Serum based clinical metabolomics analysis by NMR revealed abnormalities in mannose and myo-inositol metabolism in Scleroderma

<strong>Background and hypothesis</strong> Systemic sclerosis (SSc) is a chronic autoimmune disorder characterized by fibrosis of the skin and internal organs, as well as vascular damage. Recent research has suggested that abnormal metabolism of certain sugars, such as mannose and myo-inositol, may contribute to the development and progression of SSc. Clinical studies have found that SSc patients have increased mannose and decreased myoinositol levels in their blood plasma/serum samples compared to those of healthy controls, suggesting that abnormal mannose and myoinositol metabolism may contribute to SSc pathogenesis. However, further research on patient cohorts of different ethnicity is imperative to validate these findings and understanding the mechanisms underlying these metabolic abnormalities and to develop new therapies that target these pathways. The present hypothesis-free NMR based clinical metabolomics study is an effort in this direction to compare the circulatory levels of mannose, myo-inositol and other endogenous metabolites in the sera of scleroderma patients and control subjects so that to validate the proposed abnormalities in mannose and myo-inositol metabolic pathways and association between them. The selected metabolic features and the metabolic ratio i.e. myo-inositol to mannose ratio (MMR) were further evaluated for their clinical potential in diagnostic and prognostic screening. <strong>Methods: </strong>The serum sample from 83 SSc patients meeting ACR 1980 criteria for Systemic Sclerosis, and 43 age and sex matched normal controls, were analyzed using one dimensional (1D) ¹H NMR spectroscopy coupled with multivariate statistical analysis such as Partial Least Square-Discriminate Analysis (PLS-DA). The NMR spectra were analysed using NMR suite of commercial software CHENOMX (www.chenomx.com/) and the metabolic concentrations were measured with respect to endogenous metabolite formate (as an internal calibration standard and concentration was set to 30 µM). For evaluating serum metabolic disparity between the study groups, the machine learning model was generated using random forest (RF) classification method and the Mean decrease accuracy (MDA) scores were used to identify the distinctive metabolic abnormalities in SSc. Univariate receiver operator characteristic (ROC) curve analysis was used to evaluate the diagnostic potential of the selected metabolic features. <strong>Results: </strong>There was clear distinction between SSc and healthy controls on the PLS-DA score plots [R²= 0.98] and aberrant metabolic changes were evident in the sera of SSc patients. The sera of SSc patients were characterized by decreased serum levels of alanine, valine, <strong>myoinositol</strong>, creatinine, pyruvate, and lactate; whereas the serum levels of, acetate, and 3-hydroxybutyrate were found to be significantly elevated. Contrary to expectation, the serum levels of mannose found to be insignificantly different (SSc: 46.0 µM and NC: 43.8 µM). The majority of these metabolic alterations found to be well consistence with those reported previously in other clinical metabolomics studies [1,2]. The mean values for circulatory levels of Myoinositol in SSC patients and NC subjects were 29.2 µM and 67.4 µM, respectively. Further, the circulatory MMR levels were estimated as [Myo-inositol/ Mannose] and compared between the study groups. Like myo-inositol, the MMR levels were also found to be significantly decreased in SSc patients (mean value =0.71 ± 0.32 compared to 1.85 ± 0.73 as observed for NC subjects. <strong>Conclusion: </strong>The altered levels of myo-inositol and other endogenous metabolites in the sera of SSc patients suggested abnormalities in myo-inositol metabolism in SSc patients and future studies are warranted to underscore its role in the pathobiology of scleroderma. <strong>References:</strong> <strong>[1] </strong>Federica Murgia, Silvia Svegliati, Simone Poddighe, Milena Lussu, Aldo Manzin, Tatiana Spadoni, Colomba Fischetti, Armando Gabrielli, and Luigi Atzori. "Metabolomic profile of systemic sclerosis patients." Scientific Reports 8, no. 1 (2018): 7626. <strong>[2].</strong> Thomas Bögl, Franz Mlynek, Markus Himmelsbach, Norbert Sepp, Wolfgang Buchberger, and Marija Geroldinger-Simić. "Plasma metabolomic profiling reveals four possibly disrupted mechanisms in systemic sclerosis." Biomedicines 10, no. 3 (2022): 607.**研究背景与假说** 系统性硬化症（Systemic sclerosis, SSc）是一种以皮肤及内脏器官纤维化、血管损伤为特征的慢性自身免疫性疾病。近期研究提示，甘露糖（mannose）与肌醇（myo-inositol）等糖类的代谢异常可能参与SSc的发生与进展。临床研究发现，与健康对照相比，SSc患者血浆/血清样本中的甘露糖水平升高、肌醇水平降低，提示甘露糖与肌醇代谢异常可能参与SSc的发病机制。然而，针对不同种族患者队列的进一步研究，对于验证上述发现、阐明此类代谢异常的潜在机制，并开发靶向这些通路的新型治疗手段至关重要。本项基于无假说核磁共振（Nuclear Magnetic Resonance, NMR）的临床代谢组学研究正是朝着这一方向开展，旨在比较硬皮病患者与对照受试者血清中甘露糖、肌醇及其他内源性代谢物的循环水平，以验证甘露糖与肌醇代谢通路的异常假说及其二者间的关联。本研究还进一步评估了筛选出的代谢特征及代谢比值——即肌醇与甘露糖比值（myo-inositol to mannose ratio, MMR）——在诊断与预后筛查中的临床应用潜力。 **研究方法** 本研究纳入符合1980年美国风湿病学会（American College of Rheumatology, ACR）系统性硬化症诊断标准的83例SSc患者，以及43例年龄与性别匹配的健康对照者，采用一维（1D）氢核磁共振（¹H NMR）光谱技术结合偏最小二乘判别分析（Partial Least Square-Discriminate Analysis, PLS-DA）等多元统计分析方法对其血清样本进行检测分析。采用商业化软件CHENOMX的NMR分析套件（www.chenomx.com/）对NMR光谱进行解析，并以内源性代谢物甲酸酯作为内标校准物（浓度设定为30 μM）对代谢物浓度进行定量。为评估两组研究对象的血清代谢差异，本研究采用随机森林（Random Forest, RF）分类法构建机器学习模型，并通过平均准确率下降值（Mean Decrease Accuracy, MDA）评分识别SSc患者中具有显著差异的代谢异常特征。采用单变量受试者工作特征（Receiver Operator Characteristic, ROC）曲线分析评估筛选出的代谢特征的诊断效能。 **研究结果** PLS-DA得分图显示SSc患者与健康对照存在显著区分度[R²=0.98]，且SSc患者血清中存在明显的代谢异常改变。SSc患者血清的代谢特征为丙氨酸、缬氨酸、肌醇、肌酐、丙酮酸及乳酸水平降低；而乙酸盐及3-羟基丁酸水平显著升高。与预期相反，本研究中SSc患者与健康对照的血清甘露糖水平无显著差异（SSc组：46.0 μM，健康对照组：43.8 μM）。上述多数代谢改变与此前其他临床代谢组学研究[1,2]的报道结果基本一致。SSc患者与健康对照的循环肌醇水平均值分别为29.2 μM与67.4 μM。此外，本研究计算了循环MMR值[肌醇/甘露糖]并在两组间进行比较。与肌醇的变化趋势一致，SSc患者的MMR水平同样显著降低（SSc组均值为0.71±0.32，健康对照组为1.85±0.73）。 **研究结论** SSc患者血清中肌醇及其他内源性代谢物的水平异常，提示SSc患者存在肌醇代谢异常，未来需开展更多研究以明确其在硬皮病病理生物学过程中的作用。 **参考文献** [1] Federica Murgia, Silvia Svegliati, Simone Poddighe, Milena Lussu, Aldo Manzin, Tatiana Spadoni, Colomba Fischetti, Armando Gabrielli, and Luigi Atzori. "Metabolomic profile of systemic sclerosis patients." Scientific Reports 8, no. 1 (2018): 7626. [2] Thomas Bögl, Franz Mlynek, Markus Himmelsbach, Norbert Sepp, Wolfgang Buchberger, and Marija Geroldinger-Simić. "Plasma metabolomic profiling reveals four possibly disrupted mechanisms in systemic sclerosis." Biomedicines 10, no. 3 (2022): 607.