Lnk inhibits erythropoiesis and Epo-dependent JAK2 activation and downstream signaling pathways

Abstract

Erythropoietin (Epo), along with its receptor EpoR, is the principal regulator of red cell development. Upon Epo addition, the EpoR signaling through the Janus kinase 2 (JAK2) activates multiple pathways including Stat5, phosphoinositide-3 kinase (PI-3K)/Akt, and p42/44 mitogen-activated protein kinase (MAPK). The adaptor protein Lnk is implicated in cytokine receptor signaling. Here, we showed that Lnk-deficient mice have elevated numbers of erythroid progenitors, and that splenic erythroid colony-forming unit (CFU-e) progenitors are hypersensitive to Epo. Lnk(-/-) mice also exhibit superior recovery after erythropoietic stress. In addition, Lnk deficiency resulted in enhanced Epo-induced signaling pathways in splenic erythroid progenitors. Conversely, Lnk overexpression inhibits Epo-induced cell growth in 32D/EpoR cells. In primary culture of fetal liver cells, Lnk overexpression inhibits Epo-dependent erythroblast differentiation and induces apoptosis. Lnk blocks 3 major signaling pathways, Stat5, Akt, and MAPK, induced by Epo in primary erythroblasts. In addition, the Lnk Src homology 2 (SH2) domain is essential for its inhibitory function, whereas the conserved tyrosine near the C-terminus and the pleckstrin homology (PH) domain of Lnk are not critical. Furthermore, wild-type Lnk, but not the Lnk SH2 mutant, becomes tyrosine-phosphorylated following Epo administration and inhibits EpoR phosphorylation and JAK2 activation. Hence, Lnk, through its SH2 domain, negatively modulates EpoR signaling by attenuating JAK2 activation, and regulates Epo-mediated erythropoiesis.

Figure 1.

Enhanced erythropoietic recovery and Epo/EpoR signaling after erythropoietic stress in Lnk-deficient mice. Wild-type and Lnk-deficient mice were injected with phenylhydrazine (PHZ) at days 0 and 1 (indicated by arrows). Hematocrits (A) and reticulocyte counts (B) were measured at indicated intervals (mean ± standard deviation [SD]). The corrected reticulocyte count (also known as reticulocyte index) was calculated assuming a normal hematocrit of 45%, as follows: corrected reticulocyte count (%) = reticulocyte count (%) × (hematocrit/45). There were 8 to 9 mice used in 2 independent studies. Student t tests were performed and P values are shown on top of each pair of data points. (C) At days 4 and 5 after PHZ treatment, spleen cells were isolated. CD71⁺ populations were purified using magnetic beads, and stimulated with 0, 1, 10, or 100 U/mL Epo for 10 minutes. Protein lysates from equal cell numbers were subjected to Western blotting analysis. Phosphorylated and total protein levels of p42/44MAPK and Akt are shown. Results shown here are representatives of 3 independent experiments, and 3 wild-type and 3 Lnk-deficient mice were used and pooled in each experiment.

Figure 2.

Lnk-deficient CFU-e progenitors are hypersensitive to Epo. Spleen cells isolated from wild-type and Lnk-deficient mice were plated in methylcellulose with varying concentrations of Epo. CFU-e numbers (mean ± standard error [SE]) were determined 2 days later. Five mice from each genotype were used. Comparing the 2 curves, at all Epo concentrations the P values are less than .05.

Figure 3.

The SH2 domain of Lnk is required to inhibit Epo-dependent growth and signaling pathways in 32D/EpoR cells. (A) Overexpression of Lnk in 32D/EpoR cells inhibits cell growth in response to Epo. We introduced either vector alone or wild-type Lnk into 32D/EpoR cells and determined the proportion of infected cells as those that express GFP, 2 days later. We then measured the GFP⁺ percentages and counted total cell numbers every day. The numbers of vectoror Lnk-expressing GFP⁺ cells were then calculated and plotted (mean ± SD). (B) We introduced the wild-type or the mutant Lnk cDNA constructs into 32D/EpoR cells. Two days after infection, we cultured the cells in 1 U/mL Epo, and measured the GFP⁺ fraction every 3 days thereafter. The percentage of GFP⁺ cells relative to that at 2 days after infection was plotted. Results shown are representative of more than 5 independent experiments. (C-D) 32D/EpoR cells infected with either vector alone or the wild-type or the mutant forms of Lnk were purified, stimulated with Epo, and lysed at indicated intervals followed by Western blotting analysis. (C) Stat5 phosphorylation and protein levels after Epo administration. (D) P42/44 MAPK phosphorylation and protein levels after Epo administration. Representatives of 3 independent experiments are shown.

Figure 4.

The Lnk SH2 domain is required to inhibit Epo-dependent erythroblast survival in in vitro culture. Ter119^- fetal liver erythroid progenitor cells were purified, infected, and cultured as described in “Materials and methods.” We then introduced either the control vector or the wild-type or mutant Lnk cDNA constructs into Ter119^- erythroid progenitors. (A) Average cell numbers (± SD) at the end of 48-hour culture. The numbers were normalized to the starting culture density of 1 × 10⁵/mL. (B) At 40 hours after initiation of the culture, the erythroblasts were stained for annexin V and 7-AAD, and subjected to FACS analysis. hCD4⁺ fractions (ie, infected cells) were gated in all the FACS plots. The apoptotic cell fractions (annexin V⁺ and 7-AAD^-) and dead cell fractions (annexin V⁺ and 7-AAD⁺) are indicated in each panel.

Figure 5.

The Lnk SH2 domain is required to inhibit Epo-dependent erythroid differentiation. (A) CD71/Ter119 FACS profile of freshly isolated e14.5 fetal liver cells. About 15% of the cells are Ter119^- (R1 and R2). (B) Differentiation profile of erythroid cells immediately after Ter119^- purification. Ter119^- population is enriched to 96%. Erythroid progenitor cells were subsequently infected and cultured as described in Figure 4. (C) Median FITC-CD71 fluorescence (± SD) after 16 to 18 hours of culture. (D-E) Differentiation of vector-infected erythroblasts cultured in Epo (D), vector-infected erythroblasts in the absence of Epo (E), and Lnk-infected erythroblasts cultured in Epo (F). The left panels show the CD71/Ter119 FACS profiles and cytology of cells after 16 to 18 hours of culture. The right panels show the FACS profiles and cytology of cells after 42 to 48 hours of culture. Dashed line outlined regions (R0) indicate nonerythroid cells. The FACS plots were gated for the hCD4⁺ population, that is, for expression of the transduced genes, in all the plots. For cytology the cells were centrifuged onto slides and stained with benzidine (brown) and Giemsa (purple for the nuclei). The arrowheads indicate benzidine-positive enucleated reticulocytes, and the thin arrows point to nonerythroid cells. Scale bars are 20 μm.

Figure 6.

The Lnk SH2 domain is required to inhibit Epo-induced signaling pathways in primary erythroid cells. Ter119^- erythroid progenitors were purified, infected with either vector alone, wild-type Lnk, or the Lnk SH2 mutant, and directly cultured in Epo for 14 to 16 hours. Subsequently, the cells were starved for 2 hours and stimulated with Epo for 10 minutes. Protein lysates from equal cell numbers were subjected to Western blotting analysis. Phosphorylation and total protein levels of Stat5 (A), Akt (B), and p42/44MAPK (C) are shown. Lnk protein levels are shown in the bottom panel of A. Five independent experiments were performed and representative results are shown here.

Figure 7.

The Lnk SH2 domain is required to inhibit Epo-induced JAK2 and EpoR phosphorylation. Primary erythroblasts transduced with either vector alone, or wild-type Lnk or the SH2 mutant (R364E) were immunoprecipitated with anti-EpoR (A), or anti-JAK2 (B) or anti-Lnk (C) antibodies, and probed with antiphosphotyrosine (4G10) antibodies. Total protein levels are shown in the bottom panels.