Early metformin treatment: An effective approach for targeting fragile X syndrome pathophysiology

Abstract

Fragile X syndrome (FXS) is the most common genetic cause of autism spectrum disorder engendered by transcriptional silencing of the fragile X messenger ribonucleoprotein 1 (FMR1) gene. Given the early onset of behavioral and molecular changes, it is imperative to know the optimal timing for therapeutic intervention. Case reports documented benefits of metformin treatment in FXS children between 2 and 14 y old. In this study, we administered metformin from birth to Fmr1^−/y mice which corrected up-regulated mitogen-2 activated protein kinase/extracellular signal-regulated kinase and mammalian/mechanistic target of rapamycin complex 1 signaling pathways and specific synaptic mRNA-binding targets of FMRP. Metformin rescued increased number of calls in ultrasonic vocalization and repetitive behavior in Fmr1^−/y mice. Our findings demonstrate that in mice, early-in-life metformin intervention is effective in treating FXS pathophysiology.

Fragile X syndrome (FXS) is the leading monogenic cause of intellectual disability and autism spectrum disorder (ASD) due to the CGG repeat expansion causing transcriptional silencing of the fragile X messenger ribonucleoprotein 1 (FMR1) gene (1). In FXS patients and Fmr1 knockout mice (Fmr1^−/y), the loss of FMRP engenders increased activity of multiple cellular pathways including mammalian/mechanistic target of rapamycin complex 1 (mTORC1) and mitogen-2 activated protein kinase/extracellular signal-regulated kinase (MAPK/ERK) leading to elevated mRNA translation and several behavioral deficits (2, 3).

Metformin corrected core ASD phenotypes after a 10-d treatment of adult Fmr1^−/y mice and reduced MAPK/ERK signaling and matrix metalloproteinase 9 (MMP-9) mRNA translation but not phosphorylated S6 ribosomal protein (p-S6) (3). Human case studies reported that metformin treatment in FXS children aged 2 to 14 y improved speech and cognition and alleviated macroorchidism (4–7). Since FXS is a neurodevelopmental disease (8), it was relevant to determine whether metformin administration from birth would correct FXS pathophysiology and investigate the underlying biochemical and behavioral abnormalities in the FXS mouse model.

Results

Metformin administration from birth restored the levels of phosphorylated mitogen-activated protein kinase kinase (MEK), ERK, eukaryotic translation initiation factor 4E (eIF4E), S6, and total MMP-9 in the hippocampus (HIP) and prefrontal cortex (PFC) of Fmr1^−/y mice (Fig. 1 A–J). Similar changes were observed in the striatum as increased levels of p-ERK and p-S6 were corrected (Fig. 1 K and L). These results demonstrate that administration of metformin from birth reversed up-regulated ERK and mTORC1 signaling in HIP, PFC, and striatum of Fmr1^−/y mice.

Fig. 1.

Metformin from birth reduced up-regulated ERK and mTORC1 signaling. Representative immunoblots from vehicle and metformin-treated WT and Fmr1^−/y mice of phosphorylated and total levels of MEK (A and B), ERK (C and D), eIF4E (E and F), MMP-9 (G and H) and S6 (I and J) in HIP and PFC. Representative immunoblots from vehicle and metformin-treated WT and Fmr1^−/y mice of phosphorylated and total levels of ERK (K) and S6 (L) in the striatum. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as loading control.

Selective synaptic FMRP-binding targets are up-regulated in Fmr1^−/y mice (9). Metformin treatment from birth reduced microtubule-associated protein 2 (MAP2), eukaryotic translation elongation factor 2 (eEF2), pumilio RNA-binding family member 2 (PUM2), and synapsin (Fig. 2 A–D).

Fig. 2.

Correction of synaptic FMRP-binding target levels and behavioral deficits. Representative immunoblots of MAP2 (A), eEF2 (B), PUM2 (C), and synapsin (D) in HIP of vehicle and metformin-treated WT and Fmr1^−/y mice. USV was measured as number of calls per minute (E) and call length (F). Repetitive behavior was assessed by measuring the time spent grooming (G) and the number of grooming bouts (H).

Metformin treatment from birth rescued core ASD behaviors in Fmr1^−/y mice, specifically ultrasonic vocalizations (USV) tested in 3-wk-old mice and grooming behavior assessed in 11- to 12-wk-old mice.

Language development and communication are delayed in FXS and are studied in mice through USV (4). Increased number of calls was observed in young Fmr1^−/y mice, which was fully corrected by metformin (Fig. 2E). No differences were found in call length (Fig. 2F).

Prolonged exposure to metformin also corrected repetitive behavior, a core deficit in ASD (1), in Fmr1^−/y mice with reduced time spent grooming (Fig. 2G) and number of grooming bouts (Fig. 2H).

Discussion

Metformin, an antidiabetic FDA-approved drug, is used worldwide by more than 200 million patients daily (10). In light of the encouraging preclinical results in adult mice (3), metformin is undergoing double-blind-controlled trials as a treatment for FXS patients aged between 6 and 58 y (11). Clinical case studies reported that metformin is beneficial in FXS when treatment is started at a young age (4, 5).

Choosing an optimal time window for medical intervention to better address FXS pathophysiology is of crucial importance. For instance, arbaclofen, a GABA-B receptor agonist, had no efficacy in adolescents and adults with FXS, whereas children aged 5 to 11 y showed improvements in several behavioral tests (12).

MAPK/ERK and mTORC1 pathways play a major role in mRNA translation and synaptic formation (13, 14). Both pathways show increased signaling in FXS, leading to excessive protein synthesis (3). Specifically, increased mTORC1 signaling early in development was reported to cause impairments in behavior and synaptic transmission in FXS and ASD (2, 14). Metformin treatment in adult Fmr1^−/y mice corrected the up-regulated ERK pathway but not elevated p-S6, a downstream target of mTORC1 signaling (3). In this study, metformin treatment from birth corrected both MAPK/ERK and mTORC1 signaling pathways. A plausible reason for this lack of effect in adult mice is that p-S6 is elevated during embryonic development and gradually declines after 2 wk postnatally (15). This time-dependent expression pattern can explain why metformin administered from birth corrects both pathways. This reduction in signaling cascades normalized elevated levels of specific synaptic FMRP-binding targets, which was not observed upon treatment in adult Fmr1^−/y mice (3).

In addition, early metformin treatment corrected selective behaviors addressing impairments in vocalization in young Fmr1^−/y mice, mirroring speech difficulties in FXS children (5, 6), and repetitive behavior in adult mice as previously reported (3).

In conclusion, our study indicates that metformin administration should be started early in development in FXS patients, once diagnosed. In comparison to metformin treatment in adult Fmr1^−/y mice, more favorable cellular outcomes and early behavioral impairments were observed. Together, these data offer potential future clinical approaches for early metformin intervention in FXS patients.

Materials and Methods

Metformin Administration.

Mice received metformin (5 mg/mL) or vehicle (water) daily, starting no later than 1 d after birth, through drinking water. Procedures were in compliance with the Canadian Council on Animal Care guidelines and approved by the McGill University Animal Care Committee.

Biochemical and Behavioral Assays and Statistical Analysis.

Biochemical and behavioral assays and statistical analysis are provided in SI Appendix and Dataset S1. For all graphs, values are shown as mean ± SEM. ***P < 0.001, **P < 0.01, *P < 0.05, ns = not significant.

Supplementary Material

Appendix 01 (PDF)

Dataset S01 (XLSX)

Data, Materials, and Software Availability

All study data are included in the article and/or supporting information.