Designing a Diagnosis of Krabbe Disease using PCR to detect a 30 kb deletion on the GALC Gene of Human Epithelial Cells

THE ORANGE CRUSH

"30 Days" Experiment Still Film

Abstract

The 30 kb deletion found on the galactocerebrosidase (GALC) gene, is found in all cases of infantile Krabbe disease and causes death before the age of two (Wenger et al, 1997). Additionally, this mutation accounts for nearly 40-50% of the possible mutant alleles for this disease (Wenger and Chen, 1993). In this paper we discuss primers designed for use in a polymerase chain reaction (PCR) protocol, modeled from Bruce White’s manual, with the purpose of creating a valid diagnostic test for Krabbe Disease. Based on previous studies, we hypothesized that by using designed primers and control primers we would be able to successfully detect if the 30kb deletion was present through PCR detection techniques and gel electrophoresis. Specific primers were designed to anneal upstream of the deletion, downstream of the deletion, and where the deletion should be. Annealing primers used in the successful diagnosis of Krabbe disease were taken from the results of Paola Luzi and acted as experimental controls. (Luzi et al, 1995). We predicted to see an appearance of a 509 bp fragment for mutant homozygous individual, a 994 bp and 509 bp fragment for heterozygous individuals, and a 994 bp fragment alone for a wild-type homozygous individual. The results obtained did not match what we predicted. After testing several annealing temperatures and different PCR cocktails we were unable to detect the 30kb deletion. The only bands that appeared in the agarose gels were below 100 bp, for both controls and designed primers. These bands are a result of nonspecific binding, where the primers anneal to themselves instead of the DNA strand.

To further our study, physical symptoms that commonly afflict those with Krabbe Disease were simulated to obtain quantitative results on society’s responses. We predicted that many people would react to our experiment with frustration, a sense of dominance, or complete avoidance; three behaviors that often occur towards physically disabled children by their peer group, and can cause severe sociological distress (Barker, 1948). A correlation test was developed to see if there was correlation between perceived acceptability of symptoms by society and actual difficulty of the symptom. Our findings support the original prediction that society does react differently towards those with a physical impairment, but the data from the correlation test was 0.666 this does not support the idea that society reacts more severely to conditions that are harder for us to live with. This data helped group members connect the PCR based portion of the experiment to psychological and sociological aspects of Krabbe Disease.

Discussion

Experiment Summary

Globoid Cell Leukodystrophy, commonly known as Krabbe disease, results from a 30 kb deletion in the GALC gene, the gene responsible for the production of galactocerebrosidase (Luzi et al, 1995). This leads to a reduction in galactocerebrosidase, which results in the deterioration of the protective coating of nerve cells. The 30kb deletion is a specific variation of Krabbe Disease, which presents itself in 50% of patients with this disease (Kleijer, 1997). Through previous studies using PCR diagnostics, it was determined the deletion begins in the middle of intron 10 in chromosome 14 and proceeds past the gene by 9 kb (Wenger, 2000). We hypothesized that by using PCR techniques and gel electrophoresis, the presence of the 30 kb deletion on the GALC gene of adherent epithelial cells could be detected through use of primers designed to anneal to specific areas upstream of the deletion, within the range of the deleted section, and downstream of the deletion. These primers should amplify different lengths of DNA: a 994 bp long fragment if the individual was homozygous wild type, a fragment 504 bp long if the individual was homozygous mutant, and bands 994 bp and 504 bp long if the individual was heterozygous.

Genomic Preparation

The DNA used in all PCR cocktails was extracted from 200 uL of S9 human epithelial cells by the Generation Capture Column Kit made by Qiagen. The purity of extracted DNA was measured numerous times with the spectrophotometer and yielded an average concentration of 1.798 ng/uL meaning the ratio of pure DNA to proteins and other impurities was high enough to use in PCR. 1 uL of the extracted DNA was used in each cocktail.

PCR

To ensure the cocktail and methods being utilized could work, preliminary testing on Lambda DNA was done. When run in the PCR machine with an annealing temperature of 54° C, bands were observed around the 500 bp mark, where the Rz amplified region was meant to occur (Figure 1). The presence of the band is indicative of a successful result, which supports the validity of our cocktail in its ability to produce results through PCR amplification and supports that our designed protocol should be sufficient.

Primers were developed to run PCR diagnostic tests in an effort to determine if the specimen has the 30 kb deletion. As a control, primers from a previous study done by Paola Luzi were used to yield an expected result that would determine if the protocol we had designed could work (Luzi et al, 1995). For the control experiment the sense primer CP1 (5’-CCTATATG

GAAAACAATGTGG-3’) is located upstream of the deletion, the antisense primer CP2 (5'-AAGGAGCTAACATTT-CAGGC-3') is located within the 30 kb deletion at a location, and the antisense primer CP3 (5'-TCAAGTCCTTGATGATCACC-3') is located after the deletion. According to Paola Luzi’s experiment, these results should yield a 329 bp fragment in an individual with Krabbe disease, indicating the presence of a homozygous mutation, and a 615 bp fragment for a healthy, wild homozygous individual. Additionally, if an individual was a carrier, or heterozygous, a 329 bp fragment and a 615 bp fragment would be observed on the agarose gel. The protocol of the experiment was followed, but did not produce the expected results. Since the gene should have been intact in the cells used, CP2 should have bound to its complementary sequence within the GALC gene and should have transcribed past the origin of deletion, yielding a 615 bp DNA fragment (Luzi et al,1995). The reasoning behind this is that in a patient homozygous for this disease, CP2 will have nothing complementary to anneal to since that particular area has been deleted. This should have caused a 329 bp fragment to appear once gel electrophoresis was conducted, due to the section amplified when the primer annealed after the deletion (White, 1993). However, for a patient with wild type DNA, CP2 will anneal to the sequence normally deleted in patients with Krabbe disease, alternatively producing a fragment 615 bp long (Luzi et al, 1995). Figure 2 presents a 1% agarose gel run at 56^o C containing the control primers given in the experiment. The gel produced bands below 100 bp in length, which indicates nonspecific binding occurred. Nonspecific binding usually indicates primers that are too complementary and bind to each other instead of the DNA strand. It can also indicate incomplete primer binding, which is most likely the case in this scenario. Incomplete primer binding could have been the result of too low a concentration of MgCl₂.

We predicted that a patient could be successfully diagnosed with Krabbe disease through PCR diagnosis using specific primers we created to anneal to particular regions on and after the GALC gene because of the success previous researchers have had using primers they placed in a pattern similar to ours (Luzi et al, 1995). Our designed sense primer, DP1 (5’-CGGGATCTT

GTCAAGTTCACC-3’), was designed to anneal 174 bp before the deletion. Our second primer, an antisense primer DP2 (5’-GATGAGGAGAAGGAAAGGACGG-3’), was designed complementary to a sequence located within the deletion. This primer should have been placed 820 bp into the deletion at a position 88,441,889 bp into chromosome 14. Primer 3, the antisense primer DP3 (5’-GAGTTACTCTTCCCACACGC-3’), was designed complementary to a sequence located 509 bp after the deletion at a position 88,447,404 bp into chromosome 14. After electrophoresis was completed, following the same logic used to decipher the appearance of the Luzi control bands, we should have observed a band in the agarose gel 994 bp long if the individual was homozygous wild type, a band 504 bp long if the individual was homozygous mutant, and bands 994 bp and 504 bp long if the individual was heterozygous. Again, since the gene should still have been intact in the adherent human epithelial cells being used, DP2 should have bound to its complementary sequence within the GALC gene and should have yielded a 994 bp DNA fragment.

Because the melting point values found for each primer were different, we used a wide range of temperatures: from 49° C to 56° C in increments of 1° C. After completing PCR with all of those temperatures, the bands observed were all below 100 bp, like those of the control’s. Figure 3 shows a 1% agarose gel containing PCR cocktails run at 52° C, 54° C, and 56° C, all resulting in nonspecific binding. Again, this nonspecific binding could have resulted in primers annealing to each other, or incomplete/nonspecific primer binding due insufficient MgCl₂.

Sociological Experiment

To try to understand the physiological and sociological aspect of this very rare disease, group members designed an experiment to replicate some of the symptoms associated with Krabbe disease. These included loss of hand control, impairment of limbs, and speech impairment, and vision impairment. Loss of hand control was replicated by wearing gloves; impairment of limbs was replicated by wrapping a bandage around a piece of wood placed at our joints and the use of a wheelchair; loss of speech was replicated by wearing an athletic mouth guard; and vision impairment was replicated by two group members removing their contacts and the other two wearing prescription glasses not meant for their eyes, and later by wearing dark sunglasses and using a cane. Group members rated how difficult those symptoms were to live with on a scale of zero to ten based on time it took members to complete everyday tasks with the impairment as opposed to without the impairment. The results shown in Figure 5 display an average of how difficult group members perceived the symptoms to be. While results initially appeared to follow the trend, speech impairment, the third symptom that presents in the disease, was ranked lower than that of the first two symptoms. Experimenters believe this is because, when presented with speech impairment, words are much harder to understand, but communication was still possible if people could understand us or if gestures were used. Therefore, results do not give strong support to the original prediction.

Group members also recorded society’s reactions to the replicated symptoms. Based on findings from previous studies we predicted that society would react to the symptoms of Krabbe Disease in a particular fashion- that is, with confusion or intentional disregard and sometime even with a subconscious hostility (Barker, 1948). Figure 4 displays the types of reactions from society to our symptoms. Our predictions were correct, based on the larger percentages of society that intentionally ignored us (28%) or stared (30%) at us. The percentage of hostile reaction portrayed (3%) was much lower than expected. Experimenters believe this is because, according to the Barker study, hostility is usually presented subconsciously as opposed to provoking a physical reaction (Barker, 1948).

Figure 5 displays a correlation test of the reactions of society and group member’s difficulty rating of each symptom. The correlation value was 0.666, which indicates that the two sets of data are relatively unrelated, and does not agree with the Westbrook Hierarchy presented (Dear et al, 1997). We believe this may due to how physically obvious the symptoms were to society. For instance, vision impairment was originally difficult for society to notice, when no sunglasses or cane was used. Group members found this symptom extremely difficult, but society did not always realize our symptom or view our behavior as very abnormal. This part of the experiment has, undoubtedly, given the group a better understanding of how much the public knows about the disease and reacts to those with symptoms of the disease.

Troubleshooting

In all of the gels ran during the experiment, nonspecific binding was observed. As stated earlier, this could have been due to primers annealing to each other because they were too complementary. In the case of the control primers this is an implausible explanation, because those specific primers worked in previous experiments. Too low of a concentration of MgCl₂ or using different machines than those in the control study is a more plausible explanation. The previous study used Amplitaq Gold Fast PCR Master mix from Applied Biosystems instead of our lab-made cocktail. This could have resulted in our group using the wrong measurements of MgCl₂ or TAQ polymerase for that cocktail.

Future Directions

Several portions of the experiment could have been modified if more time was available. Ordering mutant DNA would have helped with the experiment, but it was not possible due to time and the expense of the DNA. Additionally, if more time was available, a wider range of temperatures would have been used with the design primers, including using increments of 0.5 ^o C. This would have been useful because some primers may have only annealed at exact temperatures, resulting in the desired bands. Also, different amounts of MgCl₂ and DNA template would have been used. MgCl₂ aides primers during the annealing step of PCR, which could have helped us get the desired bands. Another error that could have been remedied if more time was available was being more cautious with our materials. For instance, TAQ Polymerase is very temperature sensitive, so we may have been more diligent about keeping it cool or on ice during the entire experiment. Another approach we would have tried if given more time is Touchdown PCR. In Touchdown PCR, each annealing cycle runs at different temperatures, starting with very hot and decreasing each cycle. This allows the primers to anneal at the best temperature possible, which reduces nonspecific binding (Don, 1991). If more time was allotted for the experiment, we would have also redesigned our primers. We would have looked for regions whose fitting primers would have less complementary sections. Finally, if we had more time, we would have also changed the concentration of ethidium bromide used in the gels. This could have allowed us to more easily view any bands that could have been hidden by an incorrect concentration of ethidium bromide.

Figure 2: 1% agarose gel, from gel electrophoresis, showing attempted results of PCR amplification of the 30 kb deletion in the GALC gene from control primers, ran with an annealing temperature of 56^o C. Lanes 1 and 4 contain the invitrogen 1 kb Plus Ladder. The protocol from the Luzi article was used in an attempt to replicate results to utilize as a control for our designed primers. Lane 2 contained CP1 and CP2, while lane 3 contains CP1 and CP3. The PCR was run with wild type DNA; as such, lane two should have had a product that would not leave the well since it would have been over 30,000 base pairs long due to the lack of the 30 kb deletion, while lane three should have yielded a fragment 994 base pairs long since the primers would have annealed before and within the area where the deletion would be present in mutant DNA. 3% DMSO was used along with 2.5 uL of MgCl₂, as was specified in the Luzi article, but only nonspecific binding below 100 bp was observed.