• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Immunol. Author manuscript; available in PMC Jun 1, 2012.
Published in final edited form as:
PMCID: PMC3215299
NIHMSID: NIHMS331383

Donor CD8 T Cells and IFN-γ Are Critical for Sex-Based Differences in Donor CD4 T Cell Engraftment and Lupus-Like Phenotype in Short-Term Chronic Graft-Versus-Host Disease Mice

Abstract

The transfer of unfractionated DBA/2J (DBA) splenocytes into B6D2F1 (DBA → F1) mice results in greater donor CD4 T cell engraftment in females at day 14 that persists long-term and mediates greater female lupus-like renal disease. Although donor CD8 T cells have no demonstrated role in lupus pathogenesis in this model, we recently observed that depletion of donor CD8 T cells prior to transfer eliminates sex-based differences in renal disease long-term. In this study, we demonstrate that greater day 14 female donor CD4 engraftment is also critically dependent on donor CD8 T cells. Male DBA → F1 mice exhibit stronger CD8-dependent day 8–10 graft-versus-host (GVH) and counter-regulatory host-versus-graft (HVG) responses, followed by stronger homeostatic contraction (days 10–12). The weaker day 10–12 GVH and HVG in females are followed by persistent donor T cell activation and increasing proliferation, expansion, and cytokine production from days 12 to 14. Lastly, greater female day 14 donor T cell engraftment, activation, and cytokine production were lost with in vivo IFN-γ neutralization from days 6 to 14. We conclude the following: 1) donor CD8 T cells enhance day 10 proliferation of donor CD4 T cells in both sexes; and 2) a weaker GVH/HVG in females allows prolonged survival of donor CD4 and CD8 T cells, allowing persistent activation. These results support the novel conclusion that sex-based differences in suboptimal donor CD8 CTL activation are critical for shaping sex-based differences in donor CD4 T cell engraftment at 2 wk and lupus-like disease long-term.

Human systemic lupus erythematosus exhibits a strikingly female predominance with a female-to-male ratio of ~8–10:1 during the childbearing years (1). This well-established observation has been an important, but as yet incompletely understood clue regarding the potential role for sex hormones in disease expression (2). Although murine models have been of enormous benefit in unraveling the disordered immuno-regulation characteristic of lupus, many models do not exhibit female skewing of disease. One of the few models exhibiting greater female disease severity is an induced model of lupus, the parent-into-F1 (p → F1) model of chronic graft-versus-host (GVH) disease (GVHD), in which the transfer of parental CD4 T cells into normal semiallogeneic F1 mice results in B cell hyperactivity, autoantibody production, and lupus-like renal disease (3, 4). Donor CD4, but not CD8, T cells are critical in mediating lupus-like disease by providing cognate help to MHC class II-disparate host B cells (57). Sex-based differences in this model are best documented transferring unfractionated splenic DBA/2J (DBA) donor cells into B6D2F1 (BDF1) hosts (DBA → F1) (8). Early studies demonstrated that nephrotic syndrome-like features could be induced in female DBA→F1 mice using multiple transfers of unfractionated splenic and lymph node lymphocytes; however, disease severity in females in comparison with males was not examined (9, 10). Subsequent studies demonstrated that following a single transfer of 80 × 106 unfractionated DBA splenocytes (containing ~10–12 × 106 CD4 T cells) into BDF1 mice, female transfers (female into female [f→F]) exhibit greater elevations of lupus-specific autoantibodies [i.e., anti-dsDNA, anti-poly(ADP-ribose) polymerase-1] and more severe renal disease than in male transfers (male into male [m→M]) (8, 11). Importantly, sex-based differences could be observed as early as 2 wk after donor cell transfer manifested by 2- to 3-fold greater engraftment of donor CD4 T cells in f→F versus m→M mice. Because renal disease severity is directly related to the number of donor CD4 T cells transferred (10, 12), these results support the idea that sex-based differences in donor CD4 T cell engraftment at 2 wk can serve as a surrogate marker for long-term differences in renal disease severity.

In previous work, no sex-based differences in donor CD4 or CD8 T cells were observed prior to day 7 in DBA→F1 mice; however, during the second week after transfer, male donor CD4 T cell proliferation significantly declined relative to that of females and was associated with greater female engraftment of donor CD4 T cells both at 2 wk and long-term (8). Thus, differences in donor T cell activation kinetics during the second week after transfer appear to be central to sex-based differences in lupus-like disease severity long-term. Based on these results, we characterized donor and host lymphocyte kinetics from days 8 to 14 to determine the mechanism involved in greater female donor CD4 T cell engraftment at day 14. We demonstrate that sex-based differences in donor CD4 T cell engraftment are critically dependent on coinjection of donor CD8 T cells and on IFN-γ production.

Materials and Methods

Mice

Six- to 8-wk-old male and female DBA (H-2d) and BDF1 (H-2b/d) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). All animal procedures were preapproved by the Institutional Animal Care and Use Committee at the Uniformed Services University of Health Sciences.

Induction of GVHD

Single-cell suspensions of DBA splenocytes were prepared as described (13) and transferred into BDF1 hosts by tail vein injection. Donors and hosts were age and sex matched such that male donors were transferred into male hosts (m → M) and female donors into female hosts (f → F). Donor populations were analyzed by flow cytometry to determine the percentage of CD4 and CD8 T cells in the donor inoculum prior to transfer. The numbers of donor CD4 and CD8 T cells injected are indicated in the text and respective figure legends. For CD8 depletion studies, CD8 T cells were positively selected and removed from the donor population using magnetic beads (Dynabeads, mouse CD8, Lyt2) purchased from Invitrogen (Carlsbad, CA), according to the manufacturer’s instructions. For CD8-depleted DBA into BDF1 (CD8-depleted → F1) transfers, flow cytometry analysis confirmed that the donor inoculum contained <1% contaminating CD8+ T cells.

Flow cytometric analysis

Spleen cells were first incubated with anti-murine FcγRII/III mAb, 2.4G2, for 10 min and then stained with saturating concentrations of Alexa Fluor 488-conjugated, Alexa Fluor 700-conjugated, allophycocyanin-conjugated, biotin-conjugated, PE-conjugated, FITC-conjugated, PerCPCy5.5-conjugated, and Pacific Blue-conjugated mAb against CD4, CD8, B220, H-2Kb, I-Ab, Foxp3, KI-67, CD107a, and NK1.1 purchased from either BD Biosciences (San Jose, CA), BioLegend (San Diego, CA), eBioscience (San Diego, CA), or Invitrogen (Carlsbad, CA). Biotinylated primary mAb were detected using streptavidin-allophycocyanin (BD Biosciences). Cells were fixed in 1% paraformaldehyde before reading. Intracellular staining for Foxp3 and KI-67 was performed using the Foxp3 buffer staining set from eBio-science, according to manufacturer’s protocol. Briefly, cells were permeabilized overnight in fixation/permeabilization solution. Permeabilized cells were stained with allophycocyanin-conjugated anti-mouse/rat Foxp3 and PE-conjugated mouse anti-human Ki-67 in permeabilization buffer for 30 min, washed, and read immediately.

Intracellular staining for perforin, granzyme B, IFN-γ, and TNF was performed using Abs and reagents purchased from BD Pharmingen (San Diego, CA) or BioLegend, and staining was performed according to the manufacturer’s instructions. Importantly, there was no in vitro restimulation or use of Golgi blocking agents. Following completion of the staining protocol, cells were analyzed by flow cytometry immediately. Staining for apoptotic T cells was performed using PE Annexin V Apoptosis Detection Kit I (BD Pharmingen), according to manufacturer’s protocol, and read by flow cytometry within 1 h of staining.

Multicolor flow cytometric analyses were performed using a BD LSRII flow cytometer (BD Biosciences). Lymphocytes were gated by forward and side scatter, and fluorescence data were collected for a minimum of 10,000 gated cells. Studies of donor T cells were performed on a minimum of 3,000 cells collected using a lymphocyte gate that were positive for CD4 or CD8 and negative for MHC class I of the uninjected parent (H-2Kb negative). Host B cells were gated as positive for B220 and positive for MHC class II of the uninjected parent (I-Ab positive). Host dendritic cells (DC) and macrophages were identified as I-Ad positive and CD11c or CD11b positive, respectively, using a broad forward and side scatter gate.

Cytokine expression by real-time PCR

RNase-free plastic and water were used throughout the assay. Splenocytes (1 × 107) were homogenized in 1 ml RNA-STAT-60 (Tel-Test, Friends-wood, TX). RNA samples were reverse transcribed with Moloney murine leukemia virus reverse transcriptase (Applied Biosystems International, Foster City, CA) (13). The 18S rRNA was used as an internal control. All primers and probes (MX-1, OAS, IP-10, IL-2, IFN-γ, IL-21, TNF, 18S rRNA) were purchased from Applied Biosystems International.

Statistical analysis

Statistical comparisons (t test and linear regression) were performed using Prism 4.0 (GraphPad Software).

Results

Greater donor CD4 T cell engraftment in female DBA→F1 mice at day 14 is correlated with greater persistence of donor CD8 T cells versus males

To determine the mechanism by which f→F mice exhibit greater donor CD4 T cell engraftment at 2 wk after transfer and, as a consequence, more severe renal disease long-term compared with m→M mice (8), we examined donor T cell engraftment kinetics. Because no sex-based differences in donor T cell homing, activation, or engraftment were demonstrated during the first week after transfer (8), we examined donor and host lymphocyte kinetics during the second week after transfer using the previously published single-dose protocol shown to result in sex-based differences in engraftment and renal disease, that is, a single dose of ~80–90 × 106 unfractionated DBA splenocytes typically containing ~10–14 × 106 DBA CD4 T cells. Because severity of lupus-like disease in the DBA → BDF1 model is directly related to the number of donor CD4 T cells injected (10, 12), we reduced potential variability by standardizing donor inocula based on the number of CD4 T cells transferred rather than by total splenocyte number. We observed in preliminary experiments that a donor cell inoculum containing 107 DBA CD4 T cells was less than or equal to the threshold for GVHD induction and resulted in high mouse-to-mouse variability in engraftment to include occasional engraftment failures. We therefore examined the postinjection kinetics for two higher doses within the previously published range of donor splenocytes shown to result in greater female disease severity (8, 14), as follows: 1) 12 × 106 CD4 donor cells (80–90 × 106 unfractionated splenocytes) and containing ~3.9 × 106 DBA CD8 T cells (Fig. 1A, 1C), or 2) 14 × 106 DBA CD4 donor cells (90–100 × 106 unfractionated splenocytes and containing ~4.3 × 106 DBA CD8 T cells) (Fig. 1B, 1D). Days 8, 10, 12, and 14 were examined for both doses.

FIGURE 1
Sex-based differences in donor engraftment in DBA→F1 mice are observed during the second week following transfer. Male and female BDF1 mice received unfractionated DBA splenocytes containing either 12 × 106 CD4 T cells and ~3.9 × ...

Donor CD4 T cell engraftment kinetics for days 8, 10, 12, and 14 are shown in Fig. 1A and 1B. For both donor cell doses, donor CD4 T cell engraftment is slightly, but significantly greater in m→M mice versus f→F mice at days 8 and 10; however, between days 10 and 12, these curves cross as male numbers decline, but female numbers do not. By day 14, f→F mice exhibit a mean donor CD4 T cell engraftment that is ~2.5-fold greater at both the 12 × 106 dose (p < 0.005) (Fig. 1A) and the 14 × 106 dose (p < 0.01) (Fig. 1B). These results are similar to the ~3-fold increase previously reported at day 14 for the 12 × 106 dose (8).

The engraftment kinetics of donor CD8 T cells (Fig. 1C, 1D) parallel those of donor CD4 T cells in that there is significantly greater engraftment in m→M mice at days 8 and 10; however, this difference is greater at the 14 × 106 dose (~2.5-fold) (Fig. 1D) versus the 12 × 106 dose (<2-fold) (Fig. 1C). Nevertheless, as with donor CD4 T cell engraftment, the CD8 engraftment curves for m→M and f→F cross between days 10 and 12 as male CD8 T cells exhibit greater and more consistent homeostatic contraction than females such that by day 14, females exhibit low-level, but significantly greater (~4-fold greater, p < 0.05) donor CD8 T cell engraftment at the 12 × 106 dose versus males (Fig. 1C). At the 14 × 106 dose, females exhibit a ~2.9-fold greater average donor CD8 engraftment; however, due to high mouse-to-mouse variability, these differences were not significant (Fig. 1D).

To further address a possible relationship between engraftment of donor CD4 and CD8 T cells, linear regression analyses were performed for m→M and f→F mice at the 14 × 106 dose for days 8–14 shown in Fig. 1B and 1D. Both male and female groups exhibited weak correlation (r2 < 0.4) at day 8 (Fig. 2A, 2B), but a strong correlation (r2 > 0.8) was seen in both groups at day 10 (Fig. 2C, 2D). After day 10, a strong correlation between engrafted CD4 T cells and CD8 T cells persists only in f→F mice (r2 > 0.8) (Fig. 2E–H). Taken together, these results support the idea that greater donor CD4 T cell engraftment at day 14 in females is associated with greater donor CD8 T cell engraftment. Because of the greater engraftment of both donor T cell subsets seen at the higher donor cell dose, the following studies examine the mechanism of sex-based differences seen using the 14 × 106 dose of donor CD4 T cells, unless otherwise noted.

FIGURE 2
Female DBA→F1 mice exhibit prolonged correlation between donor CD4 and CD8 T cell engraftment compared with males. Experimental groups are as described in Fig. 1, and linear regression analysis was performed, as outlined in Materials and Methods ...

Donor CD8 T cells are a major contributor to sex-based differences in donor CD4 T cell engraftment in DBA→F1 mice at day 14

The foregoing data raise the possibility that sex-based differences in donor CD4 T cell engraftment at day 14 have a CD8 component, that is that coinjection of donor CD8 T cells influences donor CD4 T cell engraftment at day 14, particularly for f→F mice. To directly address this question, we compared the kinetics of donor CD4 engraftment following the transfer of DBA splenocytes depleted only of CD8 T cells (CD8-depleted→F1) for both m→M and f→F mice at a dose of 14 × 106 CD4 T cells. As shown in Fig. 3A, the engraftment kinetics of donor CD4 T cells for m→M and f→F mice closely resemble each other at all time points. Although female CD4 engraftment was significantly greater at days 10 and 14, the striking 2- to 3-fold greater day 14 female CD4 engraftment previously published (8) and also shown in Fig. 1B is reduced to 1.2-fold (f→F versus m→M) when CD8 T cells are depleted from the donor inoculum. This near equalization is due primarily to changes in female CD4 engraftment. Specifically, donor CD8 depletion in male mice results in a slight, but non-significant increase in day 14 mean donor CD4 T cell engraftment (CD8-intact m→M mice = 4.2 × 106 ± 0.5 versus CD8-depleted m→M mice = 5.2 × 106 ± 0.3, p = NS). In contrast, donor CD8 depletion in female mice results in a significant reduction (1.6-fold) in day 14 mean donor CD4 T cell engraftment (CD8-intact f→F mice = 10.3 ± 1.5 × 106 versus CD8-depleted f→F mice = 6.3 ± 0.3 × 106, 1.6-fold, p < 0.05, one tail). Although donor CD4 engraftment for CD8-depleted f→F mice remains slightly, but significantly greater than that of CD8-depleted m→M mice, depletion of donor CD8 T cells results in a narrowing in the differences between male and female day 14 engrafted CD4 T cells primarily by reducing engraftment in f→F mice.

FIGURE 3
Sex-based differences in donor CD4 T cell engraftment are attenuated if donor CD8 T cells are depleted from the inoculum. Male and female BDF1 mice received DBA splenocytes depleted of CD8 T cells and containing 14 × 106 DBA CD4 T cells. The kinetics ...

Donor CD4 T cells were further analyzed as either effector (CD4+, Foxp3) or T regulatory cell (Treg; CD4+, Foxp3+) subsets, and the kinetics of engraftment are shown in Fig. 3B and 3C, respectively. As with total donor CD4 T cells, sex-based differences were observed in both effector and Treg subsets at day 14 for undepleted DBA splenocytes into BDF1 (CD8-intact→F1) mice. Specifically, female CD8-intact→F1 mice exhibited significantly greater CD4 Treg (Fig. 3B) and effector (Fig. 3C) numbers at day 14 compared with male CD8-intact→F1 mice or compared with either male or female CD8-depleted→F1 mice. For males, depletion of donor CD8 T cells resulted in no significant changes in either Tregs or effector CD4 T cell numbers compared with CD8-intact→F1 mice. By contrast, CD8 depletion in female mice resulted in a significant drop in both CD4 Tregs (CD8-intact→F1 = 1.9 ± 0.5 × 106 versus CD8-depleted→F1 = 0.63 ± 0.1 × 106, p < 0.05) and effector CD4 T cells (CD8-intact→F1 = 8.9 ± 1.2 × 106 versus CD8-depleted →F1 = 5.4 ± 0.3 × 106, p < 0.05). For CD8-depleted→F1 mice, sex-based differences were lost for CD4 Treg numbers; however, CD4 effector numbers were still significantly greater in females versus males (CD8-depleted→F1: females = 5.4 ± 0.3 × 106 versus males = 4.5 ± 0.2 × 106, p < 0.05). Thus, the increase in total donor CD4 T cells at day 14 in female CD8-intact→F1 mice (Fig. 1B) reflects a proportionate increase in both donor CD4 T cell Treg and effector subsets and does not reflect skewing toward the Treg subset, supporting the idea that more severe disease in female CD8-intact→F1 mice versus males is due to greater numbers of donor CD4 T cell effectors.

Coinjection of donor CD8 T cells increases proliferation of donor CD4 T cells in both sexes

To address the mechanism by which donor CD8 T cells promote sex-based differences in day 14 donor CD4 T cell engraftment, we examined donor T cell proliferation by flow cytometric assessment of KI-67 expression, a cell cycle protein found only in cells that are actively undergoing cellular proliferation (G1, S, G2, and mitosis) and not detectable in resting (G0) cells (15). KI-67 staining was performed on the cohorts receiving a donor cell dose of 14 × 106 CD4 T cells (shown in Figs. 1B, 1D, ,3)3) at the times indicated. Two major trends were observed. First, coinjection of donor CD8 T cells boosts the proliferative response of donor CD4 T cells at days 8 and 10. This effect is best seen for m→M mice, in which a significant increase in both the percentage (Fig. 4A) and number (Fig. 4B) of proliferating donor CD4 T cells is seen at days 8 and 10 for CD8-intact→F1 versus CD8-depleted→F1 mice. A similar boosting effect is seen for f→F mice, but is less pronounced. Specifically, CD8-intact f→F mice exhibit significantly greater percentage (Fig. 4A), but not number (Fig. 4B) of proliferating donor T cells at days 8, 10, and 12 versus CD8-depleted→F1 mice.

FIGURE 4
Coinjection of donor CD8 T cells is associated with increased donor CD4 T cell proliferation. Experimental protocol and groups are as outlined for Figs. 1 and and3,3, respectively, for mice receiving either of the following: 1) unfractionated ...

Secondly, sex-based differences in donor CD4 proliferation are seen primarily with coinjection of CD8 T cells. No significant sex-based differences in donor CD4 T cell proliferation were observed at days 8 or 12 for CD8-depleted→F1 mice (Fig. 4A, 4B); however, there was a low-level, but significant (p < 0.05) 1.2-fold increase in female versus male CD8-depleted→F1 mice at day 10 for both percentage (1.2-fold) and numbers (1.4-fold) of KI-67+ donor CD4 T cells. By contrast, male CD8-intact→F1 mice exhibit significantly greater proliferation (both percentage and number) at day 8 compared with female CD8-intact→F1 mice. By day 10, however, sex-based differences in donor CD4 T cell proliferation are lost. From days 10 to 12, the male and female proliferation curves cross for CD8-intact→F1 mice, and at day 12 the percentage of proliferating CD4 T cells in CD8-intact f→F mice is significantly greater than CD8-depleted f→F mice, whereas day 12 proliferating donor CD4 T cells in CD8-intact m→M mice are not significantly different from male or female CD8-depleted→F1 mice.

Regarding donor CD8 proliferation, CD8-intact m→M mice exhibit greater proliferation at day 8 compared with CD8-intact f→F mice, as shown by both the percentage (Fig. 4C) and numbers (Fig. 4D) of proliferating cells. By day 10, significant sex-based differences are lost, however, from days 10–12, the curves cross, and by day 12, the sex-based differences seen at day 8 are beginning to be reversed in that females now exhibit a nonsignificant trend toward greater donor CD8 T cell proliferation (both percentage and numbers).

Together these data demonstrate that the presence of coinjected donor CD8 T cells boosts donor CD4 proliferation and engraftment best seen for males from days 8 to 10 and for females from days 10 to 12. The greater proliferation of male donor CD8 T cells at day 8 is consistent with a stronger donor antihost CTL response in m→M mice versus f→F mice. These results support the conclusion that sex-based differences in donor CD8 activation and engraftment are associated with sex-based differences in donor CD4 T cell engraftment.

Sex-based differences in host T and B cell kinetics mirror the sex-based differences in donor T cell kinetics

The kinetics of donor T cell activation can be assessed either directly, as in Figs. 14, or indirectly by examining donor T cell effects on host T and B cell numbers. For example, previous work has demonstrated that in the absence of an effective donor CD8 CTL response, F1 mice exhibit an expansion of host T cells, B cells, and APC (seen as early as day 3 for B cells) that is sustained long-term over the initial 2 wk (16). By contrast, in the presence of a strong effector CD8 CTL response, a curtailment of host cell expansion is as donor CTL effectors mature and eliminate host cells. Elimination of host B cells and CD4 T cells occurs earlier (days 7–10) than elimination of host CD8 T cells or DC (days 10–14) (16). Mature effector donor CD8 CTL contract between days 12–14, and if host B cell elimination is incomplete at the time of donor CD8 contraction, there is the potential for re-expansion of residual host B cells as a result of continued donor CD4 T cell help (13, 16). Thus, if CD8-intact m→M mice exhibit an earlier donor antihost response, as suggested by Figs. 1B, 1D, and and4,4, then m→M mice should also exhibit greater elimination of host cells, particularly B cells at day 8 compared with CD8-intact f→F mice. The kinetics of host B and T cell numbers from days 8 to 14 are shown in Fig. 5 for male and female CD8-intact→F1 and CD8-depleted→F1 cohorts receiving 14 million cells and correspond to the cohorts shown in Figs. 1B, 1D, and and3,3, respectively.

FIGURE 5
Sex-based differences in elimination of host B cells and CD4 T cells are seen at day 8 after transfer. Experimental protocol and groups are as outlined in Figs. 1 and and3.3. Quantitation of host B cells (A), host CD4 T cells (B), and host CD8 ...

As shown in Fig. 5A, both male and female CD8-depleted→F1 mice exhibit significant expansion of host B cells over control uninjected F1 values at all points from days 8 to 14 (range ~1.5- to 3-fold > control). Similarly, significant elevations in host CD4 and CD8 T cells are seen from days 10 to 14 for both male and female CD8-depleted→F1 mice compared with control uninjected mice. Thus, in the absence of transferred donor CD8 T cells, host cells exhibit the expected expansion out to day 14, and there is no evidence of donor T cell elimination of host B and T cells consistent with the absence of effector donor antihost CD8 CTL. There were no significant sex-based differences for CD8-depleted→F1 mice from days 8 to 14 with the exception of a small (1.1-fold), but significant increase in host CD8 numbers at day 10 for female versus male CD8-depleted→F1 mice.

By contrast, both male and female CD8-intact→F1 mice exhibit differences compared not only with CD8-depleted→F1 mice, but also compared with each other (i.e., sex-based differences). For males, CD8-intact→F1 mice exhibit a significant reduction in host B cells (Fig. 5A) and CD4 T cells (Fig. 5B) at days 8, 10, and 12 and in host CD8 T cells at days 8 and 10 (Fig. 5C) compared with the expansion of host cells seen for CD8-depleted m→M mice. Moreover, host cell values for CD8-intact m→M mice are not significantly elevated over uninjected control values during days 8–10, further supporting mild, but significant host cell elimination by donor antihost CD8 CTL.

Similarly, female CD8-intact→F1 mice also exhibit significant reductions in host cells compared with female CD8-depleted→F1 mice; however, there is a delay until day 10 for significant reductions in host B cells (Fig. 5A) and CD8 T cells (Fig. 5C). Host CD4 T cells are significantly reduced in female CD8-intact→F1 mice versus CD8-depleted→F1 mice as early as day 8 (Fig. 5B) (days 8, 10, and 12), consistent with mild, but significant host cell elimination by donor antihost CD8 CTL.

Sex-based differences for CD8-intact→F1 mice are best seen at day 8 in that CD8-intact m→M mice exhibit significantly greater elimination of host B cells and host CD4 T cells (but not host CD8) compared with f→F CD8-intact mice. However, by day 10, elimination of all host B and T cell populations in CD8-intact f→F mice also has declined to levels that do not differ significantly from those of CD8-intact m→M mice consistent with delayed female donor antihost CD8 CTL-mediated elimination. The delay in elimination of host B and T cells for f→F mice versus m→M mice further supports an earlier donor antihost CD8 CTL response in males.

Lastly, as shown in Fig. 1D, donor CD8 T cells in CD8-intact m→M mice undergo contraction from days 10 to 14, whereas donor CD8 T cell engraftment generally increases in CD8-intact f→F mice during that same period, although with high intragroup variability. However, host B cells, CD4, and CD8 rebound in both male and female CD8-intact→F1 groups between days 10 and 14 (Fig. 5A–C) such that by day 14, host B cells and CD4 and CD8 T cells are no longer significantly reduced compared with values for sex-matched CD8-depleted→F1 mice. These results indicate that elimination of host cells is transient and incomplete for both male and female CD8-intact→F1 mice and is followed by evolution to a typical chronic GVHD phenotype by day 14. We found no significant sex-based differences in host Tregs at any of the four time points. Moreover, an analysis at day 14 of donor and host Treg populations for low dose (12 × 106 CD4) CD8-intact→F1 mice (Fig. 1A, 1D) also demonstrated no significant sex-based skewing toward the Treg subset.

Sex-based differences in F1 antiparent responses parallel sex-based differences in the parent–anti-F1 response

A counterregulatory F1 antiparent or host-versus-graft (HVG) response is well described in the p→F1 model not only for acute GVHD (16), but also for the DBA→F1 model of chronic GVHD in which elimination of donor cells by host CD8 CTL and NK cells has been demonstrated (16). The relatively weaker F1 antiparent response mitigates, but does not prevent the stronger parent anti-F1 response (16). Because male CD8-intact→F1 mice exhibit an earlier GVH response compared with females, as shown by earlier donor CD8 T cell proliferation (Fig. 4) and earlier elimination of host B cells and CD4 T cells (Fig. 5A, 5B), it might be expected that there would be an earlier reciprocal HVG response in males in the form of earlier proliferation of host CD8 T cells that are less sensitive to elimination than are host CD4 T cells. As shown in Fig. 6, CD8-intact m→M mice exhibit significantly greater day 8 host CD8 T cell proliferation, either by percentage (Fig. 6A) or numbers (Fig. 6B) compared with any of the following: CD8-intact f→F mice; CD8-depleted→F1 mice (male and female); or control F1 mice. By day 10, sex-based differences are lost for CD8-intact→F1 mice as host CD8 proliferation (both numbers and percentages) increases in f→F mice to levels equivalent for CD8-intact m→M mice. Representative flow cytometry tracings are shown in Fig. 6C–F. These sex-based differences (i.e., earlier male response) in host antidonor CD8 proliferation mirror the sex-based differences in donor antihost CTL elimination of host cells (Fig. 5), and support the conclusion that m→M mice exhibit an earlier parent anti-F1 response (day 8 males versus day 10 females) that in turn elicits an earlier F1 antiparent response. Importantly, at day 10 both m→M and f→F CD8-intact mice exhibit significantly greater host CD8 T cell proliferation (both percentage and numbers) compared with either CD8-depleted→F1 mice or uninjected control mice, indicating that host CD8 T cell proliferation is strikingly and significantly dependent on the injection of donor CD8 T cells. Nevertheless, a small, but significant increase in both percentage and numbers of proliferating host CD8 T cells is also seen at days 8–12 for CD8-depleted→F1 mice compared with uninjected control F1 mice. Thus, the day 8–12 host CD8 T cell antiparent response consists of a low-level donor CD4-dependent response and a significantly greater donor CD8-dependent response.

FIGURE 6
Donor CD8 T cells induce significantly greater host CD8 T cell proliferation at days 8 and 10. Experimental groups and flow cytometry staining protocol are as outlined for Fig. 4. The percentage (A) and number (B) of host CD8 T cells staining KI-67high ...

From days 10 to 12, both intact→F1 and depleted→F1 mice exhibit an increase in the number of proliferating host CD8 T cell numbers (Fig. 6B) simultaneously with contraction of donor CD8 T cell numbers in intact m→M mice (Fig. 1) and the transition in phenotype to chronic GVHD in both sexes (13, 16).

Confirmatory evidence of sex-based differences in the F1 anti-parent response was sought in the cohort receiving 12 × 106 donor CD4 T cells as measured by host NK cell numbers and host CD8 expression of CD107a, a marker of recent degranulation and CTL activity (17). As shown in Supplemental Fig. 1A, CD8-intact m→M mice exhibit significantly greater numbers of host NK cells at day 8 compared with either CD8-intact f→F or uninjected control F1 mice. Male CD8-intact→F1 mice also exhibit significantly greater CD107a expression on both host (Supplemental Fig. 1B) and donor CD8 T cells (Supplemental Fig. 1C) at days 8 and 10 compared with females.

Taken together, these results support the conclusion that sex-based differences in CD8-intact→F1 mice result in earlier GVH and HVG responses in males compared with females.

Sex-based differences in expression of cytokines important in CTL response

Cytokines important in CD8 CTL activation were examined for sex-based differences during days 8–14. Cytokine gene expression was evaluated in whole spleen preparations (Fig. 7) and the kinetics of IFN-α–inducible genes Mx-1, OAS, and IP-10 (both IFN-α and IFN-γ inducible) determined for both CD8-intact→F1 (Fig. 7A, 7C, 7E) and CD8-depleted→F1 (Fig. 7B, 7D, 7F) mice. Three observations can be made from Fig. 7. First, for both sexes, CD8-intact→F1 mice exhibit a peak at day 8 for all three genes, followed by rapid downregulation by day 12. Secondly, males exhibit a significantly higher day 8 peak for all three genes and a significantly greater day 10 value for Mx-1 and OAS. Lastly, CD8-depleted→F1 mice of both sexes exhibit flattened curves for these three genes with either a nonexistent or very slight peak from day 10 to day 12. Although low-level (<2-fold) sex-based differences were observed within these flattened curves, a more striking effect is the dramatic effect of donor CD8 depletion. Specifically, the elevated peak day 10 expression seen for both sexes in CD8-intact→F1 mice is reduced to values ≤2-fold over control for both sexes in CD8-depleted→F1 mice. Because of the important role of IFN-α in CTL effector generation (18), the greater expression in CD8-intact males versus females for Mx-1 and OAS is consistent with the greater male GVH and HVG CTL response shown in Figs. 4 and and5.5. Moreover, in the absence of donor CD8 T cells (CD8-depleted→F1 mice), no significant elevation of IFN-α–inducible genes over control is detectable from days 8 to 12, indicating that sex-based differences in IFN-α–inducible gene expression at these times are absolutely dependent on the transfer of donor CD8 T cells for either sex.

FIGURE 7
Greater expression of IFN-α–inducible genes seen in male CD8-intact→F1 mice is lost with donor CD8 depletion. Cytokine gene expression was measured at the indicated times by RT-PCR, as described in Materials and Methods, for MX-1 ...

Consistent with increased donor T cell activity at days 8 and 10 in CD8-intact m→M mice, IL-2 (Fig. 8A, 8B) expression at days 8 and 10, although low, is increased in males versus females and sex-based differences are reduced, but not eliminated in CD8-depleted→F1 mice. These results further support the conclusion that the GVH and HVG reactions from days 8 to 10 are stronger in males than females for CD8-intact→F1 mice and are markedly attenuated in CD8-depleted→F1 mice.

FIGURE 8
Greater expression of IL-21 gene expression in females at day 14 is seen for CD8-intact→F1 mice and lost with donor CD8 depletion. Experimental protocol, mean values, and p value comparisons are as described for Fig. 8. Cytokine gene expression ...

Although IFN-γ gene expression exhibits a different kinetic curve from the IFN-α–inducible genes, greater IFN-γ expression is seen at days 8 and 10 for males versus females for CD8-intact→F1 mice and little IFN-γ expression is seen in either sex for CD8-depleted→F1 mice (Fig. 8C, 8D).

IL-21 has both CD8 CTL-promoting and B cell-promoting properties (19). The kinetics of IL-21 gene expression in CD8-intact→F1 mice differ from those of IFN-α–inducible genes in that from days 8 to 10, CD8-intact→F1 mice exhibit consistent IL-21 expression without a day 8 peak or sex-based differences (Fig. 8E). From days 12 to 14, both sexes exhibit an increase in IL-21 expression that is significantly greater for females versus males. For CD8-depleted→F1 mice (Fig. 8F), the kinetics of IL-21 gene expression exhibit a reversal of the sex-based differences seen for the other cytokine genes tested and exhibit greater expression females than males at days 8 and 10. Values then decline in both groups and no further sex-based differences are observed. Importantly, by day 14, sex-based differences in IL-21 gene expression are seen only in CD8-intact→F1 mice, that is, are increased in f→F mice (Fig. 8E) and correspond to the significantly greater day 14 donor CD4 engraftment in CD8-intact f→F versus m→M mice (Fig. 1B). Taken together, the results indicate that cytokines important in cytotoxic responses are elevated in a pattern that supports the conclusion that the cytotoxic HVG and GVH response from days 8 to 10 is stronger in males compared with females, and sex-based differences are lost if CD8 T cells are depleted from the donor inoculum. The GVH and HVG responses terminate by ~day 12 and are followed by an increase in IL-21 expression from days 12 to 14 that is significantly greater for CD8-intact f→F versus m→M mice. It should be noted that although CD8-depleted→F1 mice do not exhibit sex-based differences in IL-21 gene expression, values are still elevated ~40-fold over control for both groups at day 14.

Donor T cell homeostatic contraction from days 10 to 12 is significantly greater in males versus females

In the B6→BDF1 model of acute GVHD, upregulation of Fas on donor T cells at ~day 10 plays a major role in their subsequent homeostatic contraction, particularly for donor CD8 T cells (13, 20). Moreover, Fas upregulation is critically dependent on IFN-γ in this model, and serum IFN-γ levels from days 8 to 12 exhibit striking 10- to 100-fold elevations over control (16, 20, 21). By contrast, DBA→F1 chronic GVHD male mice exhibit only modest (~2-fold) elevations in serum IFN-γ over control and minimal upregulation of Fas on donor T cells (16, 21). Because contraction of male donor T cells in DBA→F1 mice is observed from days 10 to 12 (Fig. 1), we tested the possibility that male donor T cells may in turn exhibit greater upregulation of molecules important in homeostatic contraction other than IFN-γ–dependent Fas to include PD-1 and CD80, the latter previously shown to limit peak expansion of donor T cells in this model (22). New F1 cohorts were injected using the same protocol in Fig. 1B and 1D, that is, unfractionated (CD8-intact) DBA donor splenocytes normalized to contain ~14 × 106 CD4 T cells. Mice were examined at days 10 and 12. As shown in Fig. 9A and 9B, engrafted donor CD8 and CD4 T cells are significantly greater for males versus females at day 10; however, by day 12 these differences are no longer significant due primarily to contraction of male donor T cells down to female levels. These results reproduce the day 10 and 12 donor T cell engraftment kinetics at the same donor cell dose for the cohorts shown in Fig. 1B and 1D.

FIGURE 9
Males exhibit greater ho-meostatic contraction than females from days 10 to 12. Two separate cohorts of male and female BDF1 mice received same sex donor cells (m→M, f→F) consisting of un-fractionated DBA splenocytes normalized for donor ...

Sex-based differences in molecules important in homeostatic contraction are best seen at day 10 after donor cell transfer. Specifically, males exhibit significantly greater day 10 numbers of donor CD8 and CD4 T cells that have upregulated PD-1 (Fig. 9C, 9D) and CD80 (Fig. 9E, 9F). Similar trends were seen for Fas (Fig. 9G, 9H); however, both the day 10 differences and numbers were smaller (although significant), most likely reflecting the very low levels of IFN-γ in DBA→F1 mice compared with that seen in B6→F1 acute GVHD mice (16). Lastly, we examined donor T cell apoptosis as shown by annexin-positive, 7-aminoactinomycin D–negative staining. Males exhibited significantly greater numbers of donor CD8 T cells undergoing recent apoptosis at day 10, but not at day 12 (Fig. 9I). No significant differences in apoptosis were observed between males and females for donor CD4 T cells, and total numbers for both were very low (data not shown). Representative staining profiles are shown in Supplemental Figs. 2 and 3 for donor CD8 T and CD4 T cells, respectively.

Taken together, these results support the conclusion that greater day 10 engraftment of donor T cells in males is accompanied by greater day 10 upregulation of molecules important in limiting expansion and terminating the immune response, followed in turn by greater male homeostatic contraction and, in the case of donor CD8 T cells, greater apoptosis.

Sex-based differences in donor T cell cytokine production following homeostatic contraction

In the B6→BDF1 model of acute GVHD, both TNF and IFN-γ are part of a robust donor antihost CD8 CTL response, and each cytokine positively influences the production of the other. For example, IFN-γ production is required for optimal TNF and perforin expression (21) and TNF is required for optimal IFN-γ production (23). The role of these cytokines in DBA→F1 chronic GVHD is less clear. We therefore examined donor T cells for intracellular production of IFN-γ, TNF, and perforin expression from the cohorts shown in Fig. 9 at days 10 and 12. Results represent direct ex vivo measurement without a secondary restimulation phase or Golgi blocking agents. At day 10, low levels of IFN-γ– and TNF-expressing donor T cells were observed, and although some minor sex-based differences were seen, values were typically less than or equal to values for control uninjected donor DBA or host F1 levels (data not shown). However, by day 12, consistent elevations over controls were observed along with sex-based differences (Fig. 10). For example, females exhibit a significantly greater percentage of IFN-γ–expressing donor CD4 T cells (mean percentage of m versus f = 11.6 versus 19.5%, p < 0.05), although the total number of IFN-γ–expressing donor CD4 T cells was not significantly different (Fig. 10A–D). Similarly, female donor CD4 T cells exhibit significantly greater intracellular TNF production versus males, as shown by both percentage (mean percentage of m versus f = 5.2 versus 19.2%, p < 0.001) and numbers (Fig. 10E–H). Females also exhibit significantly greater perforin-expressing donor CD4 T cells both by percentage (mean percentage of m versus f = 2.9 versus 13.4%, p < 0.001) and by actual number (Fig. 10I–L). Lastly, females exhibit a significantly greater percentage (mean percentage of m versus f = 20.6 versus 34.2%, p < 0.05) and number of proliferating (KI-67–positive) donor CD4 T cells at day 12 (Fig. 10M–P). These latter results confirm the results of Fig. 4A, in which females also exhibited greater numbers of day 12 proliferating donor CD4 T cells versus males; however, in that experiment the differences did not reach statistical significance.

FIGURE 10
Female donor CD4 T cells exhibit increased cytokine production and proliferation at day 12 compared with males. Experimental details are as described in Fig. 9 for the cohort examined at day 12. Intracellular cytokine production was assessed by flow cytometry, ...

By contrast, donor CD8 T cells did not mirror the results for donor CD4 T cells. Specifically, there were no significant sex-based differences in the percentage or number of TNF- or perforin-positive donor CD8 T cells at either day 10 or day 12. Male donor CD8 T cells, however, did exhibit a significantly greater percentage of IFN-γ–producing cells versus females (mean value m versus f = 33.3 versus 23.7%, p > 0.001) at day 12; however, values for the total number of male versus female IFN-γ–producing donor CD8 T cells did not reach statistical significance (data not shown). No significant sex-based differences were observed for donor CD8 KI-67 expression for either percentage or number.

Taken together, the data in Figs. 9 and and1010 demonstrate that compared with females, males exhibit greater day 10 donor T cell numbers and exhibit greater T cell upregulation of markers important in downregulation, followed by strong homeostatic contraction from days 10 to 12. By day 12, activation parameters are at or near control levels. In contrast, upregulation of molecules important in homeostatic contraction is less pronounced in female donor T cells, and by day 12, female donor CD4 T cells exhibit evidence of persistent proliferation, activation, and cytokine secretion.

Greater female severity in 2-wk chronic GVHD phenotype is IFN-γ dependent

The foregoing data demonstrate that sex-based differences in donor CD4 T cell engraftment in DBA→F1 mice are critically dependent on donor CD8 T cells. Moreover, sex-based differences in gene expression of cytokines important in CD8 T CTL responses (Mx-1, OAS, IFN-γ, and IP-10) were seen at days 8–10 and are lost with donor CD8 T depletion. Because donor CD8 T cells play an important role in IFN-γ production and homeostatic contraction in the p→F1 model (13, 20, 21), it is possible that IFN-γ plays an important role in greater day 14 female donor CD4 T cell engraftment despite the relatively low levels of IFN-γ gene expression and Fas upregulation seen in CD8-intact DBA→F1 chronic GVHD mice compared with B6→F1 acute GVHD mice. To test this, we addressed the role of sex-based differences in IFN-γ production during the effector phase (days 8–10) rather than during the initial activation phase (day 0) by treating DBA→F1 GVHD mice (induced as described in Figs. 9, ,10)10) with neutralizing doses of anti–IFN-γ or equivalent doses of isotype control mAb (GL117) at day 6, that is, just prior to peak IFN-γ gene expression and serum protein production (16). Chronic GVHD phenotype was assessed at day 14 (Figs. 11, ,12).12). Females exhibited greater engraftment of both donor CD4 (Fig. 11A) and CD8 (Fig. 11B) T cells versus males for GVHD mice receiving control mAb, confirming the day 14 results shown in Fig. 1B and 1D. Importantly, sex-based differences in both donor CD4 and CD8 day 14 engraftment are lost in mice receiving anti–IFN-γ mAb. On a minor note, the greater donor CD8 T cell engraftment for females seen in Fig. 1D now reaches statistical significance in Fig. 11B for control mAb-treated GVHD mice due to reduced mouse-to-mouse variability in the latter figure despite the lower mean value and the loss of the high outliers seen in Fig. 1D.

FIGURE 11
Sex-based differences (greater female severity) in day 14 chronic GVHD phenotype are IFN-γ dependent. Male and female BDF1 mice received transfers same sex donor cells consisting of unfractionated DBA splenocytes normalized for donor CD4 T cells ...
FIGURE 12
Sex-based differences (greater female severity) in day 14 cytokine production are IFN-γ dependent. Experimental protocol is as described in Fig. 10. Splenocytes were examined at day 14 by flow cytometry for the following: A, donor CD4 IFN-γ ...

Female DBA→F1 mice receiving control mAb also exhibit greater expansion of host CD4 T cells (Fig. 11C), host CD8 T cells (Fig. 11D), host B cells (Fig. 11E), and host CD11c+ DC (Fig. 11F) consistent with greater donor CD4 T cell activation in females. Lastly, control mAb-treated f→F GVHD mice exhibit significantly greater numbers of donor and CD4 T (Fig. 11G) and CD8 T cells (Fig. 11H) that are actively proliferating. All of these sex-based differences are lost with IFN-γ mAb treatment (Fig. 11). Together, these results support the idea that greater donor and host activation in female DBA→F1 mice is IFN-γ dependent.

This conclusion is further supported by analysis of intracellular cytokine production, demonstrating that control mAb-treated female GVHD mice exhibit a greater number of donor CD4 T cells expressing intracellular IFN-γ (Fig. 12A), TNF (Fig. 12B), perforin (Fig. 12C), and granzyme B (Fig. 12D) compared with males at day 14. Control mAb-treated female GVHD mice also exhibited significantly greater numbers of donor CD8 T cells expressing intracellular IFN-γ (Fig. 12E) and granzyme B (Fig. 12F) versus males. No significant differences in TNF or perforin were seen (data not shown). All of these sex-based differences are lost with anti–IFN-γ treatment (Fig. 12). We were unable to detect any significant increase in whole spleen TNF gene expression by PCR for either males or females compared with uninjected normal F1 spleens for the cohorts shown in Figs. 912 (data not shown).

These results demonstrate that IFN-γ production is critical to sex-based differences in day 14 donor CD4 T cell engraftment and host expansion. Secondly, at day 14, control mAb-treated female DBA→F1 mice exhibit significantly greater ongoing donor T cell activation, particularly of donor CD4 T cells, as measured by both cytokine secretion and proliferation.

Discussion

CD4 T cells play a central role in the pathogenesis of both human and murine lupus (reviewed in 24). In the p→F1 model of lupus, not only are donor CD4 T cells both necessary and sufficient for disease induction (6, 7), but also the number of transferred donor CD4 T cells correlates directly with disease parameters (10, 13). Moreover, in DBA→BDF1 lupus mice, females exhibit greater long-term disease severity compared with males despite the transfer of equal numbers of donor splenocytes (8). This sex-based difference in disease severity is preceded at 2 wk by significantly greater donor CD4 T cell engraftment in females compared with males (8), supporting the idea that the number of engrafted CD4 effector/Th cells at 2 wk may be a valid surrogate marker for long-term disease severity in this model. Our study sought to determine the mechanism responsible for sex-based differences in CD4 engraftment at 2 wk.

Our experiments focused on the second week after donor transfer because previous work in DBA→F1 mice transferring equal numbers of DBA splenocytes demonstrated the following: 1) no sex-based differences in donor CD4 T cell engraftment could be detected during the first week after transfer; and 2) female donor CD4 T cells exhibited greater in vivo proliferation (i.e., BrdU incorporation) compared with males during the second week after transfer (8). In the current study, we transferred unfractionated DBA donor splenocytes containing equal numbers of CD4 T cells rather than simply equal numbers of DBA splenocytes. Using this approach, we observed that females receiving unfractionated splenocytes equalized for CD4 T cells (CD8-intact→F1 mice) exhibited significantly greater (≥2.5-fold) day 14 engraftment of donor CD4 T cells compared with males, thereby confirming previous published work using donor cell transfers normalized only for unfractionated splenocytes (8). Surprisingly, we also observed that the greater day 14 CD4 engraftment in females was preceded at days 8–10 by greater male: 1) donor CD4 and CD8 proliferation; 2) elimination of host cells; 3) host CD8 T cell proliferation and CD107 expression; 4) host NK numbers; and 5) expression of cytokine genes relevant to CTL maturation and function, all of which are characteristics of a GVH and HVG cytolytic response. The stronger male day 8–10 GVH and HVG reactions were followed in males by a strong homeostatic contraction of donor CD8 T cells from days 10 to 12. Specifically, from days 10 to 12, we observed two different kinetics for donor T cell engraftment in males versus females. Females exhibited essentially a flat curve, whereas males exhibited a day 10 peak of donor CD4 and CD8 T cell engraftment that was significantly greater than female values, followed by a steep decline during the homeostatic contraction phase to levels at day 12 that did not differ significantly from females. Despite similar donor T cell numbers at day 12, females now exhibited significantly greater donor CD4 T cells expressing IFN-γ, TNF, and perforin, a trend that continued to day 14, at which time females also exhibit significantly greater numbers of proliferating donor CD4 and CD8 T cells and greater numbers of CD4 T cells expressing IFN-γ, TNF, perforin, and granzyme B. At day 14, females also exhibit greater numbers of CD8 T cells that are proliferating and expressing IFN-γ and granzyme B. Thus, by day 12, when donor engraftment is relatively equal, male donor T cells are terminating the GVH/HVG response, whereas female donor T cells exhibit continuing activation and are beginning the evolution to a CD4 effector cytokine phenotype that by day 14 clearly differs from males.

It is clear from our in vivo neutralization studies that IFN-γ is critical for day 14 sex-based differences in chronic GVHD phenotype; however, TNF also may play an important role. TNF is well recognized for its ability to promote T cell proliferation and IFN-γ production in vitro (25). Moreover, Brown et al. (26), using an irradiated recipient model of GVHD, demonstrated that donor CD4 T cell production of TNF in vivo is critical for alloproliferation and IFN-γ production. Similarly, in the p→F1 model of acute GVHD, optimal TNF production and perforin expression require IFN-γ (21). Conversely, TNF is critical for CTL development that is a major producer of IFN-γ in this model (23). Whether IFN-γ and TNF exert functions other than promoting greater female alloproliferation at day 14 will require further studies.

Taken together, our results demonstrate that greater day 14 engraftment of female donor CD4 T cells is preceded at day 10 by greater male donor T cell engraftment as part of a stronger GVH and HVG in males that are associated with greater and longer lasting homeostatic contraction compared with females. We have previously demonstrated that greater female CD4 engraftment correlates with the sex of the host and not the donor (8). Thus, the mechanism we propose is that in the male environment, the donor CD8 CTL response is more intense and in turn elicits a stronger downregulatory response, both extrinsic in the form of HVG response and intrinsic as shown by greater donor T cell upregulation of molecules important in terminating the immune response (PD-1, CD80). The weaker response in females allows persistence of both CD4 and CD8 T cells that then exhibit increasing activation from days 12 to 14 versus male donor cells. Importantly, these sex-based differences are IFN-γ dependent.

Although the day 14 engraftment of donor CD8 T cells in females was characterized in some experiments by high intragroup variability, analysis of individual female mice demonstrated that greater persistence of donor CD8 T cells from days 12 to 14 was significantly and positively correlated with greater donor CD4 T cell engraftment, whereas no similar correlation was seen in males during this same time due in large part to the low level of donor CD8 T cells present. However, at days 8 and 10, males exhibited significantly greater donor CD8 T cell engraftment than did females, and this also was associated with significantly greater donor CD4 engraftment compared with females. Thus, in both sexes, greater donor CD8 T cell engraftment is associated with greater donor CD4 T cell engraftment, regardless of whether it be greater male engraftment at day 8 or greater female engraftment at day 14.

The question arises then as to which event is primary, that is, do greater CD8 persistence and engraftment promote greater CD4 T cell engraftment, as suggested by Fig. 1, or do intrinsic sex-based differences in donor CD4 T cell expansion promote corresponding differences in donor CD8 T cell engraftment? This question was mechanistically addressed by transferring donor splenocytes selectively depleted of CD8 T cells (CD8-depleted→F1). If sex-based differences in day 14 CD4 engraftment reflect primary CD4-intrinsic properties, then depletion of donor CD8 T cells should not alter day 14 engraftment. Conversely, if sex-based differences in donor CD4 engraftment are secondary to sex-based differences in donor CD8 T cell survival after day 10, then donor CD8 depletion should eliminate or mitigate these differences due to a reduction in female day 14 CD4 T cell numbers rather than from an enhancement of male numbers. This latter outcome was in fact the case. Specifically, the greater female:male day 14 engraftment was narrowed from ~2.5-fold in CD8-intact→F1 mice to ~1.2-fold in CD8-depleted→F1 mice. Moreover, this near equalization in engraftment resulted from a significant drop in day 14 female donor CD4 engraftment rather than from a significant increase in male CD4 engraftment, supporting a primary role for donor CD8 T cells in promoting sex-based differences in donor CD4 T cell engraftment.

Interestingly, CD8-depleted→F1 mice also exhibited a loss of the greater day 8 GVH and HVG reactions in males. For example, sex-based differences from days 8 to 12 are lost or significantly attenuated in CD8-depleted→F1 mice for the following: 1) donor CD4 (GVH) or host CD8 (HVG) T cell proliferation; 2) the kinetics of host B cell, or CD4 or CD8 T cell expansion; and 3) expression of cytokine genes important in CTL responses. CD8-depleted→F1 mice exhibit a loss of both the day 8 greater male GVH/HVG and the day 14 greater female donor CD4 engraftment seen in CD8-intact→F1 mice, supporting the conclusion that the two events are functionally linked, that is, that donor CD8 T cells are required for sex-based differences. Further supporting this idea is the observation that values for both donor CD4 and host CD8 T cell proliferation in both sexes of CD8-depleted→F1 mice are significantly reduced compared with CD8-intact→F1 mice (best seen at day 10), indicating that the presence of donor CD8 T cells induces a stronger GVH and HVG for both sexes.

Taken together, our results support the following conclusions: 1) the presence of donor CD8 T cells enhances donor CD4 T cell expansion and engraftment; 2) sex-based differences in donor CD8 engraftment are associated with corresponding sex-based differences in CD4 engraftment; 3) prolonged survival of donor CD8 T cells in females is causally related to greater day 14 engraftment of donor CD4 T cells in females; and 4) IFN-γ is critical for sex-based differences in CD8-intact→F1 mice. We hypothesize that the weaker day 8 GVH/HVG in females results in weaker donor CD8 T cell contraction and greater persistence after day 10 that in turn are associated with the production of cytokines important in alloproliferation (TNF and IFN-γ), resulting in greater day 14 female donor CD4 engraftment.

Our results demonstrate a novel role for donor CD8 T cells in shaping sex-based differences in lupus severity. In contrast to the well-documented central role of donor CD4 T cells in inducing a lupus phenotype in the p→F1 model, a role for donor CD8 T cells in lupus induction has not been previously described. To the contrary, donor CD8 T cells typically prevent a lupus-like phenotype by inducing an acute GVHD phenotype characterized by maturation of donor CD8 T cells into effector CTL specific for host MHC I and elimination of host B cells, thereby ending the potential for sustained humoral autoimmunity. Conversely, a failure of donor CD8 T cells to mature into effector CTL and completely eliminate host B cells converts an expected acute GVHD in B6→BDF1 mice to a lupus-like disease and can be seen in F1 mice receiving perforin-defective B6 donor cells (27) or anti-TNF mAb in vivo (23). Impaired donor CTL function is also characteristic of the DBA→BDF1 transfer in that DBA donor cells exhibit defects in both CD8 anti-F1 precursor CTL frequency (28, 29) and CD4 helper ability (30). As demonstrated in this study and previously (16), donor antihost and host antidonor CTL do in fact develop in DBA→F1 mice; however, donor CD8 CTL activity is defective compared with that of B6→BDF1 mice. As a result, host B cells reach a day 10 nadir in both males and females that is only slightly reduced (80–87% of control F1 values), whereas CD8-depleted→F1 mice exhibit expansion of host B cells at day 10 (180–190% of control F1 values) due to the absence of donor CD8 antihost CTL. Simultaneous with day 12–14 contraction of donor CD8 T cells in males, the day 10 nadir in CD8-intact→F1 mice is followed at days 12–14 by a rebound expansion in host B cells to levels comparable to those of CD8-depleted→F1 mice (Fig. 5A).

By comparison, the host B cell nadir in acute GVHD mice (B6→BDF1) is typically ~10% of control F1 values and is sustained rather than transient, thereby preventing long-term lupus development (13, 16). Thus, sex-based differences in day 14 donor CD4 T cell engraftment require the presence not simply of donor CD8 T cells, but of CD8 T cells that exhibit impaired antihost CTL activity. We have recently observed that CD8-depleted→F1 mice do not exhibit significant sex-based differences in long-term lupus-like renal disease severity, further corroborating 2-wk CD4 engraftment as a surrogate marker for long-term disease severity (14). Together, these results support the novel conclusion that sex-based differences in suboptimal donor CD8 CTL activation are critical for shaping sex-based differences in donor CD4 T cell engraftment at 2 wk and lupus-like disease long-term.

Lastly, IL-21 has been recently identified as a cytokine important in both CD8 CTL responses and T cell-driven B cell hyperactivity in lupus models (19). The kinetics of IL-21 expression in our study differ from the kinetics of the IFN-α–related genes. IFN-α–related gene expression in CD8-intact→F1 mice is elevated at day 8 particularly in males and declines over the next 6 d. By contrast, CD8-intact→F1 mice exhibit significant and stable elevations in IL-21 gene expression during days 8–10 (~20- to 30-fold above control). From days 12 to 14, IL-21 levels increase and are significantly greater in females versus males and correspond to the significantly greater female day 14 donor CD4 engraftment in CD8-intact→F1 mice. These results raise the possibility that greater female IL-21 expression may contribute to greater female disease severity for CD8-intact→F1 mice and support testing of IL-21 blockade in this model given its role in humoral autoimmunity (19).

Extrapolating our results in mice to humans with lupus, it is possible that following exposure to a lupus-inducing agent and activation of Ag-specific CD4 T cells, development of lupus is prevented in the setting of an optimal downregulatory CD8 T cell response, possibly CTL. By contrast, development of lupus may be facilitated in the setting of an absent or suboptimal downregulatory CD8 response in both sexes. Based on our results, we postulate that in the presence of a suboptimal CD8 T cell response, females may exhibit both a less intense initial activation and secondary downregulatory response, allowing greater persistence of suboptimal Ag-specific CD8 T cells that in turn promote greater expansion of Ag-specific CD4 T cells and with it the potential for greater B cell hyperactivity, autoantibody production, and clinical lupus expression. Our results support the therapeutic potentiation of CD8 T cells in lupus (reviewed in Ref. 31) not only to enhance direct elimination of activated autoreactive B cells, but also to induce a stronger postactivation contraction phase of CD8 T cells, thereby removing a stimulus to greater CD4 T cell expansion and the possibility of greater help to autoreactive B cells.

Supplementary Material

Fig 1

Fig 2

Fig 3

Acknowledgments

This work was supported by National Institutes of Health Grant AI047466 (to C.S.V.). R.P. is the recipient of an Engalticheff Fellowship.

Abbreviations used in this article

BDF1
B6D2F1
CD8-depleted→F1
CD8-depleted DBA into BDF1
CD8-intact→F1
undepleted DBA splenocytes into BDF1
DBA
DBA/2J
DC
dendritic cell
f→F
female into female
GVH
graft-versus-host
GVHD
GVH disease
HVG
host-versus-graft
m→M
male into male
p→F1
parent-into-F1
Treg
T regulatory cell

Footnotes

The online version of this article contains supplemental material.

Disclosures

The authors have no financial conflicts of interest.

References

1. McCarty DJ, Manzi S, Medsger TA, Jr, Ramsey-Goldman R, LaPorte RE, Kwoh CK. Incidence of systemic lupus erythematosus: race and gender differences. Arthritis Rheum. 1995;38:1260–1270. [PubMed]
2. Lahita RG. The influence of sex hormones on the disease systemic lupus erythematosus. Springer Semin Immunopathol. 1986;9:305–314. [PubMed]
3. Gleichmann E, Gleichmann H. Pathogenesis of graft-versus-host reactions (GVHR) and GVH-like diseases. J Invest Dermatol. 1985;85:115s–120s. [PubMed]
4. Eisenberg R. The chronic graft-versus-host model of systemic autoimmunity. Curr Dir Autoimmun. 2003;6:228–244. [PubMed]
5. Rolink AG, Gleichmann E. Allosuppressor- and allohelper-T cells in acute and chronic graft-vs.-host (GVH) disease. III. Different Lyt subsets of donor T cells induce different pathological syndromes. J Exp Med. 1983;158:546–558. [PMC free article] [PubMed]
6. Morris SC, Cohen PL, Eisenberg RA. Experimental induction of systemic lupus erythematosus by recognition of foreign Ia. Clin Immunol Immunopathol. 1990;57:263–273. [PubMed]
7. Morris SC, Cheek RL, Cohen PL, Eisenberg RA. Autoantibodies in chronic graft versus host result from cognate T-B interactions. J Exp Med. 1990;171:503–517. [PMC free article] [PubMed]
8. Lang TJ, Nguyen P, Papadimitriou JC, Via CS. Increased severity of murine lupus in female mice is due to enhanced expansion of pathogenic T cells. J Immunol. 2003;171:5795–5801. [PubMed]
9. van Rappardvan der Veen FM, Rolink AG, Gleichmann E. Diseases caused by reactions of T lymphocytes towards incompatible structures of the major histocompatibility complex. VI. Autoantibodies characteristic of systemic lupus erythematosus induced by abnormal T-B cell cooperation across I-E. J Exp Med. 1982;155:1555–1560. [PMC free article] [PubMed]
10. Van Rappard-Van Der Veen FM, Radaszkiewicz T, Terraneo L, Gleichmann E. Attempts at standardization of lupus-like graft-vs-host disease: inadvertent repopulation by DBA/2 spleen cells of H-2-different non-irradiated F1 mice. J Immunol. 1983;130:2693–2701. [PubMed]
11. Grader-Beck T, Casciola-Rosen L, Lang TJ, Puliaev R, Rosen A, Via CS. Apoptotic splenocytes drive the autoimmune response to poly (ADP-ribose) polymerase 1 in a murine model of lupus. J Immunol. 2007;178:95–102. [PubMed]
12. Via CS, Shearer GM. Murine graft-versus-host disease as a model for the development of autoimmunity: relevance of cytotoxic T lymphocytes. Ann N Y Acad Sci. 1988;532:44–50. [PubMed]
13. Puliaeva I, Puliaev R, Shustov A, Haas M, Via CS. Fas expression on antigen-specific T cells has costimulatory, helper, and down-regulatory functions in vivo for cytotoxic T cell responses but not for T cell-dependent B cell responses. J Immunol. 2008;181:5912–5929. [PMC free article] [PubMed]
14. Foster AD, Haas M, Puliaeva I, Soloviova K, Puliaev R, Via CS. Donor CD8 T cell activation is critical for greater renal disease severity in female chronic graft-vs.-host mice and is associated with increased splenic ICOS(hi) host CD4 T cells and IL-21 expression. Clin Immunol. 2010;136:61–73. [PMC free article] [PubMed]
15. Gerdes J, Li L, Schlueter C, Duchrow M, Wohlenberg C, Gerlach C, Stahmer I, Kloth S, Brandt E, Flad HD. Immunobiochemical and molecular biologic characterization of the cell proliferation-associated nuclear antigen that is defined by monoclonal antibody Ki-67. Am J Pathol. 1991;138:867–873. [PMC free article] [PubMed]
16. Puliaev R, Puliaeva I, Welniak LA, Ryan AE, Haas M, Murphy WJ, Via CS. CTL-promoting effects of CD40 stimulation outweigh B cell-stimulatory effects resulting in B cell elimination and disease improvement in a murine model of lupus. J Immunol. 2008;181:47–61. [PMC free article] [PubMed]
17. Betts MR, Brenchley JM, Price DA, De Rosa SC, Douek DC, Roederer M, Koup RA. Sensitive and viable identification of antigen-specific CD8+ T cells by a flow cytometric assay for degranulation. J Immunol Methods. 2003;281:65–78. [PubMed]
18. Xiao Z, Casey KA, Jameson SC, Curtsinger JM, Mescher MF. Programming for CD8 T cell memory development requires IL-12 or type I IFN. J Immunol. 2009;182:2786–2794. [PMC free article] [PubMed]
19. Spolski R, Leonard WJ. Interleukin-21: basic biology and implications for cancer and autoimmunity. Annu Rev Immunol. 2008;26:57–79. [PubMed]
20. Shustov A, Nguyen P, Finkelman F, Elkon KB, Via CS. Differential expression of Fas and Fas ligand in acute and chronic graft-versus-host disease: up-regulation of Fas and Fas ligand requires CD8+ T cell activation and IFN-gamma production. J Immunol. 1998;161:2848–2855. [PubMed]
21. Puliaev R, Nguyen P, Finkelman FD, Via CS. Differential requirement for IFN-gamma in CTL maturation in acute murine graft-versus-host disease. J Immunol. 2004;173:910–919. [PubMed]
22. Puliaeva I, Soloviova K, Puliaiev M, Lang T, Puliaev R, Via CS. Enhancement of suboptimal CD8 cytotoxic T cell effector function in vivo using antigen-specific CD80 defective T cells. J Immunol. 2011;186:291–304. [PMC free article] [PubMed]
23. Via CS, Shustov A, Rus V, Lang T, Nguyen P, Finkelman FD. In vivo neutralization of TNF-alpha promotes humoral autoimmunity by preventing the induction of CTL. J Immunol. 2001;167:6821–6826. [PubMed]
24. Horwitz DA, Stohl W, Gray JD. T lymphocytes, natural killer cells and immune regulation. In: Wallace DJ, Hahn BH, editors. Dubois’ Lupus Erythematosus. Lippincott Williams & Wilkins; Philadelphia: 2002. pp. 157–185.
25. Yokota S, Geppert TD, Lipsky PE. Enhancement of antigen- and mitogen-induced human T lymphocyte proliferation by tumor necrosis factor-alpha. J Immunol. 1988;140:531–536. [PubMed]
26. Brown GR, Lee EL, Thiele DL. TNF enhances CD4+ T cell alloproliferation, IFN-gamma responses, and intestinal graft-versus-host disease by IL-12-independent mechanisms. J Immunol. 2003;170:5082–5088. [PubMed]
27. Shustov A, Luzina I, Nguyen P, Papadimitriou JC, Handwerger B, Elkon KB, Via CS. Role of perforin in controlling B-cell hyperactivity and humoral autoimmunity. J Clin Invest. 2000;106:R39–R47. [PMC free article] [PubMed]
28. Via CS, Sharrow SO, Shearer GM. Role of cytotoxic T lymphocytes in the prevention of lupus-like disease occurring in a murine model of graft-vs-host disease. J Immunol. 1987;139:1840–1849. [PubMed]
29. Tschetter JR, Mozes E, Shearer GM. Progression from acute to chronic disease in a murine parent-into-F1 model of graft-versus-host disease. J Immunol. 2000;165:5987–5994. [PubMed]
30. De Wit D, Van Mechelen M, Zanin C, Doutrelepont JM, Velu T, Gérard C, Abramowicz D, Scheerlinck JP, De Baetselier P, Urbain J, et al. Preferential activation of Th2 cells in chronic graft-versus-host reaction. J Immunol. 1993;150:361–366. [PubMed]
31. Puliaeva I, Puliaev R, Via CS. Therapeutic potential of CD8+ cytotoxic T lymphocytes in SLE. Autoimmun Rev. 2009;8:219–223. [PMC free article] [PubMed]
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...