Inconclusive Diagnostic Assay to Identify the G542X Mutation of the CFTR Gene in Human Epithelial Cells Using ASPCR
By:
Olivia Baylis, Dyelle Frederick, Megan Gildersleeve, and Ali Kadouh
LB145 Cell and Molecular Biology
Tuesday 10:20 AM
Allison Vlk and Matt VanDer Ploeg
11/21/2017
Abstract
The G542X mutation of the cystic fibrosis transmembrane conductance regulator (CFTR) gene is a class I mutation that ultimately results in a truncated version of the CFTR protein (Schloesser et al, 1991). The purpose of our experiment was to develop an assay to identify the G542X mutation using allele specific polymerase chain reaction. Custom primers were developed to amplify the target segment of the DNA. We hypothesized that an assay which reduces pseudopositive results of the G542X mutation could be conducted using custom allele specific primers with a single base pair mismatch three base pairs from the 3’ end. This will increase primer discrimination against nonspecific binding because the intentional mismatch will create a bump in the primer that will prevent the 3’ end from binding to the DNA if the base pair closest to the 3’ end does not match the template DNA (Yaku et al, 2008). It was predicted that the use of our custom F-Primer 1 and F-Primer 2 used in conjunction with R-primer, will create a band length of 250 bp because the 3’ end of both forward primers binds to the 1626th nucleotide, while the 3’ end of the reverse primer binds to the 1876th nucleotide (Perone et al, 2010). The annealing temperature is predicted to be 58℃ because melting temperature was calculated using the following formula, Tm=64.9℃ + 41℃ x (#G’s and C’s in the primer - 16.4)/N, where N is the length of the primer, and annealing is 3℃ lower (Lorenz, 2012). Positive control experiments were conducted to amplify the Rz gene in Lambda virus, and published primers, to establish an optimal PCR cocktail and protocol. Our central thesis was that this PCR protocol and our custom primers, would accurately detect the presence of the G542X mutation, allowing for more accessible diagnosis of patients with this mutation. Inconclusive results of detecting the G542X mutation, along with developing an optimal PCR cocktail and protocol for wild type published primers, transpired.
Discussion
Experiment Summary
Cystic fibrosis, the most common autosomal recessive disease among Caucasians, affects over 70,000 individuals worldwide (Hurley et al, 2014). The second most common mutation that causes cystic fibrosis is the G542X mutation; a class one variant that results from a point mutation from guanine to thymine on the 1756th nucleotide of exon 11(Kristidis et al, 1992). This error inhibits the synthesis of the amino acid Glycine 542, and codes for a premature translational stop codon, causing the improper synthesis of the CFTR protein (Schloesser et al, 1991). This mutation results in little to no functioning CFTR causing excessive mucus production, impaired liver and pancreatic functions, and increased susceptibility to bacterial infection (Rowe et al, 2011). To treat individuals with these symptoms, doctors must be able to diagnose which of the over 1,500 mutations said individual possesses (Perone et al, 2010). Allele specific polymerase chain reaction (ASPCR) is one method in which a specific mutation can be identified within a genome. Accurate tests using ASPCR for identifying mutations, such as the G542X, is essential in ensuring that patients are diagnosed correctly and can subsequently be treated in accordance to that mutation. We hypothesized that an assay which reduces false positive results of the G542X mutations could be conducted by using custom allele specific primers with an intentional single base pair mismatch three base pairs from the 3’ end of the primer, because this will increase primer discrimination against nonspecific binding (Yaku et al, 2008). Our central thesis was that our custom Yaku-Bonczyk primers and PCR protocol, would accurately detect the presence of the G542X mutations, allowing for more accessible diagnosis of patients with this mutation.
Original Predictions
Two custom forward primers (Fprimer-1 and Fprimer-2) were designed with two intentional nucleotide mismatches for the non-intended DNA strand and a single intentional nucleotide mismatch for the intended DNA strand in accordance to the Yaku-Bonzyck method (Yaku et al, 2008). This allows for the forward primers to discriminate between the wild-type and mutant form of the CFTR gene. We predicted that after successful PCR with FPrimer-2 and RPrimer for the mutated DNA sample, gel electrophoresis would result in a band of 408 base pairs. PCR, with FPrimer-1 and RPrimer, of the wild-type DNA, was predicted to also show a band of 408 base pairs, but show no bands when F-Primer 2 was used, because the intentional mismatch in each primer (FPrimer-1 and FPrimer-2) renders them effective only to their intended strand (Yaku et al, 2008). If each respective primer annealed only to their intended strand, amplifying the target sequence, then the results ultimately support our hypothesis that allele specific primers would provide for an accurate assay for the G542X mutation.
To gain support for our hypothesis, we conducted two positive control experiments to establish optimal PCR protocol for our custom primer test. The first control involved amplifying the Rz gene of the Lambda virus using 1Rz1F and 1-Rz 1R primers. We predicted that the amplified DNA segment would be 395 base pairs in length because of research done by Powell et al (Powell et al, 1994). The second control involved using published primers to amplify the locus of the G542X mutation in wild-type DNA. We predicted that using published primers would result in an amplified sequence of DNA that is 217 bp in length because of previous research conducted by Perone et al (Perone et al, 2010).
Results and Ultimate Findings
After conducting our lambda control experiment, several weaknesses in our experimental design arose that had an effect on our results. Our original PCR cocktail design lacked a source of Mg2+, which acts as a cofactor for taq polymerase, resulting in a lack of visible bands other than the 1 kb ladder (Lorenz, 2012). This was corrected with the addition of 2 µl of MgSO4, which allowed for taq polymerase to function properly, as shown by a visible band.
Our first trial results were also largely affected by improper gel electrophoresis protocol. The casting tray was placed into the buffer with the wells on the positive end, which resulted in no movement of the PCR cocktail. DNA is negatively charged, thus the wells would need to be placed near the cathode to allow for the electroosmotic flow to move the strands towards the anode (N. Stellwagen and E. Stellwagen, 2009). The casting tray also had residue along its outermost ridges, resulting in the tray adhering to the system’s lid and splitting the gel when removed. This was corrected by cleaning the tray and removing the lid with more caution.
After the initial two unsuccessful attempts, our positive control Lambda virus Rz gene amplification successfully resulted in the production of a 385 base pair band. The 385 bp band had an 2.5% variance in the number of base pairs from the predicted band length (Powell et al, 2014). The band itself (figure 1) was not as bright as a non smeared single band that is ideal in PCR amplification (Lee et al, 2012). We predicted that the band was not as brightly colored and distinct because the annealing temperature was not high enough to fully anneal the primers to the DNA. To improve this we conducted another trial following the same procedure but the annealing temperature was increased from 52°C to 62°C. This resulted in 2 bands (figure 2) that were approximately equal in length (388 bp) to the previous trial but brighter and more distinct.
The repeated success of this experiment supported the effectiveness of our testing protocol to be used in subsequent experiments. More specifically, this solidified that our PCR cocktail contained the correct concentration of each reagent necessary to amplify target segments of DNA. Also that we had calculated the correct annealing temperature, and configured the thermocycler accurately. The procedural findings of this experiment were used as reference when errors arose in the testing of our published primers, in order to determine which variable in the experimental procedure needed to be altered in order to yield successful amplification of a target sequence.
Following the successful amplification of the Lambda virus RZ gene Four trials were conducted to amplify a 217 bp sequence of DNA containing the wild-type 1,756th nucleotide of exon 11. This was done to serve as a second positive control for experiments using our designed primers. We were not able to obtain a 217 bp band to support successful amplification. There are a few factors that may have accounted for this. One issue that we repeatedly faced was the appearance of bands that were shorter than 100 bp (figure 5). This was likely the result of primer dimers which is when primers anneal to each other and not to the DNA. We worked to troubleshoot this error by varying our annealing temperatures and including a larger volume of DNA and a smaller volume of primers in our PCR cocktail. These changes continued to result in unsuccessful PCR amplification. Our inability to successfully amplify the target segment of DNA using published primers suggested we did not have a PCR protocol that was developed well enough to create an assay using our designed primers.
Figure
Figure 1. Recreation of lambda RZ gene amplification using PCR and gel electrophoresis with an annealing temperature of 62°C. A.) Two polymerase chain reaction (PCR) trials were conducted using the same MgSO4 volume (2 µL) as used in the successful Rz gene amplification. Both trial were placed together in the thermocycler, beginning with an initial denaturation phase at 95°C for 3 min, followed by 25 cycles of a denaturation phase of 45 s at 95°C, annealing phase of 30 s at 62°C, extension phase of 1 min at 72°C, followed by a final 10 min elongation at 72°C. Trials 1 and 2 were placed in their own respective lanes, one well apart from another (lanes 1 and 3) in a 0.8% agarose gel, where electrophoresis was conducted at 140V for 20 min. The gel indicated the presence of a prominent band in both lanes 1 and 3, as well as faint bands in each lane below the ladder. B.) The log of the migration distance (cm) of observed bands in the 1kb plus ladder were plotted against the known molecular size (bp) of each band to create an exponential function. This was used to solve for the molecular size of both bands, using the migration distance. The intercept of the horizontal and vertical red lines on the scatter plot indicates both the migration distance and molecular size of the amplified DNA (x=388 bp, y=2.38 cm). The correlation coefficient of 0.97125 indicates a strong linear relationship between the ‘x’ and ‘y’ variables.
