| IntroductionAccurate diagnosis of maturity onset diabetes of the young（MODY）is required in order to select appropriate treatment options and to assess prognosis.The aim of this study was to construct clinical prediction model that could be used to differentiate MODY2,MODY3,and type 1 diabetes mellitus（T1DM）in young subjects.MethodsTwelve patients with MODY3 and 29 patients with MODY2 were characterized and compared to 26 patients with T1DM.These three groups were matched for age and gender.Clinical profiles of the 67 patients were collected.Receiver operating characteristic（ROC）curves were used to identify the optimal cutoff values of clinical indicators.ResultsCompared to patients with T1DM,subjects with MODY3 had higher fasting C-peptide levels（1.34±1.51 vs.0.29±0.22 ng/mL;P<0.001）and lower high-sensitivity C-reactive protein（hsCRP）levels（0.18±0.15 vs.1.22±1.49 mg/L,P=0.004）;patients with MODY2 had lower hsCRP（0.37±0.39 vs.1.22 ± 1.49 mg/L;P=0.003）,total cholesterol（4.12±0.68 vs.4.61±0.81 mmol/L;P=0.034）,and low-density lipoprotein cholesterol（LDL-C）（2.24±0.68 vs.2.67±0.79 ng/L,P=0.002）levels and higher fasting C-peptide levels（0.96± 0.42 vs.0.29±0.22 ng/mL,P=0.002）.The ROC-derived hsCRP values for discriminating MODY2 from T1DM,MODY3 from T1DM,and MODY3 from MODY2 were 0.675,0.833,and 0.763,respectively.The ROC derived fasting C-peptide levels for discriminating MODY2 from T1DM and MODY3 from T1DM were 0.951 and 0.975,respectively.The ROC-derived total cholesterol and LDL-C values for discriminating MODY2 from T1DM were 0.670 and 0.662,respectively;the ROC-derived triglyceride value for discriminating MODY3 from MODY2 was 0.756.Additionally,a combination of indicators permitted better discrimination of MODY subtypes than any single parameter.ConclusionOur findings suggest that fasting C-peptide,hsCRP,and lipid levels permit good discrimination among MODY2,MODY3,and T1DM.These clinical indicators could be used as markers of MODY2 and MODY3 in young patients with diabetes.BackgroundMaturity-onset diabetes of the young（MODY）constitutes 1-5%of all diabetic patients.Accurate diagnosis of MODY is required to select appropriate treatment.The patients who were diagnosed as MODY showed distinct clinical manifestations and had individualized treatment.However,the mechanism is still not clear.In recent years,metabolomics research has focused on the identification of small molecule contour maps in vivo,which has greatly promoted the research on the biomarkers and mechanism of diabetes.Here,we aimed to explore the role of the mutations in GCK and HNF1A in metabolites by detecting the differences of the small metabolic molecules between MODY and healthy controls.Further,by comparing metabolic small molecules among patients with MODY2,MODY3 and type 1 diabetes mellitus（T1DM）,and analyzing the correlations between metabolites and clinical indicators,we aimed to partly explain the underlying mechanism of low risk of complications in patients with MODY2 and the effectiveness of low-dose sulfonylureas in patients with MODY3.This study may help with improving prognosis of diabetes and its complications.Additionally,the present study evaluated the feasibility of using small metabolic molecules as diagnostic markers for differentiating HNF1A/GCK mutation carriers from T1DM.MethodsFasting serum samples were collected from MODY2（n=33）,MODY3（n=17）,T1DM（n=34）,and healthy controls（n=30）.Sanger sequencing was used to detect mutations in GCK and HNF1A genes.Metabolomics was performed using non-targeted ultra performance liquid chromatography mass spectrometry（UPLC-MS）.Results（1）Clinical investigation:compared to T1DM and normal controls,patients who were diagnosed as MODY showed more obvious of family history of diabetes and lower levels of high-sensitivity C-reactive protein（hsCRP）.Patients with T1DM had the highest levels of fasting blood glucose,glycated hemoglobin,and glycated albumin and lowest levels of fasting C-peptide.Among the three diabetic groups,patients with MODY2 showed the lowest levels of triglyceride.The above correlations were remaind after age grouping.（2）Carrying of genetic mutations:In this study,34 patients with MODY2 carried a total of 19 genetic mutations,of which 4 were the first reported mutations,namely c.263T>G,c.553C>T,c.584T>C and c.803804insA.A total of 17 patients with MODY3 were included in this research,carrying altogether 7 gene mutations,of which 2 were first reported,namely c.616T>A and c.758G>A.These 6 mutations were proved to be disease-causing mutations after verification by prediction software.（3）Metabolic pathways analysis:The different metabolites in the 4 groups were enriched in three pathways,namely glycerophospholipid metabolism,sphingolipid metabolism and phenylalanine metabolism pathway.Among them,glycerophospholipid metabolism is the most relevent pathway of MODY2 and normal controls,and is also the most relevent pathway of T1DM and two subtypes of MOD Y.Sphingolipid metabolism is the most relevent pathway of MODY3 and normal controls and MODY2 subjects.Additionally,phenylalanine metabolic pathway is involved in the regulation of glucose metabolism and is the enriched pathway of the different metabolites in the 4 groups.（4）Differential metabolites:Five metabolites were selected from the above three pathways,namely glycerophosphocholine,LysoPC（18:2（9Z,12Z））,sphinganine,galactosylceramide（d18:1/16:0）and L-Phenylalanine.Comparison with healthy controls,the levels of glycerophosphocholine,LysoPC（18:2（9Z,12Z））,galactosylceramide（d18:1/16:0）and L-Phenylalanine were significantly reduced in patients with MODY2,and all the five metabolites were decreased in MODY3.Comparison with T1DM patients,the levels of glycerophosphocholine,LysoPC（18:2（9Z,12Z））,sphinganine and L-Phenylalanine were significantly decreased in both MODY2 and MODY3.（5）The relationship between metabolites and clinical indicators:After adjuesting for gender and age,glycerophosphocholine was negatively correlated with fasting C peptide（r=-0.236,P=0.019）levels,and positively correlated with HbA1c（r=0.355,P=0.001）,LDL-C（r=0.202,P=0.042）and Hs-CRP（r=0.288,P=0.003）levels;LysoPC（18:2（9Z,12Z））was negatively correlated with fasting C peptide（r=-0.304,P=0.002）levels and positively correlated with HbAlc（r=0.229,P=0.042）levels;Sphinganine was negatively correlated with fasting C peptide（r=-0.233,P=0.021）levels;L-Phenylalanine was significantly positively correlated with HbAlc（r=0.297,P=0.008）levels.（6）Potential biomarkers among the three diabetic groups:Among the 5 metabolites,one metabolite was selected as candidate biomarker for the distinction between MOD Y2 and MODY3,namely,glycerophosphocholine.The area under the curve（AUC）was 0.795.Two metabolites were selected as candidate biomarkers for the distinction between MODY2 and T1DM,namely,glycerophosphocholine and L-Phenylalanine,and the AUC for the two biomarkers were 0.989 and 0.885,respectively.Four metabolites were selected as candidate biomarkers for the distinction between MODY3 and T1DM,namely,glycerophosphocholine,Ly soPC（18:2（9Z,12Z））,sphinganine and L-Phenylalanine,and the AUC for these biomarkers were 1.0,0.933,0.888 and 0.882,respectively.Additionally,the combination of clinical indicators（triglyceride and hypersensitive C-reactive protein）and metabolic indicators（glycerophosphocholine）can help with the distinction of the two MODY subtypes.ConclusionsThis study suggested that the glycerophospholipid metabolism may associate with MODY2 and its low risk of complications;the sphingolipid metabolism may associate with MODY3 and its properties of the effectiveness of sulfonylureas.The metabolites glycerophosphocholine,LysoPC（18:2（9Z,12Z））,sphinganine and L-Phenylalanine might be potential biomarkers for the differentiation among MODY2,MODY3 and T1DM.This study reported the metabolite composition of common monogenic diabetes（MODY2 and MODY3）,which provides a basis for further understanding of the pathogenesis and characteristic clinical manifestations of MODY2 and MODY3. |
PDF Full Text Request