Paradoxical elevation of Ki-67 labeling with protein kinase inhibition in malignant gliomas.

The monoclonal antibody Ki-67 recognizes a nuclear antigen expressed in the G1, S, G2, and M phase of the cell cycle and has been used extensively as an indicator of cellular proliferation in malignant gliomas, both in the laboratory and clinically. Recently, protein kinase C (PKC) inhibitors have been demonstrated to inhibit malignant glioma growth both in in vitro and in vivo. This study was undertaken to determine whether Ki-67 could function as an indicator of cellular proliferation rate after PKC inhibition in gliomas and to explore cell cycle specificity of such inhibition. Both established and low-passage malignant glioma cell lines have previously been shown to be sensitive to growth inhibition by the PKC inhibitors staurosporine and tamoxifen in vitro (IC50 in the nanomolar and micromolar ranges, respectively), as measured by cell numbers, [3H]thymidine uptake, and flow-cytometric DNA analysis. However, in the same cells that are inhibited by staurosporine and tamoxifen on these assays, and on the 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl tetrazolium bromide (MTT) assay in the present study, the Ki-67 labeling index paradoxically increased in a dose-related manner with the same treatments, as measured by immunohistochemistry and confirmed by flow cytometry. For example, in established line U-87, a 20.5% decrease in thymidine uptake and a 28.5% decrease in absorbance on the MTT assay produced by tamoxifen at 1 microM was associated with an increase in Ki-67 labeling from 42% to 62%; staurosporine, which produces a 78.8% decrease in thymidine uptake in cell line A-172 at 10 nM, produced an increase in Ki-67 labeling from 19% to 32%. In this regard, Ki-67 labeling of glioblastoma tissue from a patient treated with high-dose tamoxifen yielded results within the range of 10% to 15% (consistent with values seen in untreated glioblastoma), despite tumor regression with treatment. The authors' interpretation of these results is that these PKC inhibitors are halting the cell cycle in the G1 phase or the G1-S transition (beyond G0 but before S-phase), resulting in a paradoxical increase in labeling while arresting growth. Two important implications from these observations are that Ki-67 is not a reliable indicator of cellular proliferation after treatment with PKC inhibitors and that these inhibitors used at the doses given above halt cell growth in a phase-specific manner.