The Relationship Between Cereal Fiber Intake And Risk Of Gestational Diabetes Mellitus

As one of the most popular complications, gestational diabetes mellitus (GDM) has been defined as any degree of glucose intolerance with onset or first recognition during pregnancy. Depending on the studied population and employed diagnostic tests, ranging from1%to14%pregnacies are complicated by GDM, resulting in more than200000cases annually. The incidence of GDM in Chinese population is estimated to be5%. More than21million live births were affected by diabetes during pregnancy in2013. The health risk originated from GDM is not suspended with the ending of pregnancy, but throughout the whole life course. GDM is associated with increased risk of adverse perinatal events, including pregnancy-induced hypertension, macrosomia, primary cesarean delivery, and preterm birth. Meanwhile, GDM women are more prone to developing type2diabetes mellitus (T2DM). Maternal diabetes mellitus also has long-term consequences for obesity and impaired glucose tolerance in offspring during childhood and early adulthood.The etiology and pathogenesis of GDM is complex, with advanced maternal age, a family history of T2DM and prepregnancy obesity as its well-recognized risk factors. With the development of economy and society, major changes occurred in people’s eating habits. The westernized dietary habits, including high-energy diet and low physical activity, have been undisputedly considered contributing to the prevalence of GDM. Therefore, the associations between dietary factors and GDM arise great attention.Dietary fiber (DF), especially cereal fiber, has been suggested to participate in reducing excessive adiposity, which is the most commonly modifiable risk factor of GDM. However, the actual intake of DF is quitely low compared to the recommended intake level. In China, DF intakes decreased from1989to2006, and the reduction of cereal fiber intake may explain this trend. Recently, some epidemiologic studies suggested that higher whole grain intakes were associated with lower risk of chronic diseases, such as GDM. The plenty of phytochemicals and cereal fiber are deemed to play great role in the redued risk established in human studies. Therefore, the study on DF or cereal fiber for GDM is urgently needed. To our knowledge, Zhang et al. first attempted to investigate whether pregravid intakes of DF was related to GDM risk in Nurses’Health Study Ⅱ, a cohort containing U.S. women of reproductive age. After multiple adjustment, higher dietary total fiber, cereal and fruit fiber were associated with a reduced risk of GDM.However, in nutritional epidemiology, relatively low accuracy in measuring the intake of foods and nutrients is a major problem. To some extent, using a biomarker of DF or cereal fiber could overcome the problem associated with dietary assessment methods, as well as validating other methods. Therefore, researchers have made great effort to seek specific biomarkers of dietary nutrients. Alkylresorcinols (AR) are amphiphilic1,3-dihydroxy-5-alkylbenzene homologs with odd-numbered alkyl side chains containing from C17:0to C25:0. They are mainly found in the outer parts of wheat and rye. Currently, intact AR homologs have been detected in human plasma and erythrocyte. In addition, the two most important AR metabolites,3,5-dihydroxybenzoic acid (DHBA) and3-(3,5-dihydroxyphenyl)-propanoic acid (DHPPA), were also identified in human plasma and urine. Moreover, sample collection time exerts less influence on AR metabolites for the reason that the half-lives of DHBA and DHPPA were longer than intact AR homologs. Therefore, AR metabolites are more suitable to be employed in epidemiologic studies.Till now, studies focused on the association between cereal fiber intake and GDM in Chinese population are not available. Therefore, based on the previous studies, we aimed to investigate the above relationship in a case-control study. In order to overcome the defects of nutritional epidemiology, we utilized plasma AR metabolites as a potential biomarker of cereal fiber intake. The main contents were as follows: Part One Quantification of plasma AR metabolites by LC-MS/MSObjective:On the basis of previous detection methods, we aimed to establish a liquid chromatography-tandem mass spectrometry (LC-MS/MS) method, with higher sensitivity and shorter running time, to quantificate plasma AR metabolites levels.Methods:Agilent1200HPLC system and6460Triple Quadrupole Mass Spectrometer equipped with electrospray ionization (ESI) were used to the synchronous detection of DHBA, DHPPA and syringic acid (SyrA). Optimized sample preparation methods and chromatography as well as mass spectrometry conditions were employed to improve sensitivity and other indicators. The method was validated according to the Guidelines on Bioanalytical Method Validation issued by European Medicine Agency. In addition, the comparative analysis between LC-MS/MS and HPLC-CEAD was conducted.Results:(1) Only three steps, including enzymatic hydrolysis, liquid-liquid extraction, concentration and reconstitution, were used for sample preparation after optimization. The required sample volume was50μL, and the internal standard (SyrA) was added at a concentration of50ng/mL. DHBA data was collected under the negative ionization mode with multiple reaction monitoring (MRM), while DHPPA and SyrA data were collected under the positive ionization. In addition, the optimized mass transition ion pairs (m/z) for quantitation were observed at m/z153.1to109.1,183.1to165.1, and199.1to140.1for DHBA, DHPPA and SyrA, respectively.(2) Using our method, AR metabolites and SyrA were well seperated, without obvious interference of endogenous substances. The correlation coefficient of calibration curve was greater than0.999and the lower limit of quantitation was established at0.4ng/mL. Additionally, the accuracy was between85%and115%, while the precision was less than10% in different concentrations of quality controls. Moreover, the rescoveries of both analytes were90% to115%.(3) Compared to HPLC-CEAD, this method required reduced sample volume, but revealed higher sensitivity and shorter running time.Conclusion:The established LC-MS/MS method in our study was accurate and reliable. With improved sensitivity and running time, it is more suitable to be employed in epidemiologic studies. Part Two The relationship between plasma AR metabolites level and cereal fiber intakeObjective:To explore the relationship between plasma AR metabolites level and cereal fiber intake, in order to verify whether it could be considered as a potential biomarker of cereal fiber intake in the recruited volunteers.Methods:25healthy women volunteers,20-35years old, were recruited. All subjects agreed to consume their normal diet during the study. Basic information was collected by questionnaire or anthropometric measurements. Fasting bloods were collected immediately after dietary survey. Information of cereal fiber intakes came from24-hour dietary recalls and plasma AR metabolites (DHBA and DHPPA) were measured by LC-MS/MS.Results:(1) The data of5volunteers were deleted due to insufficient calorie intake (<600kcal/d). The mean age and BMI of the study subjects were23.80years old and20.20kg/m2. The mean intake of cereal fiber was1.66g/d. In addition, the median levels of plasma DHBA, DHPPA and total metabolites (DHBA+DHPPA) were0.46(0.41-0.51) μg/L,1.40(1.11-2.23) μg/L, and2.00(1.36-2.66) μg/L, respectively.(2) There was no notable correlation between plasma DHBA level and cereal fiber intake (P>0.05). After multiple adjustment, plasma DHPPA level was significantly correlated with cereal fiber intake (r=0.568, P=0.043). As for total AR metabolites, it exerted significant correlation with cereal fiber intake, with or without adjustment (P<0.05). Additionally, age, BMI, energy intake and plasma lipid status strengthened the above relationship to a certain degree (r=0.579, P=0.038).(3) Plasma DHBA, DHPPA and total AR metabolites did not notably correlate with dietary total fiber or vegetable and fruit fiber.Conclusion:Plasma total AR metabolites levels were significantly correlated with cereal fiber intakes. This correlation remained notable after adjustment for potential confounding factors. Therefore, plasma total AR metabolites could be served as a potential biomarker of cereal fiber intake in our study. Part Three Plasma total AR metabolites, biomarkers of cereal fiber intake, and risk of GDMObjective:To compare plasma total AR metabolites levels in GDM patients and non-GDM controls. To assess the correlation between plasma total AR metabolites levels and lipid or other related parameters and its association with GDM, in order to find out the potential relationship between cereal fiber intake and GDM.Methods:We performed a case-control study involving all eligible pregnancy women who took oral glucose tolerance test (OGTT) during a period of August2012to March2013at the endocrine laboratory of Tongji Hospital affiliated to Tongji Medical College, Huazhong University of Science and Technology. ADA2011criteria was utilized to distinguish cases and controls. Basic information was collected by questionnaire or anthropometric measurements. Plasma glucose and lipid parameters were detected. Plasma AR metabolites levels (DHBA and DHPPA) were measured by LC-MS/MS.Results:(1)66GDM cases, as well as66age-, gestational age-and location-matched controls were included. Compared to controls, GDM patients had higher prepregnancy BMI and higher prevelance of family history of diabetes. When looking for age, gestational age, plasma lipid status, prevelance of smoking, alcohol drinking and hypertension disgnosed, no significant differences were observed.(2) Plasma total AR metabolites levels were4.58(3.35-6.02) μg/L and5.95(4.80-7.89) μg/L, respectively in GDM patients and controls, and displayed notable difference (P<0.001).(3) Plasma total AR metabolites were significantly correlated with fasting plasma glucose (FPG).(4) Reduced odds ratios (OR) for GDM with elevated plasma total AR metabolites levels were observed. The multivariable OR of GDM associated with a1μg/L higher plasma total AR metabolites level was0.630(0.497-0.798). Participants in the higher group of plasma total AR metabolites levels, compared with the lower group, had a significantly reduced risk for GDM. After adjustment for age, gestational age, prepregnancy BMI, smoking, alcohol drinking, hypertension diagnosed, family history of diabetes and plasma lipid status, the adjusted OR was0.329(95%CI0.146-0.742; P for trend=0.007).(5) When plasma total AR metabolites levels were added to a conventional model (including age, prepregnancy BMI, smoking, alcohol drinking, hypertension diagnosed, family history of diabetes and plasma lipid status), the area under curve (AUC) was increased from0.702(95%CI0.612-0.791) to0.779(95%CI0.701-0.857), with significant difference (P=0.048).Conclusion:Plasma total AR metabolites levels were significantly lower in GDM patients compared to controls. Elevated plasma total AR metabolites levels were associated with a reduced risk of GDM, with a strong dose-response trend. The above association remained significant after adjustment for potential confounding factors. Our results indicated that there exist a relationship between cereal fiber intake and risk of GDM. Innovations of our study:1. Based on the previous studies, we established an improved LC-MS/MS method, with higher sensitivity and shorter running time. In addition, we applied this method to the detection of plasma AR metabolites for the first time.2. We verified the correlation between plasma AR metabolites levels and cereal fiber intakes in the recruited volunteers, and suggested that plasma total AR metabolites could be used as a potential biomarker of cereal fiber intake.3. To our knowledge, this is the first attempt to compare plasma AR metabolites levels in GDM patients and controls. In addition, we found that plasma total AR metabolites exerted significant association with risk of GDM. Instead of dietary assessment, we evaluated the potential association between cereal fiber intake and GDM by biomarkers detection for the first time.Limitations and future directions of our study:1. Intact AR homologs have been detected in human red blood cells using a gas chromatography/tandem mass spectrometry (GC-MS/MS) method. Therefore, whether intact AR homologs and AR metabolites levels in erythrocytes possess the ability of being specific biomarkers of cereal fiber intakes is urgently needed.2. The sample size of part three is relatively small. To overcome the inherent weakness of case-control studies (casual relationship could not be clarified), the above association needed to be assessed in prospective cohorts with larger populations.3. Data from dietary survey was not available in part three. As a result, we could not evaluate the correlation between plasma total AR metabolites levels and cereal fiber intakes. In addition, using a combination of self-reported nutrients intake and biomarkers to estimate the risk of GDM could not be conducted. In future studies, data from dietary survey or biomarker detection should be provided in the same individual, separate or combined, to assess the relationship between dietary nutrients intake and risk of chronic diseases.4. Plasma total AR metabolites levels should be utilized in other chronic diseases to assess cereal fiber intakes for their significant correlations.