Jak3, STAT3, and STAT5 inhibit expression of miR-22, a novel tumor suppressor microRNA, in cutaneous T-Cell lymphoma

Abstract

Aberrant activation of Janus kinase-3 (Jak3) and its key down-stream effectors, Signal Transducer and Activator of Transcription-3 (STAT3) and STAT5, is a key feature of malignant transformation in cutaneous T-cell lymphoma (CTCL). However, it remains only partially understood how Jak3/STAT activation promotes lymphomagenesis. Recently, non-coding microRNAs (miRNAs) have been implicated in the pathogenesis of this malignancy. Here, we show that (i) malignant T cells display a decreased expression of a tumor suppressor miRNA, miR-22, when compared to non-malignant T cells, (ii) STAT5 binds the promoter of the miR-22 host gene, and (iii) inhibition of Jak3, STAT3, and STAT5 triggers increased expression of pri-miR-22 and miR-22. Curcumin, a nutrient with anti-Jak3 activity and histone deacetylase inhibitors (HDACi) also trigger increased expression of pri-miR-22 and miR-22. Transfection of malignant T cells with recombinant miR-22 inhibits the expression of validated miR-22 targets including NCoA1, a transcriptional co-activator in others cancers, as well as HDAC6, MAX, MYCBP, PTEN, and CDK2, which have all been implicated in CTCL pathogenesis. In conclusion, we provide the first evidence that de-regulated Jak3/STAT3/STAT5 signalling in CTCL cells represses the expression of the gene encoding miR-22, a novel tumor suppressor miRNA.

Keywords: JAK3; STAT3; STAT5; cutaneous T-cell lymphoma (CTCL); miR-22; mycosis fungoides (MF).

Figure 1. Expression of mature and primary miR-22 in CTCL as determined by qPCR

a. miR-22 expression in non-malignant (MyLa1850, MySi) and malignant (MyLa2000, MyLa2059, SeAx, PB2B) CTCL T cell lines as well as one T cell line derived from psoriasis vulgaris patient (PSOR), reference U6, n = 3. b. primary miR-22 (pri-miR-22) expression in non-malignant (MyLa1850) and malignant (MyLa2059) CTCL T-cell line. Reference GAPDH, n = 3 c. pri-miR-22 expression in primary Peripheral Blood Mononuclear Cells (PBMCs) derived from two healthy donors relative to one patient diagnosed with Sézary Syndrome, reference GAPDH.

Figure 2. Effect of the T cell growth factor, IL-2, on miR-22 expression

Expression of miR-22 in IL-2 sensitive, non-malignant, CTCL T cells (MyLa1850 and MySi). Cells were depleted of IL-2 for 48 hours (– IL-2) or depleted of IL-2 for 24 hours, followed by 24 hours of IL-2 supplementation (+ IL-2). miR-22 expression was determined by qPCR using U6 as a reference n = 3.

Figure 3. Curcumin treatment increases expression of mature and primary miR-22

miR-22 a. and pri-miR-22 b. expression measured by qPCR in non-malignant (MyLa1850) and malignant (MyLa2059, SeAx) CTCL T cells subjected to 24h treatment with 20μM curcumin or DMSO (control).a. Reference U6, n = 2. b. Reference GAPDH, error bars reflect variation in technical triplicates.

Figure 4. Inhibition of JAK3 increases expression of mature miR-22 in malignant CTCL cell line MyLa2059

miR-22 expression in MyLa2059 following 24 hours treatment with Jak3iII (40ug/mL) or DMSO control. Measured by qPCR, reference U6, n = 3.

Figure 5. Transient knockdown of STAT3 and STAT5 genes increases expression of mature miR-22 in malignant CTCL cell line, Myla2059

a. miR-22 expression in MyLa2059 48h following transient transfection with siSTAT3, siSTAT5a, siSTAT5b or non-target (NT) control. Reference U6, n = 3, error bars reflect variation in technical triplicates. b. Representative Western Blot showing knockdown efficiency of siRNA transfections, 48h.

Figure 6. Binding of STAT transcription factors to the the miR-22HG C17orf91 and upstream promoter regions

a. primary miR-22 expression in MyLa2059 24h following transient transfection with siSTAT3, siSTAT5a, siSTAT5b or non-target (NT) control. Reference GAPDH, error bars reflect variation in technical triplicates. b. Knockdown efficiency is evaluated by Western Blot. c. Binding of p(y)STAT3 and p(y)STAT5 to an oligonucleotide sequence designed to mimic a putative STAT binding side inside the miR-22HG, C17orf91. Malignant CTCL cell line, Myla2059, was treated for 24h with JAK inhibitor, CP690550 (50μM), DMSO or untreated (control). Protein extracts were subjected to oligonucleotide pulldown and investigated for p(y)STAT3 and p(y)STAT5 presence by Western Blotting before (input) and after (IP) pulldown. d. ChIP-seq reads from the C17orf91 promoter region in malignant MyLa2059 cells. Reads obtained from immunoprecipitation with STAT5, STAT3, RelB, RelA and a negative control (Rabbit IgG, bottom). The chromosomal position of C17orf91 and upstream WDR81 refer to hg19. Forward reads are indicated in green and reverse reads are shown in red. e. PCR analysis of ChIP samples using the primer set indicated in Supplementary Figure 1. For the C17orf91 promoter the 120bp amplicon was detected in the STAT5- and to a lesser degree the STAT3-precipitated samples. For the BIC (miR-155HG) promoter 190bp amplicon was detected only in the STAT5-precipitated sample.

Figure 7. HDAC inhibitor, SAHA, induces time- and concentration-dependent increase in primary miR-22 expression

Malignant CTCL cell line, MyLa2059, was treated for 0, 6 or 24 hours with 2μM or 15 μM SAHA or DMSO control. Relative expression of pri-miR-22 was determined by qPCR, reference GAPDH. Error bars reflect variation in technical triplicates.

Figure 8. Expression of previously verified miR-22 targets are decreased by miR-22 mimic in malignant CTCL cell lines

a. Expression of CDK6, HDAC4, HDAC6, MYX, MYCBP, NCoA1 and PTEN following transient transfection with miR-22 mimic relative to a scrambled control in malignant CTCL cell lines (MyLa2059, SeAx) as determined by qPCR. Reference GAPDH, n = 7. b. qPCR comparison of relative NCoA1-(left panel) and pri-miR-22-(right panel) mRNA expression levels in malignant versus non-malignant cell lines (upper panel: MyLa2059 vs. MyLa1850, lower panel: SeAx vs. MyLa1850). Reference GAPDH, n = 3 c. Western Blot showing NCoA1 protein expression levels in non-malignant (MyLa1850) and malignant (MyLa2059, SeAx) CTCL cell lines. d. Relative expression of NCoA1 protein in malignant CTCL cells (left: MyLa2059, right: SeAx) following transfection with miR-22 mimic or scrambled control. Upper inset is a representative of the Western Blots results for NCoA1 and GAPDH quantification, n = 3.