Rapamycin is efficacious against primary effusion lymphoma (PEL) cell lines in vivo by inhibiting autocrine signaling

Abstract

The antitumor potency of the mTOR inhibitor rapamycin (sirolimus) is the subject of intense investigations. Primary effusion lymphoma (PEL) appears as an AIDS-defining lymphoma and like Kaposi sarcoma has been linked to Kaposi sarcoma-associated herpesvirus (KSHV). We find that (1) rapamycin is efficacious against PEL in culture and in a murine xenograft model; (2) mTOR, its activator Akt, and its target p70S6 kinase are phosphorylated in PEL; (3) rapamycin inhibits mTOR signaling as determined by S6 phosphorylation; (4) KSHV transcription is unaffected; (5) inhibition of IL-10 signaling correlates with drug sensitivity; and (6) addition of exogenous IL-10 or IL-6 can reverse the rapamycin growth arrest. This validates sirolimus as a new treatment option for PEL.

Figure 1

Rapamycin-induced inhibition of cell proliferation as measured by MTT assay. Shown in each panel is the absorption at 570 nm in absence (black diamonds) or presence (gray circles) of 50 nM rapamycin on the vertical axis, and time in days after drug addition on the horizontal axis. Also shown is the percentage of inhibition at 96 hours after addition.

Figure 2

Response to rapamycin in cell culture. (A) BC-1 or (B) BCBL-1 cells were seeded in medium containing the indicated concentrations of rapamycin. Plotted is the cell number on the vertical axis and time after seeding on the horizontal axis. This data represents 1 of 2 biologic replicates. (C-F) Cell-cycle analysis using propidium iodide stain at 4 days after exposure to 50 nM rapamycin or mock. In each panel plotted on the horizontal axis is the relative DNA content and on the vertical axis the number of cells. Also shown is the percentage of cells with ≤ G₁, G₁, S, G₂ and ≥ G₂ DNA content for each of the indicated cell lines and conditions.

Figure 3

Response to rapamycin of experimental tumors. (A) Shown on the vertical axis is the tumor volume in mm³ and on the horizontal axis the time since inoculation with 10⁶ BC-1 cells and simultaneous treatment with rapamycin (n = 5) or vehicle (n = 4). (B) Shown on the vertical axis is the tumor volume in mm³ and on the horizontal axis the time since inoculation with 10⁶ BC-1 cells and treatment with rapamycin (n = 5) or vehicle (n = 4) animals after the establishment of tumors (day 14). Error bars indicate the SD for each group of animal. For rapamycin-treated animals the error bar is smaller than the symbol. (C) Expression of IL-6 in a mock-treated (−) BC-1 tumor (red color), which is reduced by rapamycin treatment (+). Expression of IL-10 in a mock-treated BC-1 tumor (−) (red color), which is reduced by rapamycin treatment (+). No staining in the absence of the specific primary antibody (no 1°). All sections are counterstained with hematoxilin (blue) and are at 400 × magnification. (D) Representative immunohistochemistry of mouse xenograft tumors using antibodies specific for phospho-Akt (Akt-P), phospho-mTOR (mTOR-P), and phospho-p70S6 kinase (p70S6-P). The phospho Akt staining was performed on BCBL-1 tumors at either 100 × or 400 × magnification. No 1° indicates the no primary antibody control reaction. The phospho-mTOR and phospho-p70S6K staining was performed on 2 different tumors (BC-3a and BC-3b) derived from the BC-3 cell line; No 1° indicates the no primary antibody control reaction. Pictures here were taken at 400 × magnification. (E) Western lot analysis of protein extracts for the indicated cell lines exposed to rapamycin or vehicle using anti–phospho-S6 and anti-p70S6 kinase antibody. (F) Western blot analysis of protein extracts for the indicated cell lines exposed to rapamycin or vehicle using anti-mTOR and antiactin antibody.

Figure 4

Rapamycin does not affect KSHV transcription but cellular cytokine levels. Shown for each panel A to F on the vertical axis are fold changes relative to mRNA levels at t = 0 on a log10 scale and on the vertical axis days after addition to 50 nM rapamycin or vehicle control (mock). Rapamycin-treated cultures are represented by black squares, mock-treated cultures by gray circles. The results of rapamycin-resistant BCBL-1 cells are shown on the left column and of rapamycin-sensitive BC-1 cells on the right column. Individual panels refer to mRNA levels for the mRNAs actin, LANA, and vCYC. LANA refers to the latency-associated nuclear antigen (orf73) and vCYC to the viral cyclin homolog orf72. Shown for each panel are the concentrations in picogram per milliliter for the indicated cytokines in the vertical axis and time after addition of rapamycin (■) or vehicle (▩) on the horizontal axis for the cell lines BC-1 and BCBL-1: (G) hu-IL-6, (H) hu-IL-10, (I) IFN-γ, (J) IP-10, (K) RANTES, (L) MIP-1α, (M) IL12p40. These data represent 1 of 2 biologic replicates. Panel N shows a representative (IL-6) standard curve.

Figure 5

IL-6 and IL-10 counteract the rapamycin-induced growth arrest. (A) Western blot analysis of 4E-BP1 and eIF4E after immunoprecipitation with ⁷methyl-GTP Sepharose. Samples are drawn at indicated times after exposure of BC-3 cells to 50 nM rapamycin. (B-C) Shown is the concentrations in picogram per milliliter for the indicated cytokine in the vertical axis and dilution of spent medium of 4 days after addition of rapamycin (open symbols) or vehicle (closed symbol) on the horizontal axis for the cell lines BC-1 (circles) and BCBL-1 (squares) as determined by ELISA. (D-E) Relative growth in percentage after 4-day culture in 50 nM rapamycin and the indicated concentrations of either IL-6 or IL-10 as determined by MTT assay.

Figure 6

Cytokine mRNA levels predict PEL class membership. (A) Relative mRNA levels for CCR5, IL-6, and IL-10 as determined by real-time QPCR and expressed as the percentage of HPRT mRNA levels for BC-1 and BCBL-1 cells treated with rapamycin (rapa) or mock-treated; ntc indicates nontemplate control. (B) Summary of rapamycin signaling pathways in PEL (adapted from Science Slides 2005, VisiScience, Chapel Hill, NC). (C) Secreted cytokine levels in picogram per milliliter (log-scale) for virus-negative BJAB and DG75 and KSHV-positive BC-1 and BCBL-1 cell lines. Cytokines are listed on the horizontal axis. (D) Relative levels (vertical axis) of all mRNAs that can be used to predict class membership to PEL of a set of 101 lymphomas (horizontal axis). From this set the relative mRNA for vitamin D receptor (E) and IL-10 (F) levels were mean-centered across n = 101 Affymetrix datasets and averaged for PEL (n = 9), AIDS-associated Burkitt lymphoma (AIDS-BL) (n = 7), AIDS-associated diffuse large B-cell lymphoma (DLBCL) (n = 9), classic, non–AIDS-associated BL (n = 6), non–AIDS-associated DLBCL of immunoblast type (n = 8), non–AIDS-associated DLBCL of centroblast type (n = 23). Histograms show mean and standard deviation. Gray bars indicated 18S RNA levels as control.