Improving Primer Design in Allele Specific PCR Diagnostic Assays Seeking Deletion Mutations Using CFTR ΔF508 Locus

By: Sean Cutinha, Robert Rotondo, Aaron Stebbins, Vikram Venkatesan

Abstract

Conventional primers with one base pair (bp) mismatch at the 3’ end occasionally show false positives in allele specific PCR due to non-specific binding. To attempt to reduce false positives in detecting the presence or absence of the ΔF508 deletion, a primer for the ΔF508 gene with two mismatches was tested against the single mismatch primer. We predict that the primer with two mismatches will bind to the mutated template with higher specificity at a wider range of temperatures compared to the single mismatch primer because more mismatches closer to the three end of a primer lead to more specificity in allele specific PCR as demonstrated by Yaku et. al (Yaku 2008). Purified IB3 cell DNA tested with 2 mutant primers, 1 and 2 mismatches at 3’ end (MF1 and MF2 respectively) and a wild-type primer at a range of temperatures (45-54.2 Celsius) with a range of 25-35 cycles. Results indicate highest PCR product concentrations from trials using 35 cycles, primer binding at a range of temperatures with the wild type and MF1 primers binding with the most specificity. From our tests, we conclude that all designed primers are able to bind to purified template DNA, and that further research into the proposed primer design method is warranted.

Discussion

Experiment Overview

Cystic Fibrosis is an autosomal recessive disease caused by the abnormal transportation of sodium and chloride ions across epithelial cells of the lungs, pancreas, liver, and intestines. The ΔF508 mutation results in a deletion of a cytosine pair at the 507 position and two thymine pairs at the 508 position on the cystic fibrosis transmembrane conductance regulator protein (Riordan et. al, 2009). The improper folding of the channel protein is subsequently detected by the cell’s quality control system as defective and recycled before it can be situated in its proper position in the cell. This inhibits the sodium and chloride ions from passing through these channels causing a build-up of mucus and other secretions in the cells. A polymerase chain reaction (PCR) can be used to design an assay that will be able to test an individual for the ΔF508 mutation when run on an agarose gel through gel electrophoresis. (Czerny 1996).

The objective of our investigation was to find an optimal and user-friendly primer design method that can be widely utilized in clinical settings. We hypothesized that if IB3 human bronchial epithelial cells from a cystic fibrosis patient were tested in allele-specific PCR with mutant ΔF508 primers that were designed in such a way that there were two mismatched base pairs at the hydroxyl (3’) end of the primer, then that primer would correctly bind to a mutant DNA template. It would also reduce false positives when testing a wild-type DNA template at a range of annealing temperatures as two mismatched base pairs would not allow the Taq polymerase enzyme to bind to the primer and extend the primer strand.

Original Predictions

Over the course of our study, we first had to replicate an assay for the E.coli bacterium on an agarose gel that served as a control and precursor to our main investigation. Using the 8F and 529R primers of the 16S E.coli gene, we predicted that successful amplification should result in a band length of 521 base pairs. An additional experiment from a previously published paper was conducted as another control using primers with the ΔF508 mutation that we predicted should successfully amplify a 147/150 base pair product (Wittwer et. al, 1993). Our primary investigation involved our healthy primer and two mutant primers with one and two mismatches, respectively that we predicted should amplify a 149 base pair region.

Results and Ultimate Findings

Using the E.coli 8F and 529R primers, a 515 base pair product was obtained on a Tris-Borate-EDTA (TBE) gel. In order to confirm that the base pair length was at the appropriate range, a log-plot was created based on the migration distances from the wells to the bands using the DNA ladder: a gel marker with bands that correspond to specific base pair lengths. The appearance of bright bands with minimal streaking indicates the ideal, proper conditions of the assay. The successful amplification was representative of the facts that contamination when making the PCR cocktail and the agarose gel was minimal, the concentrations of the ingredients in the PCR cocktail were appropriate, and the range of annealing temperatures was acceptable. In addition, the number of cycles run in the thermocycler and conditions of each phase was successful, the completed cocktails and loading dye were aliquoted in a suitable fashion, and the appearance of a crisp ladder is indicative of an accurate band. In this sense, the purpose of these results shows that the components of the protocols works as well as to provide a checkpoint where errors can be pinpointed and variables can be eliminated if successive trials do not run successfully. In this case, there was varying levels of non-specific binding, caused by primers annealing to other regions along DNA strand, across the temperature gradient. The brightest band observed in this gel was observed at a temperature of 59.6 degrees Celsius and our results in this stage of our study support our predictions. In our other control experiment, we were able to acquire bands at the correct range of 147 base pairs despite the contamination of the ladder. All the markers in the ladder could still be verified based on ladder marker diagram in the coursepack and a suitable log-plot was able to be constructed to support that prediction (Luckie et. al, 2013). This control experiment was imperative in analyzing a discrepancy in the form of a single base pair mismatch present in the reverse primers from the template on the protein database and the template found in the published paper. We were able to support our prediction that the single change would not affect annealing.

The testing of our designed healthy and mutant primers resulted in both binding to the IB3 cells. We hypothesized that the IB3 cells had to be heterozygous for the ΔF508 mutation. Upon further investigation, we discovered that this could be supported (Schneider 1999) as the other strand has the nonsense W1282X mutation. When heterozygous DNA is used, the healthy and the mutant primers will both bind to the DNA which will cause bands to appear in both scenarios (Vestheim et. al, 2008). It was also supported that banding with both primers will occur because the heterozygous DNA is unable to discriminate between the healthy and mutant strand and as a result it binds to both primers (Friedman et. al, 1991). With the completion of the experiment, we were able to support the first part of our hypothesis. We were able to successfully replicate an experiment using the primers that scientists had researched and published in addition to successful bands with the healthy and mutant primers to the IB3 cells. Though the bands we received for the control primers and the mutant primers were very faint, we deduced that the greater number of cycles run in the thermocycler, the more visible the bands.

Weaknesses in Experimental Design

Time and cost limitations occurred well before the inception of the research. Ordering additional primers for further testing proved to be costly. As much of the duration of the study was dedicated to the control experiments to understand the variables involved in producing acceptable bands, we were rushed when it came to time to our own research question. Sterilization was an important factor in making the PCR cocktails and loading the mixes into the gels. Contamination of the ladder was observed in a few of our gels and it delayed the progress of our experiment so members made sure to wear gloves to ensure better results. Discrepancies in the primer design and our initial approach towards our project delayed our progress in getting approval for procurement of the proper primers. Precautions were taken after information did not circulate around members of the group and we resorted to email and group meetings to assist each other with the parameters of the experiment. With better communication, each student was better able to understand the protocols during the lab sessions. An increased number of cycles in thermocycler and higher annealing temperatures were utilized in later weeks and were found to be more successful on average. Many steps could be taken if research were to be continued. As the IB3 cells were heterozygous for the ΔF508 mutation, we could test mutant primers against healthy cell lines that do not contain the mutation. Furthermore, we could also test healthy primers against homozygous ΔF508 mutation. These tests would allow us to assess additional controls as healthy primers do not normally bind to mutant DNA and vice-versa and we would be able to gain a better understanding of how mismatches on the 3’ end affect annealing.

Figure 4: IB3 DNA template with Wild Type forward primer. A) A target sequence of 149 base pairs was amplified in this experiment using purified ib3 cell DNA, our reverse primer, and our wild type forward primer. The PCR master mix cocktail contained a total of 50 microliters consisting of 40.5 microliters of nuclease free water, 5 microliters of 10x PCR buffer (200 mM Tris-HCl at pH 8.4 with 500 mM KCl), 1 microliter of 10 mM dNTPs, 1 microliter of 100 micromolar wild type forward primer, 1 microliter of 100 micromolar reverse primer, 1 microliter of purified ib3 cell DNA, 0.5 microliters of Taq polymerase supplied by Invitrogen (Taq Polymerase, 10x PCR buffer, and MgCl). Thermocycler conditions were a three minute initial denaturation stage at 95 Celsius, then 35 cycles of 30 second denaturation at 95 Celsius, 30 second annealing stage at 45.0 Celsius (well 1), 46.9 Celsius (well 2), 48.9 Celsius (well 3), 52.9 Celsius (well 4), and 54.2 Celsius (well 5), and a Taq extension stage at 72 Celsius for 1 minute. At the end of the 35 cycles a 5 minute extension stage at 72 Celsius was implemented. A 1kb plus ladder was used to analyze the PCR products in gel electrophoresis. The gel was a 0.8 agarose gel using 1xTBE (Tris-borate-EDTA) buffer and 2 microliters GloGreen dye. Electrophoresis was run for 30 minutes at 120 Volts. Amplified bands appeared at approximately 150 base pairs in wells 2, 3, and 4. Well 5 showed a weak band at around 150 base pairs, while well 1 showed no bands. B) A logarithmic plot is shown on the graph on the right side. The graph was made using measurements of the 1 kb plus ladder in the gel. The logarithmic trend-line was added and an equation for that line was derived using Microsoft excel (R²=0.98961). The y-axis represents the length in base pair of the DNA strand, while the x-axis represents the migration distance in cm.