研究背景：冠脉微循环障碍（coronary microvascular dysfunction, CMD）是指冠脉微循环结构和功能的紊乱，它是糖尿病心脏病、高血压心脏病和肥厚型心肌病等多种心血管疾病的共同病理生理基础。然而，CMD的病理特点和病理生理机制尚未阐明，且缺乏针对性的治疗手段。钠-葡萄糖共转运蛋白2（sodium glucose cotransporter 2, SGLT2）抑制剂可明显改善糖尿病患者的心血管结局，但其是否通过改善糖尿病冠脉微循环障碍保护心脏正常结构和功能，尚不清楚。

研究目的：本研究拟使用2型糖尿病小鼠模型，观察糖尿病CMD的特点，明确SGLT2抑制剂恩格列净对糖尿病冠脉微循环障碍的改善作用，并初步探索其机制。

研究方法：将自发2型糖尿病db/db小鼠随机分为db/db组与db/db+EMPA（empagliflozin，恩格列净）组，db/m小鼠作为正常对照组小鼠。从8周龄开始，给予db/db+EMPA组小鼠恩格列净10 mg/（kg·d）灌胃处理，另外两组则给予相应体积的饮用水灌胃，持续8周。动态观察小鼠体重、空腹血糖、血压和24小时尿量等指标，使用小动物超声检测小鼠心脏收缩舒张功能和冠脉微循环功能。在实验结束时取材，通过组织学染色观察心肌肥大和纤维化等情况，通过免疫荧光标记周细胞和血管内皮及激光共聚焦显微镜成像，观察冠脉微循环结构和周细胞分布。

研究结果：相较于db/m小鼠，db/db小鼠体重增加、空腹血糖升高、尿量增加。db/db小鼠左室射血分数、左心室短轴缩短率和舒张早期二尖瓣血流最大流速（E）与心房收缩期二尖瓣血流最大流速（A）的比值（E/A）均无显著性差异。与db/m小鼠相比，db/db小鼠基础冠脉血流速度加快，最大充血时的冠脉血流速度减慢，冠脉血流储备（coronary flow reserve, CFR）显著下降（1.8±0.1比3.4±0.4，p＜0.001）。病理染色结果表明，db/db小鼠心肌细胞肥大，无显著纤维化。免疫荧光染色结果显示，db/db小鼠冠脉微血管数量减少（47±4/高倍镜视野比64±6/高倍镜视野，p=0.036）、管径增加（9.2±0.3μm比7.6±0.2μm，p<0.001），并可见微血管迂曲度增加、毛细血管瘤、基底膜桥等微血管病理表现。此外，db/db小鼠心脏中周细胞数量减少（3.6±0.4/HPF比5.5±0.3/HPF，p<0.001）、微血管周细胞数量覆盖率降低（28%±3%比40%±3%，p=0.005）、出现无核周细胞“鬼影”。相较于db/db组，db/db+EMPA组小鼠的空腹血糖显著降低，接近于db/m组小鼠血糖水平。恩格列净不影响db/db小鼠的心脏收缩和舒张功能，但显著增加小鼠的CFR（2.8±0.1比1.9±0.1，p＜0.001）。免疫荧光染色结果表明，相比于db/db组小鼠的冠脉微血管稀疏、异常扩张，db/db+EMPA组小鼠的冠脉微血管数量增加（82±2/HPF比73±2/HPF，p=0.015）、毛细血管管径减小（11.3±0.2μm比12.5±0.2μm，p＜0.001）。此外，相较于db/db组小鼠，db/db+EMPA组小鼠冠脉微循环周细胞数量（9.4±0.7/HPF比7.6±0.3/HPF，p=0.035）和覆盖率（61%±2%比53%±1%，p=0.003）均有所增加。

研究结论：糖尿病小鼠冠脉微循环结构和功能异常，同时心脏周细胞减少、微血管周细胞覆盖率下降。恩格列净可改善糖尿病小鼠冠脉微循环功能和结构的异常，并抑制心脏周细胞的丢失，可能是糖尿病CMD的有效治疗药物。

Background：Coronary microvascular dysfunction (CMD) refers to the disorder of coronary microcirculation structure and function, which is the common pathophysiological mechanism of various cardiovascular diseases such as diabetic heart disease, hypertensive heart disease, and hypertrophic cardiomyopathy. However, the pathological characteristics and pathophysiological mechanisms of CMD have not yet been elucidated, and targeted treatment strategies are lacked. Sodium-glucose cotransporter 2 (SGLT2) inhibitor can significantly improve cardiovascular outcomes in patients with diabetes mellitus, but it remains unclear whether it protects the cardiovascular system by alleviating coronary microvascular dysfunction.

Objective：This study aims to use the mouse model of type 2 diabetes to observe the characteristics of diabetic coronary microcirculation dysfunction. Moreover, we aim to clarify the protective effects of SGLT2 inhibitor empagliflozin on diabetic coronary microcirculation, and to preliminarily explore the possible mechanisms.

Methods：Spontaneous type 2 diabetic db/db mice were randomly divided into db/db group and db/db+EMPA group. The db/m mice were served as the control mice. At the age of 8 weeks, the mice in the db/db+EMPA group were given empagliflozin 10 mg/(kg·d) by gavage for 8 weeks. Similarly, the other two groups were given the isometric volume of drinking water by gavage. The body weight, fasting blood glucose, blood pressure, 24-hour urine volume, and other indicators of the mice were dynamically observed. The cardiac systolic, diastolic function, and coronary microcirculation function of the mice were detected by small animal ultrasound system. At the end of the experiment, the hearts of mice were collected. The cardiomyocyte hypertrophy and fibrosis were observed using histological staining. The coronary microcirculation structure and distribution of cardiac pericytes were observed using immunofluorescence and the laser confocal microscopy.

Results：Compared with db/m mice, db/db mice had higher body weight, fasting blood glucose level, and urine volume. The left ventricular ejection fraction, fractional shortening, and the maximal peak blood flow in early diastole of mitral valve (peak E) and maximal peak blood flow during mitral atrial systole (peak A) ratio (E/A) were not significant different between the db/db group and db/m group. In db/db mice, baseline coronary flow velocity (CFV) increased and hyperemic CFV decreased, therefore coronary flow reserve (CFR) decreased. Pathological staining showed that the cardiomyocytes of db/db mice were hypertrophied without significant cardiac fibrosis. Immunofluorescence staining showed that the number of coronary microvessels in db/db mice decreased (47±4/HPF vs. 64±6/HPF, p=0.036) and the diameter of microvessels increased (9.2±0.3μm vs. 7.6±0.2μm, p<0.001). At the same time, the tortuosity of microvessels increased and microvascular pathological manifestations such as capillary hemangioma and basement membrane bridge were observed in db/db mice. In addition, the number and microvascular coverage of cardiac pericytes were reduced (pericyte number: 3.6±0.4/HPF vs. 5.5±0.3/HPF, p<0.001; pericyte coverage: 28%±3% vs. 40%±3%, p=0.005), and anucleated pericyte ghosts were observed.

Compared to the mice in the db/db group, the fasting blood glucose of the mice in the db/db+EMPA group was significantly lower, which was close to the level of the db/m mice. Empagliflozin did not affect cardiac systolic and diastolic function, but significantly improved the CFR of db/db mice (2.8±0.1 vs. 1.9±0.1, p＜0.001). The results of immunofluorescence staining showed that, compared to the sparse and abnormal expansion of coronary microvessels in the db/db group, the number of coronary microvessels increased (82±2/HPF vs. 73±2/HPF, p=0.015) and the capillary diameter decreased (11.3±0.2μm vs. 12.5±0.2μm，p＜0.001) in the db/db+EMPA group. In addition, compared to the db/db group, the number and microvascular coverage of cardiac pericytes were increased in the db/db+EMPA group (pericyte number: 9.4±0.7/HPF vs. 7.6±0.3/HPF, p=0.035; pericyte coverage: 61%±2% vs. 53%±1%, p=0.003).

Conclusion：Diabetic mice had abnormal coronary microvascular structure and function, and the number and microvascular coverage of cardiac pericytes were reduced. Empagliflozin improved the function and structure of coronary microcirculation and inhibited the loss of cardiac pericytes in diabetic mice, therefore it might be an effective treatment for diabetic CMD.