Evaluation of LMP1 of Epstein-Barr virus as a therapeutic target by its inhibition

Abstract

Background: The latent membrane protein-1 (LMP1) encoded by Epstein-Barr virus (EBV) is an oncoprotein which acts by constitutive activation of various signalling pathways, including NF-kappaB. In so doing it leads to deregulated cell growth intrinsic to the cancer cell as well as having extrinsic affects upon the tumour microenvironment. These properties and that it is a foreign antigen, lead to the proposition that LMP1 may be a good therapeutic target in the treatment of EBV associated disease. LMP1 is expressed in several EBV-associated malignancies, notably in Hodgkin's lymphoma and nasopharyngeal carcinoma (NPC). However, the viral protein is only detected in approximately 30%-50% of NPC samples, as such its role in carcinogenesis and tumour maintenance can be questioned and thus its relevance as a therapeutic target.

Results: In order to explore if LMP1 has a continuous function in established tumours, its activity was inhibited through expression of a dominant negative LMP1 mutant in tumour cell lines derived from transgenic mice. LMP1 is the tumour predisposing oncogene in two different series of transgenic mice which separately give rise to either B-cell lymphomas or carcinomas. Inhibition of LMP1 activity in the carcinoma cell lines lead to a reduction in clonagenicity and clone viability in all of the cell lines tested, even those with low or below detection levels of LMP1. Inhibition of LMP1 activity in the transgenic B-cell lines was incompatible with growth and survival of the cells and no clones expressing the dominant negative LMP1 mutant could be established.

Conclusions: LMP1 continues to provide a tumour cell growth function in cell lines established from LMP1 transgenic mouse tumours, of both B-cell and epithelial cell origin. LMP1 can perform this function, even when expressed at such low levels as to be undetectable, whereby evidence of its expression can only be inferred by its inhibition being detrimental to the growth of the cell. This raises the possibility that LMP1 still performs a pro-oncogenic function in the 50% to 70% of NPC tumours wherein LMP1 protein expression cannot be detected. This reinforces the basis for pursuing LMP1 as a therapeutic target in EBV associated LMP1-expressing malignancies.

Figure 1

LMP1 expression in transgenic mouse derived cell lines. (A) LMP1 was immunoprecipitated from 300 μg of urea buffer-extracted proteins from seven PyLMP1 transgene-positive line 53 carcinoma cell lines and one transgene-negative cell line (as denoted 53.217neg) and 100 μg of protein from control cell lines: Raji (EBV +ve, containing 50-60 copies of EBV) and Ramos (EBV -ve), using the S12 antiserum. Immunoprecipitated proteins were separated by 10%SDS-PAGE and blotted alongside two total lysate (boiling mix extracted) controls, BL2B95-8 (EBV +ve) and BJAB (EBV -ve) (1 × 10⁵ cells each). The blot was probed with anti-LMP1 antibody 1G6 followed by goat α-rat IgG HRP. The bands corresponding to LMP1 and the immunoglobulin heavy (H) and light (L) chains are indicated on the left. (B) Protein was extracted from B-cell lines 39.415 and 3959.48 with controls Raji and BJAB (the number of cells indicated below each track) using boiling mix and samples western blotted for LMP1 using IG6. The transgenic LMP1 (tgLMP1) is slightly smaller than Raji LMP1 (rLMP1) [3].

Figure 2

GFPdnLMP1 inhibits the clonogenicity of PyLMP1 transgenic carcinoma cell lines. Cell lines 53.217 (transgene negative control) and 53.234a (PyLMP1 transgenic) were transfected with 5 μg of either pGFP or GFPdnLMP1 or mock transfected (no DNA) in duplicate. After 24 hrs cells were passaged 1 to 8 and cultured for two weeks under G418 selection before fixing and staining with crystal violet. Colony numbers for these and further cell lines treated in the same way are given in Table 1.

Figure 3

Expression of GFP and GFPdnLMP1 in transfected carcinoma cells. Transgene negative control cell lines 53.217 and CarbB, and transgenic PyLMP1 cell lines 53.234a and 53.278a, transfected with pGFP or pGFPdnLMP1 were sub-cloned to give clones denoted GFP and dnL (respectively). Protein samples from each cell line sub-clone (indicated by -number) or pool of clones (-p) were examined by western blotting, sequentially using anti-LMP1 (top of each panel), α-GFP (middle) and α-beta-tubulin (bottom), as indicated. (A) 100 μg of protein extract from sub-clones or pools of clones from the cell lines 53.217, CarB and 53.234a. (B) and (C) 50 μg protein extract from sub-clones or pools of clones (pI and PII) from the cell lines 53.278a and control CarB. Non-transfected parental cell line (nt).

Figure 4

Growth curves of GFP and GFPdnLMP1transfected carcinoma clones. Clones of the transfected cell lines 53.217 (transgene negative; left graph), 53.234a (PyLMP1 transgene; middle graph) and 53.278a (Py LMP1 transgene; right graph) that showed the highest expression of GFP or GFPdnLMP1 were analysed for growth by neutral red assay and compared to the parental cell line in each case. Means of the four replicates (with SD) are plotted. Clone 53.234adnL-1 shows significantly different values from the parental cell line 53.234a (day 3: p < 0.0001, day 4: p < 0.0001, day 5: p < 0.0001, day 6: p = 0.0009). Similarly 53.278a clone values differ significantly from the parental 53.278a cell line (eg. for clone 53.278adnL-8: day 3: p = 0.0003, day 4: p < 0.0001, d5: p = 0.002).

Figure 5

GFPdnLMP1 expression is lost from EμLMP1 transgenic B-cell lymphoma cultures. Transgenic EμLMP1 cell lines 39.415 and 3959.48, along with an EBV negative Akata cell line sub clone (AK31), transfected with pGFP or pGFPdnLMP1 were assayed for transfectant expression. (A) Protein extract from 5 × 10⁵ 39.415 transfected cells (and non-transfected control: nt) were examined by western blotting sequentially using anti-LMP1 (top panel), α-GFP (middle) and α-beta-tubulin (bottom), as indicated. Cell aliquots were collected after completion of selection at 3 weeks post transfection (post-tx) and then at weekly intervals (maintaining G418 selective pressure). (B) Bright field (left panel) and green fluorescence (right panel) visualized in pGFPdnLMP1 (top panel) or pGFP (bottom panel) transfected 39.415 cells at 3 weeks post transfection. (C) 40 μg of protein extract from 3959.48 and control AK31 transfected cells (and non-transfected control: nt) were examined by western blotting sequentially using anti-LMP1 (top panel), α-GFP (middle) and α-beta-tubulin (bottom), as indicated. Cell aliquots were collected at 2, 5 and 21 days post transfection (post-tx) for 3959.48 cells and at 12 weeks post-tx for AK31 cells (all under G418 selection). (D) At four weeks post pGFP or pGFPdnLMP1 transfection 3959.48 cells stained with propidium iodide (viable cells exclude staining, apoptotic cells stain) were analysed by flow cytometry, gating on GFP positive fluorescent cells only. Histograms show y axis (cell counts) and x axis (FL2-H, PI staining). The percentage of PI positive cells (of the GFP positive population) is indicated.

Figure 6

GFPdnLMP1 expression is not compatible with the survival of EμLMP1 transgenic B-cell lymphoma lines. Flow cytometric analysis of pGFP (left) or pGFPdnLMP1 (right) transfected 39.415 cells and 3959.48 cells as indicated. Cells were analysed two days and five days post-transfection (under G418 selection), dot plots showing y axis: side scatter (SSC) and x axis: FL1-H: green fluorescence (GFP). The gated population represents the GFP positive cells with the percentage indicated.