Constitutive JAK3 activation induces lymphoproliferative syndromes in murine bone marrow transplantation models

Abstract

The tyrosine kinase JAK3 plays a well-established role during normal lymphocyte development and is constitutively phosphorylated in several lymphoid malignancies. However, its contribution to lymphomagenesis remains elusive. In this study, we used the newly identified activating JAK3A572V mutation to elucidate the effect of constitutive JAK3 signaling on murine lymphopoiesis. In a bone marrow transplantation model, JAK3A572V induces an aggressive, fatal, and transplantable lymphoproliferative disorder characterized by the expansion of CD8(+)TCRalphabeta(+)CD44(+)CD122(+)Ly-6C(+) T cells that closely resemble an effector/memory T-cell subtype. Compared with wild-type counterparts, these cells show increased proliferative capacities in response to polyclonal stimulation, enhanced survival rates with elevated expression of Bcl-2, and increased production of interferon-gamma (IFNgamma) and tumor necrosis factor-alpha (TNFalpha), correlating with enhanced cytotoxic abilities against allogeneic target cells. Of interest, the JAK3A572V disease is epidermotropic and produces intraepidermal microabscesses. Taken together, these clinical features are reminiscent of those observed in an uncommon but aggressive subset of CD8(+) human cutaneous T-cell lymphomas (CTCLs). However, we also observed a CD4(+) CTCL-like phenotype when cells are transplanted in an MHC-I-deficient background. These data demonstrate that constitutive JAK3 activation disrupts T-cell homeostasis and induces lymphoproliferative diseases in mice.

Figure 1

Expansion of lymphoid cells in JAK3-AV animals. (A) Comparison of total number of white blood cells (top left panel), spleen weights (top right panel), liver weights (bottom left panel), and total cells after RBC lysis in different organs (bottom right panel) in recipients of JAK3-WT–transduced bone marrow (■) versus JAK3-AV–transduced bone marrow (□). (B) Flow cytometric analysis of the different lymphoid compartments. All analyses are gated on GFP+ cells; plots are representative of at least 5 independent experiments. Percentages are indicated. (C) Subpopulation analysis of lymphocytes migrating to the different lymphoid organs in JAK3-AV or JAK3-WT recipients. BM indicates bone marrow; Spl, spleen; and Thy, thymus. Bar graphs represent the average and standard deviation obtained from at least 5 animals.

Figure 2

Immunophenotypic characterization of the CD8⁺ T-cell population. Flow cytometric analysis of JAK3-WT versus JAK3-AV thymic (A) or splenic (B) CD8⁺CD4⁻ T cells. (C) Flow cytometric staining for splenic T cells (CD3⁺) and B cells (CD19⁺) in JAK3-AV versus JAK3-WT mice. All analyses are gated on GFP⁺ cells and are representative of at least 5 independent experiments. Numbers indicate the percentage of cells. (D) Southern blot analysis from splenic genomic DNA of 1 JAK3-WT (lane 1) and 5 different JAK3-AV mice (lanes 2-6) using a GFP-specific probe. (E) PCR for the different variable TCRβ regions from splenocyte-derived cDNA of 1 JAK3-WT and 2 JAK3-AV animals (right panel). Each lane represents 1 PCR reaction with a different forward primer for the TCRβ variable regions (1-19 from left to right) and the same reverse primer for the constant region.

Figure 3

JAK3-AV confers enhanced proliferative capacities. (A) ³H-thymidine incorporation after 60 hours of stimulation with PMA/ionomycin of purified CD8⁺ T cells from JAK3-WT (■) and JAK3-AV (□) mice. Rightmost bars show effect of pharmacologic inhibition of JAK3 with JAK inhibitor I (JAK-I). Bar graphs represent the average and standard deviation of 3 independent experiments performed in triplicate. (B) Assessment of phosphorylation status of downstream effectors of JAK3 by intracellular staining of JAK3-WT (shaded histogram) and JAK3-AV (open histogram) splenocytes with phosphospecific antibodies. Histograms display a representative experiment gated on CD8⁺ T cells (n = 6). Bar graphs represent the average and standard deviation of 3 independent experiments for each phosphoprotein. (C) Western blot from bone marrow lysates from JAK3-WT or JAK3-AV animals shows constitutive phosphorylation of JAK3 downstream targets in the latter. The last lane shows inhibition of phosphorylation by JAK-I.

Figure 4

JAK3-AV–expressing T cells show decreased apoptosis and increased Bcl-2 expression. (A) Flow cytometric analysis of intracellular Bcl-2 levels in 3 different thymocyte populations (DP and SP) of JAK3-WT (shaded histogram) versus JAK3-AV (open histogram) animals. Bottom panel shows mean fluorescence intensity of the Bcl-2 signal normalized against an isotype control antibody in thymus and blood of JAK3-WT (■) and JAK3-AV (□) mice. All analyses are gated on GFP; data represent the average and standard deviation of 4 independent experiments. (B) Effect of JAK inhibitor I on Bcl-2 levels in CD8 SP thymocytes shown as a histogram and as bar graphs representing the mean and standard deviation of 2 independent animals for each group performed in duplicate. (C) Annexin-V staining for assessment of apoptosis levels in JAK3-WT versus JAK-3 AV CD8⁺ thymocytes. Analyses are gated on GFP⁺CD8⁺7-AAD⁻ thymocytes. Bar graphs represent the mean and standard deviation of at least 2 independent experiments each performed in duplicate.

Figure 5

JAK3-AV CD8⁺ T cells produce more inflammatory/cytotoxic cytokines and display enhanced cytotoxic activity. (A) Redirected ⁵¹Cr release assay of CD8⁺ “effector” T cells incubated with allogeneic (P815) target cells at different ratios in the presence (left panel) or absence (middle panel) of anti-CD3. Syngeneic EL4 target cells (right panel) were used as a negative control. Graphs display a representative of 4 experiments performed in triplicate. Percentage of killing was calculated as decribed in “Methods.” *P < .05. (B) ELISA assay for production of IFNγ in unstimulated (unstim) or PMA/ionomycin-treated (P+I) purified CD8⁺ T cells of JAK3-WT (■) and JAK3-AV (□) mice. Data (mean ± SD) represent the average of 5 experiments performed in duplicate. (C) Intracellular staining of splenic GFP⁺CD8⁺ T cells for IFNγ and TNFα confirm the increased production of these 2 cytokines in JAK3-AV (open histogram) compared with JAK3-WT cells (shaded histogram) at a single-cell level. Bar graphs indicate the mean and SD of 3 independent experiments.

Figure 6

Constitutive JAK3 activation induces a T-cell lymphoproliferation with prominent cutaneous involvement. (A) H&E-stained skin tissue sections from JAK3-AV (top row and rightmost panel) or JAK3-WT (bottom row) animals showing a dense atypical dermal infiltrate comprised of pleomorphic lymphoid cells tagging along the dermal-epidermal junction in the former. (B) H&E-stained sections of skin lesions from secondary recipients display even more pronounced cutaneous disease, with significant involvement of the dermis and extension into subcutaneous adipose tissue. Sections of the epidermis highlight frequent collections of atypical intraepidermal lymphocytes resembling Pautrier microabcesses (indicated by). (C) Immunohistochemistry of skin sections from secondary recipients with anti-CD3 (left panel), anti-B220 (middle panel), or anti-CCR10 (right panel) antibodies. Insets show staining of intraepidermal lymphocyte collections highlighted in panel B.

Figure 7

JAK3A572V mutation is present in human CTCL. (A) DNA samples from 30 patients with CTCL were PCR amplified and resequenced using M13-tailed primers flanking exon 12 of the human JAK3 gene. Forward (For.) and reverse (Rev.) sequence traces from Patient 7 (Pt.7) sample as well as wild-type trace are shown. (B) PCR products obtained in (A) were cloned and individually sequenced. Eleven of 44 (25%) products showed the mutant JAK3A572V allele. (C) Hematoxylin&Eosin staining of the skin biopsy shown in (A). (D) Sequence traces from Patient 7 (Pt.7) biopsies collected during disease progression. SK1 indicates skin biopsy at diagnosis prior to large-cell transformation shown in panel A; PB0, peripheral blood at diagnosis; PB1, before diagnosis of large cell transformation; and PB2, after diagnosis of large cell transformation.