Following injury and refractive surgery, corneal wound healing can initiate a protracted fibrotic response that interferes with ocular function. This fibrosis is related, in part, to the myofibroblast differentiation of corneal keratocytes in response to transforming growth factor beta 1 (TGF-β1). Previous studies have shown that changes in the mechanical properties of the extracellular matrix (ECM) can regulate this process, but the mechanotransductive pathways that govern stiffness-dependent changes in keratocyte differentiation remain unclear. Here, we used a polyacrylamide (PA) gel system to investigate how mechanosensing via focal adhesions (FAs) regulates the stiffness-dependent myofibroblast differentiation of primary corneal keratocytes treated with TGF-β1. Soft (1 kPa) and stiff (10 kPa) PA substrata were fabricated on glass coverslips, plated with corneal keratocytes, and cultured in defined serum free media with or without exogenous TGF-β1. In some experiments, an inhibitor of focal adhesion kinase (FAK) activation was also added to the media. Cells were fixed and stained for F-actin, as well as markers for myofibroblast differentiation (α-SMA), actomyosin contractility phosphorylated myosin light chain (pMLC), focal adhesions (vinculin), or Smad activity (pSmad3). We also used traction force microscopy (TFM) to quantify cellular traction stresses. Treatment with TGF-β1 elicited stiffness-dependent differences in the number, size, and subcellular distribution of FAs, but not in the nuclear localization of pSmad3. On stiff substrata, cells exhibited large FAs distributed throughout the entire cell body, while on soft gels, the FAs were smaller, fewer in number, and localized primarily to the distal tips of thin cellular extensions. Larger and increased numbers of FAs correlated with elevated traction stresses, increased levels of α-SMA immunofluorescence, and more prominent and broadly distributed pMLC staining. Inhibition of FAK disrupted stiffness-dependent differences in keratocyte contractility, FA patterning, and myofibroblast differentiation in the presence of TGF-β1. Taken together, these data suggest that signaling downstream of FAs has important implications for the stiffness-dependent myofibroblast differentiation of corneal keratocytes.

No abstract

Gene silencing through RNA interference (RNAi) has been touted as a boon for identifying potential therapies for difficult-to-treat pathologies. In this regard, siRNA-mediated gene silencing for tackling the multifaceted pathophysiology of spinal cord injury seemed promising. The genes caspase 3 and sarm1 were targeted in the present study, using siRNAs in a rodent model of spinal cord injury, as the feasibility of concomitant silencing of more than one gene had not been previously attempted. The results indicated meager benefits in terms of functional recovery and tissue preservation. Interestingly, differential transfection efficiencies due to the heterogeneous nature of cells in the spinal cord along with variability in efficacy based on time of intervention affected the reproducibility of this approach. Complex gene interactions and inadequacies in molecular evaluation strategies further complicated the interpretation of the outcome. If these glitches are resolved through further research, gene therapy in general and RNAi, in particular, may become a mainstay approach for treating contusion spinal cord injury. Impact statement Gene therapy has reached the level of clinical trials. However, safety and efficacy are yet to be confirmed. The present study tested the prospects of gene silencing using siRNAs in a rat model of spinal cord injury. Some noteworthy observations include the effective and long-lasting silencing effects of siRNAs, inhibition of one gene's expression resulting in silencing of multiple genes in associated pathways, possibility of targeting more than one gene through siRNA cocktails, and differential gene silencing effects based on temporal changes in their expression patterns. It is argued that differential uptake of siRNAs by cells as observed and limitations in the analysis methods available can skew interpretations. Thus, this study may serve as a cautionary tale indicating that gene silencing using siRNAs for spinal cord injury can be a potential therapy, but practical issues are to be addressed in order to ensure consistency and safety.

Experimental studies found extracts of Mucuna pruriens (L.) DC, a plant used in the traditional medical systems to possess strong anti-inflammatory and anti-oxidant properties among a wide range of other beneficial effects. Hypothesizing the possibility for a multifaceted beneficial effect, the therapeutic potential of the ethanolic extract of Mucuna pruriens (MP) in treating spinal cord injury (SCI) was tested using the highly standardized Multicenter Animal Spinal Cord Injury Study (MASCIS) animal model of contusive SCI. Male Wister rats with SCI were treated with the ethanolic extract of MP at a dose of 200 mg/kg/day for 10 weeks. The outcome was assessed through molecular, biochemical, histological, and behavioral scoring parameters. Results indicated anti-apoptotic, anti-necrotic, anti-inflammatory, and neuroprotective effects of MP. Nevertheless, many of these beneficial effects were not statistically significant and there was no functional improvement due to MP treatment. MP at comparable doses was reported to be significantly effective in mitigating oxidative stress and/or inflammation under various other pathological conditions. Lack of significance in the present study could be due to the presence of blood-spinal cord-barrier; that might have prevented the components of MP from reaching spinal cord tissue in adequate quantities. Additionally, heterogeneity in the function of the cells typical to the CNS environment could be the reason for the failure of recovery. To conclude, the limitations posed by the structural and functional distinctions in the spinal cord environment in comparison to other non-CNS tissue environments deserve attention while adjudicating the efficacies of herbal remedies for SCI.

The manifold assumption for high-dimensional data assumes that the data is generated by varying a set of parameters obtained from a low-dimensional latent space. Deep generative models (DGMs) are widely used to learn data representations in an unsupervised way. DGMs parameterize the underlying low-dimensional manifold in the data space using bottleneck architectures such as variational autoencoders (VAEs). The bottleneck dimension for VAEs is treated as a hyperparameter that depends on the dataset and is fixed at design time after extensive tuning. As the intrinsic dimensionality of most real-world datasets is unknown, often, there is a mismatch between the intrinsic dimensionality and the latent dimensionality chosen as a hyperparameter. This mismatch can negatively contribute to the model performance for representation learning and sample generation tasks. This paper proposes relevance encoding networks (RENs): a novel probabilistic VAE-based framework that uses the automatic relevance determination (ARD) prior in the latent space to learn the data-specific bottleneck dimensionality. The relevance of each latent dimension is directly learned from the data along with the other model parameters using stochastic gradient descent and a reparameterization trick adapted to non-Gaussian priors. We leverage the concept of DeepSets to capture permutation invariant statistical properties in both data and latent spaces for relevance determination. The proposed framework is general and flexible and can be used for the state-of-the-art VAE models that leverage regularizers to impose specific characteristics in the latent space (e.g., disentanglement). With extensive experimentation on synthetic and public image datasets, we show that the proposed model learns the relevant latent bottleneck dimensionality without compromising the representation and generation quality of the samples.Preprint. Under review.