miRNA‐23a/27a attenuates muscle atrophy and renal fibrosis through muscle‐kidney crosstalk

Background The treatment of muscle wasting is accompanied by benefits in other organs, possibly resulting from muscle–organ crosstalk. However, how the muscle communicates with these organs is less understood. Two microRNAs (miRs), miR‐23a and miR‐27a, are located together in a gene cluster and regulate proteins that are involved in the atrophy process. MiR‐23a/27a has been shown to reduce muscle wasting and act as an anti‐fibrotic agent. We hypothesized that intramuscular injection of miR‐23a/27a would counteract both muscle wasting and renal fibrosis lesions in a streptozotocin‐induced diabetic model. Methods We generated an adeno‐associated virus (AAV) that overexpresses the miR‐23a∼27a∼24‐2 precursor RNA and injected it into the tibialis anterior muscle of streptozotocin‐induced diabetic mice. Muscle cross‐section area (immunohistology plus software measurement) and muscle function (grip strength) were used to evaluate muscle atrophy. Fibrosis‐related proteins were measured by western blot to monitor renal damage. In some cases, AAV‐GFP was used to mimic the miR movement in vivo, allowing us to track organ redistribution by using the Xtreme Imaging System. Results The injection of AAV‐miR‐23a/27a increased the levels of miR‐23a and miR‐27a as well as increased phosphorylated Akt, attenuated the levels of FoxO1 and PTEN proteins, and reduced the abundance of TRIM63/MuRF1 and FBXO32/atrogin‐1 in skeletal muscles. It also decreased myostatin mRNA and protein levels as well as the levels of phosphorylated pSMAD2/3. Provision of miR‐23a/27a attenuates the diabetes‐induced reduction of muscle cross‐sectional area and muscle function. Curiously, the serum BUN of diabetic animals was reduced in mice undergoing the miR‐23a/27a intervention. Renal fibrosis, evaluated by Masson trichromatic staining, was also decreased as were kidney levels of phosphorylated SMAD2/3, alpha smooth muscle actin, fibronectin, and collagen. In diabetic mice injected intramuscularly with AAV‐GFP, GFP fluorescence levels in the kidneys showed linear correlation with the levels in injected muscle when examined by linear regression. Following intramuscular injection of AAV‐miR‐23a∼27a∼24‐2, the levels of miR‐23a and miR‐27a in serum exosomes and kidney were significantly increased compared with samples from control virus‐injected mice; however, no viral DNA was detected in the kidney. Conclusions We conclude that overexpression of miR‐23a/27a in muscle prevents diabetes‐induced muscle ca