Effects of in utero delta-9-tetrahydrocannabinol (THC) exposure on fetal and infant musculoskeletal development in a preclinical nonhuman primate model

Abstract

The endocannabinoid system (ECS) plays a major role in the maintenance of bodily homeostasis and adaptive response to external insults. It has been shown to regulate crucial physiological processes and behaviors, spanning nervous functions, anxiety, cognition, and pain sensation. Due to this broad activity, the ECS has been explored as a potential therapeutic target in the treatment of select diseases. However, until there is a more comprehensive understanding of how ECS activation by exogenous and endogenous ligands manifests across disparate tissues and cells, discretion should be exercised. Previous work has investigated how endogenous cannabinoid signaling impacts skeletal muscle development and differentiation. However, the effects of activation of the ECS by delta-9-tetrahydrocannabinol (THC, the most psychoactive component of cannabis) on skeletal muscle development, particularly in utero, remain unclear. To address this research gap, we used a highly translational non-human primate model to examine the potential impact of chronic prenatal THC exposure on fetal and infant musculoskeletal development. RNA was isolated from the skeletal muscle and analyzed for differential gene expression using a Nanostring nCounter neuroinflammatory panel comprised of 770 genes. Histomorphological evaluation of muscle morphology and composition was also performed. Our findings suggest that while prenatal THC exposure had narrow overall effects on fetal and infant muscle development, the greatest impacts were observed within pathways related to inflammation and cytokine signaling, which suggest the potential for tissue damage and atrophy. This pilot study establishes feasibility to evaluate neuroinflammation due to prenatal THC exposure and provides rationale for follow-on studies that explore the longer-term implications and functional consequences encountered by offspring as they continue to mature.

Fig 1. Experimental design and differential gene expression workflow.

A) Delta-9-tetrahydrocannabinol dosing timeline for fetal and infant rhesus macaque cohorts. B) RNA sample processing and differential gene expression workflow using NanoString nCounter. Created with BioRender.com.

Fig 2. In utero THC-exposure shifts gene expression at the fetal and infant timepoints.

A) Heatmaps of DEG between control and THC-exposed offspring in the fetal (left) and infant (right) cohorts. Values were normalized in ROSALIND by dividing gene counts within a lane by the geometric mean of normalization probes from the same lane. Differential gene expression was assessed using a generalized linear model, assuming a negative binomial distribution. B) PCA plots of control (red) vs THC-exposed (blue) differential gene expression. C) Contributions of DEG to differences between control and THC-exposed groups. The red line denotes variability contribution percentage if each gene contributed equally.

Fig 3. Gene enrichment analysis using STRING pathway and Enrichr.

A) Gene enrichment analysis using STRING to highlight the pathways most affected by in utero THC exposure in fetal (left) and infant (right) cohorts. Fetal DEG involved in PI3k-Akt signaling pathway are highlighted in red. Infant DEG involved in inflammatory response, IL-10 signaling, and IL-4 and IL-13 signaling are highlighted in red, blue, and green, respectively. B) Gene enrichment analysis using Enrichr. Bar charts of the top 10 enriched terms using the WikiPathway 2023 Human gene set library for fetal (top) and infant (bottom) cohorts.

Fig 4. Chromatic staining of fetal and infant rhesus macaque rectus femoris tissue.

A) Representative images of Masson-Goldner Trichrome staining of fetal (top) and infant (bottom) rectus femoris tissues. Whole tissue scale bars measure 2000 μm, zoomed image scale bars measure 100 μm. B) Quantification of collagen in Masson-Goldner Trichrome stain in fetal and infant cohorts. Horizontal lines represent the mean of each group. Statistical analysis was performed using unpaired t-test on GraphPad Prism 10.