Researchers at Peking University International Cancer Center and Peking University School and Hospital for Stomatology have made a groundbreaking discovery in the quest to combat osteoporosis. Utilizing a deep learning algorithm, they pinpointed the antimalarial drug dihydroartemisinin (DHA) as a potential remedy for the debilitating bone condition.
DHA has been shown to effectively reverse bone loss related to osteoporosis by preserving the “stemness” of bone marrow mesenchymal stem cells (BMMSCs). This breakthrough has the potential to revolutionize osteoporosis treatment, offering hope to millions of affected individuals.
A novel approach to Osteoporosis treatment
Osteoporosis, a degenerative disease impacting the skeletal system, poses a significant health challenge, especially for older adults. It is characterized by the loss of bone density and deterioration of the bone microstructure. In healthy individuals, there exists a delicate equilibrium between osteoblasts, responsible for bone formation, and osteoclasts, tasked with bone resorption. However, when osteoclasts become excessively active, it leads to bone loss, a hallmark of osteoporosis.
Crucially, bone marrow mesenchymal stem cells (BMMSCs), the precursors of osteoblasts, play a pivotal role in osteoporosis. They ensure a constant supply of functional osteoblasts for bone repair by both differentiation and steady proliferation. This equilibrium is essential for maintaining bone health and regeneration.
Yet, during osteoporosis, BMMSCs tend to transform into fat-producing adipocytes, significantly reducing their regenerative potential. Hence, restoring the function of these multipotent cells is paramount in addressing osteoporosis.
Harnessing deep learning to discover DHA’s potential
The research team, led by Zhengwei Xie, PhD, at Peking University, harnessed the power of deep learning to predict cellular responses to drug treatments. This pioneering deep learning-based efficacy prediction system (DLEPS) has previously been successful in identifying new drugs for various diseases, including obesity, hyperuricemia, and NASH.
For their current study, the team employed this innovative algorithm to find a new therapeutic strategy for osteoporosis focusing on BMMSCs. The algorithm analyzed the profiles of differently expressed genes (DEGs) in newborn and adult mice, revealing dihydroartemisinin (DHA) as one of the top-ranked compounds. DHA, a traditional Chinese herbal extract and a key component of malaria treatments, was identified for its ability to promote BMMSC stemness, a crucial factor in maintaining healthy bone homeostasis.
In vivo studies demonstrated that administering DHA extract for six weeks to an ovariectomized (OVX) mouse model of osteoporosis significantly reduced bone loss in the animals’ femurs and preserved bone structure almost entirely. This effect was achieved by rescuing endogenous BMMSC stemness in OVX mice and correcting the bias toward adipogenesis over osteogenesis.
To further enhance drug delivery efficiency, the researchers designed a robust system using DHA-loaded nanoparticles designed to target bone. These mesoporous silica nanoparticles (MSNs) were conjugated with bone-targeting alendronate (ALN) to deliver DHA effectively. Subsequent in vivo experiments yielded promising results, with mice receiving nanoparticle-delivered MSN-ALN@DHA nanoparticles showing bone characteristics similar to the control group, and no signs of toxicity.
Additionally, the team elucidated that DHA interacted with BMMSCs, preserving their stemness and ultimately generating more osteoblasts. Furthermore, MSN-ALN exerted its own limited antiosteoporotic effect.
The collective findings from this groundbreaking research indicate that dihydroartemisinin (DHA) holds immense promise as a therapeutic agent for osteoporosis. Notably, this work underscores the potential of deep learning approaches to accelerate drug development and advance precision medicine.
