|
|
||||||||
CUTTING EDGE |

* Immunology Program, Sackler School of Graduate Biomedical Sciences, Tufts University, Boston, MA 02111; and
Department of Pathology, Tufts University School of Medicine, Boston, MA 02111
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
During B cell development, the Igh locus becomes accessible in discrete domains as the DH-Cµ region is hyperacetylated and activated for recombination before the VH region (3). In addition, subregions within the large Igh-V domain are regulated independently at the level of accessibility. Accessibility of the large J558 VH gene family that occupies the D-distal region of Igh-V is down-regulated at the pro- to pre-B cell transition, whereas the more D-proximal VH gene families continue to be germline transcribed through the mature B cell stage (4, 5). The independent regulation of the D-distal and D-proximal VH domains is also reflected in the relative usage of these segments during development because the D-distal segments are underused in both fetal development and in early stages of adult B cell lymphopoiesis (6, 7).
Recent evidence has led to the suggestion that two independently functioning pathways participate in controlling the accessibility and recombination of the D-distal VH gene segments (8, 9). The first pathway involves the repositioning of the entire Igh locus to the center of the nucleus in committed pro-B cells followed by a second Pax5-mediated step that induces contraction of the Igh locus and is uniquely required for rearrangement of the D-distal VH segments (8, 9). The second pathway mediates changes in the local chromatin surrounding individual VH gene segments (10) and is dependent on IL-7R signaling. Disruption of this pathway in the IL-7R-deficient mouse leads to a marked defect in germline transcription and rearrangement of the D-distal, but not D-proximal, VH families (11). Furthermore, IL-7 induces histone acetylation at several D-distal VH genes in RAG-2-deficient pro-B cells (3). The mechanism through which IL-7 signaling leads to these local chromatin changes has not been elucidated. To dissect this mechanism, we developed a model for IL-7-induced accessibility of D-distal VH genes. We demonstrate that in response to IL-7 signaling, Stat5 is recruited to a D-distal VH gene where it induces accessibility.
| Materials and Methods |
|---|
|
|
|---|
1B4 cells (gift from Henry Wortis, Tufts University, Boston, MA) were obtained by immortalizing BALB/c splenic B cells with the J2 retrovirus containing v-raf and v-myc. This cell line was found to have an IgMhighIgDlow493+IL-7R+ surface phenotype and to express RAG-1, RAG-2, and Pax5 mRNA (data not shown) and thus mirrors a recently emigrating transitional B cell (12). The Ab-murine leukemia virus (MLV)3-transformed pro-B cell line 22D6 was a gift from Naomi Rosenberg (Tufts University). 1B4 cells were treated with 20 ng/ml IL-7 (PeproTech), pretreated for 2 h with AG-490 (Calbiochem) before IL-7 stimulation, or simultaneously treated with 100 nM trichostatin A (TSA) (Calbiochem) and IL-7.
Reverse transcription and real-time PCR
Total RNA was isolated using TRIzol (Invitrogen Life Technologies), DNased, and reverse transcribed using SuperScript II (Invitrogen Life Technologies) with random hexamers. Resulting cDNA was amplified by real-time PCR with custom TaqMan primers and probes or rodent GAPDH control reagents (Applied Biosystems). The levels of A1, V11, and GAPDH transcripts were each determined by comparison to standard curves of serially diluted DNA containing the relevant target gene. The levels of A1 and V11 germline transcripts in each sample were normalized to the endogenous control gene, GAPDH. The fold change in germline transcripts was determined by dividing the levels of A1 or V11 transcripts by the levels in the untreated sample. The same TaqMan primers and probes were also used to analyze precipitated DNA in chromatin immunoprecipitation (ChIP) experiments (sequences available upon request).
ChIP
ChIP assays were performed using a ChIP Assay kit (Upstate Biotechnology). Briefly, formaldehyde-cross-linked chromatin was sonicated to generate a mean fragment size of 0.21.2 kb and divided into aliquots containing fragments from 106 cells. DNA from one aliquot was isolated following reverse cross-linking and represents the total input DNA available for immunoprecipitation. The other aliquots were precleared and incubated with 1 µg of anti-Stat5a (Santa Cruz Biotechnology), 4 µg of anti-acetyl-H3 (Upstate Biotechnology), 3 µl of anti-acetyl-H4 serum (Upstate Biotechnology), 1 µg of normal rabbit IgG (Santa Cruz Biotechnology), or no Ab and precipitated with ssDNA/protein A agarose. After washing, the chromatin was eluted, and cross-links were reversed. Purified DNA from the immunoprecipitates, as well as from total input DNA, was analyzed by real-time PCR. Results are represented as the percentage of total input DNA that was precipitated in each sample. Shown are representative experiments from at least three independent stimulations.
Western blot analyses and immunoprecipitations
For Western blot analyses, 60 µg of whole cell lysate were separated by 8% SDS-PAGE, transferred to nitrocellulose, and probed with anti-Stat5 (Santa Cruz Biotechnology) or anti-actin (Santa Cruz Biotechnology). For immunoprecipitations, lysates were prepared after 30 min of IL-7 stimulation and then incubated with 1 µg of anti-Stat5a or anti-Stat5b (Santa Cruz Biotechnology). Immune complexes were precipitated with protein G-Sepharose beads (Amersham Biosciences), washed, eluted, and then separated by SDS-PAGE. The blot was probed with anti-phosphotyrosine (pTyr) (Upstate Biotechnology) and then reprobed with the appropriate Stat5-specific Abs.
Retrovirus-mediated gene transfer
The wild-type (WT), dominant negative (DN) (
749 C-terminal truncation), and constitutively active (CA) (H299R and S711F) Stat5a MSCV-IRES-GFP retroviral constructs have been described previously (13, 14, 15). Retrovirus was prepared from 293T fibroblasts using the calcium phosphate method. 1B4 cells were sorted 24 h after infection and expanded for 48 h before experimentation.
| Results and Discussion |
|---|
|
|
|---|
IL-7 rapidly induces germline transcription and histone H4 acetylation of a D-distal V gene but not a D-proximal V gene
IL-7 signaling has been demonstrated to specifically induce accessibility of the D-distal, but not D-proximal, VH genes (3, 11). To demonstrate that a similar response occurred in 1B4 cells, we designed gene-specific real-time PCR assays to detect germline transcripts from a representative D-distal gene, the J558 family A1 gene, and a representative D-proximal gene, the S107 family V11 gene (both are abundantly transcribed in nontransformed pro-B cells (5)). In the absence of IL-7 signaling, the level of A1 transcripts was 10- to 20-fold lower than the level of V11 transcripts (data not shown), consistent with our prior observation that only D-proximal VH genes continue to be actively transcribed in late stages of B cell development. In response to IL-7 signaling, there was a dose-dependent 10- to 20-fold increase in A1 transcripts, whereas the V11 transcripts were unaffected (Fig. 1A). Thus, IL-7 signaling is sufficient for the specific induction of germline transcription of a D-distal, but not a D-proximal, VH gene.
|
We then analyzed by ChIP assay whether 1 h of IL-7 stimulation also led to increased A1 histone acetylation, another chromatin modification associated with accessibility. Accessible D-distal VH gene segments are hyperacetylated at histones H3 and H4, but their H4 acetylation status is associated more closely with accessibility both in pro-B cells and in allelically excluded B cells (10). Similarly, we found that IL-7 induced a consistent 2-fold increase in the acetylation of H4, but not H3, associated with A1 (Fig. 1C). No significant change in H3 or H4 acetylation was observed at the D-proximal V11 gene (data not shown). These results indicate that multiple chromatin modifications are induced at the A1 locus directly following IL-7R triggering, as evidenced by increased transcription and H4 acetylation.
Activated Stat5 is recruited to the D-distal A1 gene in IL-7-stimulated B cells and Ab-MLV-transformed pro-B cells
Stat5 has been implicated in the regulation of TCR
accessibility via induction of germline transcription and histone hyperacetylation (16), raising the possibility that the IL-7-induced accessibility of the D-distal VH genes is mediated similarly by Stat5. This is supported by the finding that both D-distal VH gene accessibility and Stat5 activation are down-regulated in the pro- to pre-B cell transition (4). If Stat5 does mediate the induction of A1 accessibility, we reasoned that its activation in 1B4 cells would be IL-7 dependent. We immunoprecipitated both isoforms of Stat5, Stat5a and Stat5b, from IL-7-stimulated 1B4 cells and probed the immunoprecipitated proteins with a pTyr-specific Ab to determine their activation state. We found that the activation of both Stat5 isoforms was dependent on IL-7 (Fig. 2A).
|
We also tested whether Stat5 was recruited to the A1 gene in the Ab-MLV-transformed pro-B cell line, 22D6, that expresses high levels of D-distal VH germline transcripts and has CA Stat5 (data not shown). Using ChIP analysis, we demonstrated that A1 was enriched specifically in the anti-Stat5 sample, compared with precipitation using nonspecific rabbit IgG (Fig. 2C). Thus, in two cell lines representing different stages of B cell development in which the D-distal Igh-V locus becomes accessible in response to IL-7 signaling, Stat5 is specifically recruited to a D-distal and not a D-proximal, VH gene.
IL-7 induced A1 germline transcription, and Stat5 activation is inhibited by AG-490
IL-7 stimulation of B lineage cells has been shown to activate Stat5a and Stat5b through the JAK1 and JAK3 kinases (17). Therefore, we reasoned that if A1 transcriptional activation is mediated via the JAK3/Stat5 pathway, then IL-7-induced A1 transcription would be inhibited by AG-490, a JAK2 and JAK3 inhibitor (18). We found that AG-490 did inhibit both IL-7-induced Stat5 phosphorylation and A1 germline transcription in a dose-dependent manner (Fig. 3A), without affecting the transcription of the V11 gene. This result supports the hypothesis that Stat5 mediates IL-7-induced A1 transcription and provides evidence that signaling through JAK3 (JAK2 being nonresponsive to IL-7) may be necessary for this pathway.
|
We then directly tested whether Stat5 activation is required for the regulation of A1 accessibility by introducing a DN Stat5a into 1B4 cells by retroviral-mediated transfer and examining the IL-7 response. The DN Stat5a competes for DNA binding but lacks the transactivation domain (14). Cells infected with the DN Stat5a retrovirus coexpress GFP, allowing cells expressing both low and high levels of the mutant protein to be sorted. Uninfected (GFP) cells were also sorted from each infection and served as internal controls for the IL-7-induction of germline transcription. We showed that while infection with either a vector- or WT Stat5a-containing virus had no effect on A1 transcription, expression of the DN Stat5a led to a dose-dependent inhibition of IL-7-induced A1 transcription (Fig. 3B). The level of V11 transcript was unaffected by retroviral expression of either protein (data not shown). The correlation of GFP expression with expression of the truncated DN Stat5a protein was confirmed by Western blot analysis (Fig. 3B). In addition, the defect in A1 transcription is not due to decreased IL-7R expression because cells expressing the DN Stat5a displayed similar levels of IL-7R compared with cells expressing ectopic WT Stat5a (data not shown). These results demonstrate that Stat5 is necessary for IL-7-induced A1 germline transcription. Despite sorting cells expressing the highest level of GFP, we were not able to achieve >70% inhibition, possibly because DN Stat5a was expressed at levels comparable to the endogenous protein and may not have been stoichiometrically able to completely quench WT activity.
CA Stat5a induces A1 germline transcription in the absence of cytokine signaling
To determine whether activated Stat5 was sufficient to induce accessibility of the D-distal VH gene, A1, we infected 1B4 cells with vector-, WT Stat5a-, and CA Stat5a-containing retroviruses and used RT-PCR to measure germline transcripts in unstimulated cells. The CA Stat5a protein contains two point mutations (H299R and S711F) that result in its constitutive tyrosine phosphorylation, nuclear localization, and transcriptional activity (16). Cells expressing the CA Stat5a showed a 6- to 8-fold increase in the level of A1 transcripts, compared with cells expressing either the vector or WT Stat5a (Fig. 3C). The level of V11 transcripts remained constant despite ectopic expression of WT or CA Stat5a. These results demonstrate that Stat5 activation is sufficient for the increase in A1 germline transcription in this system.
Induction of A1 germline transcription by Stat5 requires a deacetylase activity
Transcriptional activation of Stat5 target genes requires a deacetylase activity, and thus, deacetylase inhibitors block Stat5-mediated transcription (19). Therefore, we tested whether the deacetylase inhibitor, TSA, would similarly block the IL-7 induction of A1 transcription. We treated 1B4 cells with IL-7 and TSA and measured the level of A1 and V11 germline transcripts. After 4 h, IL-7 induced a 20-fold increase in A1 transcripts, which was inhibited by 75% in the presence of TSA (Fig. 4A). A similar inhibitory effect of TSA was apparent at 1 h (data not shown). Neither IL-7 nor TSA affected the expression of V11 transcripts at either time point.
|
The deacetylase activity required for Stat5 transactivation controls the proper assembly and stability of the basal transcription machinery (19). Because Stat5 activation of the D-distal VH gene is inhibited similarly by a deacetylase inhibitor, we suggest that Stat5 acts in part by recruiting a deacetylase to the local chromatin that affects transcription initiation by stabilizing the basal transcription machinery. This is in contrast to a mechanism in which Stat5 acts primarily to recruit histone acetylases (HAT) to the local chromatin, and as a secondary by-product of increased accessibility, germline transcription is activated.
The requirement for a deacetylase activity does not preclude the contribution of Stat5-mediated recruitment of HAT coactivator proteins, and accordingly, we observed a specific increase in H4 acetylation concurrent with the recruitment of Stat5 to the A1 gene. Stat5 interacts with the HAT coactivator CBP/p300, and these two proteins were demonstrated to cooperate in the activation of TCR
J segment accessibility (16). It is conceivable that the IL-7 induction of A1 H4 acetylation results from the Stat5-mediated recruitment of CBP/p300 and that local histone acetylation is a prerequisite for Stat5-mediated germline transcription. Conversely, histone acetylation at A1 could be the result of transcription per se because HAT coactivator proteins can associate with RNA polymerase II and acetylate histones during elongation (20).
Stat5 activation of D-distal VH germline transcription does not require Pax5 because Pax5-deficient pro-B cells have acetylated and transcriptionally active D-distal VH genes, although they are inefficiently recombined (9). However, Stat5 activation is insufficient in the absence of other regulatory mechanisms to establish accessibility because Stat5 is activated in many cell types, including T cells (13, 16), but D-distal VH gene accessibility is strictly limited to B cells. Busslinger and colleagues (8) recently demonstrated that when Pax5 is expressed ectopically in pro-T cells, the Igh locus is relocated to a central nuclear position, and D-distal VH germline transcription is activated. Although not tested, it is likely that these pro-T cells have activated Stat5 because IL-7 signaling is active during the early stages of T cell development (16, 21). Therefore, because the convergence of central localization of Igh loci and Stat5 activation appears to result in D-distal VH germline transcription in both pro-B cells and Pax5-expressing pro-T cells, it is possible that central nuclear positioning of Igh is the prerequisite for Stat5-mediated D-distal VH germline transcription that normally limits this activity to B cells.
| Disclosures |
|---|
|
|
|---|
| Acknowledgments |
|---|
| Footnotes |
|---|
1 This work was supported by National Institutes of Health Grant GM36064 and National Institutes of Health Training Grant 5T32AI07077. ![]()
2 Address correspondence and reprint requests to Dr. Peter Brodeur, Department of Pathology, Tufts University School of Medicine, 150 Harrison Avenue, Boston, MA 02111. E-mail address: peter.brodeur{at}tufts.edu ![]()
3 Abbreviations used in this paper: MLV, murine leukemia virus; TSA, trichostatin A; ChIP, chromatin immunoprecipitation; pTyr, phosphotyrosine; WT, wild-type; DN, dominant negative; CA, constitutively active; HAT, histone acetylase. ![]()
Received for publication November 15, 2004. Accepted for publication January 15, 2005.
| References |
|---|
|
|
|---|
-helical region is required for transcriptional activities and proteasome-dependent turnover of the tyrosine-phosphorylated Stat5. EMBO J. 19:392.[Medline]
locus by Stat5 and histone acetylation. Immunity 15:813.[Medline]
This article has been cited by other articles:
![]() |
J. G. Bates, D. Cado, H. Nolla, and M. S. Schlissel Chromosomal position of a VH gene segment determines its activation and inactivation as a substrate for V(D)J recombination J. Exp. Med., December 24, 2007; 204(13): 3247 - 3256. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Pawlitzky, C. V. Angeles, A. M. Siegel, M. L. Stanton, R. Riblet, and P. H. Brodeur Identification of a Candidate Regulatory Element within the 5' Flanking Region of the Mouse Igh Locus Defined by Pro-B Cell-Specific Hypersensitivity Associated with Binding of PU.1, Pax5, and E2A. J. Immunol., June 1, 2006; 176(11): 6839 - 6851. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. M. Will, J. D. Aaker, M. A. Burchill, I. R. Harmon, J. J. O'Neil, C. A. Goetz, K. L. Hippen, and M. A. Farrar Attenuation of IL-7 Receptor Signaling Is Not Required for Allelic Exclusion J. Immunol., March 15, 2006; 176(6): 3350 - 3355. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Harnaha, J. Machen, M. Wright, R. Lakomy, A. Styche, M. Trucco, S. Makaroun, and N. Giannoukakis Interleukin-7 Is a Survival Factor for CD4+ CD25+ T-Cells and Is Expressed by Diabetes-Suppressive Dendritic Cells Diabetes, January 1, 2006; 55(1): 158 - 170. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. E. Johnson, N. Shah, A. Panoskaltsis-Mortari, and T. W. LeBien Murine and Human IL-7 Activate STAT5 and Induce Proliferation of Normal Human Pro-B Cells J. Immunol., December 1, 2005; 175(11): 7325 - 7331. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |