1. Peripheral insulin resistance (IR) is a precursor to prediabetes, metabolic syndrome, and type 2 diabetes, but it remains unclear how IR manifests structurally in adipose tissue. Traditional metrics such as glucose uptake assays are a systemic average, and thus cannot represent tissue heterogeneity. We hypothesized IR causes a shift in the fraction of responsive plasma membrane. We developed large-scale imaging assays of adipose tissue to provide functional measurements of the activation of ex vivo human adipose tissue. We tested if AKT (Thr308) phosphorylation (pAKT), a signaling node of the insulin-signaling pathway, is a biomarker reflecting the number and distribution of insulin-responsive or refractory membranes in human adipose tissue. Subcutaneous adipose tissue, obtained from adult subjects at the NIH Clinical Center, was fixed within one hour of biopsy. Two plasma membrane populations were observed: low pAKT, primarily seen in the absence of insulin stimulation, and high pAKT, primarily seen in tissue from healthy subjects following insulin stimulation, and described by a mixture model (basal-like and insulin-responsive). In response to maximal insulin stimulation the high pAKT membrane fraction is significantly correlated (Pearson Rho = 0.7, p < 0.0001) with plasma membrane resident glucose transporter GLUT4. Additionally, IR, as calculated by the insulin-modified frequently sampled intravenous glucose tolerance test, was inversely correlated with the fraction of insulin-responsive membrane within adipose tissue, in agreement with previous work from isolated adipocytes. Our validation of a bi-modal membrane model also revealed large contiguous refractory membrane regions adjacent to other refractory regions, suggesting local paracrine signaling. Understanding the spatial nature of IR in primary tissue will lead to improved methods that modulate local signaling and increase the number of insulin-responsive regions in prediabetic individuals. 2. Maintaining spatial resolution and signal at large sample depth is a key challenge in fluorescence microscopy. Optical aberrations significantly degrade imaging performance at distances far from the coverslip, particularly in super-resolution imaging. Here we address this problem in multiphoton structured illumination microscopy by simple modifications to the underlying microscope optical path. Optical aberrations are determined using direct wavefront sensing with a nonlinear guide star jargon (multiphoton-excited fluorescence emitted either from the labeled sample or an exogenous marker) and subsequently corrected using a deformable mirror, restoring super-resolution information that is otherwise lost. We demonstrate the flexibility of our adaptive optics approach on a variety of semi-transparent samples, including bead phantoms, cultured cells in collagen gels, Drosophila brain, live zebrafish and live nematode larvae. The performance of our super-resolution microscope is improved in all of these samples, as peak intensity is increased (up to 40-fold) and resolution recovered (up to 17610 nm laterally and 72939 nm axially) at depths up to 250 m from the coverslip surface.
Showing the most recent 10 out of 40 publications
