Gene Expression Changes in Response to In Vitro Culture and Anti-IgM Stimulation

The observation that cultured hbcl-2 transgenic B cells essentially maintained mature B cell surface marker phenotypes and showed comparable functional responses to freshly isolated transgenic and wild-type B cells indicated that the cultured cells were a stable population. To further examine the suitability of cultured hbcl-2 transgenic B cells as an alternative signaling model for primary B cells, we employed microarray analyses of gene expression profiles in these cells. Gene expression changes induced by culture and the effect of prolonged culture on transcriptional responses to anti-IgM treatment were assessed. A few probes showed reproducible changes in expression between day 0 and day 1 or day 0 and day 3 of culture. Of these, we identified two clusters of differentially expressed probes with statistically significant associations with biological functions or processes, using CLASSIFI analysis. One of the clusters was enriched with probes from hemoglobin genes, which were downregulated following one and three days of culture (Fig. 16A and C). The downregulation of hemoglobin genes might be due to the gradual disappearance of contaminating reticulocytes from B cell cultures with time. The second cluster of probes, mostly from genes encoding enzymes in the glycolysis pathway, showed consistent upregulation only in hbcl-2 transgenic B cells cultured for three days (Fig. 16A and B). Glycolysis regulation is required for cell survival (20-22). Indeed, the proapoptotic BCL-2 family member, BAD (which can be sequestered by antiapoptotic BCL-2 or BCL-X_L), has recently been shown to occupy an unanticipated role in integrating signaling pathways for glucose metabolism and apoptosis (23). In conclusion, the upregulation of genes encoding glycolysis pathway enzymes in hbcl-2 transgenic B cells after three days could be contributing to the survival of these cells during prolonged culture.

Click on each section of the figure to enlarge.

Fig. 16. Gene expression changes in hbcl-2 transgenic B cells during in vitro culture. (A) A dendrogram of hierarchically clustered gene expression changes of hbcl-2 transgenic B cells after one and three days of in vitro culture. The gene expression changes were calculated as the ratio of expression level at day 1 or day 3 (normalized and background subtracted cy5 intensity) versus that at day 0 (normalized and background subtracted cy3 intensity) in log2. Only probes with duplicate measurements and an average of twofold or more changes in response to one or three days of culture were included. Two or three replicate data sets were presented for each time point in culture. Each replicate data set is labeled as Expt. Data sets for B cells cultured for one and three days are labeled as D1/D0 and D3/D0, respectively. CLASSIFI analysis identified two clusters of probes that showed significant association with glycolysis and hemoglobin. (B) The cluster enriched with probes from genes encoding glycolysis enzymes was upregulated only in hbcl-2 transgenic B cells cultured for three days. Glycolysis enzyme genes are indicated by red arrows. (C) The cluster enriched with probes from hemoglobin genes was downregulated in both hbcl-2 B cells cultured for one day and those cultured for three days.

Human bcl-2 transgenic B cells had robust, and generally consistent, transcriptional responses to anti-IgM stimulation following prolonged culture. A common set of 228 array probes showed at least an average of twofold expression changes in response to 2 hours of anti-IgM treatment in B cells that had been cultured for 0 days, 1 day, 3 days, or 6 days. Of the 121 probes from known genes, 80 array probes representing 65 genes were upregulated, while 41 array probes representing 38 genes were downregulated (see Table 2, supplemental Excel file, for complete gene list). To evaluate the effect of prolonged culture on the transcriptional response to anti-IgM, we compared anti-IgM-induced expression changes of 1007 array probes in B cells cultured for different numbers of days. These probes exhibited at least twofold expression changes when B cells were stimulated with anti-IgM for 2 hours at one or more time points during culture. The majority of the probes showed similar expression changes whether the cells were stimulated on day 0, 1, 3, or 6, as visualized by hierarchical clustering (data not shown). Only 10 out of the 1007 probes were differentially expressed at a false discovery rate of 8% between different time points during culture, as determined by Multiclass SAM analysis (Fig. 17). Many of these 10 probes, such as those representing cAMP responsive element modulator, calmodulin 1, golgi phosphoprotein 3, and fatty acid binding protein 5, showed different expression responses to anti-IgM stimulation in B cells cultured for 0 days versus in B cells cultured for 6 days. Whether these differences are the results of prolonged culture is unknown, and their relevance to cellular response to anti-IgM is undetermined.

Fig. 17. Only a few probes showed different expression responses to anti-IgM stimulation in hbcl-2 transgenic B cells cultured for different numbers of days. The 10 probes were identified by SAM analysis as differentially responded to anti-IgM (AIG) treatment among hbcl-2 transgenic B cells that had been cultured for 0 days (D0), 1 day (D1), 3 days (D3), and 6 days (D6) at a false discovery rate of 8%. A dendrogram of hierarchically clustered expression changes of the probes in hbcl-2 transgenic B cells in response to AIG stimulation is shown. The gene expression change was calculated as the ratio of the expression level in stimulated cells (normalized and background subtracted cy5 intensity) versus that in the time-matched control cells (normalized and background subtracted cy3 intensity) in log2. There are two or three replicate data sets for each time point, where each replicate data set is labeled as Expt. The color scale of expression changes is the same as that in Fig. 16. The gray spot represents blank measurements.

To determine whether anti-IgM-induced gene expression changes in hbcl-2 transgenic B cells were similar to those induced in freshly isolated wild-type B cells, we compared the data with those in wild type B cells stimulated for 1 hour, 3 hours, and 6 hours with anti-IgM measured in another experiment. Gene expression changes relative to time-matched control cells were determined in a similar manner as for the hbcl-2 transgenic B cells. A total of 1416 probes showed twofold or more anti-IgM-induced expression changes in either hbcl-2 transgenic or wild-type B cells, and their expression changes were generally similar in direction (up- or downregulation) in both types of B cells. By examining the dendrogram of hierarchically clustered expression changes of the 1416 probes (data not shown), we found that the magnitude of expression change in hbcl-2 transgenic B cells (2 hour-stimulated with anti-IgM) was generally greater than that in 1 hour-stimulated wild-type B cells and comparable to that observed in 3 or 6 hour-stimulated wild type B cells, as would be expected if hbcl-2 transgenic and wild-type B cells have similar transcriptional responses to anti-IgM. Finally, the results regarding anti-IgM-stimulated gene expression changes for both cell types were generally consistent with historical studies of wild-type B cells (24-26).