Reduced metabotropic glutamate receptor subtype 5 in mice and men with fragile X syndrome

Measurement of metabotropic glutamate receptor subtype 5 (mGluR₅) density in heterozygous fmr1 knockout (KO) mouse models

A major neuroimaging problem in the examination of mGluR₅s in the fmr1 KO mouse model of fragile X syndrome (FXS) (Dahlhaus R., 2018) is the small volume of the striata and other structures in these mice. Visualizing mGluR₅s in the tiny structures in the brains of the fmr1 KO models is further hindered by the recessive effects of inbreeding or loss of Fragile X Mental Retardation Protein (FMRP). To counteract reduced expression of mGluR₅s in these models, heterozygous offspring from a KO parent and a wild type parent exhibit the robust expression of mGluR₅s required for functional imaging studies with relatively poor resolution. Therefore, in order to reduce or eliminate the recessive effects from inbred genetic loci (Yan QJ et al., 2005) and foster expression of mGluR₅s, we employed heterozygous male offspring of an fmr1 KO FVB strain female and a male of wild-type (C57) (Denman RB 2012a,b; Zupan B et al. 2012) because fmr1 KOs and heterozygotes exhibit greater mGluR₅ protein levels (Chen et al. 2012)].

While several techniques exist to estimate the concentration of glutamate in the living brain, including magnetic resonance imaging (MRI) and brain biopsy, positron emission tomography (PET) uniquely provides the optimal means to measure mGluR₅s. For these reasons, radiotracers that bind to mGluR₅ in the living brain and can be visualized with PET are promising tools to quantify the density and the distribution of mGluR₅s in Fmr1 KO mouse (Hamill TG et al., 2005) and in humans with FXS.

[¹¹C]ABP688

3-(6-methyl-pyridin-2-ylethynyl)-cyclohex-2-enone-O-[¹¹C]methyl-oxime ([¹¹C]ABP688), a noncompetitive and highly selective antagonist for mGluR₅s (Ametamey SM et al. 2006) has been demonstrated to confirm target engagement of mGluR₅s in mice ((Ametamey SM et al. 2006; Kim J et al. 2015), baboons (Mathews WB et al. 2014), and humans (DeLorenzo C et al. 2017). 

Materials and Methods

Participants

fmr1 KO mouse model

To maximize the expression of mGluR₅s in the fmr1 KO mouse model and to minimize the recessive effects of a full knockout mouse model, we compared and contrasted the density and distribution of mGluR₅s in six wild-type C57 male mice and six heterozygous FVBxC57 male mice. The heterozygotes were obtained by crossing female fmr1 KO mice (bred in FVB as the background strain) with male wild-type C57 mice (FVBxC57) (Yan QJ et al. 2005; Denman RB 2012a,b; Zupan et al. 2012; Kazdoba TM et al. 2014; Berman RF et al. 2014). To test the ability of [¹¹C]ABP688 PET to accurately estimate mGluR₅ density, we compared C57 wild-type and FVBxC57 mice heterozygous for the Fmr1 knockout.

MicroPET

After the induction of anesthesia with isoflurane (0.5-1%; approximately 1 L/min), 6 FVBxC57 male mice and 6 wild type C57 male mice underwent PET on a SuperArgus micro PET/CT (Sedecal, 2020) for 60 min after the intravenous bolus injection of 7.4 to 11.1 MBqs (200 to 300 microcuries) [¹¹C]ABP688 via the lateral tail vein.

Mice were scanned in pairs of one FVBxC57 and a matched littermate wild type C57 control, placed side by side in the scanner bed aligned so that their heads were upright and centered in the field of view. All scans occurred between 1 PM and 3 PM to avoid the possible effects of diurnal variations on mGluR₅ (DeLorenzo et al. 2017; Castañeda TR et al. 2004; Fuller PM et al. 2006; Meng T et al. 2015). Following each PET scan, a 5-min CT scan was performed for anatomical reference. PET images were analyzed using standard volumes of interest (VOIs) derived from the mouse brain atlas of the Laboratory for Neuro Imaging (Mark and Mary Stevens Neuroimaging and Informatics Institute. 2020; Van Essen DC 2002). 

Both PET and CT images were specially aligned and normalized to PET space with a perpendicular mid-plane using skull-to-skull and CT-to-standard CT spatial normalization parameters with the SPM5 module (Ashburner J et al., 2004a,b; Nandi et al. 2017). The prepared VOIs were applied to PET frames to obtain tissue time activity curves (TACs) to estimate the nondisplaceable binding potential (BP_ND) (Innis etal 2007) using the simplified reference tissue method (SRTM) (Lammertsma AA et al. 1996).

mGluR₅s in an fmr1 KO mouse model

In contrast to wild type C57 male mice, FVBxC57 heterozygous male mice exhibited significantly greater uptake of [¹¹C]ABP688 in the striatum (Two-sample t test, P < 0.05) (Figure 1), but not the cortex, the hippocampus, or the cerebellum.

The procedure provides measurements of mGluR₅ in mouse models of FXS.

mGluR₅s in a fmr1 KO mouse model

We showed that PET provides a quantitive measurement of mGluR₅s density in an fmr1 KO model of FXS. Unlike the fmr1 KO mouse model, our heterozygous fmr1 mouse model exhibited increased mGluR₅ density in the striatum, but not in the cerebellum, the cortex, or the hippocampus (Two-sample t test).

Different radiotracers were used for the animal and human studies after the completion of the animal studies with [¹¹C]ABP688, as we showed that 3-[¹⁸F]fluoro-5-(2-pyridinylethynyl)benzonitrile  [¹⁸F]FPEB) (Wong DF et al. 2013)exhibits better reproducibility and higher regional BP_ND values than [¹¹C]ABP688 (Kuwabara H et al. 2011). Since the human studies were performed after the animal studies with [¹¹C]ABP688, the radiotracer with optimal properties, [¹⁸F]FPEB, was employed. Using the same radiotracer for both investigations of animals and humans would improve the ability to compare and contrast findings.

Participants with fragile X syndrome (FXS), fragile X syndrome allele size mosaiscism (FXS-M), premutation of the fragile X gene (PM), autism spectrum disorder (ASD), and typical development (TD) underwent positron emission tomography to measure the density and the distribution of metabotropic glutamate receptors subtype 5 (mGluR₅) in the brain. 

Participants from the Institute for Neurodegenerative Disorders (IND), New Haven, Connecticut, USA, included 7 men with FXS aged 22.3 to 33.6 (27.1+4.7) years, a man with FXS-M aged 56.6 years, and a women with PM and TD aged 56.3 years. For the men with FXS the height ranged from 67 to 74 (70.33+3.27) inches, the weight ranged from 154 to 285 (203.5+49.61) pounds, and the BMI ranged from 22.0 to 36.6 (28.8+5.7). The man with FXS-M was 70 inches in height, 238 pounds in weight, and had a BMI of 34.1. The woman with PM TD was 65.6 inches in height, 258 pounds, and 42.3 BMI. The man with FXS-M was allele size mosaic (PM 181 CGGs, 20% methylated, and the full-mutation allele 100% methylated). The men with FXS read below the first grade level and the man with FXS-M read at the eighth grade level. All participants from IND were non-Hispanic Caucasians.

Participants from Johns Hopkins University (JHU), Baltimore, Maryland, USA included 4 men with FXS,aged 19 to 41 (27.6+9.43) years, 6 men with ASD aged 18 to 22 (20+2.1) years, and 3 individuals (1 man and 2 women) with TD aged 19 to 24 (20.67+2.89) years. 

.

Table S1 lists the concomitant medications of participants from the Institute for Neurodegenerative Disorders (IND), New Haven, Connecticut, USA.. 

The genetic and neurobehavioral assessments are presented for participants from the Institute for Neurodegenerative Disorders (IND), New Haven, Connecticut, USA, in Table S2 and for participants from the Johns Hopkins University (JHU), Baltimore, Maryland, USA in Table S3.

Clinically participants JHUFXS1 and JHUFXS2 did not exhibit ASD. Participant JHUFXS1 withdrew from the study for a family emergency before neurobehavioral testing including symptoms of ASD (Lord et al., 2012) was accomplished.  Because of his age participant JHUFSX2 was adminiered module 4 of the ADOS (Lord et al., 2012). Due to his limited speech a lower module would likely appropriately indicate the absence of autism. In other words, the symptoms demonstrated by module 4 of the ADOS (Lord et al. 2012) likely reflect an artifact of the incorrect administration of a module for higher functioning individuals. At JHU recruiting participants was challenging because research with positron emission tomography (PET) had not previously been conducted on people with FXS. Therefore, all participants with FXS at JHU were recruited regardless of the presence of ASD.  . 

The positron emission tomography data and analysis are presented for participants from the Institute for Neurodegenerative Disorders in Table S4 and for participants from the Johns Hopkins University in Table S5. 

The authors thank Flora Tassone, Ph.D., Department of Biochemistry and Molecular Medicine, School of Medicine, UC Davis Health, Sacramento, California, for providing genetic and protein data about participants.

Disclosures: We disclose the unlabeled/unapproved use of 3-(6-methyl-pyridin-2-ylethynyl)-cyclohex-2-enone-O-[¹¹C]methyl-oxime ([¹¹C]ABP688)