Figure 1.
Detection of increased glutamate secretion and glutaminase expression during de novo KSHV infection and in latently infected cells.
A) Primary HMVEC-d cells were left uninfected (UN) or infected with KSHV for different time points or with 100 µg heaprin-treated KSHV (Hep-KSHV) or UV-KSHV for 5d. Supernatants from uninfected and infected cells were collected and measured for the release of glutamate by a glutamate assay kit. B) Glutamate release in supernatants collected at various times from TIVE, TIVE-LTC, (C) BJAB, BJAB-KSHV, and BCBL-1 cells. Concentration of glutamate released is expressed in nanomoles per microliter. In a–c, error bars represent the mean ± SD of three independent experiments. D) Expression of glutaminase assessed by immunoblot analysis in primary HMVEC-d cells left uninfected or infected with KSHV for the indicated time points or UV-KSHV for 5 d. (E) Glutaminase expression in TIVE, TIVE-LTC, BJAB, BJAB-KSHV and BCBL-1 cells. β-actin was used as loading control. Fold change is relative to the uninfected control.
Figure 2.
Exogenous KSHV latency associated LANA-1 protein expression increases glutamate secretion and the expression of glutaminase and c-Myc in BJAB cells.
(A) BJAB cells were transduced with empty lentivirus vector or lentivirus LANA-1 constructs. 72 h after transduction, the media was replaced with fresh medium and cultured for a further 24 h; supernatants were collected and measured for the release of glutamate. Error bars represent the mean ± SD of three independent experiments. (B) Cell lysates from the transduced cells were Western blotted for glutaminase, LANA-1, and c-Myc expression. β-actin was used as loading control. Fold change is relative to the vector control. (C, D) Western blot analysis of c-Myc and glutaminase expression in TIVE-LTC (C) and BCBL-1 cells (D) following transduction with c-Myc shRNA or control shRNA.
Figure 3.
mGluR1 expression is upregulated in KSHV infected cells.
(A) RNA isolated from uninfected HMVEC-d or infected with either live KSHV or UV-KSHV for 5 d were analyzed for mGluR1 gene expression by RT-PCR. β-actin was used as an internal control. B) RT-PCR for mGluR1 expression in RNA isolated from TIVE, TIVE-LTC, BJAB, BJAB-KSHV and BCBL-1 cells. C) Western blot determination of mGluR1 expression in HMVEC-d cells left uninfected or infected with KSHV and UV-KSHV for 5 d. (D) Western blot for mGluR1 expression in TIVE and TIVE-LTC cells, BJAB, BJAB-KSHV and BCBL-1 cells. Fold induction was calculated relative to the uninfected control. (E, F, G) mGluR1 expression in KSHV infected cells determined by double immunostaining with mGluR1 and LANA-1 in uninfected and KSHV infected HMVEC-d cells (Magnifications 40×) (E), TIVE and TIVE-LTC cells (Magnifications 40×) (F), and BJAB, BJAB-KSHV and BCBL-1 cells (G) (Magnifications 80×). Boxed areas are enlarged. White arrows indicate mGluR1 staining and yellow arrows indicate LANA-1 staining. Nuclei were stained with DAPI.
Figure 4.
mGluR1 detection in KSHV infected patient samples.
A) Immunofluorescence detection of mGluR1 and LANA-1 expression in normal human skin and KS tissue. B) Immunofluorescence staining of mGluR1 expression in normal stomach tissue and PEL affected stomach tissue. The tissue sections were counterstained with DAPI. White and yellow arrows indicate mGluR1 and LANA-1 staining, respectively. Magnifications 20×. Boxed areas are enlarged.
Figure 5.
Cytoplasmic localization of REST in KSHV infected cells.
A) Immunofluorescence staining showing nuclear and cytoplasmic localization of REST in TIVE and TIVE-LTC cells, (B) BJAB, BJAB-KSHV, and BCBL-1 cells. Nuclei were stained with DAPI. Boxed areas are enlarged. The arrows indicate REST localization. (C and D) Subcellular fractionation and REST detection. TIVE and TIVE-LTC (C), BJAB, BJAB-KSHV, and BCBL-1 cell (D) lysates were fractionated into nuclear and cytoplasmic fractions, and the fractions were Western blotted with REST antibodies. Nuclear and cytoplasmic fractions were blotted for TBP or tubulin to validate the purity of nuclear and cytoplasmic fractions, respectively.
Figure 6.
Phosphorylation and ubiquitination of REST in the cytoplasm of KSHV infected cells.
A) Immunoprecipitation of cytoplasmic fractions from TIVE and TIVE-LTC, BJAB, BJAB-KSHV and BCBL-1 cells with anti-phosphoserine antibodies and analysis by immunoblotting with anti-REST antibody. B) Immunoprecipitation of the cytoplasmic fractions from BJAB, BJAB-KSHV and BCBL-1 cells with anti-REST antibody followed by immunoblot with anti-β-TRCP antibody. Bottom panel shows whole cell extracts Western blotted with anti-REST and β-TRCP antibodies. (C) Immunoprecipitation of cytoplasmic fractions from TIVE and TIVE-LTC, BJAB, BJAB-KSHV and BCBL-1 cells with anti-REST antibody followed by anti-ubiquitin immunoblotting. Bottom panel shows whole cell extracts subjected to Western blot using anti-REST antibody. (D) Effects of MG132 treatment on REST protein levels. BJAB, BJAB-KSHV, BCBL-1 and TIVE-LTC cells were treated with 10 µM MG132 for 6 h, and the cell lysates were Western blotted using REST antibody. As a control for loading, equivalent amount of protein samples were Western blotted with an anti-actin antibody.
Figure 7.
KSHV latency associated Kaposin A protein mediates mGluR1 expression.
A) Kaposin A (ORFK12) over expression leads to mGluR1 expression in BJAB cells. BJAB cells were transduced with control lentivirus vector or lentiviruses expressing ORFs 71, -72, -73 and Kaposin A for 72 h. The expression levels of mGluR1 in the cell lysates were determined by Western blot using anti-mGluR1 antibody. B) mGluR1 and Kaposin A expression determined by Western blot in primary HMVEC-d cells following transduction with lentivirus control or lentivirus-Kaposin A. β-actin was used as loading control. C) Colocalization of REST and Kaposin A in HMVEC-d cells transduced with lentivirus vector, or lentivirus-Kaposin A after 72 h. Boxed areas are enlarged in the right panel. Arrows indicate colocalization of REST and Kaposin A. Magnifications 40×.
Figure 8.
KSHV latency associated Kaposin A protein interacts with REST in the cytoplasm.
(A and B) Immunoprecipitation of cytoplasmic fractions from BJAB and BCBL-1 cells (A), TIVE and TIVE-LTC cells (B) with anti-REST antibody followed by anti-Kaposin A immunoblotting. BCBL whole cell lysates were used as a positive control for Kaposin A expression. C) Colocalization of REST and Kaposin A in the cytoplasm of BJAB and BCBL-1 cells determined by immunostaining with anti-REST and anti-Kaposin A antibodies. D) Colocalization of REST and Kaposin in TIVE and TIVE-LTC cells. Nuclei were stained with DAPI. Arrows indicate colocalization of REST with Kaposin A. Boxed areas are enlarged on the rightmost panel. E) HEK 293T cells were transduced with vector alone or with FLAG-WT-REST and Kaposin A for 72 h. FLAG-tagged REST interaction with Kaposin A in the cytoplasm and nuclei was assessed by immunoprecipitation with Kaposin A antibody followed by immunoblotting with FLAG antibody.
Figure 9.
Colocalization of REST and Kaposin in KSHV-infected patient samples.
(A) Immunofluorescence analysis of REST and Kaposin staining in normal human skin and KS tissue. (B) Immunofluorescence analysis of REST and Kaposin staining in normal stomach tissue and PEL affected stomach tissue. The tissue sections were counterstained with DAPI. White arrows indicate cytoplasmic localization of REST and Kaposin A, and their colocalization in KS and PEL tissues. Magnifications 80×. Boxed areas are enlarged.
Figure 10.
Effect of Kaposin A on phosphorylation of degron mutant REST.
(A) 293T cells were transduced with vector alone or retroviruses expressing FLAG-WT-REST or Kaposin A and FLAG-WT-REST or Kaposin A and FLAG-mutant REST. Upper panel shows cytoplasmic fractions immunoprecipitated with anti-phosphoserine antibody followed by Western blot with anti-FLAG antibody. Middle panel shows cytoplasmic fractions immunoprecipitated with anti-FLAG antibody followed by Western blot with anti-β-TRCP. Bottom panel shows whole cell extracts subjected to Western blot using anti-β-TRCP antibody. (B) 293T cells were transduced with vector alone or Kaposin A for 2 days and the Kaposin A transduced cells were transfected with FLAG tagged REST-WT or FLAG-REST individually mutated at serine 1024, (FLAG-REST S1024A), 1027 (FLAG-REST S1027A), or 1030 (FLAG-REST S1030A) for 48 h. Cell lysates were IPed with anti-phosphoserine antibody followed by immunoblot with anti-FLAG antibody (First panel) or IP-ed with anti-FLAG antibody followed by immunoblot with anti-phosphoserine antibody (Second panel) or immunoprecipitated using anti-FLAG antibody and Western blotted with poly-ubiquitin antibody (Third panel). The fourth and fifth panels show cell lysates subjected to Western blot using anti-FLAG, and anti-β-actin, respectively. (C) Effect of Kaposin A siRNA on REST phosphorylation: 293T cells transduced with vector alone or Kaposin A were transfected with FLAG-REST S1027A or FLAG-REST WT followed by transfection with control siRNA (C siRNA) or a pool of Kaposin A specific siRNA (Kap A siRNA). After 48 h post-transfection, cell lysates were prepared and immunoprecipitated with anti- FLAG antibody followed by immunoblot with anti-phosphoserine antibody (First panel). The membrane was striped and reprobed with anti-FLAG antibody as shown in the second panel. Whole cell extracts were subjected to Western blots analysis using anti-Kaposin A for Kaposin A expression (Third panel). β-actin was used as loading control.
Figure 11.
Glutamate release inhibitor and mGluR1 antagonists efficiently blocked KSHV infected cell proliferation.
A) HMVEC-d cells were left uninfected or infected with KSHV for 3 d. The cells were then cultured in the absence or presence of riluzole (10 µM), A841720 (50 µM) or Bay36-7620 (50 µM) for 48 h followed by BrdU pulse labeling for 2 h. BrdU incorporation was detected by staining with a BrdU antibody. Bottom panel shows merged images of BrdU and DAPI. (B and C) BrdU incorporation determined by ELISA in TIVE-LTC (B), and BCBL-1 cells (C) treated in the absence or presence of different concentrations of riluzole, A841720, or Bay36-7620 for 48 h followed by BrdU pulse labeling for 2 h. BrdU incorporation was analyzed by a BrdU cell proliferation ELISA kit. Error bars represent the mean ± SD of three independent experiments. **p<0.001, ***p<0.0001 compared with DMSO treatment. (D and E). MTT assay for cell proliferation. TIVE-LTC (D) and BCBL-1 cells (E) were incubated with riluzole, A841720, or Bay36-7620 for 48 h. At the end of incubation cell growth was measured with Vibrant MTT cell proliferation assay kit as described in materials and methods. Data are mean ± SD. (F and G) mGluR1 shRNA transduction: TIVE-LTC cells transduced with control shRNA lentivirus or with mGluR1 shRNA lentivirus were selected using puromycin hydrochloride. Cell lysates Western blotted with anti mGluR1 antibody showed 80% knockdown of mGluR1 expression in mGluR1 shRNA transduced cells. β-actin was used as a loading control. (G) BrdU incorporation determined by a BrdU cell proliferation ELISA in control and mGluR1 shRNA transduced TIVE-LTC cells. Error bars represent the mean ± SD.
Figure 12.
Schematic diagram showing the pathways for glutamate generation and mGluR1 upregulation in KSHV infected cells.
During latent infection, KSHV LANA-1 protein activates c-Myc, which leads to the upregulation of glutaminase and induction of glutamate release. Dashed line denotes c-Myc regulation and glutaminase expression. KSHV latent viral protein Kaposin A binds and sequesters REST in the cell cytoplasm, which in turn relieves the REST mediated suppression of the mGluR1 gene and upregulates expression of the mGluR1 receptor. Binding of glutamate to mGluR1 induces signaling and proliferation of infected cells. These studies show that proliferation of cancer cells latently infected with KSHV in part depends upon glutamate and glutamate receptor and therefore could potentially be used as therapeutic targets for the control and elimination of KSHV associated cancers.