Double JAK3 mutants show enhanced signaling and transformation potential. (A) Crystal structure of JAK3 JH1 (kinase domain; Protein Data Bank [PDB]: 1YVJ) and a model of JAK3 JH2 (pseudokinase domain) obtained by SWISS-MODEL server using JAK1 JH2 (PDB: 4L00) as a template. The 2 domains were assembled together by superimposition of the TYK2 JH1-JH2. (B) Western blot analysis of whole-cell lysates after reconstitution of the IL-7 receptor signaling complex in 293T cells. The 293T cells were transiently transfected with the constructs as indicated. (C) Phospho-flow analysis of phospho-STAT5 in Ba/F3 cells expressing single or double JAK3 mutants after 3 hours without IL-3 stimulation. Results are compared with BaF3 wild-type cells after 3 hours without IL-3 stimulation. (D-F) Growth of primary mouse pro–T cells in the absence of IL-7. (D) STAT5 (N642H) confers IL-7–independent growth. Significance was calculated compared with empty vector control using the Mann-Whitney test. (E) Analysis of single or double JAK3 mutants; only double mutants confer IL-7–independent growth to pro–T cells. Significance was calculated compared with wild-type control using the Kruskal-Wallis test and Dunn’s multiple comparisons correction. (F) Comparison of proliferation of pro–T cells transduced with 2 JAK3-mutant constructs or 1 double JAK3-mutant construct; only the JAK3 double mutant confers IL-7–independent growth. Significance was calculated compared with JAK3 A572T using the Kruskal-Wallis test and Dunn’s multiple comparisons correction. (G) Graph showing relative percentage of Ba/F3 cells expressing wild-type JAK3 (green fluorescent protein positive [GFP⁺]) in cells transformed by STAT5 (N642H), JAK3 (L875H), JAK3 (M511I+A573V), or JAK3 (M511I). Competition between wild-type JAK3 and mutant JAK3 results in the disappearance of wild-type JAK3–expressing cells, observed only in the case of JAK3 (M511I). **P ≤ .01, *P ≤ .05.