Antisense therapy restores fragile X protein production in human cells

An antisense therapy developed by Joel D. Richter, Ph.D., Sneha Shah, Ph.D., and Jonathan K. Watts, Ph.D., at UMass Chan Medical School and Elizabeth Berry-Kravis, MD, Ph.D., at RUSH University Medical Center, restores production of the protein FMRP in cell samples taken from patients with fragile X syndrome.

Published in the Proceedings of the National Academy of Sciences, this breakthrough was possible because of the novel findings, also presented in the study, that aberrant alternative splicing of messenger RNA (mRNA) plays a principal role in fragile X syndrome, the most common form of inherited intellectual disability and the most frequent single-gene cause of autism.

"This discovery offers real hope that a therapy to mitigate fragile X syndrome may be possible and could be translated to the clinic sooner than we once thought," said Dr. Richter, the Arthur F. Koskinas Chair in Neuroscience and professor of molecular medicine. "These findings are unconventional and weren't something we were expecting. If you do good basic science, believe in your data and follow where it takes you, the results can change our fundamental understanding of biology and disease."

Fragile X syndrome is a genetic condition resulting from a CGG repeat expansion in the DNA sequence of the fragile X (FMR1) gene. People with fragile X suffer from intellectual disability as well as behavioral and learning challenges. Cognitive disabilities can range from mild to severe and afflict boys more frequently than girls.

There is no cure for fragile X syndrome although interventions such as special education, speech therapy, physical therapy or behavioral therapy and drugs providing symptomatic relief can provide the opportunity for optimizing a full range of skills.

When viewed under the microscope, the FMR1 gene containing the repeat expansion is detected as a narrowed band pinching the tip of one arm of the X chromosome (identified as the fragile site). The main function of the protein product of the FMR1 gene (FMRP) is to bind as many as 1,000 different mRNAs and inhibit their translation.

When FMRP is absent, as in fragile X syndrome, there is excess production of hundreds of different proteins in the brain. Although it's not fully understood how, FMRP control of mRNA translation plays a critical role in synaptic plasticity and higher brain function. Without FMRP, normal neurological development doesn't occur.

Normally, humans have between five and 55 CGG repeats in the FMR1 gene. Fragile X syndrome occurs when an individual has more than 200 CGG repeats in the FMR1 gene sequence. The conventional model of the disease holds that once a CGG repeat length reaches 200 or more, the gene becomes methylated and shut down, and does not produce FMR1 RNA or FMRP.

Utilizing blood samples from males with fragile X provided by Dr. Berry-Kravis, professor of pediatrics, neurological sciences and anatomy and cell biology, Drs. Richter and Shah found something unexpected.

"We had reason to believe that there were defects in a number of the mRNAs being produced by fragile X patients," said Dr. Shah, assistant professor of molecular medicine. "We ran the experiments and began looking at the various RNA readouts, however, we were surprised to find that the cells were producing fragile X mRNA even though no protein was being made. They shouldn't have been producing any fragile X mRNA. This wasn't supposed to be happening. It made us rethink how the disease was occurring on a basic biological level."

Looking closely at the mutation-carrying fragile X mRNA, Shah found a little-known abnormal splice isoform, a sequence variation, referred to as FMR1-217. Before mRNA can be translated by the ribosome into a functioning protein it undergoes a process called splicing.

This intermediary process removes all the non-coding regions of the RNA (introns) and splices back together the protein coding regions (exons). It's believed that variations in this splicing mechanism, called alternative splicing, allows a single gene to create different RNA isoforms. These isoforms, because they each contain different coding regions, allow a single gene to make multiple proteins.

The CGG repeats found in the FMR1 gene mutation, however, were causing a mis-splicing event that left a crucial piece of an intron (a pseudo-exon) in the mature mRNA. This simple splicing error was the reason the FMRP wasn't being made, not methylation of the gene, as had been previously believed. Richter and Shah hypothesized that if this mis-splicing could be corrected or avoided, then normal fragile X protein production could be restored.

One way to alter RNA splicing is to create an antisense oligonucleotide (ASO), a short piece of DNA with a complementary sequence, which will bind to the target mRNA. This binding causes the splicing machinery to skip over the improper splice sites on the RNA, resulting in normal splicing and mature mRNA formation. It's also a technique that is already being employed in the clinic to treat the neuromuscular disorder spinal muscular atrophy (SMA) and is in clinical trials for other neurological diseases.

To design an ASO targeting the fragile X mRNA, Richter and colleagues turned to Dr. Watts, an ASO expert who also works on neurological diseases such as Huntington's disease and ALS. Watts, professor of RNA therapeutics, designed 11 ASOs attempting to find one that would block mis-splicing of the fragile X RNA and restore FMRP production.

A combination of two ASOs developed by Watts successfully inhibited the aberrant splicing and rescued proper FMR1 mRNA splicing in patient-derived cells. This led to production of normal levels of FMRP in these cells.

"We never would have found this using a mouse model of fragile X," said Richter. "The mouse model is a gene knockout. Because it simply doesn't have the fragile X gene, there is no mRNA that is made. The FMR1 mRNA mis-splicing is a gene regulation mechanism that depends on the CGG expansion, which may be unique to human and primates. We only discovered this mis-splicing because we were working in human cells."

Richter and colleagues hope that translating this discovery to the clinic can be expediated because current treatments for SMA are based on a similar technology. The only difference between the two is the genetic sequence of the ASO used to treat the fragile X mis-splicing.

"This is a very exciting finding that has high therapeutic potential," said Berry-Kravis.

"It is very early in development, however, and much work is needed to determine how effectively the ASO strategy can restore FMRP, in what percent of brain cells and in which individuals with fragile X. If the ASO strategy turns out to be successful in cells from a significant percent of individuals with fragile X, this may provide a genetic reversal of disease that could have high clinical impact and improve the functional level of people living with fragile X and reduce the burden on their caregivers."