Autoantibodies against cell surface GRP78 promote tumor growth in a murine model of melanoma

Abstract

Autoantibodies that react with GRP78 expressed on the cell-surface of many tumor cell lines occur in the sera of patients with prostate cancer, melanoma, and ovarian cancer. These autoantibodies are a negative prognostic factor in prostate cancer and, when purified, stimulate tumor cell proliferation in vitro. It is unclear, however, whether these immunoglobulin Gs are merely a biomarker, or whether they actually promote the tumor growth in vivo. We immunized C57Bl/6 mice with recombinant GRP78 and then implanted the B16F1 murine melanoma cell line as flank tumors. We used the antisera from these mice for in-vitro cell signaling and proliferation assays. The immunodominant epitope in patients with cancer was well represented in the antibody repertoire of these immunized mice. We observed significantly accelerated tumor growth, and shortened survival in GRP78-immunized mice compared with controls. Furthermore, antisera from these mice, and purified anti-GRP78 immunoglobulin G from similarly immunized mice, stimulate Akt phosphorylation and proliferation in B16F1 and human DM6 melanoma cells in culture. These studies show a causal link between a humoral response to GRP78 and the progression of cancer in a murine melanoma model. They support the hypothesis that such autoantibodies are involved in the progression of human cancers and are not simply a biomarker. As GRP78 is present on the surface of many types of cancer cells, this hypothesis has broad clinical and therapeutic implications.

Figure 1. Cell-Surface Presentation of GRP78

Formalin-fixed non-permeabilized B16F0, B16F1, B16F10, 1-LN, and PC3 cells were labeled with either 0.5µg/ml sheep anti-GRP78 polyclonal antibody or with non-immune sheep IgG. The secondary antibody was labeled with a near-IR 800nm fluor and read on a Li-Cor Odyssey® detector. The upper panel shows a histogram of the integrated signal from scans of 24-well plates after subtraction of the background non-immune signal. Below is an example of this cell-based ELISA.

Figure 2. Flow Cytometry

Live, non-permeabilized cells were labeled with 5µg/10⁶ cells of N20 or C20 antibody, or else normal goat IgG. Data shown are representative of several such experiments, and are gated only on live cells.

Figure 3. Murine Anti-GRP78 Autoantibodies

Panel A. Endpoint titers of mouse sera taken before tumor implantation. Samples were diluted and then assayed on plates coated with rGRP78. Endpoints were defined as the highest dilution yielding absorbance threefold higher than background. Panel B. ELISA data from adjuvant-only (N=9) and GRP78-immunized (N=10) mice. EIA/RIA plates were coated with CNVSDKSC peptide conjugated to KLH. Sera were diluted 1:100 in 1%BSA and incubated overnight at 4°C. Data are presented as absorbance units at 450nm and P value here is derived from the nonparametric Mann-Whitney test.

Figure 4. Effects of GRP78 Immunization on B16F1 Flank Tumor Growth

Panel A. Growth of 5×10³ B16F1 cells implanted in the flanks of C57BL/6 mice. Average tumor volumes in GRP78-immunized mice (N=10) are represented by squares while those of adjuvant-only mice (N=9) are represented by circles. Data shown represent mean tumor volume in cm³ ± SEM versus day post-implantation. Asterisks represent statistically significant differences in mean tumor volume at a given timepoint (P<0.05, Student’s t-test). Panel B. Kaplan-Meier analysis of mouse survival compares GRP78-immunized mice to controls. The median survival for the two groups is 26 and 29 days, respectively (P<0.05, log-rank test). The GRP78-immunized group is represented by a solid line, while the adjuvant control group is represented by a dashed line.

Figure 5. Murine Anti-GRP78 Autoantibodies Stimulate Akt Activation

Panel A. ELISA-based analysis of Akt signaling in murine B16F1 cells. Cells were grown to 90% confluency, deprived of serum for five hours, and and then stimulated for 15 minutes with a variety of reagents. Data shown represent mean ±SD of experiments performed in triplicate on two different days. Phospho-Akt results were normalized to the amount of total Akt in each sample. Asterisks represent significant increases in phospho-Akt over baseline, normal sera, and adjuvant-control sera levels (P<0.05) Panel B. ELISA-based analysis of Akt signaling in human DM6 cells. This experiment was performed exactly as in 4A.

Figure 6. Murine Anti-GRP78 Autoantibodies Stimulate Tumor Cell Proliferation

Panel A. Effect of murine GRP78 antisera on B16F1 proliferation. Cells in 96-well plates were deprived of serum for five hours and then stimulated for 16 hours in the presence of various stimuli. Data shown represent mean ±SD of experiments performed in triplicate on two different days. Asterisks represent significant increases in proliferation over baseline and normal sera stimulation (P<0.05). Panel B. Effect murine GRP78 antisera on proliferation in human DM6 cells. This experiment was performed exactly as in 5A.