UDC 616.379-008.64:617.735:576.3:611-018

Abstract

Background. Diabetic retinopathy (DR) is increasingly recognized as a heterogeneous neurovascular disorder, yet preclinical models often fail to distinguish between the distinct pathophysiological clusters of type 2 diabetes. Recent clinical evidence demonstrates that patients with different metabolic phenotypes, specifically Severe Insulin-Resistant Diabetes (SIRD) versus Severe Insulin-Deficient Diabetes (SIDD), exhibit divergent patterns of retinal complications. However, the specific cytomorphometric signatures that distinguish these phenotypes at the cellular level remain poorly characterized.

Aim: to determine whether specific systemic metabolic phenotypes, SIRD versus SIDD, induce distinguishable cytomorphometric signatures of retinal neurodegeneration in experimental models.

Materials and Methods. Male Wistar rats were stratified into three metabolic models: (1) Chronic High-Fat Diet (HFD, 150-180 days) mimicking the SIRD phenotype; (2) HFD combined with low-dose streptozotocin (STZ, 30 days post-induction) mimicking the SIDD phenotype; and (3) HFD with supplemental glucose loading. Systemic phenotyping included longitudinal profiling of lipids, liver function, and protein status. Retinal cytoarchitecture was quantified using a computational pathomics pipeline (QuPath/StarDist) with deep learning-based segmentation of >90,000 nuclei, analyzed via Bayesian hierarchical modelling.

Results. The HFD model exhibited a "lipotoxicity-first" trajectory where dyslipidemia and hepatic steatosis preceded hyperglycemia. This SIRD-like phenotype was characterized by progressive retinal atrophy, specifically an 18-46% reduction in Inner Nuclear Layer neuronal density, accompanied by nuclear hyperchromasia indicative of epigenetic condensation. In contrast, the HFD+STZ model, characterized by acute hyperglycemia and significant hypoproteinemia, exhibited a paradoxical 1.54-fold increase in Ganglion Cell Layer density and thickening. This SIDD-like signature reflected cytotoxic edema driven by osmotic and oncotic failure, confirmed morphologically by severe irregular karyomegaly and profound chromatin dissolution (hypochromasia).

Conclusions. Retinal neurodegeneration is not a monolithic process but mirrors specific systemic metabolic trajectories. Chronic insulin resistance (SIRD) drives progressive atrophy via lipotoxicity, while insulin deficiency (SIDD) precipitates acute edema driven by hemodynamic oncotic failure. These findings establish retinal morphometry as a sensitive biomarker for systemic metabolic phenotyping, supporting a precision medicine approach to diabetic retinopathy.

Introduction: There is growing evidence from animal and clinical studies suggesting probiotics can positively affect type 2 diabetes (T2D). In a previous randomized clinical study, we found that administering a live multistrain probiotic and absorbent smectite once a day for eight weeks to patients with T2D could reduce chronic systemic inflammatory state, insulin resistance, waist circumference and improve the glycemic profile. However, there is a lack of evidence supporting the efficacy of probiotic co-supplementation with absorbent smectite on pancreatic b-cell function in T2D.

Aim: This secondary analysis aimed to assess the effectiveness of an alive multistrain probiotic co-supplementation with absorbent smectite vs placebo on b-cell function in T2D patients.

Material and methods: We performed a secondary analysis on a previously published randomized controlled trial (NCT04293731, NCT03614039) involving 46 patients with T2D. The main inclusion criteria were the presence of b-cell dysfunction (%B<60%) and insulin therapy alone or combined with oral anti- diabetic drugs. The primary outcome was assessing b-cell function as change C- peptide and %B.

Results: We observed only a tendency for improving b-cell function (44.22 ± 12.80 vs 55.69 ± 25.75; р=0.094). The effectiveness of the therapy probiotic- smectite group was confirmed by fasting glycemia decreased by 14% (p=0.019), HbA1c – 5% (p=0.007), HOMA-2 – 17% (p=0.003) and increase of insulin sensitivity by 23% (p=0.005). Analysis of the cytokine profile showed that statistical differences after treatment were in the concentration of both pro- inflammatory cytokines: IL-1b (22.83 ± 9.04 vs 19.03 ± 5.57; p=0.045) and TNF-a (31.25 ± 11.32 vs 26.23 ± 10.13; p=0.041).

Conclusion: Adding a live multistrain probiotic and absorbent smectite supplement slightly improved b-cell function and reduced glycemic-related parameters in patients with T2D. This suggests that adjusting the gut microbiota could be a promising treatment for diabetes and warrants further investigation through more extensive studies.