December 18, 2013. Over 20 years ago, several genetics groups simultaneously linked a region of chromosome 22, containing about 45 genes, to schizophrenia. This evidence was bolstered by the fact that people with deletions of the region, who often have a developmental disorder called velocardiofacial syndrome (VCFS), are more likely to develop schizophrenia (Pulver et al., 1994), and the finding that several patients with schizophrenia had 22q11.2 deletions (Karayiorgou et al., 1995).
Since then, researchers have been working out how this rare deletion elevates risk for schizophrenia. “Even as we find other deletions or duplications elsewhere in the genome associated with schizophrenia, 22q11.2 deletions remain the standard bearer,” Anne Bassett of the University of Toronto, Canada, told SRF. “It’s the only proven molecular genetic subtype of schizophrenia.”
In this research roundup, Schizophrenia Research Forum reviews some of the latest findings in 22q11.2 deletion syndrome research, including insights from across the lifespan in people with the deletion, as well as from the affected brain circuits in mouse models.
A big clue
People missing all or much of a 3 Mb stretch on one strand of DNA in the region of 22q11.2 develop a variable syndrome, including heart defects, facial abnormalities, and cognitive impairments, which sometimes convene to create the classic VCFS. But up to one-quarter also go on to develop schizophrenia, perhaps accounting for up to 1 percent of all schizophrenia cases.
Bassett has followed patients with the deletion for over 20 years, with an interest in understanding how the full spectrum of phenotypes develops. As a rule, people with 22q11.2 deletion syndrome (22q11.2DS) vary in terms of which features they have, and their severity. But the schizophrenia observed in some of these patients is the real thing, Bassett says: “They have negative symptoms, they have positive symptoms, they have disorganized thought—the whole DSM symptom list is in these people.”
Their intelligence quotients (IQs) are lower on average than those measured for people with schizophrenia, however, but most people with 22q11.2DS do not meet the criterion for intellectual disability. This means they constitute a naturally occurring human model for common variety schizophrenia, Bassett argues. More recently, a connection to schizophrenia has been bolstered by the report that duplications of the 22q11.2 region protect from schizophrenia (see SRF related news story).
“It is so gratifying that people have finally recognized how much the rare can inform the common,” Bassett says.
Still, sifting through the different genes in the region to find the key instigators of schizophrenia remains a formidable task. Among the early, obvious candidates were the genes for catechol-O-methyltransferase (COMT), involved in dopamine metabolism, and proline dehydrogenase (PRODH), which helps regulate glutamate levels (see SRF related news story). More recently, mouse models carrying a similar deletion have highlighted a possible role for microRNAs, which regulate gene expression. Within the 22q11.2 region lies the gene DGCR8, which prepares microRNAs for action. The loss of DGCR8 (see SRF related news story), and more recently, of a specific microRNA-185 in the region (see SRF related news story), have been linked to schizophrenia-like signs in mice.
These microRNAs might normally help the brain compensate for mutations elsewhere in the genome, either in the intact 22q11.2 region on the counterpart chromosome or beyond. When microRNA buffering capacity is reduced through a 22q11.2 deletion, then, this might unmask the effects of these other mutations and explain the variable effects of 22q11.2 deletions (Brzustowicz and Bassett, 2012). An international consortium has just been funded by the National Institute of Mental Health to sequence the entire genomes of people with 22q11.2 deletions to find genetic factors that influence whether someone develops schizophrenia, Bassett said.
Across the lifespan
Most 22q11.2 deletions begin as mistakes during meiosis, when chromosomes misalign prior to exchanging bits of DNA. This means egg or sperm cells carry the deletion, whereas the parents themselves do not. A majority of 22q11.2 deletions stem from mothers, according to a study published in the American Journal of Human Genetics in March. Bernice Morrow of Albert Einstein College of Medicine, Bronx, New York, teamed up with other researchers around the world to survey a large enough sample of people with 22q11.2 deletions and their parents. First author Maria Delio and colleagues reported that 57 percent of subjects with 22q11.2 deletions acquired them from their mother’s egg, whereas 43 percent were from sperm, which suggests that the 22q11.2 region is more vulnerable to meiotic mishaps in females than in males.
Psychiatric problems emerge early in many children with 22q11.2 deletions, but these do not result from their cognitive impairments, according to a study published in the British Journal of Psychiatry from Marianne van den Bree of Cardiff University, Wales, United Kingdom, and colleagues. Assessing 80 children with 22q11.2 deletions, first author Maria Niarchou reported that 54 percent qualified for a psychiatric disorder, including anxiety disorders and attention-deficit hyperactivity disorder, compared to only 10 percent of their siblings. Though 22q11.2DS subjects also had lower IQ scores than their siblings, their IQ did not explain the risk for these psychiatric disorders. This bolsters the idea that childhood psychiatric disorders are directly related to the deletion, rather than being non-specific fallout of cognitive impairments.
As if the risk for psychiatric disorders in childhood and early adulthood wasn’t enough, a study published in JAMA Neurology in November from Bassett and colleagues reported an increased risk for Parkinson's disease (PD) later in life. First author Nancy Butcher and colleagues found that adults with 22q11.2 deletions more frequently developed PD than expected, with 70 times the number seen in the general population. Furthermore, they developed PD before 50 years of age, which points to a new, rare form of early-onset PD. Postmortem brain samples confirmed that this was true PD, with loss of dopamine-containing substantia nigra neurons and other signs of neurodegeneration. Why 22q11.2 deletions would increase risk for a disorder associated with too much dopamine (schizophrenia) and one associated with too little dopamine (PD) is “a sweet mystery,” Bassett says, and suggests that the conception of schizophrenia as primarily a disorder of dopamine is incomplete.
Early signs of psychosis
Following the development of children with 22q11.2 deletions could divulge early predictors of psychosis, and three studies report comparisons of features in the same people across two time points. One study from Stephan Eliez and colleagues at the University of Geneva School of Medicine, Switzerland, points to anxiety: As reported in the Journal of the American Academy of Child and Adolescent Psychiatry, first author Doron Gothelf and colleagues found that, of 10 people with 22q11.2DS who developed psychosis, nine had an anxiety disorder as children. Another study from the same group, published in European Child and Adolescent Psychiatry, found that low-grade psychotic symptoms preceded the onset of full-blown psychosis in people with 22q11.2 deletions, similar to the prodrome described for schizophrenia in general. First author Maude Schneider and colleagues did not find concomitant changes in cognition: Though those who developed psychosis had lower IQs to begin with, the slope of the change in IQ over time did not differ from those who did not develop psychosis. They suggest low IQ as a marker for increased risk for psychosis.
A third study, published in Research in Developmental Disabilities, also found that adolescents with 22q11.2 deletions displayed evidence of the schizophrenia prodrome—mild psychotic symptoms not reaching a diagnostic level. Led by Vandana Shashi at Duke University, Durham, North Carolina, the study found that these subjects did not show the expected gains in attention and general function as they matured. First author Stephen Hooper and colleagues also noted that lower IQ at baseline was associated with the development of schizophrenia 3.5 years later.
As for genetic factors that might tilt a person with a 22q11.2 deletion toward psychosis, the COMT gene has been a go-to candidate. Lying within the 22q11.2 region, COMT encodes an enzyme that degrades dopamine and other monoamines. In people with a 22q11.2 deletion, any consequences of any variant in the remaining, intact COMT may be unmasked. Though the well-known Val158Met variant has not been consistently linked to schizophrenia in subjects with 22q11.2DS, a study in Biological Psychiatry suggests other COMT variants may influence COMT function and the development of psychosis. Led by Daniel Weinberger of the Lieber Institute in Baltimore, Maryland, the study analyzed blood cells from 53 people with 22q11.2 deletions. First authors Doron Gothelf and Amanda Law reported that combinations of single nucleotide polymorphisms (SNPs) could modulate the effect of Val158Met on COMT expression and activity. Moreover, one SNP in an untranslated region of COMT was associated on its own with low COMT activity and occurred more frequently in those who developed psychosis than in those who did not.
In the brain
As a group, people with 22q11.2 deletions display deficits in sensorimotor gating and mismatch negativity (MMN), both neurophysiological signs associated with schizophrenia. According to a study in the Journal of Psychiatric Research led by Doron Gothelf at Tel Aviv University, Israel, these measures are modulated by the Val158Met variant in COMT, as well as by SNPs in PRODH, a modulator of glutamatergic and dopaminergic neural transmission. First author Omer Zarchi also found that the degree of the deficit in MMN correlated with schizophrenia-like symptoms and measures of executive function; however, these measures did not differ between psychotic and non-psychotic people with 22q11.2 deletions, suggesting that MMN deficits flag general risk for psychosis.
Two neuroimaging studies led by Stephan Eliez of the University of Geneva, Switzerland, also revealed schizophrenia-like patterns of connectivity in people with 22q11.2 deletions. As reported in PLoS One, diffusion tensor imaging (DTI) used to characterize the trajectories of white matter bundles throughout the brain found sparser connections in 30 people with 22q11.2 deletions compared to 30 controls. First author Marie-Christine Ottet and colleagues reported that this reduction was pronounced in connections linking the frontal and temporal cortices, as well as those within and between limbic areas, and suggests that the state of frontotemporal connections may serve as a biomarker for vulnerability to psychosis. The second study, also from Ottet, Eliez, and colleagues, explored the network of brain connections in 46 people with 22q11.2 deletions who also had hallucinations, and so may be at even higher risk of developing schizophrenia than 22q11.2DS subjects without hallucinations. In general, connections in the brain are organized like airplane flight routes, with many connections stopping through key hub regions, and in schizophrenia, the brain has been reported to contain fewer hubs with fewer connections (see SRF related news story). Writing in Frontiers of Human Neuroscience, first author Ottet reports a similarly disrupted network in the 22q11.2DS population, with fewer connections between hubs, suggesting that this fragmented network reflects a predisposition for schizophrenia.
Researchers have modeled the 22q11.2 deletion in mice by knocking out the equivalent region, which resides on mouse chromosome 16. Several models exist that show a variety of schizophrenia-related phenotypes ranging from disruptions of brain development to deficits in sensory gating. The advent of small animal brain imaging will supplement the comparisons between mice and humans (see SRF related news story). In a study published in Molecular Psychiatry, Joseph Gogos and Maria Karayiorgou teamed up with researchers in Toronto to scan their mouse model of 22q11.2 deletion syndrome, called Df(16)A+/-, which lacks 27 genes. Using a 7 Tesla scanner, first authors Jacob Ellegood and Sander Markx report 13 different regions that had significantly smaller or larger volumes than wild-type mice, and several of these overlapped with changes observed in humans with 22q11.2 deletions, including the frontal lobe, the striatum, the third ventricle, and the cerebellum.
Another study of this mouse model suggests that the brain’s ability to change may be compromised by multiple factors within 22q11.2 deletions. Published in the Journal of Neuroscience in September, the study, led again by Gogos and Karayiorgou, evaluated synapses in the medial prefrontal cortex (mPFC), a region key for cognition. First author Karine Fénelon and colleagues found impaired short- and long-term plasticity in mPFC synapses, meaning that the synapses did not adequately scale their responses up or down according to input. The impaired short-term plasticity, as well as working memory, could be blamed on the loss of DGCR8, as similar deficits are seen in mice lacking that single gene; however, the long-term plasticity deficits and accompanying changes to dendritic spine shapes suggests other genetic culprits in the region, which may involve miRNA processing.
Finally, however, a study led by Jennifer Linden at University College London, United Kingdom, sounds a note of caution in evaluating mouse models of 22q11.2 deletions or other large genetic hits. Published in PLoS One, the study reports that the Df1/+ mouse model, which lacks 18 genes in the human 22q11.2 region, showed hearing loss arising from chronic ear infections. First author Jennifer Fuchs and colleagues found that the hearing loss often occurred in one—not both—ears and mimics a susceptibility to ear infections found in humans with 22q11.2 deletions. The results, however, suggest that previous reports of impaired sensorimotor gating in 22q11.2 deletion mouse models may stem from deficits in hearing rather than problems with the interface between sensory processing and motor responses.—Michele Solis.
Delio M, Guo T, McDonald-McGinn DM, Zackai E, Herman S, Kaminetzky M, Higgins AM, Coleman K, Chow C, Jalbrzikowski M, Bearden CE, Bailey A, Vangkilde A, Olsen L, Olesen C, Skovby F, Werge TM, Templin L, Busa T, Philip N, Swillen A, Vermeesch JR, Devriendt K, Schneider M, Dahoun S, Eliez S, Schoch K, Hooper SR, Shashi V, Samanich J, Marion R, van Amelsvoort T, Boot E, Klaassen P, Duijff SN, Vorstman J, Yuen T, Silversides C, Chow E, Bassett A, Frisch A, Weizman A, Gothelf D, Niarchou M, van den Bree M, Owen MJ, Suñer DH, Andreo JR, Armando M, Vicari S, Digilio MC, Auton A, Kates WR, Wang T, Shprintzen RJ, Emanuel BS, Morrow BE. Enhanced maternal origin of the 22q11.2 deletion in velocardiofacial and DiGeorge syndromes. Am J Hum Genet. 2013 Mar 7;92(3):439-47. Abstract
Niarchou M, Zammit S, van Goozen SH, Thapar A, Tierling HM, Owen MJ, van den Bree MB. Psychopathology and cognition in children with 22q11.2 deletion syndrome. Br J Psychiatry. 2013 Oct 10. Abstract
Butcher NJ, Kiehl TR, Hazrati LN, Chow EW, Rogaeva E, Lang AE, Bassett AS.
Association between early-onset Parkinson disease and 22q11.2 deletion syndrome: identification of a novel genetic form of Parkinson disease and its clinical implications. JAMA Neurol. 2013 Nov 1;70(11):1359-66. Abstract
Gothelf D, Schneider M, Green T, Debbané M, Frisch A, Glaser B, Zilkha H, Schaer M, Weizman A, Eliez S. Risk factors and the evolution of psychosis in 22q11.2 deletion syndrome: a longitudinal 2-site study. J Am Acad Child Adolesc Psychiatry. 2013 Nov;52(11):1192-1203. Abstract
Schneider M, Schaer M, Mutlu AK, Menghetti S, Glaser B, Debbané M, Eliez S.
Clinical and cognitive risk factors for psychotic symptoms in 22q11.2 deletion syndrome: a transversal and longitudinal approach. Eur Child Adolesc Psychiatry. 2013 Sep 3. Abstract
Hooper SR, Curtiss K, Schoch K, Keshavan MS, Allen A, Shashi V. A longitudinal examination of the psychoeducational, neurocognitive, and psychiatric functioning in children with 22q11.2 deletion syndrome. Res Dev Disabil. 2013 May;34(5):1758-69. Abstract
Gothelf D, Law AJ, Frisch A, Chen J, Zarchi O, Michaelovsky E, Ren-Patterson R, Lipska BK, Carmel M, Kolachana B, Weizman A, Weinberger DR. Biological Effects of COMT Haplotypes and Psychosis Risk in 22q11.2 Deletion Syndrome. Biol Psychiatry. 2013 Aug 27. Abstract
Zarchi O, Carmel M, Avni C, Attias J, Frisch A, Michaelovsky E, Patya M, Green T, Weinberger R, Weizman A, Gothelf D. Schizophrenia-like neurophysiological abnormalities in 22q11.2 deletion syndrome and their association to COMT and PRODH genotypes. J Psychiatr Res. 2013 Nov;47(11):1623-9. Abstract
Ottet MC, Schaer M, Cammoun L, Schneider M, Debbané M, Thiran JP, Eliez S. Reduced fronto-temporal and limbic connectivity in the 22q11.2 deletion syndrome: vulnerability markers for developing schizophrenia? PLoS One. 2013;8(3):e58429. Abstract
Ottet MC, Schaer M, Debbané M, Cammoun L, Thiran JP, Eliez S. Graph theory reveals dysconnected hubs in 22q11DS and altered nodal efficiency in patients with hallucinations. Front Hum Neurosci. 2013 Sep 5;7:402. Abstract
Ellegood J, Markx S, Lerch JP, Steadman PE, Genç C, Provenzano F, Kushner SA, Henkelman RM, Karayiorgou M, Gogos JA. Neuroanatomical phenotypes in a mouse model of the 22q11.2 microdeletion. Mol Psychiatry. 2013 Sep 3. Abstract
Fuchs JC, Zinnamon FA, Taylor RR, Ivins S, Scambler PJ, Forge A, Tucker AS, Linden JF. Hearing loss in a mouse model of 22q11.2 deletion syndrome. PLoS One. 2013 Nov 14;8(11):e80104. Abstract