Advertisement

Cyclin-dependent kinase 5 activity is required for allogeneic T-cell responses after hematopoietic cell transplantation in mice

David Askew, Tej K. Pareek, Saada Eid, Sudipto Ganguly, Megan Tyler, Alex Y. Huang, John J. Letterio and Kenneth R. Cooke

Data supplements

Article Figures & Data

Figures

  • Figure 1.

    Immune reconstitution is intact in Cdk5−/−Chematopoietic chimeras. The spleen, LNs, and thymus were harvested from naive, fully engrafted, Cdk5+/+C (WT), and Cdk5−/−C (KO) mice and either sectioned and stained for anatomical evaluation (spleen and LN [A]) or enzymatically digested, dispersed into single-cell suspensions, stained for CD4 and CD8, and examined by flow cytometry (spleen and thymus [B]). No differences between groups were noted in architecture, cellularity, or CD4+ and CD8+ phenotypes. The numbers of nTregs in spleen were determined by staining single-cell suspension with CD4 and CD25 on surface followed by intracellular staining of FoxP3 (C). Vβ usage in T-cell populations was determined by staining single-cell suspensions of splenocytes with Vβ antibodies in combination with anti-CD4 and anti-CD8 (D). No differences in splenic Treg numbers or Vβ usage were noted. N = 3 to 6 mice per group. CD4+ and CD8+ T cells from or Cdk5−/−C mice were cultured with B6D2F1 splenic DCs for 96 hours and examined for proliferative capacity (E) and cytotoxicity (F) when 3H-thymidine–labeled, allogeneic (P815) or syngeneic (EL4) tumor cells were added. Data are representative of 1 of at least 3 replicate experiments. *P < .05. CPM, counts per minute.

  • Figure 2.

    Cd5k activity is critical for GVHD induction. Lethally irradiated B6D2F1 mice received HCT from either syngeneic B6D2F1 or allogeneic (B6) mice as described in supplemental Methods. Cdk5 kinase activity was significantly increased in the spleen and small intestine by day 10 after HCT (A). Allo-HCT using Cdk5−/−C (Allo Cdk5 ko) donors results in significant reduction in GVHD severity as measured by survival (B), clinical score (C), splenic T-cell expansion (D), serum IFNγ levels (E), and target organ histopathology (F) compared with recipients of Cdk5+/+C (Allo wt) donors. Data are from 2 to 3 comparable experiments: n = 3 to 4 mice per group (A), n = 12 to 24 mice per group (B-C), and 4 to 8 mice per group (D-F). *P < .01 for all comparisons.

  • Figure 3.

    Cdk5 regulates migrational and proliferative capacity of donor T cells after allo-HCT. Isolated Cdk5+/+C (WT) T cells were stained with 2.5 μL of CFSE and Cdk5−/−C (KO) T cells were stained with 5 μL of SNARF-1. Equal numbers of WT and KO T cells (3-5 × 106) were coinjected into lethally irradiated B6D2F1 mice. Spleens and LNs were isolated from recipient mice at 24 and 48 hours after injection and cells were stained with CD4 or CD8 fluorescent antibodies and examined by flow cytometry. The percentage of CD4+ and CD8+ from WT (CFSE+) and KO (SNARF-1+) and percentage reduction of cells from KO mice were determined (A-B). In vivo T-cell proliferation was assessed in parallel experiments. T cells from Cdk5+/+C or Cdk5−/−C were labeled with CFSE and 5 × 106 T cells were separately injected IV into lethally irradiated B6D2F1 mice. Seventy-two hours later, mice were sacrificed and the spleens were removed. Proliferating cells from WT or KO donors were identified based upon decreased staining for CFSE (C-D). Percentage reduction at 72 hours likely reflects effects of Cdk5 on both migration to SLOs and subsequent proliferation once present. N = 4 mice per group and represent 1 of at least 3 replicate experiments.

  • Figure 4.

    Loss of Cdk5 expression results in fewer donor-derived, cytokine-producing T cells early after HCT. Lethally irradiated B6D2F1 received HCT from either syngeneic (B6D2F1) or allogeneic Cdk5+/+C (Cdk5 wt) or Cdk5−/−C (Cdk5 ko) donors as described in Figure 2. HCT recipients were sacrificed and single-cell suspensions were prepared from spleen of mice 3, 7, 14 days posttransplant. Total numbers of donor-derived, CD4+ and CD8+ T lymphocytes were determined in Kd or Kd+ cell populations in spleens of mice receiving allo- or syngeneic HCT, respectively (A). Cells were also incubated on anti-CD3–coated plates for 6 hours in RPMI1640 + 10% fetal bovine serum (FBS) containing 1 µg/mL brefeldin A, subsequently stained with anti-CD4 and anti-CD8, and incubated with fluorescent anti-cytokine antibodies. Total numbers of donor-derived, CD4+ and CD8+ T cells were counted and examined for the production of IFNγ, IL-2, and TNFα (B). Data are representative of 1 of 3 replicate experiments. N = 3 to 4 mice per group. *P < .01, #P < .05 or as otherwise noted.

  • Figure 5.

    Patterns of phosphorylation of known intermediates of CCR7 intracellular signaling are altered in CDk5-deficient T cells. Cell surface expression of CCR7 on peripheral blood CD4+ and CD8+ T cells was examined by flow cytometry after labeling cells with CCL19-Fc followed by anti-human immunoglobulin G (IgG)-phycoerythrin (PE) and either CD4-APC or CD8-APC (A). Next, purified T cells from Cdk5+/+C (Cdk5 WT) or Cdk5−/−C (Cdk5 KO) mice were incubated with 100 ng/mL CCL19 for 0 to 4 minutes and then lysed with Triton X-100 buffer containing phosphatase/protease inhibitors. Total cell lysate (4 μg) was separated on a 4% to 12% Bis-Tris gel and transferred to nitrocellulose. Cell lysates were examined by western blot using antibodies against pERK1/2 to measure levels of phosphorylated protein and ERK1/2 to measure levels of pERK1/2 and total ERK1/2 loaded (B1). The relative increase in expression of pERK/ERK in lysates from either WT or KO T cells was determined at each time point using time 0 for respective samples (eg, 100%) as baseline (B2). The percentage reduction of pERK expression in KO T cells was also determined at each time point (B3). In separate experiments, purified T cells from WT or KO mice were again incubated with 100 ng/mL CCL19 and cell lysates were examined for expression of pMEK1/2 and GAPDH (C). Data shown are from 1 of 3 replicate experiments.

  • Figure 6.

    Effects of Cdk5 gene deletion on GVL activity. Lethally irradiated B6D2F1 mice received HCT from either syngeneic (B6D2F1) or allogeneic Cdk5+/+C (Allo Cdk5 WT) or Cdk5−/−C (Allo Cdk5 KO) donors as described in Figure 4. Consistent with previous experiments, animals receiving HCT from Cdk5−/−C donors have significantly reduced mortality from GVHD (A). In subsequent experiments, groups of HCT recipients received 250 or 500 P815 tumor cells at time of transplant. Recipients of BM and T cells from Cdk5 KO donors effectively eliminate low-dose tumor challenge and show improved leukemia-free survival (B). At higher tumor dose, significant GVL activity remains, but some Cdk5 KO HCT recipients succumb to tumor (C). The expression of Cdk5 and p35 was determined on P815 and EL4 tumor cell lysate using western blot analysis (D). Tumor cells (104 cells per well) were incubated with 3H-Thy for 24 hours in the presence or absence of roscovitine or CIP and cell proliferation was determined (E-F). In parallel experiments, tumor cells incubated with roscovitine or CIP for 24 hours were stained with Annexin V and 7AAD and the percentage of early apoptotic cells was determined (G). Data are from at least 2 combined experiments: n = 8 to 12 per group (A-C), or 1 of at least 2 comparable experiments (E-F). *P < .01.

  • Figure 7.

    Pharmacologic inhibition of Cdk5 after HCT reduces clinical GVHD severity and while maintaining potent GVL effects. Lethally irradiated B6D2F1 mice received HCT from either syngeneic B6D2F1 or allogeneic (B6) mice as described in Figure 2. Groups of mice received 500 P815 cells previously transduced using a lentiviral vector carrying a luciferase reporter that allows visualization of proliferating cells using bioluminescence imaging (BLI). Syngeneic and allo-HCT recipients were monitored for tumor progress (A) and clinical GVHD severity (B). Roscovitine, at the dose and schedule administered after HCT effectively inhibited Cdk5 activity measured in spleens of mice collected on day 7 posttransplant (C). Data shown are representative of 2 replicate experiments: n = 5 to 12 mice per group; P < .01. Note: A single mouse in the allo-HCT group receiving roscovitine unexpectedly died following anesthesia for BLI and was censored “C”. Radiance is expressed as p/sec/cm2/sr. Color scale: Min = 2.00e4, Max = 2.00e5.

Tables

  • Table 1.

    Cdk5 activity in donor T cells is primarily responsible for reductions in GVHD

    GroupGVHD scoreSurvival, %
    Week 4Week 5Week 6
    F1 → F10.2 ± 0.20.0 ± 0.00.0 ± 0.0100
    Allo WT T + Allo WT BM → F15.0 ± 0.45.2 ± 0.55.8 ± 0.113
    Allo WT T + Allo KO BM → F14.8 ± 0.64.6 ± 0.64.5 ± 0.533
    Allo KO T + Allo KO BM → F12.4 ± 0.3*2.5 ± 0.4*3.0 ± 0.5*73*
    Allo KO T + Allo WT BM → F13.2 ± 0.8*2.6 ± 0.5*2.7 ± 0.4*60*
    • Lethally irradiated B6D2F1 received HCT from either syngeneic B6D2F1 or allogeneic Cdk5+/+C (Allo WT) or Cdk5−/−C (Allo KO) donors as described in Figure 2. Two additional allogeneic groups were evaluated in this “mixing” experiment as described in “Methods”: allogeneic KO BM cells mixed with allogeneic WT T cells and allogeneic WT BM cells mixed with allogeneic KO T cells. Survival was monitored daily and GVHD clinical scores were assessed weekly as described; n = 8 to 16 per group.

    • * P < .01 compared with recipients of Cdk5+/+C HCT.