When I visited Ricaurte, Colombia, in 2016, I was surrounded by men with long faces and prominent ears. As we spoke, they would ask repetitive questions while mumbling and failing to maintain eye contact, and when they shook my hand, they turned their body away from me. They were interested in me but were too shy to interact. This type of anxiety-related approach-withdrawal behavior is typical of those with fragile X syndrome (FXS), a well-characterized genetic disease that is the most common inherited form of intellectual disability and the most common single-gene cause of autism. Even many of the Ricaurte women, who usually have at least one good copy of the X chromosome, showed similar social deficits. I had never seen so many individuals with FXS all together. I thought to myself: This is ground zero for FXS. ...

Likely because the founding families of this small village had one or more carriers of the causative mutation, Ricaurte has the highest known prevalence of FXS in the world. Last year, our team published the results of genetic testing of almost all of the inhabitants in this village. We found that nearly 5 percent of male and more than 3 percent of female inhabitants of Ricaurte have FXS,1 compared to around 0.02 percent of people living in the US and in Europe. In Ricaurte, the residents are supportive of these individuals, who work in the community and are well accepted. Their behavior does not seem unusual to those living in the village. Relatives who have moved away from Ricaurte and then subsequently have had a child with FXS will move back to this town for the acceptance and support they find there. This pattern further enhances the genetic cluster of FXS-causing mutations in this area.

For the foreseeable future, treating FXS will be a matter of managing symptoms and working through therapy to support healthy development.

Initial chromosome testing documented the fragile site on the X chromosome, and we saw the disease’s high prevalence as an advantage in our quest to study the hard-to-treat condition.  While research of the past three decades has detailed the underlying genetic and molecular causes of FXS, it has yet to yield a treatment that can reverse the associated neurobiological abnormalities in the brain. By analyzing the genomes of FXS patients in the village, we hope to learn more about the founders and to investigate the background allelic variations that facilitate the appearance of new FXS mutations. We are also investigating individual variability and whether environmental toxicity can exacerbate FXS patients’ conditions. 

At the same time, we are conducting several controlled trials of promising FXS drug candidates at the MIND Institute at the University of California, Davis, Medical Center. The candidates include cannabidiol as well as metformin, a type 2 diabetes medication that can also help obesity. We are hopeful that our work in Ricaurte, along with the work of other teams investigating populations with high rates of FXS in Indonesia, Israel, and on the Spanish island of Mallorca, among other places, will inform the development of better treatments for patients all over the world.

What we know about fragile X

Fragile X syndrome was named for a delicate site at the bottom of the X chromosome of patients with the disease that for unknown reasons looked “fragile” when their cells were studied in culture media deficient of folic acid. In 1991, researchers identified an expansion of the DNA sequence CGG in the promoter region of the FMR1 gene as the underlying cause of the X chromosome’s broken appearance. This gene codes for fragile X mental retardation protein (FMRP), which controls the translation of hundreds of messenger RNAs into various proteins that are important for neuronal function. FMRP is the main controller of synaptic plasticity in early brain development and is needed throughout life to make new neurons. Recent research has found that the protein also controls a protein essential for epigenetic changes across the genome, perhaps explaining FMRP’s wide-ranging effects. 

The mutated FMR1 gene itself harbors epigenetic differences. While the wildtype gene lacks methylation, sequences with more than 200 CGG repeats—considered the full mutation—are heavily methylated, which thereby blocks transcription of the gene. In males, this leads to a complete absence of FMRP, while females typically have one X with the wildtype gene that provides some amount of the protein. As a result, their symptoms are typically less severe. 

Some individuals carry what is called the premutation, characterized by a CGG expansion of 55 to 200 repeats. Premutation genes with fewer than 120 CGG repeats typically result in normal FMRP levels, while those with more repeats have lowered FMRP production, leading to relatively minor learning problems or, rarely, intellectual disabilities. Up to a normal threshold, generally the more FMRP produced, the higher your IQ. Paradoxically, premutations result in higher FMR1 mRNA levels than are found in wildtype individuals, and these RNAs can cause problems even if FMRP protein levels are normal. (See “Sidebar” on page 51.)

If women carrying the premutation pass it on, it can expand into the full mutation during egg development. Once the CGG repeats number more than 100, every time that X chromosome is passed from mother to child it will carry the full mutation, as the CGG repeats apparently expand during either oogenesis or embryonic development. The details of this expansion are still murky. Male carriers of the premutation will pass it on to all of their daughters. If a man carries a full mutation, it reverts back to a premutation in his sperm, such that all of his daughters inherit the premutation. Again, the reason for and mechanisms of this back mutation are unclear. (See illustration on page 49.)

Despite questions about how the FXS locus shuffles CGG repeats in a somewhat predictable manner, the underlying molecular mechanisms of the disease are fairly well characterized, thanks to copious research on FMRP’s role in the cell. This has led to human trials of several treatments that target the relevant molecular pathways, but unfortunately, they’ve had little success. While there are therapies to manage symptoms, there is currently no cure for FXS.

Our failure in treatment trials

There are many ways to cure a mouse of FXS. For example, inhibitors of the metabotropic glutamate receptor 5 (mGluR5) pathway, which is involved in neural signaling and is known to be overactive in FXS, can relieve mouse models of all symptoms of the disease. But when such therapies moved into human trials a few years ago, adolescents and adults with FXS experienced no benefit beyond what the placebo group enjoyed. Similarly, agonists that target the neurosignaling GABA system, which is underactive in FXS, succeeded in alleviating symptoms in mice but in humans proved no better than placebo.2,3 Regardless of whether patients got the mGluR5 inhibitor, GABA agonist, or the placebo, between 20 percent and 30 percent of patients showed improved behavior and their families were feeling positive about the outcome.4,5  

Given the high expectations among patients and their families for targeted FXS treatments, such high placebo responses are common. There is thus a need for quantitative outcomes that measure patient responses to treatment through electrophysiological measures. In the last six years, such measures have been developed. For example, event-related potentials (ERPs), patterns of activity visible on EEG readouts in response to sensory cues, can reveal the failure of FXS patients to habituate to repeated stimuli such as repetitive sounds or visual stimuli, leading to hyperarousal. Another option is to track patients’ eye movements to monitor their ability to make eye contact. There are also new molecular biomarkers that reflect a reversal of the neurobiological changes in FXS. For example, a protein called matrix metallopeptidase-9 (MMP-9) is elevated in the blood of FXS patients, and a targeted treatment such as metformin lowers the MMP-9 level in mice and patients with FXS. 

The Fragile X Mutation

Fragile X syndrome is caused by an expansion of CGG nucleotide repeats in the FMR1 gene at the end of the long arms of the X chromosome. To identify the mutation, researchers culture cells in media deficient in folic acid, which causes the ends of the X chromosome to appear as though they are about to break off. Before molecular testing, this was the only way to see the mutation.

The FMR1 gene encodes the fragile X mental retardation protein (FMRP), which regulates gene expression and protein translation in the brain. FMRP is important for maintaining synaptic plasticity and the ability to make new neurons. Levels of FMRP associated with disease severity in patients with FXS.

CGG repeats < 55
55 ≤ CGG repeats ≤ 200
Full Mutation
CGG repeats > 200

FXTAS, FXPOI, Neuropyschiatric problems

Premutation: The FMR1 gene has 55 to 200 CGG repeats that are not methylated. Premutations with fewer than 120 CGG repeats typically lead to normal FMRP levels; more repeats lead to lowered FMRP production, though this doesn’t necessarily translate to more severe disease. Premutations of any number of repeats can result in higher FMR1 mRNA levels, which can cause problems for the individual even if FMRP protein levels are normal. Collectively these problems are referred to as fragile X–associated disorders; they include fragile X–associated tremor/ataxia syndrome (FXTAS), primary ovarian insufficiency (FXPOI), and neuropsychiatric problems.

Full mutation: The FMR1 gene has more than 200 repeats of the nucleotides CGG that are heavily methylated. Males completely lack FMRP, while females typically have some protein produced from the FMR1 gene on the healthy X chromosome. Individuals carrying a copy of the full mutation present with FXS.

See full infographic: WEB | PDF

These new outcome measures are now being employed in targeted treatment trials for FXS. In 2013, Andrea Schneider at the MIND Institute, in collaboration with me and other colleagues, used ERPs to determine that auditory habituation to repetitive sounds improved in children with FXS who received minocycline, which lowers MMP-9, compared with those receiving a placebo.6 Patients on minocycline also had lower levels of an MMP-9 protein biomarker that is elevated in FXS.7  And earlier this year, the MIND’s David Hessl and colleagues demonstrated that an mGluR5 antagonist that had shown promise in mouse models but failed to beat the placebo in controlled trials that used multiple behavioral questionnaires, did in fact improve FXS patients’ ability to look into the eyes of a person’s photograph as measured by an infrared gaze-tracker.8 

Armed with more-objective measures, researchers are hoping to improve the track record of FXS clinical trials. The US pharmaceutical company Ovid, for example, is currently trialing a GABA agonist at multiple centers in the US for FXS patients ages 13 to 22 years. The outcome measures include a specific quantitation of behaviors typical for FXS, electronic measures such as eye tracking, and molecular biomarkers. 

Meanwhile, Pennsylvania-based pharma company Zynerba has been testing its transdermal ointment of pure cannabidiol (CBD), a nonpsychotropic component of marijuana, in FXS patients at multiple sites in the US, Australia, and New Zealand, using eye tracking and molecular biomarkers alongside questionnaires measuring patients’ anxiety. Preliminary studies demonstrated improvement in several behaviors on standardized questionnaires in children with FXS when the ointment was administered topically on the shoulder over a 12-week period. This positive open-label study where all patients received the medication, not placebo, is still ongoing in Australia and has led to a controlled trial of this CBD ointment, rubbed twice a day on the shoulder, at multiple clinics in the US and Australia for children ages 3 to 18. 

This follow-up study points to another trend in FXS research: the treatment of young FXS patients. As children, patients’ brains have more time, and may be better equipped, to build synaptic connections. At the MIND Institute, for example, my colleagues and I tested long-term low dosing of a selective serotonin reuptake inhibitor (SSRI) called sertraline in two- to six-year-old children with FXS. Serotonin levels are low in the brains of young children with autism spectrum disorder, and an SSRI functions to increase serotonin at the synapse. In 2016, we published results demonstrating significant benefits in fine motor, visual reception, and expressive language skills,9,10 leading doctors to begin using the drug to treat FXS. Now, the FX-LEARN study—currently enrolling patients at several US centers, including the MIND Institute—will test the use of an mGluR5 antagonist in combination with an educational intervention in three- to six-year-old children with FXS, using ERPs, eye tracking studies, and other quantitative outcome measures.


In males: Full mutations revert back to premutations during development of X chromosome–carrying sperm. Thus, men with the full mutation or the premutation pass the premutation on to their daughters. Because it is X-linked, they cannot pass it on to their sons.

In females: Premutations of 100 CGG repeats or more convert to the full mutation during egg development or in the embryo. Thus, women with the full mutation or a premutation with 100 repeats or more on one of their X chromosomes pass on the full mutation to their children approximately 50 percent of the time. Women with a premutation of fewer than 100 repeats will pass on a full mutation less often.

See full infographic: WEB | PDF

Another treatment that we’re beginning to test in children with FXS is metformin. In FXS, metformin has been shown to lower activity in the mTOR pathway, which is overactive in FXS patients. It is inexpensive and available in its generic form. In 2017, after the drug treated the behavioral, cognitive, and neuroanatomical features of FXS in FMR1 knockout mice and in a Drosophila model of FXS,11,12 my colleagues at the MIND Institute and I tested metformin in obese patients with FXS. We found that after just a few months, not only did it reduce weight gain, families said that their children could communicate better and carry on a conversation when they previously could not.13 This prompted us to launch a controlled trial with patients 6 to 25 years old with multiple electrophysiological outcome measures, including eye tracking and ERPs at the MIND and at two sites in Canada. The trial is ongoing, but already some doctors are beginning to prescribe metformin for some FXS patients.

A new way ahead

For the foreseeable future, treating FXS will be a matter of managing symptoms and working through therapy to support healthy development. There are a few early-stage trials for FXS treatments, and one, the FX-LEARN trial, involves targeted therapies that could reverse some of the pathways that are dysfunctional due to the absence or deficiency of FMRP. However, because so many pathways are problematic in FXS, it is likely that more than one targeted treatment will need to be utilized in the treatment of FXS. Thus, a cure for the disease remains a long way off.

Continued clinical work has improved the educational and behavioral interventions that can help young children with FXS. One example developed by Len Abbeduto and his team at the MIND Institute, is teaching parents to work directly with their children on language development. We are currently using the so-called parent-implemented language intervention (PILI) in our FX-LEARN study, and we have demonstrated efficacy with PILI alone to improve language skills in kids with FXS.14 

We are making progress, but we have a long road to travel and will need to involve multimodal interventions for the best effect. As we improve our treatments for FXS, we want to apply these treatments to those in Ricaurte, Colombia, and elsewhere. We have returned to Colombia and found other centers where fragile X is common, so there is much work to be done regarding further testing and treatment endeavors. We must use the tools we have today while pursuing the knowledge that will yield the cure. 

Randi Hagerman is a Distinguished Professor of Pediatrics at the MIND Institute and Department of Pediatrics, University of California, Davis, Medical Center, Sacramento.

Consequences of the fragile X premutation

Patients with the fragile X premutation have normal to low levels of the fragile X mental retardation protein (FMRP) and are usually unaffected intellectually. Counterintuitively, however, they actually exhibit increased FMR1 mRNA expression, and the excess transcripts often lead to health problems, from neuropsychiatric disorders to reproductive issues.

The RNA itself can be toxic to neurons, leading to dysregulation of calcium levels and mitochondrial dysfunction that cause the cells to die more easily in culture. Moreover, elevated levels of FMR1 mRNA can sequester proteins that are important for neuronal function in the nuclei of neurons and astrocytes throughout the brain and peripheral nervous system. This sequestration dysregulates the levels of certain miRNAs and somehow leads to increases in calcium ions in neurons, upping the likelihood that the cells will fire an action potential. These changes somehow cause mitochondrial dysfunction. Lastly, there is evidence that translation begins at an aberrant start site in the FMR1 mRNA, leading to the production of a protein called FMRPolyG that is toxic to the neuron.

How these issues relate to one another and affect patient outcomes is unclear. People with the premutation suffer from high rates of anxiety and depression, along with other neuropsychiatric conditions such as attention deficit/hyperactivity disorder, chronic fatigue syndrome, and migraines. The label of fragile X–associated neuropsychiatric disorders (FXAND) has been given to these conditions.15 The fragile X premutation is also the most common genetic cause of early menopause, before age 40, and about 20 percent of female carriers suffer from this condition, called fragile X–associated primary ovarian insufficiency (FXPOI). 

The most severe problem associated with the premutation is fragile X–associated tremor/ataxia syndrome (FXTAS), a neurodegenerative disorder that develops in more than 40 percent of males and 16 percent of females carrying the premutation.16 FXTAS typically presents in patients’ 60s as tremors followed by balance problems. In men, and very rarely in women, it can also be associated with dementia as it progresses. Patients with FXTAS show areas on an MRI scan that look extra-white in several brain regions, indicating damage to the white matter. When a sample of the brain is viewed under the microscope, the white matter looks motheaten. In addition, these patients develop spherical structures within the nucleus, called inclusions, of excess FMR1 mRNA and the proteins that it sequesters. These inclusions are a pathological marker of FXTAS. 

In the Ricaurte, Colombia,  FXS hotspot, we see more, and more-severely affected, premutation carriers than in other regions of the world. The more severe involvement may be related to exposure to environmental toxins, perhaps the pesticides used in agriculture around the village. Premutation carriers tend to be more sensitive to environmental toxins because the health of their neurons is already threatened by RNA toxicity. In addition, there may be effects of poverty or drug use. Research now aims to tease apart the genetic and environmental factors that play a role in outcomes for both premutation carriers and those with full-blown FXS.

MODIFIED FROM © istock.com, photo5963


  1. W. Saldarriaga et al., “Genetic cluster of fragile X syndrome in a Colombian district,” J Hum Genet, 63:509–16, 2018.
  2. E. Berry-Kravis, “Mechanism-based treatments in neurodevelopmental disorders: Fragile X syndrome,” Pediatr Neurol, 50:297–302, 2014.
  3. E.M. Berry-Kravis et al., “Drug development for neurodevelopmental disorders: Lessons learned from fragile X syndrome,”  Nat Rev Drug Discov, 17:280–99, 2018.
  4. E. Berry-Kravis et al., “Mavoglurant in fragile X syndrome: Results of two randomized, double-blind, placebo-controlled trials,” Sci Transl Med, 8:321ra5, 2016.
  5. E.A. Youssef et al., “Effect of the mGluR5-NAM basimglurant on behavior in adolescents and adults with fragile X syndrome in a randomized, double-blind, placebo-controlled trial: FragXis phase 2 results,” Neuropsychopharmacology, 43:503–12, 2018.
  6. A.Schneider et al., “Electrocortical changes associated with minocycline treatment in fragile X syndrome,” J Psychopharmacol, 27:956–63, 2013. 
  7. M. Dziembowska et al., “High MMP-9 activity levels in fragile X syndrome are lowered by minocycline,” Am J Med Genet A, 161:1897–903, 2013.
  8. D. Hessl et al., “Effects of mavoglurant on visual attention and pupil reactivity while viewing photographs of faces in fragile X syndrome,” PLOS ONE, 14:e0209984, 2019. 
  9. L. Greiss Hess et al., “A randomized, double-blind, placebo-controlled trial of low-dose sertraline in young children with fragile X syndrome,” J Dev Behav Pediatr, 37:619–28, 2016.
  10. K.H. Yoo et al., “Low-dose sertraline improves receptive language in children with fragile X syndrome when eye tracking methodology is used to measure treatment outcome,” J Psychol Clin Psychiatry, 7:00465, 2017.
  11. R.E. Monyak et al., “Insulin signaling misregulation underlies circadian and cognitive deficits in a Drosophila fragile X model,” Mol Psychiatry, 22:1140–48, 2017.
  12. I. Gantois et al., “Metformin ameliorates core deficits in a mouse model of fragile X syndrome,” Nat Med, 23:674–77, 2017.
  13. A.B.C. Dy et al., “Metformin as targeted treatment in fragile X syndrome,” Clin Genet, 93:216–22, 2018.
  14. S. Nelson et al., “Inferential language use by school-aged boys with fragile X syndrome: Effects of a parent-implemented spoken language intervention,” J Commun Disord, 72:64–76, 2018.
  15. R.J. Hagerman et al., “Fragile X-associated neuropsychiatric disorders (FXAND),” Front Psychiatry, 9:564, 2018.
  16. S. Jacquemont et al., “Penetrance of the fragile X-associated tremor/ataxia syndrome in a premutation carrier population,” JAMA, 291:460–69, 2004.

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!
Already a member?