Here we report SUPPORT (statistically unbiased prediction utilizing spatiotemporal information in imaging data), a self-supervised learning method for removing Poisson–Gaussian noise in voltage imaging data. SUPPORT is based on the insight that a pixel value in voltage imaging data is highly dependent on its spatiotemporal neighboring pixels, even when its temporally adjacent frames alone do not provide useful information for statistical prediction. Such dependency is captured and used by a convolutional neural network with a spatiotemporal blind spot to accurately denoise voltage imaging data in which the existence of the action potential in a time frame cannot be inferred by the information in other frames. Through simulations and experiments, we show that SUPPORT enables precise denoising of voltage imaging data and other types of microscopy image while preserving the underlying dynamics within the scene.

Chronic kidney disease (CKD) is one of the most common renal diseases manifested by gradual loss of kidney function with no symptoms in the early stage. The underlying mechanism in the pathogenesis of CKD with various causes such as high blood pressure, diabetes, high cholesterol, and kidney infection is not well understood. In vivo longitudinal repetitive cellular-level observation of the kidney of the CKD animal model can provide novel insights to diagnose and treat the CKD by visualizing the dynamically changing pathophysiology of CKD with its progression over time. In this study, using two-photon intravital microscopy with a single 920 nm fixed-wavelength fs-pulsed laser, we longitudinally and repetitively observed the kidney of an adenine diet-induced CKD mouse model for 30 days. Interestingly, we could successfully visualize the 2,8-dihydroxyadenine (2,8-DHA) crystal formation with a second-harmonics generation (SHG) signal and the morphological deterioration of renal tubules with autofluorescence using a single 920 nm two-photon excitation. The longitudinal in vivo two-photon imaging results of increasing 2,8-DHA crystals and decreasing tubular area ratio visualized by SHG and autofluorescence signal, respectively, were highly correlated with the CKD progression monitored by a blood test showing increased cystatin C and blood urea nitrogen (BUN) levels over time. This result suggests the potential of label-free second-harmonics generation crystal imaging as a novel optical technique for in vivo CKD progression monitoring.

Park I, et al. J Breast Cancer. 2021 Oct;24(5):463-473. https://doi.org/10.4048/jbc.2021.24.e40

Despite the apparent copious fluorescent probes targeting mitochondria, the development of low cytotoxic probes is still needed for improving validation of mitochondrial function assessment. Herein, we report a novel cyanine-based NIR fluorescent probe, T2, which selectively targets mitochondria with significantly low toxicity by modulating the intracellular redox status. Additionally, T2 inhibits oxidative stress-induced cell death in cortical neurons. This study provides new insight into developing low-toxic mitochondrial imaging agents by regulating redox homeostasis.

BACKGROUND: The entrance of naive T cells into lymph nodes (LNs) is a crucial step for induction of heart transplant acceptance under costimulatory blockade. Specialized blood vessels within the LN known as high endothelial venules (HEVs) mediate this process. HEVs express a key glycoprotein containing 6-sulfo sialyl Lewis X, the binding site for L-selectin on the membrane of naive T cells. The proper formation of this carbohydrate requires sulfation by N-acetylglucosamine 6-O-sulfotransferases, particularly CHST4. Given the critical role of CHST4 in HEVs and transplant immunity, we aimed to develop a first-in-class antibody-drug conjugate (ADC) targeting the CHST4/HEV axis to deliver immunoregulatory therapeutics to LNs and promote long-term cardiac transplant survival. METHODS: We used CHST4 knockout mice and inducible diphtheria toxin receptor mouse models to investigate the role of CHST4 in HEV function during heart allograft transplantation. To assess the impact of CHST4 deficiency on heart transplant outcomes, we used fully major histocompatibility complex–mismatched and single major histocompatibility complex class II–mismatched models to study acute and chronic heart transplant rejection, respectively. We developed a first-in-class, dual-payload antibody-drug conjugate aimed at promoting regulatory T cells by conjugating rapamycin and tubastatin A to the HEV-targeting MHA112 antibody. The efficacy of this ADC was tested in murine heart allograft models. RESULTS: Loss of CHST4 expression by HEVs impaired the entry of naive T cells into LNs and consequently reduced regulatory T cell formation, thereby disrupting cardiac transplant acceptance under costimulatory blockade. Treatment with the HEV-targeted ADC carrying rapamycin and tubastatin A prolonged murine heart allograft survival and suppressed chronic rejection, a major barrier to long-term graft success. Mechanistically, this ADC enhanced regulatory T cell induction, reduced proinflammatory T cells within lymph nodes, and decreased immune cell infiltration in cardiac allografts, achieving these effects with substantially lower doses of rapamycin and tubastatin A compared with administration of the free drugs. CONCLUSIONS: Our findings uncover a key mechanistic role for CHST4 in HEVs in facilitating the regulatory T cell generation within the LN and promoting heart transplant acceptance. Moreover, we introduce a promising CHST4-based, HEV-targeted ADC therapy as a potential strategy to suppress heart transplant rejection.