Decoding the impact of genetic manipulation is a convoluted task. When it comes to investigating multi-layered diseases like Parkinson’s, it is fully expected that a whole host of genes in sequence or applied in simultaneity are used to promote onset of the disease. The process of figuring out which genes are involved or, conversely, which genes may be exploited as a negative control to reduce the symptoms and causes of the disease, is no easy task.
Generally, single gene overexpression and knockouts are considered a way to find out whether the genetic functions are related to the disease in question. But since synergistic effects across multiple genes are often how neurological diseases progress, this tactic may not have much luck. Attempts with multi-gene alteration have tried to fix this oversight, but it is an incredibly slow process.
Screening For Answers
Researchers at MIT decided to look into whether CRISPR could be taken advantage of as a gene transcriptional modulator. Rather than using it to modify individual nucleotides of genes and add or remove sections, they instead engineered a version that lacked the traditional cutting and splicing capabilities of CRISPR. This new tool binds to the gene of interest and then co-opts the use of transcription factors (commonly called crisprTFs) to turn genes on and off. These factors were set up to have their modulating effects be randomized, meaning that whether they over- or underexpressed a gene wasn’t controlled for.
They named this system PRISM, Perturbing Regulatory Interactions by Synthetic Modulators.
The next step was to combine this altered CRISPR with guide RNAs (gRNAs), but they didn’t want to pick any that targeted specific genes of interest. That would totally defeat the random testing part of the PRISM screening. So they used a plasmid library for guide RNAs that would encode 20 random nucleotides into each one, along with end caps to bind them to the plasmids.
Isolating A Winner
The first screening experiment used Saccharomyces cerevisiae, the yeast model organism, that had been genetically modified to express a protein known to be one of the primary components in Parkinson’s symptoms and eventual death. Called alpha-synuclein (αSyn), this protein has been found to misfold and clump together in neurons. This toxic outcome has also been found to happen in yeast if made to express the protein.
When the PRISM screen and random gRNAs were applied to colonies of yeast, the random transcription variance in whether genes are expressed allowed the researchers to look for those that managed to survive longer or even in total from the formation of alpha-synuclein. While most of the cells did still have extreme structural defects, some colonies did survive and these had their genomes sequenced to determine what had changed.
The scientists found that two particular gRNAs were able to rescue the yeast from αSyn toxicity. They chose to work with the one capable of strong suppression, designated gRNA 9-1. Interestingly, no specific nucleotide match was found in the yeast genome when compared to the sequence of the gRNA. But when they changed the program to allow for several mutations in code that might look for off-target binding, they found 114 genes that had been expressed differently upon inclusion of gRNA 9-1.
A Human Comparison
Almost all of the discovered genes (93%) were not known to have been involved in Parkinson’s activity. 95 of the genes had already been included in the yeast expression library, so the researchers used this to test them individually via overexpression to see what impact each had. 57 of the 95 (60%) were found to have significant αSyn suppression reactions.
An intriguing finding among the genes was that one of the main functional categories that many of them fell into was the heat shock chaperones. Chaperone proteins are involved in controlling and regulating proper folding of proteins and heat shock ones are only expressed when the cell is under a condition of sudden heat increase.
This led to the researchers working with homolog genes in humans that match up with the yeast heat shock chaperones and they were then tested in a separate experiment using human neuroblastoma cells. The DJ-1, ALS2, GGA1, and DNAJB1 homolog genes were employed. After immediate expression, no αSyn toxicity was observed, though neurite retraction did begin to occur eventually. Even so, about 50% of the cells were still viable even after 6 days.
Finally, homologs of the genes TXN and TIMM9 were implemented together and they appeared to have synergistic beneficial effects leading to an 88% cell survival rate. It is possible that more extensive co-expression may increase this rate of return even further.
Some limitations the researchers made sure to make note of are as follows. The CRISPR transcription factors used were first generation ones only capable of modest levels of over- or underexpression. Future studies with stronger factors could identify better genes to test.
Secondly, they focused only on yeast cells that obtained a single gRNA, which led to them picking out gRNA 9-1 for added experimentation. The cells that had obtained multiple gRNAs were discarded. Due to this, there may have been better candidates for gene matching in those gRNAs, so future tests should take steps to ensure only singular gRNA uptake occurs, rather than multiples.
Thirdly, the study only checked the genes pointed out by gRNA 9-1 by using overexpression, even though some of the suppression events could have been caused by downregulating some genes. This will need to be something that is covered by future studies as well. Combinatorial techniques may be needed to assess how these genes each change the phenotype cell expression.
And lastly, the crisprTFs used appear to have had weak binding thanks to not having perfect matches in the genome to the gRNA 9-1 sequence. So it wasn’t possible to identify all of the genes that were likely modified by it. Better methods of screening for regulated genes will be required.
Medical Steps Forward
What’s funny to consider is that, even with all of those limitations, the experiment as a whole was an unmitigated success, finding a wide handful of unknown genes with strong anti-αSyn effects. Once some extra study had been conducted, they should be able to be utilized in gene therapy to have an even greater ability to slow or stop the onset of Parkinson’s.
Mind you, even these genes aren’t a cure, they only deal with one of the main contributors to Parkinson’s development. And they don’t stop the onset even from αSyn completely. They also likely will only prove effective if used as a treatment early in the disease’s formation, before αSyn is able to build up.
But it’s a clear step forward. With this new PRISM screening technique of CRISPR, yet another door had been opened for the medical community to find new possibilities in dealing with some of the most intractable diseases and disorders.
Photo CCs: Hippocampal neurons in culture from Wikimedia Commons