Identification of somatic JAK1 mutations in patients with acute myeloid leukemia

Zhifu Xiang, Yu Zhao, Vesselin Mitaksov, Daved H. Fremont, Yumi Kasai, AnnaLynn Molitoris, Rhonda E. Ries, Tracie L. Miner, Michael D. McLellan, John F. DiPersio, Daniel C. Link, Jacqueline E. Payton, Timothy A. Graubert, Mark Watson, William Shannon, Sharon E. Heath, Rakesh Nagarajan, Elaine R. Mardis, Richard K. Wilson, Timothy J. Ley and Michael H. Tomasson

Data supplements

  • Supplemental materials for: Xiang et al

    Files in this Data Supplement:

    • Figure S1. Sequence coverage for JAK1 gene (JPG, 4 MB) -
      All 24 JAK1 exons were sequenced using genomic DNA from 94 AML patients. Each column represents a patient, and each row represents a base pair of nucleotides. High quality double strand coverage is indicated by green shading (98.04% of exonic sequences); single strand coverage, yellow (3.14%); and failed coverage, red shading (0.83%). At the left, intron/exon boundaries are indicated by white (exon) and black (intron) stripes.

Article Figures & Data

  • Figure 1

    Somatic nonsynonymous mutations in conserved residues of the JAK1 gene. (A,C) Electropherograms of matched tumor and germline samples from 2 patients with AML. Heterozygous mutations are indicated by double peaks (Embedded Image) consistently detected in both forward and reverse sequencing reactions but not present in the germline samples. (B,D) Change of amino acid sequences as a result of mutations. (E) Schematic diagram of JAK1 protein structure. Somatic mutations are indicated by arrows. The Thr478 residue resides in the β2 strand of the SH2 (JH3-JH4) domain near the phospho-tyrosine binding site of this domain. The Val623 residue resides in the β3 strand of the pseudo-kinase (JH2) domain in close proximity to the G-loop binding site of this domain. (F) Alignment of peptide sequences of conserved JAK1 residues. Both JAK1 mutations affect residues that are highly conserved throughout evolution.

  • Figure 2

    Somatic JAK1 mutations facilitate the activation of downstream signaling pathways. (A) Autophosphorylation of mutant JAK1 proteins. Mutant, but not wild-type, JAK1 proteins are activated. (B) Growth of Ba/F3 cells expressing JAK1 and mutants in the absence of IL-3. Both mutant and wild-type JAK1-expressing cells can grow in the absence of IL-3. (C) Mutant JAK1 proteins do not affect sensitivity of Ba/F3 cells to IL-3. Cells expressing wild-type or JAK1 mutants were plated in different concentrations of IL-3, and cell growth was measured by MTT assay. The V623A mutant was mildly resistant to high doses of IL-3. Error bars represent SD. (D) Activation of downstream signaling pathways by JAK1 mutants. Protein lysates of Ba/F3 cells expressing wild type and JAK1 mutants were analyzed by Western blot using phospho-specific antibodies shown. Stat1, Stat3, Akt, and Erk signaling was activated in cells expressing each JAK1 mutation. (E) Activation of STAT1 by interferon α in cells expressing JAK1WT, JAK1T478S, and JAK1V623A. STAT1 phosphorylation 15 minutes after stimulation was consistently increased in cells expressing JAK1T478S and JAK1V623A compared with JAK1WT. (F) Densitometry of results in panel E showing increased STAT1 phosphorylation in JAK1 mutant-expressing cells. U4A parental cells that do not express JAK1 are shown as controls. Similar results were obtained in three independent experiments, and a representative example is shown.

Supplementary Materials

  • Figure S1

    Supplementary PDF file available online.