Introduction
The 22q11.2 deletion syndrome (22q11DS) is a relatively common genetic disorder, with an estimated prevalence of one in 2000–4000 births. It is characterized by a deletion on locus 22q11.2, a copy number variant that contributes significantly to the risk for psychotic disorders (Murphy et al., Reference Murphy, Jones and Owen1999; Schneider et al., Reference Schneider, Debbané, Bassett, Chow, Fung, van den Bree, Owen, Murphy, Niarchou, Kates, Antshel, Fremont, McDonald-McGinn, Gur, Zackai, Vorstman, Duijff, Klaassen, Swillen, Gothelf, Green, Weizman, Van Amelsvoort, Evers, Boot, Shashi, Hooper, Bearden, Jalbrzikowski, Armando, Vicari, Murphy, Ousley, Campbell, Simon and Eliez2014). 22q11DS has a heterogeneous phenotype including cardiac anomalies (Guo et al., Reference Guo, Repetto, McDonald McGinn, Chung, Nomaru, Campbell, Blonska, Bassett, Chow, Mlynarski, Swillen, Vermeesch, Devriendt, Gothelf, Carmel, Michaelovsky, Schneider, Eliez, Antonarakis, Coleman, Tomita-mitchell, Mitchell, Digilio, Dallapiccola, Marino, Philip, Busa, Kushan-Wells, Bearden, Piotrowicz, Hawuła, Roberts, Tassone, Simon, van Duin, van Amelsvoort, Kates, Zackai, Johnston, Cutler, Agopian, Goldmuntz, Mitchell, Wang, Emanuel, Morrow, Mcginn, Chung, Nomaru, Campbell, Blonska, Bassett, Chow, Mlynarski, Swillen, Vermeesch, Devriendt, Gothelf, Carmel, Michaelovsky, Schneider, Eliez, Antonarakis, Coleman, Tomita-mitchell, Mitchell, Digilio, Dallapiccola, Marino, Philip, Busa, Kushan-Wells, Bearden, Piotrowicz, Hawuła, Roberts, Tassone, Simon, Van Duin, Van Amelsvoort, Kates, Zackai, Johnston, Cutler, Agopian, Goldmuntz, Mitchell, Wang, Emanuel and Morrow2017) and several psychiatric problems (Schneider et al., Reference Schneider, Debbané, Bassett, Chow, Fung, van den Bree, Owen, Murphy, Niarchou, Kates, Antshel, Fremont, McDonald-McGinn, Gur, Zackai, Vorstman, Duijff, Klaassen, Swillen, Gothelf, Green, Weizman, Van Amelsvoort, Evers, Boot, Shashi, Hooper, Bearden, Jalbrzikowski, Armando, Vicari, Murphy, Ousley, Campbell, Simon and Eliez2014). Cognitive impairments (Oskarsdóttir et al., Reference Oskarsdóttir, Vujic and Fasth2004; Bassett et al., Reference Bassett, Chow, Husted, Weksberg, Caluseriu, Webb and Gatzoulis2005; Biswas and Furniss, Reference Biswas and Furniss2016; Norkett et al., Reference Norkett, Lincoln, Gonzalez-Heydrich and D'Angelo2017) are part of the core symptoms of the syndrome. Additionally, approximately one in four individuals with 22q11DS develop a psychotic disorder, making 22q11DS one of the greatest known risk factors for developing psychosis (Bassett, Reference Bassett2011).Therefore, it is suggested that 22q11DS represents a valuable model for the study of neurobiological factors underlying both cognitive impairments (Oskarsdóttir et al., Reference Oskarsdóttir, Vujic and Fasth2004; Bassett et al., Reference Bassett, Chow, Husted, Weksberg, Caluseriu, Webb and Gatzoulis2005; Biswas and Furniss, Reference Biswas and Furniss2016; Norkett et al., Reference Norkett, Lincoln, Gonzalez-Heydrich and D'Angelo2017) and psychotic disorders (Gur et al., Reference Gur, Bassett, McDonald-McGinn, Bearden, Chow, Emanuel, Owen, Swillen, Van den Bree, Vermeesch, Vorstman, Warren, Lehner and Morrow2017). Although the biological factors underlying psychotic disorders and (their) cognitive symptoms are still poorly understood, there is evidence suggesting for aberrant dopamine (DA) levels in several brain regions (Howes et al., Reference Howes, Kambeitz, Kim, Stahl, Slifstein, Abi-Dargham and Kapur2012; Fusar-Poli and Meyer-Lindenberg, Reference Fusar-Poli and Meyer-Lindenberg2013), including the prefrontal cortex (PFC) (Slifstein et al., Reference Slifstein, van de Giessen, Van Snellenberg, Thompson, Narendran, Gil, Hackett, Girgis, Ojeil, Moore, D'Souza, Malison, Huang, Lim, Nabulsi, Carson, Lieberman and Abi-Dargham2015).
Alterations in DA neurotransmission are also suggested to underlie some of the psychiatric problems typically seen in 22q11DS (Boot et al., Reference Boot, Booij, Zinkstok, Abeling, de Haan, Baas, Linszen and van Amelsvoort2008, Reference Boot, Booij, Abeling, Meijer, da Silva Alves, Zinkstok, Baas, Linszen and van Amelsvoort2011a; Evers et al., Reference Evers, Curfs, Bakker, Boot, da Silva Alves, Abeling, Bierau, Drukker and van Amelsvoort2014; de Koning et al., Reference de Koning, van Duin, Boot, Bloemen, Bakker, Abel and van Amelsvoort2015). These alterations are possibly due to haplo-insufficiency (reduced dosage of the gene due to hemizygosity) of the catechol-O-methyltransferase (COMT) gene, located on the deleted region and coding for the enzyme that catabolizes extracellular DA (Chen et al., Reference Chen, Lipska, Halim, Ma, Matsumoto, Melhem, Kolachana, Hyde, Herman, Apud, Egan, Kleinman and Weinberger2004). Especially frontal DA is thought to be affected by COMT haploinsufficiency (Yavich et al., Reference Yavich, Forsberg, Karayiorgou, Gogos and Männistö2007) in 22q11DS. This could be explained by the relatively low density of the DA transporter in the PFC (Sesack et al., Reference Sesack, Hawrylak, Matus, Guido and Levey1998), resulting in a DA dependency of COMT enzyme activity for clearance (Tunbridge et al., Reference Tunbridge, Lane and Harrison2007). It has been indicated that 50% of the prefrontal DA clearance results from COMT activity (Yavich et al., Reference Yavich, Forsberg, Karayiorgou, Gogos and Männistö2007). Since patients with 22q11DS have only one copy of the COMT gene, which is associated with reduced COMT gene expression (van Beveren et al., Reference van Beveren, Krab, Swagemakers, Buitendijk, Buitendijk, Boot, van der Spek, Elgersma and van Amelsvoort2012) and enzyme concentrations (Gothelf et al., Reference Gothelf, Law, Frisch, Chen, Zarchi, Michaelovsky, Ren-Patterson, Lipska, Carmel, Kolachana, Weizman and Weinberger2014), they may consequently be chronically exposed to abnormally high DA levels (Boot et al., Reference Boot, Booij, Zinkstok, Abeling, de Haan, Baas, Linszen and van Amelsvoort2008), particularly in the PFC. We previously showed that the COMT functional polymorphism Val158Met indeed affects DA function in 22q11DS (Boot et al., Reference Boot, Booij, Zinkstok, Baas, Swillen, Owen, Murphy, Murphy, Linszen and Van Amelsvoort2011b). 22q11DS Val-hemizygotes have higher post-synaptic striatal DA D2/3 non-displaceable receptor binding potential (D2/3R BPND) compared to carriers with the relatively unstable and less active COMT Met-allele (Boot et al., Reference Boot, Booij, Zinkstok, Baas, Swillen, Owen, Murphy, Murphy, Linszen and Van Amelsvoort2011b), further implicating altered DA neurotransmission.
The COMT Val/Met genotype has also been related to (dys)function of frontal brain regions in the psychosis continuum (Egan et al., Reference Egan, Goldberg, Kolachana, Callicott, Mazzanti, Straub, Goldman and Weinberger2001; Hernaus et al., Reference Hernaus, Collip, Lataster, Ceccarini, Kenis, Booij, Pruessner, van Laere, van Winkel, van Os and Myin-Germeys2013). Abnormalities in frontal brain DA have been hypothesized to especially underlie cognitive and negative symptoms of psychotic disorders (Howes and Kapur, Reference Howes and Kapur2009; Howes et al., Reference Howes, Kambeitz, Kim, Stahl, Slifstein, Abi-Dargham and Kapur2012), which may also be true for 22q11DS (Stoddard et al., Reference Stoddard, Niendam, Hendren, Carter and Simon2010; Schneider et al., Reference Schneider, Debbané, Bassett, Chow, Fung, van den Bree, Owen, Murphy, Niarchou, Kates, Antshel, Fremont, McDonald-McGinn, Gur, Zackai, Vorstman, Duijff, Klaassen, Swillen, Gothelf, Green, Weizman, Van Amelsvoort, Evers, Boot, Shashi, Hooper, Bearden, Jalbrzikowski, Armando, Vicari, Murphy, Ousley, Campbell, Simon and Eliez2014; Tang et al., Reference Tang, Yi, Calkins, Whinna, Kohler, Souders, McDonald-McGinn, Zackai, Emanuel, Gur and Gur2014). Frontal DA neurotransmission has also been related to (impairments in) different neuropsychological functional domains, including memory, motivation, attention, and concentration (Howes and Kapur, Reference Howes and Kapur2009; Jonas et al., Reference Jonas, Montojo and Bearden2014; Slifstein et al., Reference Slifstein, van de Giessen, Van Snellenberg, Thompson, Narendran, Gil, Hackett, Girgis, Ojeil, Moore, D'Souza, Malison, Huang, Lim, Nabulsi, Carson, Lieberman and Abi-Dargham2015). In addition, the COMT genotype is found to modulate cognitive functioning, relying on frontal DA neurotransmission, in psychotic disorder (Jonas et al., Reference Jonas, Montojo and Bearden2014; Slifstein et al., Reference Slifstein, van de Giessen, Van Snellenberg, Thompson, Narendran, Gil, Hackett, Girgis, Ojeil, Moore, D'Souza, Malison, Huang, Lim, Nabulsi, Carson, Lieberman and Abi-Dargham2015) and in 22q11DS (Gothelf et al., Reference Gothelf, Eliez, Thompson, Hinard, Penniman, Feinstein, Kwon, Jin, Jo, Antonarakis, Morris and Reiss2005; de Koning et al., Reference de Koning, Boot, Bloemen, van Duin, Abel, de Haan, Linszen and van Amelsvoort2012; Carmel et al., Reference Carmel, Zarchi, Michaelovsky, Frisch, Patya, Green, Gothelf and Weizman2014). Moreover the COMT genotype has been implicated in dopaminergic drug effects on cognitive functioning (Schacht, Reference Schacht2016).
In summary, there is evidence for abnormal frontal DA functioning in cognitive impairments, psychotic disorders, and implications for altered DA function in 22q11DS. More insight into the neurobiological factors associated with both psychotic disorder and cognitive deficits in 22q11DS can be gained, by investigating frontal DA function in 22q11DS using in vivo molecular imaging methods.
Neuroimaging techniques consistently showed both aberrant frontal brain anatomy and function as well as an effect of COMT Val/Met genotype in 22q11DS (van Amelsvoort et al., Reference van Amelsvoort, Daly, Robertson, Suckling, Ng, Critchley, Owen, Henry, Murphy and Murphy2001, Reference van Amelsvoort, Zinkstok, Figee, Daly, Morris, Owen, Murphy, De Haan, Linszen, Glaser and Murphy2008; Gothelf et al., Reference Gothelf, Eliez, Thompson, Hinard, Penniman, Feinstein, Kwon, Jin, Jo, Antonarakis, Morris and Reiss2005; Zinkstok and van Amelsvoort, Reference Zinkstok and van Amelsvoort2005; Kates et al., Reference Kates, Antshel, Abdulsabur, Colgan, Funke, Fremont, Higgins, Kucherlapati and Shprintzen2006; Howes et al., Reference Howes, Kambeitz, Kim, Stahl, Slifstein, Abi-Dargham and Kapur2012; Shashi et al., Reference Shashi, Veerapandiyan, Keshavan, Zapadka, Schoch, Kwapil, Hooper and Stanley2012; van Beveren et al., Reference van Beveren, Krab, Swagemakers, Buitendijk, Buitendijk, Boot, van der Spek, Elgersma and van Amelsvoort2012; Jonas et al., Reference Jonas, Montojo and Bearden2014).
In addition, molecular imaging techniques, including [11C]DTBZ- and [18F]fallypride positron emission tomography (PET) and [123I]IBZM single photon emission computed tomography (SPECT), have been used successfully in 22q11DS to investigate abnormalities in the striatal DA system (Boot et al., Reference Boot, Booij, Zinkstok, de Haan, Linszen, Baas and van Amelsvoort2010; Butcher et al., Reference Butcher, Marras, Pondal, Rusjan, Boot, Christopher, Repetto, Fritsch, Chow, Masellis, Strafella and Lang2017; van Duin et al., Reference van Duin, Kasanova, Hernaus, Ceccarini, Heinzel, Mottaghy, Mohammadkhani-Shali, Winz, Frank, Beck, Booij, Myin-Germeys and van Amelsvoort2018). However, no studies to date have investigated frontal DA signaling in patients with 22q11DS. This can be measured in vivo with PET, using high-affinity radioligands such as the highly selective DA D2/3 receptor (D2/3R) radioligand [18F]fallypride, successfully used to probe frontal DA functioning (Lataster et al., Reference Lataster, Collip, Ceccarini, Haas, Booij, Van Os, Pruessner, Van Laere and Myin-Germeys2011; Ceccarini et al., Reference Ceccarini, Vrieze, Koole, Muylle, Bormans, Claes and Van Laere2012; Hernaus et al., Reference Hernaus, Collip, Lataster, Ceccarini, Kenis, Booij, Pruessner, van Laere, van Winkel, van Os and Myin-Germeys2013; Nagano Saito et al., Reference Nagano-Saito, Dagher, Booij and Gravel2013).
The present study aimed to investigate, for the first time, frontal D2/3R BPND in 22q11DS using [18F]fallypride PET. Because of COMT haploinsufficiency in 22q11DS and previously described findings of SPECT and PET studies (Boot et al., Reference Boot, Booij, Zinkstok, de Haan, Linszen, Baas and van Amelsvoort2010, Reference Boot, Booij, Zinkstok, Baas, Swillen, Owen, Murphy, Murphy, Linszen and Van Amelsvoort2011b; Butcher et al., Reference Butcher, Marras, Pondal, Rusjan, Boot, Christopher, Repetto, Fritsch, Chow, Masellis, Strafella and Lang2017; van Duin et al., Reference van Duin, Kasanova, Hernaus, Ceccarini, Heinzel, Mottaghy, Mohammadkhani-Shali, Winz, Frank, Beck, Booij, Myin-Germeys and van Amelsvoort2018), we expected reduced D2/3R BPND in frontal brain regions compared to healthy controls (HCs), as a proxy marker of chronically increased extracellular frontal DA levels.
Materials and methods
Participants
Fourteen non-psychotic adult individuals (eight females and six males, mean age = 34.6 years, s.d. = 9.7 years) with 22q11DS and no family history of psychotic disorder were included. They were compared to a previously published (Kasanova et al., Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, Heinzel, Mottaghy and Myin-Germeys2017, Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, van Duin, Steinhart, Vaessen, Heinzel, Mottaghy and Myin-Germeys2018) sample of 18 HCs (12 females and six males, mean age = 38.1 years, s.d. = 15.6 years). Recruitment and exclusion criteria of HC have been described previously (Kasanova et al., Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, Heinzel, Mottaghy and Myin-Germeys2017, Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, van Duin, Steinhart, Vaessen, Heinzel, Mottaghy and Myin-Germeys2018).
All participants were capable of giving written informed consent and did so after receiving full information on the study. Participants were treated in accordance with the Declaration of Helsinki. The study was approved by the Medical Ethical Committee of Maastricht University (The Netherlands) and the RWTH Aachen University ethics committee of Universitäts Klinikum (Germany). The PET protocol was additionally approved by the national authority for radiation protection in humans in Germany (Bundesamt für Strahlenschutz, BfS). Participants received coupons with a total value of €100 for participating in the PET study.
Exclusion criteria for 22q11DS participants were: (1) lifetime history of psychosis as determined by the Mini-International Neuropsychiatric Interview (M.I.N.I.) (Sheehan et al., Reference Sheehan, Lecrubier, Sheehan, Amorim, Janavs, Weiller, Hergueta, Baker and Dunbar1998) and/or current or previous use of antipsychotic or stimulant medication, (2) contraindications for MRI and/or PET imaging, (3) pregnancy (verified on the day of the scan using a pregnancy test), (4) current drug use (verified on the day of the scan using a urine drug test).
Two HC participants were cigarette smokers. Given the well-known association between smoking (status) and DA function (Mansvelder and McGehee, Reference Mansvelder and McGehee2000), they were asked to refrain from nicotine use on the day of the imaging session. One HC was excluded due to positioning difficulties during scanning. Another HC participant was excluded based on non-compliance with the study procedures. Two 22q11DS participants used the selective serotonin reuptake inhibitors escitalopram (10 mg) or paroxetine (20 mg). Since this may influence DA functioning (Tanda et al., Reference Tanda, Carboni, Frau and Chiara1994; Damsa et al., Reference Damsa, Bumb and Bianchi-Demicheli2004) they were asked to refrain from taking their medication on the day of the imaging session. Other participants did not take any psychotropic medication. The final sample consisted of 16 HC and 14 22q11DS participants (Table 1).
Table 1. Demographics and binding potential (BPND) per region of interest (ROI)c
HC, healthy controls; IQ, intelligence quotient; PANSS, positive and negative symptom scale: total score rage min 30–max 210, positive and negative symptom score range min 7–max 49, general psychopathology score range min 16–max 112; PFC, prefrontal cortex; OFC, orbito frontal cortex; ACC, anterior cingulate cortex.
**p < 0.01 and survived Bonferroni correction for multiple testing a = t test, b = χ2 test, c = 2 participants with 22q11DS used selective serotonin reuptake inhibitors (SSRIs) escitalopram (10 mg) and paroxetine (20 mg).
Behavioral and physiological assessments
Full scale intelligence quotient (IQ) of the 22q11DS participants was determined using a shortened Dutch version of the Wechsler Adult Intelligence Scale – III (WAIS-III) (Wechsler, Reference Wechsler1997) and was assessed on the day of scanning or in a separate session before or after the PET session (mean = 52.8 days, s.d. = 49.8 days). The shortened WAIS-III consists of four subtests: arithmetic and information (verbal IQ) digit-symbol-coding and block patterns (performance IQ) (Wechsler, Reference Wechsler1997; Brooks and Weaver, Reference Brooks and Weaver2005). In the HC group, total IQ was estimated using the Dutch Adult Reading Test (DART) (Schmand et al., Reference Schmand, Bakker, Saan and Louman1991). Other assessments of the HC group were described previously (Kasanova et al., Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, Heinzel, Mottaghy and Myin-Germeys2017, Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, van Duin, Steinhart, Vaessen, Heinzel, Mottaghy and Myin-Germeys2018).
To assess the presence and severity of psychotic symptoms, the Positive and Negative Syndrome Scale (PANSS) (Kay et al., Reference Kay, Fiszbein and O1987) for psychotic disorders was used.
Image data collection
The [18F]fallypride PET data collection acquired for this research was part of a comprehensive PET acquisition protocol, previously carried out to investigate reinforcement learning task-induced striatal DA release (Kasanova et al., Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, Heinzel, Mottaghy and Myin-Germeys2017, Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, van Duin, Steinhart, Vaessen, Heinzel, Mottaghy and Myin-Germeys2018; van Duin et al., Reference van Duin, Kasanova, Hernaus, Ceccarini, Heinzel, Mottaghy, Mohammadkhani-Shali, Winz, Frank, Beck, Booij, Myin-Germeys and van Amelsvoort2018). For the current PET analyses, only the [18F]fallypride sensorimotor control and baseline conditions were considered, including the first 120 min of the scan protocol (Fig. 1). All details of the whole PET procedure and the structural MRI and PET data acquisition have been described previously (Kasanova et al., Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, Heinzel, Mottaghy and Myin-Germeys2017; van Duin et al., Reference van Duin, Kasanova, Hernaus, Ceccarini, Heinzel, Mottaghy, Mohammadkhani-Shali, Winz, Frank, Beck, Booij, Myin-Germeys and van Amelsvoort2018) and additional analyses including the control only condition (excluding the 25 min baseline scan) to confirm reliability of the used method can be found in the Supplementary Materials.
Fig. 1. PET acquisition protocol. The original PET acquisition protocol. In gray, the part of the PET acquisition protocol used for analyses in this study is highlighted. *TS = 68Ge/68Ga-transmission scan, timeline in minutes. PET control: Sensori-motor control condition: Participants conducted a sensori-motor control condition prior to the baseline and experimental condition (previously described in Kasanova et al., Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, Heinzel, Mottaghy and Myin-Germeys2017, Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, van Duin, Steinhart, Vaessen, Heinzel, Mottaghy and Myin-Germeys2018). This condition was designed to contain all features of the task of the experimental condition, without the main manipulation of the experimental condition; outcome-based associative learning. This control condition was presented on a 30-inch screen placed in the field of view of the participant. Similar to the experimental condition, images of a stimulus (photographs of actors) appeared on the screen and participants had to choose between one of two items depicted under the stimulus, for instance, indicate whether the actor was male or female, had short or long hair. The participant was instructed before the task that there was no right or wrong answer. No feedback was provided during the task. The control condition consisted of six blocks of 120 trials, in which 18 actors were presented 40 times, lasting approximately 10 min per block with intertrial intervals where the previous stimulus and items were still visible on the screen for 4 s. The sensori-motor control scan lasted 80 min and consisted of a total of 36 frames (6 × 60 s frames + 30 × 120 s frames). PET baseline condition: During the baseline condition the participants were instructed to lay down and rest in the scanner. The baseline scan lasted 25 min and consisted of 18 (120 s) frames.
Image processing – dopamine D2/3 receptor binding potential maps – and analysis
Image pre-processing procedures were performed as described previously (Kasanova et al., Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, Heinzel, Mottaghy and Myin-Germeys2017, Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, van Duin, Steinhart, Vaessen, Heinzel, Mottaghy and Myin-Germeys2018; van Duin et al., Reference van Duin, Kasanova, Hernaus, Ceccarini, Heinzel, Mottaghy, Mohammadkhani-Shali, Winz, Frank, Beck, Booij, Myin-Germeys and van Amelsvoort2018) using an automatic pipeline in the PMOD brain PNEURO tool (v. 3.8, PMOD Technologies, Zurich, Switzerland) (see Supplementary Materials). For each subject, individual voxel-wise parametric maps of DA D2/3R BPND (Innis et al., Reference Innis, Cunningham, Delforge, Fujita, Gjedde, Gunn, Holden, Houle, Huang, Ichise, Iida, Ito, Kimura, Koeppe, Knudsen, Knuuti, Lammertsma, Laruelle, Logan, Maguire, Mintun, Morris, Parsey, Price, Slifstein, Sossi, Suhara, Votaw, Wong and Carson2007) were generated in patient space using the Ichise's Multilinear Reference Tissue Model 2 (MRTM2) (Ichise et al., Reference Ichise, Liow, Lu, Takano, Model, Toyama, Suhara, Suzuki, Innis and Carson2003). The cerebellum, including the cerebellar hemispheres without the vermis, was used as the reference region, because of its relative lack of D2/3R (Hall et al., Reference Hall, Sedvall, Magnusson, Kopp, Halldin and Farde1994). The details of the MRTM2 analyses can be found in the Supplementary Materials. For the regional-based group comparison analysis (HC v. 22q11DS), a predefined prefrontal mask was generated in patient space for each subject according to the Hammers N30R83 atlas (Hammers et al., Reference Hammers, Allom, Koepp, Free, Myers, Lemieux, Mitchell, Brooks and Duncan2003). This predefined mask included composite and bilateral region of interests (ROIs), for: (1) PFC, including orbitofrontal cortex (OFC), inferior, middle, and superior frontal gyrus, (2) OFC only, including the anterior, medial, lateral, and parietal orbital gyrus, (3) anterior cingulate cortex (ACC), including only the subgenual and presubgenual ACC, and (4) anterior cingulate gyrus (Fig. 2 and online Supplementary Fig. S1).
Fig. 2. Masks for the frontal cortex. The mask is overlaid on a structural MRI scan and shown in transversal (a), sagittal (b), and coronal (c) views. MRI, magnetic resonance imaging; PFC, prefrontal cortex; OFC, orbitofrontal cortex; ACC, anterior cingulate cortex; ant cing gyr, anterior cingulate gyrus.
Statistical analyses
Statistical analyses were conducted in SPSS (IBM SPSS Statistics version 25.0). Between-group differences in demographic characteristics were investigated using χ2 and independent sample t tests. Average BPND values within each ROI (PFC, OFC, ACC, anterior cingulate gyrus) were determined and compared between the 22q11DS and HC group using analysis of variance. Post-hoc analyses were conducted to investigate group differences between HC and 22q11DS in BPND in all sub-regions of the frontal ROIs performing an analysis of variance. In the 22q11DS group, to investigate the relation between frontal BPND, IQ, and PANSS scores, Pearson correlation coefficients were calculated with two-tailed tests of significance. The analyses were corrected for n = 4 ROIs, using a Bonferroni correction (critical p value p = 0.05/4 = 0.013).
Results
Demographic data
Sociodemographic variables of the sample are summarized in Table 1. There were no significant differences between the 22q11DS and the HC group in age (t = 0.74, p = 0.48) and gender distribution (22q11DS M/F ratio 6/8; HC M/F ratio 4/12; χ2 = 1.07, p = 0.30). As expected, IQ-scores were significantly lower in the non-psychotic [PANSS (Leucht et al., Reference Leucht, Kane, Kissling, Hamann, Etschel and Engel2005) scores <58] 22q11DS group compared with the HC group (t = 6.48, p < 0.001), given that impaired cognitive functioning is a core characteristic of the syndrome (Jonas et al., Reference Jonas, Montojo and Bearden2014; Schneider et al., Reference Schneider, Debbané, Bassett, Chow, Fung, van den Bree, Owen, Murphy, Niarchou, Kates, Antshel, Fremont, McDonald-McGinn, Gur, Zackai, Vorstman, Duijff, Klaassen, Swillen, Gothelf, Green, Weizman, Van Amelsvoort, Evers, Boot, Shashi, Hooper, Bearden, Jalbrzikowski, Armando, Vicari, Murphy, Ousley, Campbell, Simon and Eliez2014; Weinberger et al., Reference Weinberger, Yi, Calkins and Guri2016).
Frontal D2/3R BPND in 22q11DS v. HC
Compared with HC, adults with 22q11DS revealed a significant lower D2/3R BPND in the PFC (F = 4.91, p = 0.035) and anterior cingulate gyrus (F = 12.07, p = 0.002) (see Table 1 and Fig. 3, individual data points are plotted in online Supplementary Fig. S2), suggesting lower receptor BPND in 22q11DS. There was no significant difference in D2/3R BPND between HC and adults with 22q11DS in the OFC and ACC (F = 1.47, p = 0.24 and F = 0.40, p = 0.53, respectively; Table 1 and Fig. 3). Results of separate sub-regions of the PFC, OFC, and ACC can be found in the online Supplementary Table S1 and Fig. S3. There was no significant association between D2/3R BPND in any of the frontal ROIs (p > 0.05) and IQ within the HC group and with IQ or PANSS scores within the 22q11DS group.
Fig. 3. Binding potential (BPND) per region of interest (ROI). Average dopamine D2/3 receptor binding potential (D2/3R BPND) (y-axis) in the prefrontal cortex (PFC), the orbitofrontal cortex (OFC), the anterior cingulate cortex (ACC), and the anterior cingulate gyrus (x-axis). The healthy control (HC) group is depicted in gray and the 22q11DS group in white. Mean D2/3R BPND was significantly (**) lower in the 22q11DS group compared with the HC group in the anterior cingulate gyrus. Error bars represent standard deviation's (s.d.s). **p < 0.013 survived Bonferroni correction for multiple testing. HC, healthy controls.
Discussion
Here we report the results of the first study investigating frontal dopaminergic neurotransmission in 22q11DS, a genetic syndrome that is considered a valuable model for the study of biomarkers of psychotic disorders and cognitive deficits. As hypothesized, we found lower frontal D2/3 receptor BPND in adults with 22q11DS compared with HCs, indicating abnormal frontal DA levels in adults with 22q11DS.
Lower frontal D2/3R BPND in 22q11DS
Lower D2/3R BPND in frontal brain regions adds to the growing evidence indicating aberrant DA neurotransmission in 22q11DS (Boot et al., Reference Boot, Booij, Zinkstok, Abeling, de Haan, Baas, Linszen and van Amelsvoort2008, Reference Boot, Booij, Zinkstok, de Haan, Linszen, Baas and van Amelsvoort2010, Reference Boot, Booij, Abeling, Meijer, da Silva Alves, Zinkstok, Baas, Linszen and van Amelsvoort2011a; de Koning et al., Reference de Koning, Boot, Bloemen, van Duin, Abel, de Haan, Linszen and van Amelsvoort2012; Evers et al., Reference Evers, Curfs, Bakker, Boot, da Silva Alves, Abeling, Bierau, Drukker and van Amelsvoort2014; Butcher et al., Reference Butcher, Marras, Pondal, Rusjan, Boot, Christopher, Repetto, Fritsch, Chow, Masellis, Strafella and Lang2017; van Duin et al., Reference van Duin, Kasanova, Hernaus, Ceccarini, Heinzel, Mottaghy, Mohammadkhani-Shali, Winz, Frank, Beck, Booij, Myin-Germeys and van Amelsvoort2018). There are several potential underlying mechanisms that could explain this novel finding.
It is thought that the radiotracer [18F]fallypride competes with endogenous DA levels for D2/3 receptor binding (Morris et al., Reference Morris, Fisher, Alpert, Rauch and Fischman1995; Ceccarini et al., Reference Ceccarini, Vrieze, Koole, Muylle, Bormans, Claes and Van Laere2012). Lower receptor BPND can therefore be the result of a higher DA concentration in the synaptic cleft, which results in lower BPND due to competition and/or a down-regulation of post-synaptic DA receptor density (Wong et al., Reference Wong, Wagner, Tune, Dannals, Pearlson, Links, Tamminga, Broussolle, Ravert and Wilson1986; Boot et al., Reference Boot, Booij, Abeling, Meijer, da Silva Alves, Zinkstok, Baas, Linszen and van Amelsvoort2011a). This adds to accumulating evidence indicating a hyperdopaminergic state as a general endophenotype of 22q11DS in their young adulthood (Boot et al., Reference Boot, Booij, Zinkstok, Abeling, de Haan, Baas, Linszen and van Amelsvoort2008; Butcher et al., Reference Butcher, Marras, Pondal, Rusjan, Boot, Christopher, Repetto, Fritsch, Chow, Masellis, Strafella and Lang2017). In line with current results, a recent PET study in non-psychotic adults with 22q11DS found higher pre-synaptic DA synthesis capacity in striatal brain regions (Butcher et al., Reference Butcher, Marras, Pondal, Rusjan, Boot, Christopher, Repetto, Fritsch, Chow, Masellis, Strafella and Lang2017). A hyperdopaminergic state could be the result of reduced frontal DA clearance compared with HCs, caused by COMT haploinsufficiency in 22q11DS (Chen et al., Reference Chen, Lipska, Halim, Ma, Matsumoto, Melhem, Kolachana, Hyde, Herman, Apud, Egan, Kleinman and Weinberger2004; Tunbridge et al., Reference Tunbridge, Weickert, Kleinman, Herman, Chen, Kolachana, Harrison and Weinberger2006). COMT hemizygosity in 22q11DS is suggested to result in reduced COMT enzyme activity and consequently higher DA levels, especially in the PFC (Tunbridge et al., Reference Tunbridge, Weickert, Kleinman, Herman, Chen, Kolachana, Harrison and Weinberger2006; Boot et al., Reference Boot, Booij, Zinkstok, Abeling, de Haan, Baas, Linszen and van Amelsvoort2008; van Beveren et al., Reference van Beveren, Krab, Swagemakers, Buitendijk, Buitendijk, Boot, van der Spek, Elgersma and van Amelsvoort2012), in line with our findings. It has been suggested that the ‘clearance role’ of COMT and the effect of COMT Val/Met genotype in (frontal) DA turnover becomes increasingly important under challenged conditions (Huotari et al., Reference Huotari, Gogos, Karayiorgou, Koponen, Forsberg, Raasmaja, Hyttinen and Männistö2002; Yavich et al., Reference Yavich, Forsberg, Karayiorgou, Gogos and Männistö2007), for instance during stress task-induced DA release paradigms (Hernaus et al., Reference Hernaus, Collip, Lataster, Ceccarini, Kenis, Booij, Pruessner, van Laere, van Winkel, van Os and Myin-Germeys2013). Future studies, possibly using a challenge condition and larger samples, are necessary to elaborate on the role of COMT genotype on frontal DA functioning in 22q11DS.
Furthermore, a chronic exposure to higher endogenous DA could have a toxic effect on dopaminergic neurons and is proposed to precede the onset of DA denervation in 22q11DS which is, amongst others, implicated in Parkinson's disease (PD) (Goldstein et al., Reference Goldstein, Kopin and Sharabi2014; Butcher et al., Reference Butcher, Marras, Pondal, Rusjan, Boot, Christopher, Repetto, Fritsch, Chow, Masellis, Strafella and Lang2017). Recent studies indeed show that 22q11DS patients older than 30–40 years have an increased risk for the development of PD (Booij et al., Reference Booij, Van Amelsvoort and Boot2010; Butcher et al., Reference Butcher, Marras, Pondal, Rusjan, Boot, Christopher, Repetto, Fritsch, Chow, Masellis, Strafella and Lang2017), further linking abnormal dopaminergic neurotransmission to 22q11DS.
It is interesting to speculate about the clinical implications of the observed lower frontal D2/3 BPND and the proposed hyperdopaminergic state. On the one hand our results may be associated with cognitive impairments often seen in 22q11DS (Oskarsdóttir et al., Reference Oskarsdóttir, Vujic and Fasth2004; Bassett et al., Reference Bassett, Chow, Husted, Weksberg, Caluseriu, Webb and Gatzoulis2005; Biswas and Furniss, Reference Biswas and Furniss2016; Norkett et al., Reference Norkett, Lincoln, Gonzalez-Heydrich and D'Angelo2017). Abnormal frontal DA levels may play a role in the induction of cognitive deficits based on the inverted U-shaped curve model (Goldman-Rakic et al., Reference Goldman-Rakic, Muly and Williams2000; Gothelf et al., Reference Gothelf, Schaer and Eliez2008). Thus the lower frontal D2/3 BPND in 22q11DS could be the result of excessive DA levels inducing cognitive deficits, including deficits in memory, attention, and reward processing (Gothelf et al., Reference Gothelf, Schaer and Eliez2008). Such cognitive domains have previously been shown (using e.g. single-cell recordings and PET imaging) to rely, amongst others, on frontal DA functioning (Goldman-Rakic et al., Reference Goldman-Rakic, Muly and Williams2000; Slifstein et al., Reference Slifstein, van de Giessen, Van Snellenberg, Thompson, Narendran, Gil, Hackett, Girgis, Ojeil, Moore, D'Souza, Malison, Huang, Lim, Nabulsi, Carson, Lieberman and Abi-Dargham2015) and several of these cognitive domains have been found to be impaired in 22q11DS (de Koning et al., Reference de Koning, Boot, Bloemen, van Duin, Abel, de Haan, Linszen and van Amelsvoort2012; Weinberger et al., Reference Weinberger, Yi, Calkins and Guri2016; Norkett et al., Reference Norkett, Lincoln, Gonzalez-Heydrich and D'Angelo2017; van Duin et al., Reference van Duin, Kasanova, Hernaus, Ceccarini, Heinzel, Mottaghy, Mohammadkhani-Shali, Winz, Frank, Beck, Booij, Myin-Germeys and van Amelsvoort2018). Future research including a comprehensive cognitive assessment tool is necessary, in order to associate cognitive functioning with frontal DA neurotransmission in 22q11DS.
Abnormal frontal DA levels could furthermore be related to the increased risk for developing psychotic disorders in 22q11DS. Problems in the cognitive domain often occur in psychotic disorders (Green and Nuechterlein, Reference Green and Nuechterlein1999; Nuechterlein et al., Reference Nuechterlein, Barch, Gold, Goldberg, Green and Heaton2004).
Moreover, the severity of (primarily) cognitive and negative symptoms of psychotic disorders relying on frontal DA function (Okubo et al., Reference Okubo, Suhara, Suzuki, Kobayashi, Inoue, Terasaki, Someya, Sassa, Sudo, Matsushima, Iyo, Tateno and Toru1997; Abi-Dargham et al., Reference Abi-Dargham, Mawlawi, Lombardo, Gil, Martinez, Huang, Hwang, Keilp, Kochan, Van Heertum, Gorman and Laruelle2002; Slifstein et al., Reference Slifstein, van de Giessen, Van Snellenberg, Thompson, Narendran, Gil, Hackett, Girgis, Ojeil, Moore, D'Souza, Malison, Huang, Lim, Nabulsi, Carson, Lieberman and Abi-Dargham2015) is likely to be associated with decreased DA release in frontal brain regions (Okubo et al., Reference Okubo, Suhara, Suzuki, Kobayashi, Inoue, Terasaki, Someya, Sassa, Sudo, Matsushima, Iyo, Tateno and Toru1997). Although a frontal hypodopaminergic state is proposed to be related to non-deleted psychosis (Slifstein et al., Reference Slifstein, van de Giessen, Van Snellenberg, Thompson, Narendran, Gil, Hackett, Girgis, Ojeil, Moore, D'Souza, Malison, Huang, Lim, Nabulsi, Carson, Lieberman and Abi-Dargham2015), we found lower frontal D2/3R BPND suggestive of a frontal hyperdopaminergic state and/or lower expression of post-synaptic DA receptor density (Wong et al., Reference Wong, Wagner, Tune, Dannals, Pearlson, Links, Tamminga, Broussolle, Ravert and Wilson1986; Boot et al., Reference Boot, Booij, Abeling, Meijer, da Silva Alves, Zinkstok, Baas, Linszen and van Amelsvoort2011a, Reference Boot, Booij, Zinkstok, Baas, Swillen, Owen, Murphy, Murphy, Linszen and Van Amelsvoort2011b) in non-psychotic adults with 22q11DS with (mild) cognitive impairments. This might be explained by the same mechanism as is proposed to result in cognitive dysfunction with the inverted U-shaped curve model (Goldman-Rakic et al., Reference Goldman-Rakic, Muly and Williams2000). This model suggests that either too much or too little frontal DA levels induce cognitive deficits, which could also be true for psychosis-related symptoms. It could additionally be explained by previously found differences in DAergic markers in 22q11DS compared with individuals with ultra-high risk (Vingerhoets et al., Reference Vingerhoets, Bloemen, Boot, Bakker, de Koning, da Silva Alves, Booij and van Amelsvoort2018). Disturbances of the DAergic system in the pathway to psychosis may be different in the 22q11DS population compared with other risk groups.
However, direct evidence for frontal dopaminergic alterations in psychotic disorders is inconsistent and previous findings are inconclusive (Kambeitz et al., Reference Kambeitz, Abi-Dargham, Kapur and Howes2014). In this study, we found results indicating a hyperdopaminergic state in non-psychotic 22q11DS individuals, suggesting that frontal dopaminergic alterations are present in this group regardless of psychopathology. Future research in a sample including also patients with psychotic symptoms with 22q11DS would be interesting to provide additional insight in the association between psychotic risk and frontal DA functioning.
Strengths and limitations
The main strength of this study is the use of a unique patient group with a well-defined genetic syndrome which is a valuable model for the study of biomarkers underlying, among others, cognitive impairments and psychotic disorders. Some limitations of the study should also be taken into account. First, the relatively small sample size of the sample and the use of antidepressant medication in some of the participants. We reanalyzed our main analyses excluding the 22q11DS subjects with medication and replicated our findings, indicating that the results were not affected by medication. Given the challenge of recruitment of (medication-naive) participants, the 22q11DS sample (size) could be considered representative, also in light of previous studies using similar paradigms (Hernaus et al., Reference Hernaus, Collip, Lataster, Ceccarini, Kenis, Booij, Pruessner, van Laere, van Winkel, van Os and Myin-Germeys2013; Kasanova et al., Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, Heinzel, Mottaghy and Myin-Germeys2017; van Duin et al., Reference van Duin, Kasanova, Hernaus, Ceccarini, Heinzel, Mottaghy, Mohammadkhani-Shali, Winz, Frank, Beck, Booij, Myin-Germeys and van Amelsvoort2018).
Secondly, given the well-known association between smoking (status) and DA function (Mansvelder and McGehee, Reference Mansvelder and McGehee2000), we reanalyzed our main analyses excluding the HC subjects that were habitual cigarette smokers and replicated our findings, indicating that the results were not affected by smoking status.
Additionally, the design of the scanning protocol may also have affected the results, and should be taken into consideration in future research. For the analysis of ‘relative resting state’ DA levels, from the original protocol, the sensorimotor control and baseline condition were analyzed, without the experimental condition (designed to induce reward-related DA release) (Kasanova et al., Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, Heinzel, Mottaghy and Myin-Germeys2017, Reference Kasanova, Ceccarini, Frank, van Amelsvoort, Booij, van Duin, Steinhart, Vaessen, Heinzel, Mottaghy and Myin-Germeys2018; van Duin et al., Reference van Duin, Kasanova, Hernaus, Ceccarini, Heinzel, Mottaghy, Mohammadkhani-Shali, Winz, Frank, Beck, Booij, Myin-Germeys and van Amelsvoort2018). This design is necessary to detect reliable task-induced changes on the [18F]fallypride uptake (Vernaleken et al., Reference Vernaleken, Peters, Raptis, Lin, Buchholz, Zhou, Winz, Rösch, Bartenstein, Wong, Schäfer and Gründer2011). A sensorimotor control task was used to control for sensorimotor influence on the experimental reward task condition and to keep subjects awake, in order to prevent unpredictable head movement. Although the subjects were well instructed before the sensorimotor control task (Fig. 1), the task might have influenced and elicited (sensorimotor-induced) DA release in frontal brain regions. However, this would have been the case for both the control and the 22q11DS group, and there is no evidence, to the best of our knowledge, to suggest that 22q11DS confers a different DA release to sensorimotor tasks compared with controls.
Furthermore, lower D2/3R BPND was found in the PFC and the anterior cingulate gyrus, however only the difference in the anterior cingulate gyrus survived the Bonferroni correction. Although D2/3R BPND seemed also lower in the OFC and ACC in 22q11DS compared with controls, this difference failed to reach significance. This could be due to a power issue and in increased sample sizes it is expected to find significant differences in these regions as well. More research is necessary to further explain the absence of significant differences in the OFC and ACC.
Conclusion
This study is the first to demonstrate lower frontal dopamine D2/3 receptor binding in adults with 22q11DS, which may represent a hyperdopaminergic state in frontal brain areas. This could be the result of haplo-insufficiency of COMT in these patients, and may play a role in their increased risk for developing cognitive impairments and psychotic disorders.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S003329171900062X.
Author ORCIDs
Esther D. A. van Duin, 0000-0002-4679-522X.
Acknowledgements
We thank the participants and their families. The authors thank Wendy Beuken, Debora op ‘t Eijnde, Dennis Hernaus, Merrit Beck, India Teunissen, Justine Lamee, Lucas Martens, Nele Soons, Lara Janssen, Fabiana da Silva Alves, Bernward Oedekoven, and Ron Mengelers for their assistance in data collection and management.
Financial support
This work was supported by an ERC consolidator grant to Prof Dr Inez Myin-Germeys (ERC-2012-StG, project 309767 – INTERACT) and by the National Institute of Mental Health of the National Institutes of Health under Award Number U01MH101722. Jenny Ceccarini is a postdoctoral fellow of the Research Foundation – Flanders (FWO).
Conflict of interest
None.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.
Open access
