Peptidyl-prolyl isomerase Pin1 markedly enhances the oncogenic activity of the rel proteins in the nuclear factor-kappaB family

Abstract

The peptidyl-prolyl isomerase Pin1 is frequently up-regulated in human cancers in which Rel/nuclear factor-kappaB (NF-kappaB) is constitutively activated, but its role in these cancers remains to be determined, and evidence is still lacking to show that Pin1 contributes to cell transformation by Rel/NF-kappaB. Rel/NF-kappaB transcriptional and oncogenic activities are modulated by several posttranslational modifications and coregulatory proteins, and previous studies showed that cytokine treatment induces binding of Pin1 to the RelA subunit of NF-kappaB, thereby enhancing RelA nuclear localization and stability. Here we show that Pin1 associates with the Rel subunits of NF-kappaB that are implicated in leukemia/lymphomagenesis and modulates their transcriptional and oncogenic activities. Pin1 markedly enhanced transformation of primary lymphocytes by the human c-Rel protein and also increased cell transformation by the potent viral Rel/NF-kappaB oncoprotein v-Rel, in contrast to a Pin1 mutant in the WW domain involved in interaction with NF-kappaB. Pin1 promoted nuclear accumulation of Rel proteins in the absence of activating stimuli. Importantly, inhibition of Pin1 function with the pharmacologic inhibitor juglone or with Pin1-specific shRNA led to cytoplasmic relocalization of endogenous c-Rel in human lymphoma-derived cell lines, markedly interfered with lymphoma cell proliferation, and suppressed endogenous Rel/NF-kappaB-dependent gene expression. Together, these results show that Pin1 is an important regulator of Rel/NF-kappaB transforming activity and suggest that Pin1 may be a potential therapeutic target in Rel/NF-kappaB-dependent leukemia/lymphomas.

Figure 1. The c-Rel and v-Rel subunits of NFκB associate with Pin1

(A) Sequence alignment of the vertebrate Rel/NF-κB family proteins, highlighting conservation of N-terminal sequences flanking the Thr254-Pro Pin1 recognition motif in RelA, whereas sequences flanking its C-terminal end are more divergent. (B) Immunoblots showing endogenous expression of Pin1 in 293T cells, v-Rel-transformed CSC and in the primary mediastinal B cell lymphoma cell line Karpas 1106. The blot was probed with anti-Pin1 and reprobed for actin. (C) Left panel. Pull-down of v-Rel or hc-Rel in extracts from transiently transfected 293T cells with GST-Pin1 or GST as control, followed by immunoblotting with anti-Rel. Where indicated, cells were treated with hTNFα prior to harvest (lanes 4–6). Right panel. Pull-down of v-Rel, hc-Rel or hRelA transfected in 293T cells with GST-Pin1, followed by immunoblotting with anti-Rel. Where indicated, cells were treated with hTNFα. Input (1/10 of lysate; 1–2, 5–6 and 9–10). The blot was reprobed for actin and Pin1. (D) Left panel. Pull-down of endogenous v-Rel from v-Rel-transformed CSC (lanes 1–3) or hc-Rel from human lymphoma-derived Karpas 1106 cells (lanes 4–6) with GST-Pin1 or GST, followed by immunoblotting with anti-Rel (v-Rel: #1691 and hc-Rel: #265). Right panel. Co-immunoprecipitation of endogenous hc-Rel with Pin1 in extracts from human lymphoma cell lines using anti-Pin1, followed by immunoblotting with anti-hc-Rel. Input (1/10 of lysate). The membrane was stained with Ponceau S (bottom panel).

Figure 2. Pin1 markedly increases the transforming activity of Rel proteins

(A) Pin1 and mutant S16A associate with v-Rel in GST pull down assays significantly more than mutant S16E. Pull downs were carried out as in Figure 1C with extracts from 293T cells transfected with Pin1-GFP, Pin1(S16E)-GFP or Pin1(S16A)-GFP and GST-v-Rel or GST, followed by western blot with anti-Pin1. Input (1/10 of lysate). (B) Immunoblots showing efficient co-expression of v-Rel or hc-Rel with GFP-tagged Pin1, Pin1(S16E), Pin1(S16A) or GFP control in CEFs used as the source of virus to infect primary CSCs. The blots were probed with antibodies to v-Rel (#2716) or hc-Rel (#265) (top panels) or GFP (middle), and reprobed for actin as control (bottom). (C) Co-expression of Pin1 or mutant S16A markedly increases the transforming activity of hc-Rel and also enhances that of v-Rel in primary CSC, in contrast to Pin1(S16E), as detected by colony formation in soft agar. The results of four independent experiments are shown along with averages and standard deviations.

Figure 3. Pin1 and Pin1(S16A) promote relocalization of v-Rel and hc-Rel to the nucleus in the absence of stimuli

(A) Immunofluorescence showing that co-expression of Pin1 or Pin1(S16A) provokes nuclear translocation of v-Rel in CEFs infected with bicistronic retroviral vectors, in contrast to Pin1(S16E). (B) Pin1 or Pin1(S16A) similarly induce accumulation of hc-Rel in the nucleus of infected CEFs. (C) Immunofluorescence showing relocalization of hc-Rel to the nucleus of transformed CSCs coexpressing hc-Rel proteins with Pin1, but not in those co-expressing Pin1(S16E).

Figure 4. Juglone induces cytoplasmic relocalization of endogenous hc-Rel in human lymphoma cells

Immunofluorescence analysis of endogenous hc-Rel subcellular localization in presence of increasing concentrations of Juglone (0.1, 1 or 10 µM) for 2hrs in human lymphoma-derived KM-H2 (A), MedB-1 (B) or Karpas 1106 (C) cell lines, compared to DMSO control, using anti-hc-Rel and a rhodamine-conjugated secondary antibody. Nuclei were stained with Hoechst 33258 dye. (D) Fractionation of Karpas 1106 cells treated with juglone for 2hrs confirmed rapid redistribution of endogenous hc-Rel from the nucleus to the cytoplasm. The blots were probed with anti-hc-Rel and reprobed for uncleaved poly-ADP-ribose polymerase (PARP; ~115 kDa) or actin as nuclear and cytosolic markers. Histograms show quantitation of hc-Rel band intensities in the nucleus and cytoplasm normalized to those of PARP and actin.

Figure 5. Juglone interferes with cell survival and/or growth in Rel-dependent human lymphoma cells

Treatment of human lymphoma-derived KM-H2 (A) or MedB-1 (B) cells with 0.1 µM juglone during a time course induces cell death compared to the DMSO control, as determined by live cell count following Trypan blue exclusion staining. Whereas this treatment did not affect the survival of Karpas 1106 cells (C), juglone significantly interfered with Karpas 1106 cell growth compared to the control (D). The average of three independent experiments is shown.

Figure 6. Pin1 knockdown prompts cytoplasmic relocalization of hc-Rel, interferes with lymphoma cell growth and suppresses expression of Rel/NF-κB target genes

(A) Immunoblot of lymphoma-derived Karpas 1106, KM-H2 or MedB-1 cells harvested at 96hrs post-nucleofection with Pin1 shRNA or a scrambled shRNA control, using monoclonal anti-Pin1 or anti-actin. (B) Immunofluorescence of endogenous hc-Rel subcellular localization in lymphoma cells lines transfected with Pin1 shRNA or shRNA control, using anti-hc-Rel and a rhodamine-conjugated secondary antibody. Nuclei were stained with Hoechst 33258 dye. (C) Pin1 knockdown with shRNA (○) significantly delays the growth of lymphoma cell lines, compared to a scrambled shRNA control (●). The average of three independent experiments is shown. (D) RNA was analyzed by RT-PCR at 48hrs post-nucleofection with Pin1 shRNA or a scrambled shRNA control using primers specific for the Rel/NF-κB-regulated genes cyclin D1, VEGF or Bfl-1. Actin mRNA was amplified as a control.