In this study, we tested the hypothesis that mTOR signalling regulates PEC activation and that mTOR inhibition may prevent PEC proliferation and subsequently glomerulosclerosis in the anti-Thy1.1 mouse model, an experimental model of collapsing FSGS.
We observed an increased mTOR activity, expressed as the phosphorylation of S6RP, in affected glomeruli of the anti-Thy1.1 mice. Especially PECs showed a striking increase in pS6RP expression. Sirolimus treatment reduced the number of affected glomeruli, which was associated with less PEC proliferation, in the early lesions seen at day 4. The reduced PEC proliferation could be a direct effect of rapamycin on PECs, since also rapamycin treated PECs in culture showed reduced pS6RP levels and reduced proliferation. In addition, the effects on PECs were independent of podocytes as the sirolimus treatment did not have an effect on albuminuria, the number of podocytes or podocyte injury. However, despite the marked effects of rapamycin in culture and in the development of the early lesions in the mice, continuation of the treatment with sirolimus could not prevent glomerulosclerosis at day 7. Also sirolimus treatment after sclerosis development could not reduce lesion formation.
Recent studies indicated that in the formation of lesions in FSGS and crescentic glomerulonephritis the PECs become activated5,9. The normal quiescent flat cells become enlarged and show different marker expression, i.e. a de novo expression of CD448,11,15. Although, we and others have established the involvement of PECs in the development of glomerular lesions, still little is known about the molecular processes driving PEC activation16. Several studies have described a role for mTOR in glomerular disease models. Although, most of these studies focused on mTOR signalling in podocytes17,18,19,20,21, a few studies have studied the pathway in PECs10,22,23. These studies indicated that mTOR signalling might be involved in PEC activation and therefore the development of sclerotic lesions.
This study shows a link between mTOR signalling and PEC activation as well. Inhibiting mTOR in the anti-Thy1.1 mice reduced PEC proliferation and sclerosis formation at day 4. However, it was striking that the beneficial effects of mTOR signalling were not seen in later phases of the disease. These contradictory findings can also be seen in other studies that investigated mTOR signalling in PECs. In general, all studies, including this work, showed that the effects of mTOR inhibition are dependent on the dose of the mTOR inhibitor, the animal model used, the route of administration and on the onset and duration of treatment.
Similar to our observations, Kurayama et al. showed that the outcome of mTOR inhibition is dependent on the onset and duration of the mTOR inhibition. The study demonstrated that in a rat model of crescentic nephritis early treatment (day of disease induction) with everolimus led to increased cellular necrosis. However, later treatment (7 days after disease induction) reduced glomerular crescent formation22. Since the anti-Thy1.1 model and the nephritis model are different models of treatment strategies and time, the observed effects are difficult to compare to our findings. Nevertheless, both studies show that the timing of treatment is critical.
Besides the timing, also the dose of rapamycin is of importance for the treatment results. From seminal studies performed by Gödel and co-workers it has become clear that mTOR signalling is tightly regulated and that an imbalance in mTOR activity may lead to adverse effects, facilitating glomerular disease17. In their study Gödel and colleagues deleted the regulatory-associated protein (Raptor), an essential adaptor molecule of mTOR, in podocytes of diabetic mice. They demonstrated that complete reduction of mTORC1 in RaptorΔpodocyte mice resulted in proteinuria and the progression of diabetic nephropathy, while deletion of only one allele of Raptor (RaptorHet podocyte) inhibited mTOR activation, leading to decreased proteinuria and glomerulosclerosis17. Hence, it was demonstrated that complete reduction of mTOR activity can be responsible for the progression of glomerular injury. This means that the reduction of mTOR activity to prevent disease progression should be done with care, as a too strong reduction can have deleterious effects. This conclusion can also be drawn from a similar study performed by Zschiedrich et al. They showed that a homozygous podocyte specific knockout (KO) of Raptor in an adriamycin induced nephropathy mouse model caused massive proteinuria and sclerosis, while a heterozygous KO resulted in a reduction of proteinuria and sclerosis21. They also showed that a high dose of administered rapamycin has a similar effect as a complete Raptor KO, while low rapamycin concentrations resembled the beneficial outcome of the heterozygous KO, again stressing the tight regulation of mTOR in the glomerulus.
The possible decrease in endothelial proliferation due to mTOR inhibition might have contributed to the recurrent sclerosis formation observed in our experiment, as reduced endothelial cell proliferation presumably impairs endothelial repair of capillaries that normally is required after glomerular damage due to anti-Thy1.1 injection24.
Due to the narrow therapeutic window of sirolimus, and in the murine studies described problems of treatment onset and duration, the implementation of sirolimus into the clinic to treat glomerulosclerosis will be challenging. Sirolimus and other rapamycin analogues (rapalogous) are approved by the Food and Drug administration (FDA) for the treatment of cancers, transplant rejection, lymphangioleiomyomatosis and tuberous sclerosis complex25. Therapeutic effects of sirolimus or rapalogous are often hampered due to recurrence of the disease assumably caused by negative feedback loops. To prevent this, treatments combining sirolimus with drugs targeting the pathways involved in the negative feedback (e.g. Akt pathway), seem more promising25. Also, ATP-competitive mTOR inhibitors are developed that prevent the feedback-mediated Akt activation26. Although the treatment with mTOR inhibitors can have beneficial effects, it is often accompanied with severe side effects such as haematological abnormalities, nephrotoxicity, dermatological diseases, impaired wound healing and metabolic disorders27, which makes long lasting treatment difficult.
In conclusion, this study shows that rapamycin can have a direct effect on parietal epithelial cells, which in vivo can lead to a short-lasting reduction of PEC proliferation and sclerotic lesions. Our findings point out that an elevated mTOR expression in parietal epithelial cells might contribute to PEC activation and the formation of glomerulosclerosis. However, literature and this study show that within the glomerulus mTOR signalling is tightly regulated, resulting in a narrow therapeutic window of rapamycin. When not used in optimal concentrations, as well as in the favourable time window for treatment, mTOR inhibition can have no beneficial or even deleterious effects. In addition, systemic administration of the inhibitor may have non-specific effects, decreasing mTOR inhibition in other cells resulting in for instance prevention of normal kidney repair functions. Whether ATP-competitive mTOR inhibitors or combination treatments of sirolimus and drugs targeting pathways involved in negative feedback loops could decrease sclerosis formation permanently evoking tolerating side effects has to be investigated.