In search of locus of additional HphI recognition site in a CF novel mutation using H. sapien DNA, PCR, and HphI analysis

 

 

 

 

By: Fadumo Ali, Renee Kinne, and Lakota Shehi

 

Abstract

Cystic fibrosis (CF) is a recessive genetic disorder that affects 1 in 2000 people (Kerem et al, 1989). There are many mutations that lead to CF, one such is the C to T 3849 + 10Kb mutation found in intron 19. The C to T 3849 + 10Kb mutation is a splice mutation that creates a new HphI restriction site that leads to a newly expressed 84-bp exon (Highsmith et al, 1994). This mutation is intriguing for several reasons. Because it is an intron mutation, the location of the specific nucleotide substitution is ambiguous in current research. It also lacks common symptoms found in CF patients making diagnosis challenging. The purpose of this study was to create an accurate assay to increase the rate of successful diagnosis of this mutation. However, because of the uncertainty, the locus of the mutation site had to be first experimentally confirmed. Because of the creation of an additional HphI restriction site, we hypothesize that through PCR amplification of the DNA suspected of containing the mutation and applying an HphI enzyme we will see an addition cut on the mutated DNA splitting a 766 bp band into a 449 bp and 317 bp band that differentiate the wild type DNA from the mutant DNA (Lander et al, 2003). PCR was used to amplify both wild-type and mutant DNA, using designed primers that bracketed the predicted mutation site (Schochetman et al, 1988). The PCR reactions were then incubated with the HphI enzyme and analyzed with gel electrophoresis to compare the count and locations of the bands shown in the gel, which indicated the length of the DNA segments (Laurell, 1965). We successfully amplified a 395 bp segment of the Rz gene of the Lambda virus (Sanger et al, 1982) and a 454 bp segment of exon 19 of the CFTR gene to serve as controls for subsequent experiments (Zielenski et. al, 1991). We then designed a successful set of primers that amplified the 766 bp band we predict to contain the mutation (Laurell, 1965).

 

Discussion

Experiment Summary

Cystic Fibrosis is a recessive autosomal disease caused by mutations in the CFTR gene, which codes for a protein that regulates chloride flow in cells. Patients with CF suffer from pancreatic exocrine insufficiency, infertility in males, and bacterial infections in airways from the lack of chloride movement that facilitates osmosis and disrupts mucus homeostasis (Knowles and Durie, 2002).  This study focused on a class V mutation found on intron 19 that causes alternative splicing, resulting an atypical 84 bp exon and the creation of a new HphI restriction site (Figure 6) (Highsmith et al, 1996). The C to T 3849 + 10Kb mutation is unusual because the common symptoms of CF, such as elevated chloride levels in sweat and azoospermia, are not present (Stern et al, 1995). The lack of conventional symptoms and the nature of intron mutations, leads to complications in successfully diagnosing and therefore, treating patients with this mutation. We utilized the additional HphI as a possible avenue for developing a diagnostic assay. However, because of ambiguity of the mutation location, we first had to confirm the exact location downstream from exon 19 of the mutation. We hypothesized that the mutation was located exactly 13,773 bp downstream from the the 3849th bp on exon 19. We predicted the DNA containing the C to T 3849 + 10Kb mutation will be visibly different from wild type DNA when amplified and incubated with the HphI enzyme and ran through gel electrophoresis allowing for successful diagnosis of the mutation because of the addition HphI recognition site created by the C to T substitution (Highsmith et al, 1994).

Original Predictions

            PCR was the major method used for the experiment, important for both establishing our controls and the diagnostic assay itself. We predicted that through PCR a 396 bp long segment of the Rz gene on the Lambda virus would be isolated and amplified based on the primers homologous bases to the lambda virus (Sanger et al, 1982).  The Rz gene was used as a positive control throughout the experiment to aid in trouble shooting when subsequent trials lacked visible or clear bands. A second control was developed through the replication of primers used to isolate a target segment of the CFTR gene from a previously published study done by Zielenski et al (1991). We predicted that PCR the forward and reverse primers would result in a band of 454 bp due to its isolation of the entire 19th exon of the CFTR gene (Zielenski et al., 1991). With two controls to ensure proper PCR and gel electrophoresis protocols, we designed two primers that would bind to both the mutated and wild type DNA around the segment of DNA we predicted the C to T mutation to be located. We predicted the set of primers would isolate a 766 bp long segment on both the mutant and wild type DNA because of the homologous bases of the primers that would allow for annealing and therefore replication through PCR (Lander et al., 2003). To the amplified wild type and mutant DNA we applied a HphI enzyme and incubated it to facilitate restriction. Because there were no HphI restriction sites found on the wild type segment, we predicted that after HphI analysis there would still be a single band seen in gel electrophoresis of 766 bp long (Laurell, 1965). However, because the mutated DNA contained a single HphI restriction site (Highsmith et al, 1994), we predicted after HphI analysis when the mutated DNA was run through gel electrophoresis there would be two bands, 317 and 449 bp long (Laurell, 1965). This would allow us to differentiate between the wild type and mutant DNA and determine if we successfully amplified the mutation as well as successfully diagnose the C to T 3849 + 10Kb mutation.

Results and Findings

            The 395 bp segment of the Rz gene on the Lambda virus was successfully amplified through PCR shown through gel electrophoresis (Figure 4). Three trials of the PCR product were ran through agarose gels to create a semi log plots that allowed for the amplified DNA segment length to be calculated to examine the PCR and gel electrophoresis protocol accuracy. Each trail had a percent error less than 5% which provided enough support that the bands seen in the agarose gel were the isolated Rz gene section. This result validated the PCR and gel electrophoresis protocol we used and allowed for us to use the lambda PCR product as a positive control through the subsequent experimental trails.

            The primers replicated from Zielenski et al’s paper successfully amplified exon 19, creating a band 454 bp in length (1991). 5 trials of the published primer PCR product were run through gel electrophoresis which allowed for a semi log plot to be created and the band length calculated. Unfortunately, two trials had a percent error over 5%. However, all the percent errors were still below 10% and the total average of percent errors was 5.5% which provided us enough support that the band seen in the gel was the 454 bp exon 19. This repetition of other scientists’ work not only added validity to Zielenski’s study but added support to the successful isolation of wild type CFTR DNA from human buccal cells. Data from the Epoch spectrometer provided an average concentration of DNA of 0.993 mg/ml which was converted to be a yield of 0.02602 mg of DNA isolated through our Chelex bead protocol. Because less that a nanogram of DNA is needed for a successful PCR reaction to happen, this was considered a high enough yield to be utilized in subsequent experiments with our designed primers (Sommer & Tautz, 1989). The purity of the isolated DNA sample was not high, with an average 260nm/280nm ratio of 1.182, meaning our protocol successfully isolated the DNA from the cell but did not successfully purify it. However, because proteins will not affect a PCR reaction, the low purity was not considered a concern and the sample of isolated DNA was used for following trials. Requiring a second confirmation that the DNA was successfully isolated, we ran the isolated DNA through an agarose gel. We predicted the DNA would glow in the well because its immense size would prevent any movement (Laurell, 1965). However, no glow appeared possibility because the concentration of DNA was too small to successfully create a visible glow (Figure 4). Nevertheless, the bands created by the published primers provided the second support of successfully isolated wild type DNA. If the DNA was not successfully isolated, the primer would have nothing to anneal to and nothing would be amplified leading to no band seen in the agarose gel.

            The primers we designed successfully isolated a 766 band of wild type DNA. While x trails of the designed primers were done, only 2 of those trials provided clear ladders for a semi log plot to be created the bands to be analyzed. This limited the data available for us to draw our final conclusions on. However, the two successful trials provided band lengths that supported our predictions. The average percent error for the bands was 4.70% which was below 5%. We considered this enough support to the conclusion the bands seen were the 766 bp segment downstream from exon 19. Unfortunately, even the successful gels contained unidentified band smears bellow each band, bringing into question the integrity of the protocol used. This was a problem we were unable to resolve and requires future experimentation. However, because each gel had a distinct band, we considered this issue not significant enough to consider our deigned primer trials unsuccessful.       

 

 

Future Direction

Because of several restraints, such as time, money, and availability, DNA that contained the C to T 3849 + 10Kb mutation was not obtained and therefore unable to be tested. This lead to several complications for our experiment, as the mutant DNA was vital for locating the mutation and designing the diagnostic assay. Therefore, in the future we would continue our experimentation protocol once we successfully obtained DNA with the C to T 3849 + 10Kb mutation.

            We would first need to apply the designed primers to the mutant DNA and successfully isolate the same 766 bp band that we isolated from the wild type DNA. Because there is no difference between the mutant and wild type DNA at the segments where our designed primers anneal, we predict that the same time and temperatures used in the PCR of the wild type will allow for successful isolation and amplification of the mutant DNA (Lander et al, 2003).     

            After successfully amplifying both the mutant DNA and wild type DNA we would then incubate both PCR products with the HphI restriction enzyme to facilitate HphI analysis. Because there are no HphI recognition site on the wild type PCR product but a single HphI recognition site on the mutant DNA we predict that when ran through an agarose gel, the wild type DNA will remain a 766 bp long band while the mutant DNA will instead be cut into two bands, 317 and 449 bp in length (Pingoud & Jeltsch, 2001). The difference in the number of bands would allow us to differentiate the mutant and wild type DNA and we predicted would be a successfully diagnostic assay for the C to T 3849 + 10Kb mutation that causes CF. 

Beyond the lack of mutated DNA, there were two aspects of our completed experiments that require continuing investigation. The first is the fact the isolated wild type DNA failed to be visible in the well when ran through gel electrophoresis. We predict that this happened because of the low concentration of the DNA we isolated. To increase the likelihood of seeing the DNA glow we would experiment with our Chelex DNA isolation and purification protocol. One possible variable we would test would be the time the mixtures were left to incubate. Because the mixtures don’t reach the correct temperature the moment it is placed into the water or sand bath, we would factor the time it takes for the mixture the reach the temperature before starting the incubation timer. This we predict would help optimize the performance of the Chelex because it is reaching it’s needed temperature for a more accurate amount of time (Ikkanda, 2004).

We were unsuccessful in eliminating the unidentified band smears that appeared below each band when using our designed primers. Our latest trial included lowering the annealing temperature to the actual calculated annealing temperature of 52 23. However, that resulted in no bands when run through gel electrophoresis. For future trials we would continue adjusting the annealing temperature to find the lowest temperature where the primers still annealed. We predict this would eliminate nonspecific binding that leads to smears forming.      

 


Figure 3. Agarose gel and semi log plot for designed primers and wild type DNA. A. Gel electrophoresis of lambda Rz and designed primers in wells 2 and 3 respectively. In well 2 there is a slight smear, but no strong, bright lambda Rz band. The gel electrophoresis was run for 30 minutes. For Lambda Rz the target sequence was formed using a forward primer of 5’ GATGTATGAGCAGAGTCACCGCGAT 3’ and a reverse primer of 5' GAGGGTGAAATAATCCCGTTCAG 3’ (Sanger et al, 1982). We designed primers for the target sequence using a forward primer of 5’ GTGAAGGCCTTCTTCCTCAC 3’ and a reverse primer of 5’ GAATTTCTCCAGCTCTCCTCC 3’ (Laurell, 1965). ). One PCR mixture contained 1 µl of Lamdba DNA while the other PCR mixture contained 1 µl isolated DNA, wild type DNA for the designed primers primers. Both also contained 1 µl 5 U/µl Taq polymerase, 1 µl 10 mM deoxy nucleotides (dNTPS), 5 µl 10x PCR buffer, 1 µl 100 µM forward primer, 1 µl 100 µM reverse primer, 1 µl magnesium sulfate, and 39 µl nucleotide free water. The thermocycler started at an initial temperature of 95 ºC for 3 minutes for the initial denaturing. It then ran for 40 cycles of 30 seconds at 95 ºC, 60 seconds at 53 ºC, and 90 seconds at 72 ºC and finished with 3 minutes of final extension at 72 ºC. A 1.5% agarose gel was used. In the first well, was a mixture of 4 µl of loading dye and 6 µl of 1 Kb+ Ladder. In the second well was a mixture of 10 µl of isolated Lambda Rz gene and 5 µl of loading dye as a control. In the third was a mixture of 10 µl of wild type PCR product with and 5 µl of loading dye. The gel was run at 180 V for 30 minutes. B. Semi-log plot of the results achieved with respect to the ladder. We measured pixel distance from well based on an image on the computer. A line of best fit was found using exponential method we found the R2 value to be 0.986. An equation derived from the line of best fit was used to determine the number base pairs in the different samples. The designed primers were found to anneal at a distance of 801 base pairs, which has a 4.63% error. The purple intercept line represents where the designed primers are on the semi-log plot.