NF-kappaB as a target for the prevention of graft-versus-host disease: comparative efficacy of bortezomib and PS-1145

Abstract

NF-kappaB is a transcription factor that controls the expression of a number of genes important for mediating immune and inflammatory responses. In this study, we examined whether bortezomib and PS-1145, each of which inhibits NF-kappaB, could protect mice from lethal graft-versus-host disease (GVHD), which is characterized by immune activation and proinflammatory cytokine production. When administered within the first 2 days after transplantation, bortezomib and PS-1145 both protected mice from fatal GVHD, did not compromise donor engraftment, and effected marked reduction in the levels of serum cytokines that are normally increased during GVHD. Extending the course of bortezomib administration or delaying the initiation of this agent for as few as 3 days after bone marrow transplantation (BMT), however, significantly exacerbated GVHD-dependent mortality because of severe pathological damage in the colon. In contrast, prolonged administration of PS-1145, which, unlike bortezomib, is a selective inhibitor of NF-kappaB, caused no early toxicity and resulted in more complete protection than that observed with an abbreviated PS-1145 treatment schedule. These results confirm a critical role for NF-kappaB in the pathophysiology of GVHD and indicate that targeted inhibition of NF-kappaB may have a superior therapeutic index and may constitute a viable therapeutic approach to reduce GVHD severity.

Figure 1.

Administration of bortezomib immediately before transplantation protects mice from lethal GVHD. Lethally irradiated (1000 cGy) B6 mice underwent transplantation with TCD B10.BR BM alone or together with B10.BR spleen cells adjusted to yield a dose of either 2 × 10⁶ (A) or 4 × 10⁶ T cells (B). Mice that received transplanted TCD BM only either were left untreated (□, n = 8) or were administered bortezomib (1 mg/kg, day 0) (○, n = 6) immediately before transplantation. Animals that received transplanted TCD BM plus B10.BR T cells also either were left untreated (▪, n = 8) or were administered bortezomib (•, n = 10) on day 0 immediately before TBI. Data are cumulative results from 2 experiments at each T-cell dose. Percentages of donor T-cell chimerism in the spleen (C) and overall spleen (D) and thymic (E) cellularity of mice described in panel A that survived 60 days after transplantation are shown (□, TCD BM; ▪, TCD BM/bortezomib; , TCD BM/spleen cells/bortezomib). Data in bar graphs are presented as the mean ± SEM. Spleen cellularity: TCD versus TCD/bortezomib, P = .53; TCD/bortezomib versus TCD/spleen/bortezomib, P = .002. Thymus cellularity: TCD versus TCD/bortezomib, P = 0.94; TCD/bortezomib versus TCD/spleen/bortezomib, P = .02.

Figure 2.

Extended posttransplantation administration of bortezomib exacerbates GVHD-dependent mortality. Lethally irradiated (1000 cGy) B6 mice underwent transplantation with TCD B10.BR BM alone (□, n = 6) or with spleen cells adjusted to yield a dose of 2 × 10⁶ B10.BR T cells. Mice that received transplanted B10.BR T cells were treated with PBS (•, n = 9) or bortezomib (▪, n = 9) (1 mg/kg twice weekly beginning on days 3-4). Survival is shown in panel A. In 2 additional experiments, cohorts of mice that underwent transplantation as in panel A were killed 6 to 7 days after BMT, and colon tissue was obtained for histologic analysis. Panel B shows pathology scores in the colons of mice that received TCD BM alone (□, n = 6) or with B10.BR T cells, followed by treatment with PBS (, n = 8) or bortezomib (▪, n = 7). In panels C and D, lethally irradiated B6 mice underwent transplantation with 10 × 10⁶ B6.SJL BM cells and then were treated with twice-weekly PBS (□, n = 10) or bortezomib (▪, n = 10) (1 mg/kg/dose) beginning 3 to 4 days after BMT. Survival is shown in panel C, and panel D shows mean percentages of pretransplantation weight loss/gain. Data are cumulative results from 2 independent experiments for each strain combination (allogeneic and syngeneic).

Figure 3.

Lethality caused by extended administration of bortezomib is attributable to gut toxicity. Lethally irradiated (1000 cGy) B6 mice underwent transplantation with TCD B10.BR BM alone or with spleen cells adjusted to yield a dose of 2 × 10⁶ B10.BR T cells. Mice that received transplanted B10.BR T cells were treated with PBS or bortezomib (1 mg/kg twice weekly beginning on day 3). Eight days after transplantation, prior to demise, mice from each of the 3 cohorts were killed, and samples of colon tissue were analyzed for evidence of pathologic damage. Hematoxylin and eosin stains of colon tissue obtained from these mice are depicted (original magnification, 200 ×). (A) Colon tissue from a representative mouse that underwent transplantation with TCD BM only showing normal-appearing colon tissue with intact crypts lined by mucin-filled enterocytes. (B) PBS-treated mouse undergoing GVHD showing increased mitotic activity at the base of crypts associated with a few crypt abscesses (solid arrows). (C) Bortezomib-treated animal demonstrating extensive crypt destruction with sloughing of the colonic mucosa (dashed arrows), goblet cell depletion, and numerous crypt abscesses (solid arrows).

Figure 4.

PS-1145 inhibits alloreactive T-cell responses in vitro but is less potent than bortezomib. B10.BR T cells (1 × 10⁵/well) were cultured with B6 CD11c⁺ dendritic cells (5 × 10⁴) in a standard mixed lymphocyte culture for 5 days. (A) Replicate wells were cultured alone (□) or in the presence of 5-fold μM dilutions of bortezomib (▪) or PS-1145 () for 5 days. Cells were pulsed with ³H-Tdr for the last 18 hours of culture, and the percentages of incorporated radioactivity were determined. (B-C) B10.BR T cells (1 × 10⁵/well) were cultured with B6 CD11c⁺ dendritic cells (5 × 10⁴) alone or with bortezomib (0.32 μM) or PS-1145 (8 μM) for 5 days. Bortezomib and PS-1145 were added to microwells on days 0, 1, or 2 after culture initiation. Cells were pulsed with ³H-Tdr for the last 18 hours of culture and the percentages of incorporated radioactivity were determined (B); in separate experiments, supernatants were collected and tested for the presence of IFN-γ (C). Data are presented as the mean ± SEM from triplicate control wells. Data shown are from 1 of 3 experiments that produced similar results. (D) B10.BR T cells were cultured with B6 CD11c⁺ dendritic cells alone or with bortezomib (0.32 μM) or PS-1145 (8 μM). Cells were harvested and pooled from triplicate microwells after 3 or 4 days and were surface stained for CD4 or CD8 and annexin V. Data are presented as the mean ± SEM. Data shown are cumulative results of 4 experiments.

Figure 5.

Administration of PS-1145 early after transplantation protects mice from lethal GVHD. Lethally irradiated (1000 cGy) B6 mice underwent transplantation with TCD B10.BR BM alone or with B10.BR spleen cells adjusted to yield a dose of 4 × 10⁶ T cells. Mice that received transplanted TCD BM only were untreated (□, n = 8) or were administered PS-1145 (50 mg/kg, days 0-2) (○, n = 5). Animals that received transplanted TCD BM plus B10.BR T cells were also left untreated (▪, n = 8) or were administered PS-1145 (•, n = 10) on days 0-2. (A) Survival percentages. (B) Weight loss percentages. Data are cumulative results from 2 experiments.

Figure 6.

Bortezomib and PS-1145 both effectively reduce serum cytokine levels in mice with GVHD. Lethally irradiated (1000 cGy) B6 mice underwent transplantation with TCD B10.BR BM alone (□) or with B10.BR spleen cells adjusted to yield a dose of 2 × 10⁶ T cells. Mice that received transplanted T cells were left untreated (▪) or were administered bortezomib (1 mg/kg, day 0) () or PS-1145 (50 mg/kg, days 0-2) (). Mice in all groups were bled on day 7 after BMT and were analyzed for the specified cytokines. Data are cumulative results from 2 independent experiments with 5 to 7 mice per group. Data are presented as the mean ± SEM. IL-2: bortezomib or PS-1145 versus GVHD control, P < .005. TNF-α: bortezomib or PS-1145 versus GVHD control, P = .02. IL-10: bortezomib or PS-1145 versus GVHD control, P < .005.

Figure 7.

Extended posttransplantation administration of PS-1145 protects animals from lethal GVHD and does not induce fatal gut toxicity. Lethally irradiated (1000 cGy) B6 mice underwent transplantation with TCD B10.BR BM alone (□, n = 6) or together with spleen cells adjusted to yield a dose of 2 × 10⁶ B10.BR T cells. (A) Animals that received transplanted B10.BR T cells were administered DMSO (▪, n = 11) or PS-1145 (•, n = 11) (50 mg/kg for 10 days beginning the day of transplantation). Survival is shown in panel A, and weight loss, in panel B. Overall cellularity (C), absolute number of B cells (D), and donor T-cell chimerism (E) in the spleens of mice that survived 60 days after transplantation are shown (□, TCD BM; , DMSO; ▪, PS-1145). Data in bar graphs are presented as the mean ± SEM. Spleen cellularity: DMSO versus PS-1145, P = .006; TCD versus PS-1145, P < .001. Total splenic B cells: DMSO versus PS-1145, P = .04; TCD versus PS-1145, P < .001. Percentage donor T cells: DMSO versus PS-1145, P = .66; TCD versus PS-1145, P < .001