The Oncoprotein Fra-2 Drives the Activation of Human Endogenous Retrovirus Env Expression in Adult T-Cell Leukemia/Lymphoma (ATLL) Patients

Abstract

Human endogenous retroviruses (HERVs) are retroviral sequences integrated into 8% of the human genome resulting from ancient exogenous retroviral infections. Unlike endogenous retroviruses of other mammalian species, HERVs are mostly replication and retro-transposition defective, and their transcription is strictly regulated by epigenetic mechanisms in normal cells. A significant addition to the growing body of research reveals that HERVs' aberrant activation is often associated with offsetting diseases like autoimmunity, neurodegenerative diseases, cancers, and chemoresistance. Adult T-cell leukemia/lymphoma (ATLL) is a very aggressive and chemoresistant leukemia caused by the human T-cell leukemia virus type 1 (HTLV-1). The prognosis of ATLL remains poor despite several new agents being approved in the last few years. In the present study, we compare the expression of HERV genes in CD8⁺-depleted PBMCs from HTLV-1 asymptomatic carriers and patients with acute ATLL. Herein, we show that HERVs are highly upregulated in acute ATLL. Our results further demonstrate that the oncoprotein Fra-2 binds the LTR region and activates the transcription of several HERV families, including HERV-H and HERV-K families. This raises the exciting possibility that upregulated HERV expression could be a key factor in ATLL development and the observed chemoresistance, potentially leading to new therapeutic strategies and significantly impacting the field of oncology and virology.

Keywords: AP-1; ATLL; Fra-2; HERVs; leukemia.

Figure 1

Prevalence of Abs against human HERV Env antigens in patients with acute ATL and the respective control groups. (A) HERV envelope (ENV) antigenemia in sera from non-infected patients NI (n = 7; black circle), HTLV-1 asymptomatic carriers (n = 7; green square), and acute ATLL patients ATLL (n = 7; red diamond). The dotted lines represent positivity thresholds calculated by ROC analysis. (B) The area under the curve (AUC) and its statistical significance are reported (ns p ≤ 0.05 and **** p < 0.0001).

Figure 2

Detection of human HERV mRNA by qRT-PCR in ATL-derived cell lines. (A–C) Expression of HERV mRNA was assessed in one HTLV-1-negative cell line (Jurkat) and three HTLV-1-derived cell lines (HUT102, C81–66, and ATL-2). (D,E) HERV gag mRNA expression was compared in Jurkat cells and HTLV-1-derived cell lines. (F) HERV-R-Pol mRNA was compared in Jurkat cells and HTLV-1-derived cell lines (** p ≤ 0.001). (G–I) Relative Fra-2, Tax, and HBZ expression in Jurkat cells and HTLV-1-derived cell lines (statistical significance was determined using a one-way ANOVA test with Dunn‘s multiple comparisons post-test ns p ≤ 0.05, ** p ≤ 0.01; *** p ≤ 0.001, **** p ≤ 0.0001).

Figure 3

Detection of human HERV mRNA in CD8⁺-depleted PBMCs from HTLV-1-infected patients. (A–D) HERV mRNA was expressed in acute ATLL patients (red square) compared to HTLV-1 asymptomatic carriers (ACs) (black dot). (E,F) HERV gag mRNA expression was compared in AC (black dot) and ATLL patients (red square) (one-way ANOVA test with Dunn‘s multiple comparisons post-test ns p ≤ 0.05, ** p ≤ 0.001; **** p ≤ 0.00001). (G) HERV-R-Pol was measured in AC and ATLL patients compared to ACs (** p ≤ 0.001). (H,I) The relative expression of Tax and HBZ in AC patients and ATL patients.

Figure 4

HBZ does not activate HERV LTR. HEK293T was co-transfected with a plasmid carrying the luciferase reporter gene under the control of an NF-kB-dependent promoter (A), the collagenase promoter (B), or different HERV LTRs (C–F) and Tax-Flag, p65-Flag, or HBZ-Myc expression vectors, in addition to pRcActin-LacZ for the normalization of transfection efficiency. Cells were harvested 48 h. post-transfection and assayed for luciferase activity. The results show a fold increase compared to the mock control (set at a value of 1) and represent the mean values of three independently transfected cells. (A,B) Western blot analyses assessed the expression of Tax, p65, and HBZs. Actin is shown as a loading control (one-way ANOVA test with Dunn‘s multiple comparisons post-test ns p ≤ 0.05, * p ≤ 0.01, ** p ≤ 0.001; *** p ≤ 0.00001).

Figure 5

Fra-2 but not cFos activate the HERV LTR. HEK293T cells were co-transfected with a plasmid carrying the luciferase reporter gene under the control of the collagenase promoter in triplicate as shown (A) or different HERV 5′LTRs (B–E), different combinations of binding partners of AP-1-related transcription factors, and pRcActin-LacZ. The cells were harvested 48 h post-transfection and assayed for luciferase activity. The results show a fold increase in the mock control and represent the mean values of three independently transfected cell samples. (F) Western blot analyses were carried out to assess the expression of AP-1 transcription factors. Actin is shown as a loading control (one-way ANOVA test with Dunn‘s multiple comparisons post-test ns p ≤ 0.05,* p ≤ 0.01, ** p ≤ 0.001; *** p ≤ 0.0001 and **** p ≤ 0.00001).

Figure 6

HBZ does not alter the activation of type I and type II HERV-H LTRs by Fra-2. (A) HEK293T cells were co-transfected with the different AP-1 in the absence or presence of HBZ and a plasmid carrying the luciferase reporter gene under the control of the collagenase promoter. (B) Western blot analyses were carried out to assess the expression of AP-1 transcription factors and HBZ. Actin is shown as a loading control. HEK293T cells were co-transfected with a plasmid carrying the luciferase reporter gene under the control of the collagenase promoter (A) or different HERV 5′LTRs (C–E), different AP-1 expression vectors, and pRcActin-LacZ in the presence or absence of HBZ. The cells were harvested 48 h post-transfection and assayed for luciferase activity. The results show a fold increase in the mock control and represent the mean values of three independently transfected cell samples (one-way ANOVA test with Dunn‘s multiple comparisons post-test ns p ≤ 0.05, * p ≤ 0.01).

Figure 7

HERV Env mRNA detected in HEK293T cells stably expressing Fra-2. (A) Western blot analyses were carried out on the lysate of two pools of HEK293-Fra2 to assess the expression of Fra-2. Beta actin is shown as a loading control. (B–E) HEK293T stably expressing Fra2 was harvested at different passages (from p3 to p8), and the expression of HERV-H Env, HERV-R Env, HERV-K Env, and HERV-E gag mRNAs was assessed by qRT-PCR (one-way ANOVA test with Dunn‘s multiple comparisons post-test ns p ≤ 0.05, * p ≤ 0.01; ** p ≤ 0.001).

Figure 8

Kinetic analysis of Fra-2 and HERV Env mRNA in CD8⁺-depleted PBMCs from asymptomatic carriers and ATL patients. CD8⁺-depleted PBMCs from five ATL patients were cultivated ex vivo for five days, and Fra-2 (closed black square) (A), HERV-H Env (closed green triangle) (B), and HERV-K Env mRNA (closed red triangle) (C) were quantified at different time points using qRT-PCR (one-way ANOVA test with Dunn‘s multiple comparisons post-test; *** p ≤ 0.0001 and **** p ≤ 0.00001 (D,E) Relevance of Fra-2 and HERV Env mRNA expression in ATL patients, as analyzed with the Pearson correlation Test.

Figure 9

Fra-2 binds to HERV-H-LTR in ATL-2 cells. ChIP assays were performed on chromatin prepared from the indicated ATL-2 cell lines using antibodies against Fra-2. Data are presented as fold enrichment relative to the IgG control. Data are an average of three independent experiments. Error bars represent the SEM (two-way ANOVA t-test, ns p ≤ 0.05 ** p < 0.01, *** p < 0.001, **** p < 0.0001).