Abstract
Background
It has been reported that human FOXP3+ CD4 Tregs express GARP-anchored surface latency-associated peptide (LAP) after activation, based on the use of an anti-human LAP mAb. Murine CD4 Foxp3+ Tregs have also been reported to express surface LAP, but these studies have been hampered by the lack of suitable anti-mouse LAP mAbs.
Methodology/Principal Findings
We generated anti-mouse LAP mAbs by immunizing TGF-β−/− animals with a mouse Tgfb1-transduced P3U1 cell line. Using these antibodies, we demonstrated that murine Foxp3+ CD4 Tregs express LAP on their surface. In addition, retroviral transduction of Foxp3 into mouse CD4+CD25− T cells induced surface LAP expression. We then examined surface LAP expression after treating CD4+CD25− T cells with TGF-β and found that TGF-β induced surface LAP not only on T cells that became Foxp3+ but also on T cells that remained Foxp3− after TGF-β treatment. GARP expression correlated with the surface LAP expression, suggesting that surface LAP is GARP-anchored also in murine T cells.
Conclusions/Significance
Unlike human CD4 T cells, surface LAP expression on mouse CD4 T cells is controlled by Foxp3 and TGF-β. Our newly described anti-mouse LAP mAbs will provide a useful tool for the investigation and functional analysis of T cells that express LAP on their surface.
Introduction
TGF-β controls immune responses by multiple mechanisms including the suppression of Th1 and cytotoxic lymphocytes, and the induction of Th17 cells depending on the context [1]. TGF-β is first synthesized as pro-TGF-β and is then intracellularly processed by furin proprotein convertase to form a latent TGF-β complex which consists of non-covalently associated dimmers of the N-terminal region of pro-TGF-β (latency-associated peptide, LAP) and of the C-terminal region of pro-TGF-β (mature TGF-β) [2]. Expression of pro-TGF-β, LAP, latent TGF-β and/or mature TGF-β (hereafter referred as LAP/TGF-β) on mouse CD4 T cells was first reported in 2001 by Nakamura et al. [3]. They proposed that CD4+CD25+ regulatory T cells (Tregs) mediated their suppressive function by presenting active TGF-β to effector cells in a cell-cell contact manner. They used a polyclonal chicken anti-TGF-β antibody and a monoclonal anti-human LAP mAb (clone 27232) for FACS staining of mouse CD4 T cells. Our laboratory has also reported the presence of surface LAP+ on mouse T cells using a polyclonal goat anti-human LAP antibody [4], [5]. However, use of a polyclonal antibody is problematical due to the inherent variance between different polyclonal preparations. The anti-LAP mAb (clone 27232) used by Nakamura, et al., was raised against recombinant human LAP (R&D Systems). Although Nakamura et al. used this antibody to stain mouse CD4 T cells [3], in our hands, we did not find that this anti-human LAP mAb cross-reacted with mouse LAP. Thus, although clone 27232 stained human TGFB1-transduced cells [6], it did not stain mouse Tgfb1-tranduced cells at all (Figure S1). To overcome these problems, a fully characterized anti-mouse LAP mAb would be required for staining mouse T cells.
Recently, by using the anti-human LAP mAb 27232 [7], [8], it was reported that human FOXP3+ Tregs express surface LAP after activation and that the surface LAP is anchored by GARP/LRRC32 [8], [9].
We raised anti-mouse LAP mAbs by immunizing TGF-β−/− mice with mouse Tgfb1-transduced cells, and used them to stain mouse CD4 T cells. We found that the majority of mouse Foxp3+ CD4 T cells expressed surface LAP after activation. Surface LAP was induced by Foxp3-transduction into mouse CD4+CD25− T cells and by addition of TGF-β to mouse CD4+CD25- T cell cultures. In contrast to human T cells [8], TGF-β induced surface LAP not only on T cells that converted to Foxp3+ but also on T cells in which Foxp3 was not expressed. GARP expression correlated with the surface LAP expression suggesting that surface LAP is anchored by GARP.
Results
Generation of anti-mouse LAP mAbs
We used mouse Tgfb1-transduced P3U1 (P3U1-muTGF-β cells) cells as an immunogen. We have recently shown that human TGFB1-transduced P3U1 (P3U1-huTGF-β) cells express LAP/TGF-β on the surface [6]. Surface expression of murine LAP was also expected on P3U1-muTGF-β cells since we found that anti-TGF-β (clone 9016) surface stained P3U1-muTGF-β cells as well as P3U1-huTGF-β cells (Figure S1). We elected to immunize TGF-β-deficient mice. TGF-β−/− mice manifest an autoimmune syndrome and die at 3–4 wks after birth [10], [11]. We attempted to prolong their life by injecting galectin-1, which has been reported to suppress other autoimmune diseases [12], starting at day 7 of birth. P3U1-muTGF-β cells were injected i.p. every other day 5 times beginning at day 8 after birth, and spleen cells were taken at day 22 after birth and fused with P3U1 myeloma cells. The hybridoma cells were grown in hypoxanthine-aminopterin-thymidine (HAT)-supplemented methylcellulose medium. Approximately 2,800 clones were picked from the plates, and transferred to hypoxanthine-thymidine (HT)-supplemented DMEM in 96-well plates. The culture supernatants were screened by surface staining of P3U1-muTGF-β cells by FACS. Thirty-six positive clones were selected and recovered (TW7 series) (Figure S2). Of the 36 clones, 32 clones were IgG and 4 clones were IgM. To check their specificity, we tested the ability of the antibodies to immunoprecipitate Flag-tagged mouse LAP (Flag-mLAP) produced by retrovirally Flag-mLAP-transduced P3U1 cells. Of the 32 IgG clones, 26 clones, including TW7-16B4 and TW7-20B9, immunoprecipitated Flag-mLAP (Figure S3, underlined) and thus were true anti-mouse LAP mAbs. Several clones, including TW7-28G11, did not immunoprecipitate Flag-mLAP (Figure S3). TW7-28G11, however, stained human latent TGF-β-coated beads, but not human LAP- or human active TGF-β-coated beads (Figure S4A). TW7-28G11 immunoprecipiated Flag-mLAP only when active TGF-β was exogenously added to Flag-mLAP solution (Figure S4B), and immunoprecipiated pro-TGF-β and latent TGF-β from the culture supernatant of P3U1-muTGF-β cells (Figure S5A). These results indicate that TW7-28G11 is a conformation specific anti-mouse/human latent TGF-β/pro-TGF-β mAb which recognize LAP and TGF-β in combination. The specificity of some clones, including TW7-16B4, TW7-20B9 and TW7-28G11, were further confirmed by testing their ability to detect mouse pro-TGF-β and/or LAP by Western blot (Figure S5B), and by their ability to immunoprecipiate pro-TGF-β and latent TGF-β from culture supernatant of P3U1-muTGF-β cells (Figure S5A).
Surface LAP expression on mouse Foxp3+ CD4 T cells
It has been reported that human FOXP3 Tregs express surface LAP after activation [7], [8] by a GARP-mediated anchoring mechanism [8], [9]. We tested our anti-LAP/TGF-β mAbs for their ability to stain pre-activated mouse CD4 T cells. CD4 T cells were stimulated with plate-bound anti-CD3/anti-CD28 for 2 days and rested for 1 day. Following this, they were surface stained with anti-LAP/TGF-β mAbs using PE-labeled anti-mouse IgG1 (for IgG1 subtype clones) or anti-mouse Igκ secondary antibody (for non-IgG1 clones), then fixed and intracellularly stained with anti-Foxp3-Alexa Fluor647. Of the 36 potential anti-LAP/TGF-β candidate clones, 31 clones surface stained Foxp3+ cells. Three representative clones (TW7-16B4, TW7-20B9, and TW7-28G11) are shown in Figure 1A and all 36 clones are shown in Figure S6. It should be noted that 24 of the 26 clones which immunoprecipitated Flag-mLAP as described above stained Foxp3+ CD4 T cells with a similar pattern. Among them, TW7-16B4 produced the highest staining signal followed by TW7-20B9. An anti-pro-TGF-β/latent TGF-β clone, TW7-28G11, also stained Foxp3+ CD4 T cells (Figure 1A and Figure S6), suggesting that surface LAP exists as pro-TGF-β and/or latent TGF-β rather than free LAP without mature TGF-β. Surface LAP staining strongly correlated with GARP expression (Figure 1B), indicating that surface LAP on mouse Foxp3+ CD4 Tregs is also anchored by GARP as on human FOXP3+ CD4 Tregs.
For further analysis we selected TW7-16B4 (IgG1, κ) and TW7-20B9 (IgG1, κ) as the highest staining anti-LAP clones, and TW7-28G11 (IgG2b, κ) as an anti-pro-TGF-β/latent TGF-β clone. These clones were used with secondary antibodies or as antibodies directly labeled with PE or Allophycocyanin (APC). We tested whether unstimulated CD4 T cells also express surface LAP using the direct conjugates. We found that freshly prepared mouse CD4+25+ T cells also weakly expressed surface LAP (Figure 2A). We also investigated the time course of surface LAP expression. We found that surface LAP expression on Foxp3+ cells peaked on days 1 and 2, and then gradually decreased when the cells were rested (days 3 and 5) (Figure 2B, upper panels). We found that GARP was co-expressed with LAP in all time points (Figure 2B, lower panels).
Foxp3-induced surface LAP expression
We then asked whether surface LAP expression is controlled by Foxp3. We found that retroviral Foxp3 transduction into mouse CD4+CD25− T cells induced surface LAP (GFP+ population vs. GFP− population in Figure 3). This result demonstrates that surface LAP is under control of Foxp3.
TGF-β-induced surface LAP expression
TGF-β converts Foxp3− CD4 T cells into induced Foxp3+ Tregs (iTregs) [1]. To determine whether iTregs also express surface LAP, we stimulated mouse CD4+CD25− T cells in the presence or absence of recombinant TGF-β and checked for surface LAP expression. As expected ∼25% of CD4+CD25- T cells were converted to Foxp3+ iTregs in presence of TGF-β (Figure 4A). We found that these iTregs expressed surface LAP. Interestingly, the Foxp3−-remaining cells also became surface LAP+ cells following culture in the presence of TGF-β. GARP expression correlated with surface LAP expression on both Foxp3+ cells and Foxp3− cells (Figure 4B), suggesting that surface LAP is GARP-dependent not only on natural Tregs and iTregs cells but also on non-Tregs.
It is possible that surface LAP expression on natural Foxp3+ Tregs might also be maintained by TGF-β produced by Tregs themselves. We found, however, that the ALK5 inhibitor or anti-TGF-β 1D11 did not affect surface LAP expression or GARP expression on Foxp3+ Tregs (Figures S7 and S8). Thus, these results suggest that surface LAP expression on Foxp3+ Tregs is independent of TGF-β.
Discussion
The existence and function of surface LAP on Tregs has been a matter of debate. Contrary to the first report by Nakamura et al. [3], Shevach's group questioned the function of TGF-β in Treg-mediated suppression [13], and their staining of mouse T cells was quite faint, if at all present [14]. As a part of our investigation of TGF-β, we found that the anti-human LAP mAb 27232 used by Nakamura et al. does not cross-react with mouse LAP (Figure S1). In this report, we raised anti-mouse LAP mAbs by immunization of TGF-β−/− mice and revisited the existence of surface LAP on mouse CD4 T cells. We found that anti-mouse LAP mAbs stained majority of Foxp3+ Tregs, but not Foxp3− T cells after activation (Figure 1A). Fresh CD4+CD25+ T cells also expressed surface LAP at a weak level (Figure 2A). Thus our results establish that mouse Foxp3+ Tregs do express surface LAP. It should be mentioned, however, that it is not yet determined whether surface LAP/TGF-β has a functional contribution to Treg-mediated suppression.
Using the anti-LAP mAb 27232 [7], [8] it was recently reported that human FOXP3+ Tregs express surface LAP and that the surface LAP is anchored by GARP [8], [9]. It appears that this is also the case with mouse CD4 T cells since GARP expression strongly correlated with surface LAP expression (Figure 1B and Figure 2B). We recently reported the occurrence of GARP-independent, GRP78-associated surface LAP on TGFB1-transduced cells [6]. It is unknown at this time whether GARP-independent surface LAP also can be seen on T cells.
In humans, TGF-β-induced FOXP3+ CD4 T cells do not express surface LAP or GARP [8]. On the contrary, in mice, not only did TGF-β-induced Foxp3+ CD4 T cells express surface LAP and GARP, but TGF-β-exposed CD4+CD25− T cells that did not become Foxp3+ CD4 T cells also expressed surface LAP and GARP (Figure 4B). Some Foxp3− CD4 T cells also expressed surface LAP/TGF-β without exogenous of TGF-β. We do not know whether this LAP/TGF-β expression was induced by TGF-β in an autocrine fashion or occurred independent of TGF-β. However, TGF-β signaling seems not absolutely required for surface LAP expression since natural Foxp3+ Tregs maintained surface LAP expression even when TGF-β signaling was blocked (Figure S7). Thus, surface LAP expression may be controlled independently by Foxp3 and TGF-β signaling.
In summary, we raised anti-mouse LAP mAbs and revisited surface LAP expression on mouse CD4 T cells. We found that Foxp3+ Tregs expressed surface LAP and that surface LAP is induced by forced expression of Foxp3 or by TGF-β irrespective of Foxp3 induction. Furthermore, surface LAP expression strongly correlated with GARP, suggesting that surface LAP is GARP-mediated. These newly described anti-mouse LAP mAbs will provide a useful tool for functional analysis of T cells that express LAP on their surface.
Materials and Methods
Generation of anti-mouse LAP mAbs (TW7 series)
Mice were housed in a pathogen-free environment and the animal protocols were approved according to the guidelines of the Committee on Animals of Harvard Medical School (Protocol No. 02683). TGF-β−/− mice [10] were injected i.p. with 20 µg galectin-1 (Sigma-Aldrich) [12] every other day starting at 7 day after birth to prevent the fatal autoimmunity seen in TGF-β−/− mice [11]. Mouse Tgfb1-transduced P3U1 (P3U1-muTGF-β) cells (clone #11) were injected i.p. at 1-4×106 cells (in 10–25 µl PBS) every other day 5 times starting at 8 days after birth. At age 22 days, the spleen cells were fused with P3U1 myeloma cells, and the hybridoma cells were plated in methylcellulose medium (ClonaCell-HY, Stemcell Technologies). Screening was conducted by surface staining of P3U1-muTGF-β cells by FACS. Anti-mouse LAP specificity was confirmed by immunoprecipitation of recombinant Flag-tagged mouse LAP (lacking C-terminal mature TGF-β sequence) (Flag-mLAP) [15], immunoprecipitation of pro-TGF-β and latent TGF-β, staining recombinant human latent TGF-β (R&D Systems)-coated polystyrene beads, and/or staining recombinant human TGF-β (R&D Systems) coated polystyrene beads.
Other antibodies and reagents
Anti-human LAP mAb clone 27232, anti-TGF-β mAb clone 9016, and biotinylated goat anti-LAP (BAF246) were obtained from R&D Systems. Anti-mouse CD3 (145-2C11), anti-mouse CD28 (37.51), Allophycocyanin (APC)-labeled goat anti-mouse Ig, PE- or APC-labeled anti-mouse IgG1 (A85-1), and PE-labeled anti-mouse Igκ (187.1) were from BD Biosciences. PE-labeled anti-mouse GARP (YGIC86), and Alexa Fluor647-labeled anti-Foxp3 (FJK-16s) were from eBioscience. Alexa Fluor488-labeled anti-Foxp3 (150D) was from Biolegend. TGF-β receptor I kinase inhibitor (ALK5 inhibitor II) was from EMD/Calbiochem. Anti-Flag mAb (M2) was from Sigma-Aldrich. (caga)12-MLP-Luc TGF-β reporter plasmid [16], [17] was kindly provided by Dr. D. Vivien (the Universite' de Caen, Daix, France). Mv1Lu cells (ATCC) were stably transfected with (caga)12-MLP-Luc plasmid and used for testing dose-response of ALK5 inhibitor II in TGF-β bioassay.
CD4 T cell stimulation and FACS staining
CD4 T cells were separated from BALB/c mice (The Jackson Laboratories) or C57BL/6 background Foxp3-GFP knock-in (Foxp3-KI) mice [18] using MACS CD4 purification kit (Miltenyi Biotec). When CD4+CD25− T cells were prepared, biotinylated anti-CD25 antibody was additionally mixed to the MCAS antibody cocktail. T cells were stimulated with plate-bound anti-CD3 and anti-CD28 for 2 days. The cells were split into non-coated wells and rested for 1 day, then stained by FACS. Surface LAP staining was conducted by either PE- or APC-directly conjugated anti-mouse LAP mAbs, or unconjugated anti-mouse LAP mAbs followed by PE- or APC-conjugated goat anti-mouse Ig, monoclonal anti-mouse IgG1 or monoclonal anti-mouse Igκ secondary antibody. Intracellular Foxp3 staining was done with Alexa Fluor647- or Alexa Fluor488-labeled anti-Foxp3 using Foxp3 Staining Buffer Set (eBioscience).
Retroviral transduction
Retroviral vector pMCs-IRES-GFP [19], ecotropic retroviral packaging cell line Plat-E [20] were kindly provided by Dr. Kitamura (Tokyo Univ., Tokyo, Japan). Foxp3 was cloned into pMCs-IRES-GFP vector, and the retroviral supernatant was produced by Plat-E. Mouse CD4+25− T cells from BALB/c mice pre-activated with plate-bound anti-CD3 and anti-CD28 for 30 hrs were infected with Foxp3 ecotropic retrovirus by centrifugation at 3,000 rpm for 1 hr. 1 day after infection, the cells were split onto a non-coated wells, and rested. The transduced cells were re-stimulated with plate-bound anti-CD3/anti-CD28 for 14 hrs, rested for 2 days, and surface stained with anti-LAP mAbs and then intracellularly with anti-Foxp3.
Supporting Information
Acknowledgments
We thank Thomas Koeglsperger for providing Tgfb1 −/+ mice.
Footnotes
Competing Interests: The authors have declared that no competing interests exist.
Funding: This work was supported by the National Institutes of Health Grants AI435801 and NS38037. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
References
- 1.Rubtsov YP, Rudensky AY. TGFβ signalling in control of T-cell-mediated self-reactivity. Nat Rev Immunol. 2007;7:443–453. doi: 10.1038/nri2095. [DOI] [PubMed] [Google Scholar]
- 2.Miyazono K, Ichijo H, Heldin CH. Transforming growth factor-β: latent forms, binding proteins and receptors. Growth Factors. 1993;8:11–22. doi: 10.3109/08977199309029130. [DOI] [PubMed] [Google Scholar]
- 3.Nakamura K, Kitani A, Strober W. Cell contact-dependent immunosuppression by CD4+CD25+ regulatory T cells is mediated by cell surface-bound transforming growth factor β. J Exp Med. 2001;194:629–644. doi: 10.1084/jem.194.5.629. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Oida T, Zhang X, Goto M, Hachimura S, Totsuka M, et al. CD4+CD25− T cells that express latency-associated peptide on the surface suppress CD4+CD45RBhigh-induced colitis by a TGF-β-dependent mechanism. J Immunol. 2003;170:2516–2522. doi: 10.4049/jimmunol.170.5.2516. [DOI] [PubMed] [Google Scholar]
- 5.Ochi H, Abraham M, Ishikawa H, Frenkel D, Yang K, et al. Oral CD3-specific antibody suppresses autoimmune encephalomyelitis by inducing CD4+CD25− LAP+ T cells. Nat Med. 2006;12:627–635. doi: 10.1038/nm1408. [DOI] [PubMed] [Google Scholar]
- 6.Oida T, Weiner HL. Overexpression of TGF-β1 gene induces cell surface localized Glucose-Regulated Protein 78-associated Latency-associated peptide/TGF-β. J Immunol. 2010;185:3529–3535. doi: 10.4049/jimmunol.0904121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tran DQ, Andersson J, Hardwick D, Bebris L, Illei GG, et al. Selective expression of latency-associated peptide (LAP) and IL-1 receptor type I/II (CD121a/CD121b) on activated human FOXP3+ regulatory T cells allows for their purification from expansion cultures. Blood. 2009;113:5125–5133. doi: 10.1182/blood-2009-01-199950. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Tran DQ, Andersson J, Wang R, Ramsey H, Unutmaz D, et al. GARP (LRRC32) is essential for the surface expression of latent TGF-β on platelets and activated FOXP3+ regulatory T cells. Proc Natl Acad Sci U S A. 2009;106:13445–13450. doi: 10.1073/pnas.0901944106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Stockis J, Colau D, Coulie PG, Lucas S. Membrane protein GARP is a receptor for latent TGF-β on the surface of activated human Treg. Eur J Immunol. 2009;39:3315–3322. doi: 10.1002/eji.200939684. [DOI] [PubMed] [Google Scholar]
- 10.Kulkarni AB, Huh CG, Becker D, Geiser A, Lyght M, et al. Transforming growth factor β1 null mutation in mice causes excessive inflammatory response and early death. Proc Natl Acad Sci U S A. 1993;90:770–774. doi: 10.1073/pnas.90.2.770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Christ M, McCartney-Francis NL, Kulkarni AB, Ward JM, Mizel DE, et al. Immune dysregulation in TGF-β1-deficient mice. J Immunol. 1994;153:1936–1946. [PubMed] [Google Scholar]
- 12.Liu FT, Rabinovich GA. Galectins: regulators of acute and chronic inflammation. Ann N Y Acad Sci. 2010;1183:158–182. doi: 10.1111/j.1749-6632.2009.05131.x. [DOI] [PubMed] [Google Scholar]
- 13.Piccirillo CA, Letterio JJ, Thornton AM, McHugh RS, Mamura M, et al. CD4+CD25+ regulatory T cells can mediate suppressor function in the absence of transforming growth factor β1 production and responsiveness. J Exp Med. 2002;196:237–246. doi: 10.1084/jem.20020590. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Andersson J, Tran DQ, Pesu M, Davidson TS, Ramsey H, et al. CD4+FoxP3+ regulatory T cells confer infectious tolerance in a TGF-β-dependent manner. J Exp Med. 2008;205:1975–1981. doi: 10.1084/jem.20080308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Young GD, Murphy-Ullrich JE. Molecular interactions that confer latency to transforming growth factor-β. J Biol Chem. 2004;279:38032–38039. doi: 10.1074/jbc.M405658200. [DOI] [PubMed] [Google Scholar]
- 16.Dennler S, Itoh S, Vivien D, ten Dijke P, Huet S, et al. Direct binding of Smad3 and Smad4 to critical TGFβ-inducible elements in the promoter of human plasminogen activator inhibitor-type 1 gene. EMBO J. 1998;17:3091–3100. doi: 10.1093/emboj/17.11.3091. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Docagne F, Colloc'h N, Bougueret V, Page M, Paput J, et al. A soluble transforming growth factor-β (TGF-β) type I receptor mimics TGF-β responses. J Biol Chem. 2001;276:46243–46250. doi: 10.1074/jbc.M010915200. [DOI] [PubMed] [Google Scholar]
- 18.Bettelli E, Carrier Y, Gao W, Korn T, Strom TB, et al. Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells. Nature. 2006;441:235–238. doi: 10.1038/nature04753. [DOI] [PubMed] [Google Scholar]
- 19.Kitamura T, Koshino Y, Shibata F, Oki T, Nakajima H, et al. Retrovirus-mediated gene transfer and expression cloning: powerful tools in functional genomics. Exp Hematol. 2003;31:1007–1014. [PubMed] [Google Scholar]
- 20.Morita S, Kojima T, Kitamura T. Plat-E: an efficient and stable system for transient packaging of retroviruses. Gene Ther. 2000;7:1063–1066. doi: 10.1038/sj.gt.3301206. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.