UVB-induced cutaneous microvascular responses and NOS-related modulation assessed by multimodal in vivo imaging and graph-based OCTA analysis

Abstract

Background: Ultraviolet B (UVB) irradiation induces inflammatory and vascular alterations in the skin, including changes in microvascular perfusion and endothelial activation. However, quantitative in vivo assessment of UVBassociated microvascular remodeling and its modulation by nitric oxide (NO) signaling remains limited. Objective: This study aimed to characterize UVB-induced cutaneous microvascular responses using multimodal in vivo imaging and graph-based vascular network analysis and to evaluate the effects of nitric oxide synthase (NOS) inhibition on these vascular responses. Methods: Seven-week-old male C57BL/6 mice were used for laser Doppler flowmetry (LDF), whereas seven-weekold female B6 nude mice were used for optical coherence tomography angiography (OCTA) and OCT structural imaging. In each modality-specific cohort, mice were allocated to control, UVB irradiation (200 mJ/cm2), UVB combined with topical N omega-nitro-L-arginine methyl ester (L-NAME, 20 mg/kg), or L-NAME alone groups. Cutaneous perfusion was measured in dorsal skin by LDF at baseline and 24, 48, 72, and 96 h after intervention. OCTA and OCT imaging of the ear were performed at baseline and 48 h. OCTA images were analyzed using a graph-based framework to quantify node number, edge number, and total vessel length. Histological and immunohistochemical analyses for CD31 and inducible nitric oxide synthase (iNOS) were performed at 48 h. Results: UVB exposure significantly increased cutaneous perfusion at 48 h compared with the control group (p = 0.0164), whereas L-NAME attenuated this response (p = 0.0439 vs. UV). OCTA demonstrated greater visibility and extent of perfused superficial microvascular structures following UVB exposure, with apparent tissue thickening on structural OCT. Whole-image graph-based analysis showed modest positive changes in node number, edge number, and total vessel length in the UV group from baseline to 48 h, whereas the UV +L-NAME group showed decreases in all three network measures. These topology-derived changes were descriptive deterministic estimates rather than group-level inferential comparisons. Histological analyses demonstrated increased CD31-positive endothelial structures and iNOS immunoreactivity in UVB-exposed tissue. Conclusions: Multimodal in vivo imaging combined with graph-based OCTA analysis enabled complementary assessment of UVB-associated changes in cutaneous perfusion and perfused microvascular network organization. The attenuation of UVB-associated hyperperfusion by L-NAME, together with the observed OCTA and histological findings, supports a contributory role of NOS-related signaling in UVB-associated cutaneous vascular dysregulation. Graph-based OCTA analysis may provide a useful approach for characterizing microvascular network alterations during cutaneous inflammatory responses.