A pitfall of using a second plasmid to determine transfection efficiency.

When comparing the ability of two sequences to regulate the expression of a reporter gene, following transfection of cells in cultures, it is necessary to control for the potential difference in transfection efficiency. This is done by either conducting the experiment several times with at least two different preparations of each DNA or by adding a second plasmid (internal control plasmid) encoding a different reporter gene whose product can be used to normalize for transfection. In using this latter approach, we have found that the interpretation of results may be affected. The activity of the internal control plasmid, pRSV/3gal, was suppressed by certain plasmids. This would, in turn, overestimate the enhancer/promoter activity of the cotransfected plasmid. To determine the activity of the upstream regulatory region (URR) of a human papillomavirus (HPV6-W50), we cloned the URR into the enhancerless plasmid, pSVEcat, in which the SV40 early promoter lies just upstream of the cat gene (1). pSVEcat or the test plasmids (pW50acat and pW50bcat, plasmids which contain the URR in the sense and anti-sense orientation, respectively) were cotransfected with pRSV/3gal (a gift from F.Thierry and M.Yaniv) into HeLa cells and the CAT activity (2) and /3gal activity (3) were determined 48 hours later. pSV2cat, which contains the SV40 promoter and enhancer driving the cat gene, was included as a positive control (4). The results indicated that the regulatory region contained enhancer activity (5). They also, however, showed that introduction of pSV2cat into cells along with pRSV/3gal decreased the amount of £gal produced in HeLa cells (Table 1). When the /3gal activity was normalized to that obtained upon cotransfection with pSVEcat, the /3gal activity in cells cotransfected with pSV2cat was consistently low. Cotransfection with the test plasmids showed that /9gal activity varied from one experiment to the next, as expected if the activity is indicative of variations in transfection efficiency. The amount of /3gal activity in cells transfected with pRSV/3gal alone was 7 to 10-fold higher than upon cotransfection with pSV2cat (data not shown). In contrast, in keratinocyte cotransfections, the /3gal activity varied from one experiment to another whether in the presence of pSV2cat or the test plasmids (Table 1). In the case of the HeLa cell transfections, normalizing CAT activity to /Sgal activity (to normalize for differences in transfection efficiency) would suggest that the activity of the SV40 enhancer/promoter was very high compared to the test plasmids. However, this would have been an experimental artifact. Therefore, one needs to be aware of the effects that one plasmid may have on another when in the same cellular milieu. Not surprisingly, such effects may be cell-specific.