S. L. Giannini, and B. K. Birshtein. 1995. Identification of 3Ј␣-hs4, a novel Ig heavy chain enhancer element regulated at multiple stages of 19. Zhang, B., A. Alaie-Petrillo, M. Ko, F. Li, and L. A. Eckhardt. 2007. Transcrip- tion of a productively rearranged Ig VDJCa does not require the presence of Hs4 B cell differentiation. Nucleic Acids Res. 23: 975–981. Ј 6. Chauveau, C., and M. Cogne. 1996. Palindromic structure of the IgH 3Ј locus in the IgH 3 regulatory region. J. Immunol. 178: 6297–6306. control region. Nat. Genet. 14: 15–16. 20. Wuerffel, R., L. Wang, F. Grigera, J. Manis, E. Selsing, T. Perlot, F. W. Alt, 7. Chauveau, C., E. Pinaud, and M. Cogne. 1998. Synergies between regulatory M. Cogne, E. Pinaud, and A. L. Kenter. 2007. S-S synapsis during class switch elements of the immunoglobulin heavy chain locus and its palindromic 3Ј locus recombination is promoted by distantly located transcriptional elements and ac- control region. Eur. J. Immunol. 28: 3048–3056. tivation-induced deaminase. Immunity 27: 711–722.

8. Gregor, P. D., and S. L. Morrison. 1986. Myeloma mutant with a novel 3Ј ﬂank- 21. Ju, Z., S. A. Volpi, R. Hassan, N. Martinez, S. L. Giannini, T. Gold, and Downloaded from ing region: loss of normal sequence and insertion of repetitive elements leads to B. K. Birshtein. 2007. Evidence for physical interaction between the immuno- decreased transcription but normal processing of the ␣ heavy-chain gene prod- globulin heavy chain variable region and the 3Ј regulatory region. J. Biol. Chem. ucts. Mol. Cell Biol. 6: 1903–1916. 282: 35169–35178. 9. Lieberson, R., J. Ong, X. Shi, and L.A. Eckhardt. 1995. Immunoglobulin gene 22. Saleque, S., M. Singh, and B. K. Birshtein. 1999. Ig heavy chain expression and transcription ceases upon deletion of a distant enhancer. EMBO J. 14: class switching in vitro from an allele lacking the 3Ј enhancers DNase I-hyper- 6229–6238. sensitive hs3A and hs1,2. J. Immunol. 162: 2791–2803. 10. Dunnick, W. A., J. Shi, K. A. Graves, and J. T. Collins. 2005. The 3Ј end of the 23. Seidl, K. J., J. P. Manis, A. Bottaro, J. Zhang, L. Davidson, A. Kisselgof, heavy chain constant region locus enhances germline transcription and switch H. Oettgen, and F. W. Alt. 1999. Position-dependent inhibition of class-switch http://www.jimmunol.org/ recombination of the four ␥ genes. J. Exp. Med. 201: 1459–1466. recombination by PGK-neor cassettes inserted into the immunoglobulin heavy 11. Shi, X., and L. A. Eckhardt. 2001. Deletional analyses reveal an essential role for chain constant region locus. Proc. Natl. Acad. Sci. USA 96: 3000–3005. the hs3b/hs4 IgH enhancer pair in an Ig-secreting but not an earlier-stage B cell line. Int. Immunol. 13: 1003–1012. 24. Qiu, G., G. R. Harriman, and J. Stavnezer. 1997. I␣ exon-replacement mice 12. Manis, J. P., N. van der Stoep, M. Tian, R. Ferrini, L. Davidson, A. Bottaro, and synthesize a spliced HPRT-C␣ transcript which may explain their ability to F. W. Alt. 1998. Class switching in B cells lacking 3Ј immunoglobulin heavy switch to IgA: inhibition of switching to IgG in these mice. Int. Immunol. 158: chain enhancers. J. Exp. Med. 188: 1421–1431. 4780–4787. 13. Pinaud, E., A. A. Khamlichi, C. Le Morvan, M. Drouet, V. Nalesso, M. Le Bert, 25. J. P. Manis, J. S. Michaelson, B. K. Birshtein, and F. W. Alt. 2003. Elucidation and M. Cogne. 2001. Localization of the 3Ј IgH locus elements that effect long- of a downstream boundary of the 3Ј IgH regulatory region. Int. Immunol. 39: distance regulation of class switch recombination. Immunity 15: 187–199. 753–760. by guest on October 1, 2021