Blockade of bulky lymphoma-associated CD55 expression by RNA interference overcomes resistance to complement-dependent cytotoxicity with rituximab

Abstract

Recently, anti-CD20 (rituximab) and anti-Her2/neu (trastuzumab) antibodies have been developed and applied to the treatment of malignant lymphoma and breast cancer, respectively. However, bulky lymphoma is known to be resistant to rituximab therapy, and this needs to be overcome. Fresh lymphoma cells were collected from 30 patients with non-Hodgkin's lymphoma, the expression of CD20 and CD55 was examined by flow cytometry, and complement-dependent cytotoxicity (CDC) assays were carried out. Susceptibility to CDC with rituximab was decreased in a tumor size-dependent manner (r=-0.895, P<0.0001), but not in a CD20-dependent manner (r=-0.076, P=0.6807) using clinical samples. One complement-inhibitory protein, CD55, contributed to bulky lymphoma-related resistance to CDC with rituximab. A decrease in susceptibility to CDC with rituximab was statistically dependent on CD55 expression (r=-0.927, P<0.0001) and the relationship between tumor size and CD55 expression showed a significant positive correlation (r=0.921, P<0.0001) using clinical samples. To overcome the resistance to rituximab by high expression of CD55 in bulky lymphoma masses, small interfering RNA (siRNA) was designed from the DNA sequence corresponding to nucleic acids 1-380 of the CD55 cDNA. Introduction of this siRNA decreased CD55 expression in the breast cancer cell line SK-BR3 and in CD20-positive cells of patients with recurrent lymphoma; resistance to CDC was also inhibited. This observation gives us a novel strategy to suppress bulky disease-related resistance to monoclonal antibody treatment.

Figure 1

Relationships between the size of extirpated tumors, susceptibility to complement‐dependent cytotoxicity (CDC), and CD20 expression. The size of tumors from 30 patients with non‐Hodgkin's lymphoma was measured and the cells were collected. After isolation of CD19‐positive cells, FACScan analysis was carried out with anti‐CD20 antibody, and CDC assay with rituximab was performed. Intensity of CD20 expression was normalized compared with a control. (a) Scatter plot and correlation analysis for size of extirpated tumor versus susceptibility to CDC. (b) Scatter plot and correlation analysis for size of extirpated tumor versus mean fluorescence intensity of CD20. (c) Scatter plot and correlation analysis for mean fluorescence intensity of CD20 versus susceptibility to CDC. All correlations were tested using the Spearman rank correlation coefficient.

Figure 2

Relationships between tumor size, CD55 expression and susceptibility to complement‐dependent cytotoxicity (CDC). The size of tumors from 30 patients with lymphoma was measured and the cells were collected. After isolation of CD19⁺/CD20⁺ cells, FACscan analysis for CDC assay and CD55 expression were carried out. The intensity of CD55 expression was normalized compared with a control. (a) Scatter plot and correlation analysis for size of extirpated tumors versus CD55 expression. (b) Scatter plot and correlation analysis for CD55 expression versus susceptibility to CDC. All correlations were tested using the Spearman rank correlation coefficient.

Figure 3

Effect of small interfering RNA (siRNA) against the 5′‐site of the CD55 gene on expression of the exogenous CD55 gene. MCF7 cells were transfected with pEGFP or pEGFP‐CD55 in the presence or absence of siRNA. After 24 h, the cells were observed by laser scanning microscopy.

Figure 4

Blockade of endogenous CD55 on breast cancer cells by small interfering RNA (siRNA). (a,b) SK‐BR3 cells were transfected with siRNA against three parts of CD55, namely CD55‐N, CD55‐M and CD55‐C, for 72 h. After transfection, the cells were stained with the anti‐CD55 antibody and DAPI, and then the complement‐dependent cytotoxicity (CDC) assay with trastuzumab was carried out with or without adding fresh human AB serum (a, left and right panels). (b) The percentage of propidium iodide‐positive cells was calculated by counting 100 cells. Data are the mean ± SD (error bars) from experiments with triplicate samples. All statistical tests were two‐sided Student's t‐tests.

Figure 5

Blockade of CD55 on primary lymphoma cells by small interfering RNA (siRNA). (a,b) Lymphoma cells from the lymph nodes of five patients with chemotherapy refractory and resistant lymphoma were transfected with siRNA against three parts of CD55, namely CD55‐N, CD55‐M and CD55‐C, for 24 h. (a) After transfection, the cells were stained with anti‐CD55 antibody and propidium iodide (PI), and then the complement‐dependent cytotoxicity (CDC) assay with rituximab was carried out with or without adding fresh human AB serum. (b) The percentage of PI‐positive cells was calculated by counting 100 cells. Data are the mean ± SD (error bars) from experiments with triplicate samples. All statistical tests were two‐sided Student's t‐tests.