Screening and identification of differentially expressed transcripts in circulating cells of prostate cancer patients using suppression subtractive hybridization

Abstract

Background: Tumor metastasis and changes in host immunosurveillance are important components in cancer development. Tumor cell invasion into the bloodstream is an essential step for systemic metastasis. Currently, the detection of tumor cells in the circulation is mainly dependent upon the utilization of known epithelial cell markers. However, expression of these molecules is not limited to cancer patients; healthy people also have a small number of epithelial cells in their circulation. Utilizing these markers to detect circulating tumor cells (CTCs) cannot adequately explain the mechanisms of tumor cell survival or their development of metastatic potential in peripheral blood. The immune system can also evolve along with the cancer, actually promoting or selecting the outgrowth of tumor variants. Unfortunately, both metastasis and immunosurveillance remain mysterious and are debatable because we have yet to define the molecules that participate in these processes. We are interested in identifying the existence of expressed genes, or mRNA species, that are specifically associated with circulating cells of cancer-bearing patients using prostate cancer (PCa) as a model.

Results: We established two comprehensive subtracted cDNA libraries using a molecular technique called suppression subtractive hybridization. This technique selectively amplifies transcripts that are specifically expressed in circulating cells of either PCa patients or healthy men. Following sequencing reaction, we showed that 17 out of 23 (73.9%) sequenced clones did not match any mRNAs in the GenBank database. This result suggests that genes associated with alterations in circulating cells of cancer-bearing patients are largely unknown. Semi-quantitative RT-PCR confirmed that two genes are up-regulated in circulating cells of PCa patients, whereas another two genes are down-regulated in the same patients.

Conclusion: The comprehensive gene expression analysis is capable of identifying differentially expressed genes in circulating cells of healthy men and PCa patients. We did not attempt to enrich specific cell types in this study because phenotypes of CTCs and subsets of leukocytes participating in immunosurveillance remain largely unknown. Continuous studies of these differentially expressed genes will eventually lead us to understand the mechanisms involved in tumor metastasis and immune modulation during cancer development.

Figure 1

Evaluation of subtraction efficiency and the presence of potential differentially expressed genes in the subtracted libraries. To determine the subtraction efficiency, β-actin was PCR amplified using a primer set located within the very 3'-end of Rsa I digested β-actin fragment following the second round of hybridization. PCR products were electrophoresed on an agarose gel. No β-actin product was detected after 40 cycles of PCR amplification in a subtracted library, whereas β-actin was detected after 25 cycles of amplification in corresponding un-subtracted library (A). To amplify differentially expressed genes in circulating cells of healthy men and PCa patients, two rounds of PCR amplification was performed following hybridization steps described in Materials and Methods. To demonstrate the presence of potential differentially expressed genes in the subtracted libraries, the final PCR products were analyzed on a 1.5% agarose gel followed by ethidium bromide staining. We detected a series of distinct bands ranging from 300 to 1,000 bp. These DNA fragments represented genes that are either present (B, lane 1) or absent (B, lane 2) in circulating cells of PCa patients.

Figure 2

Confirmation of differential gene expression in circulating cells of healthy men and PCa patients using semi-quantitative RT-PCR. RT-PCR was performed on individual samples from 8 healthy controls and 12 PCa patients to confirm the SSH results. After sequencing reaction to reveal the identities of a total of 23 clones present in the subtracted libraries, PCR primers were designed (Table 2). β-actin was also amplified from the same samples using a β-actin primer set (BD Bioscinces Clontech) to serve as an internal control for standardizing the quantity of the RNA applied in each reaction. After PCR amplification, aliquots (10 μl) of these PCR products were electrophoresed into 2% agarose gels followed by ethidium bromide staining.

Figure 3

Relative levels of target genes expression in peripheral blood circulating cells of healthy men and PCa patients. Images obtained from Figure 2 were captured and analyzed using the Quantity One^® software. For each target gene, levels of gene expression were normalized to the level of β-actin expression for each individual sample. * indicates statistical significance between healthy men and PCa patients at p < 0.05.