| ObjectivesAbout 10%-15%of SGA children fail to achieve spontaneous catch-up growth,and the risk of short stature in adulthood is significantly increased.Recombinant human growth hormone(rhGH)has become the main treatment method for SGA children with short stature.The spontaneous catch-up growth of SGA children is closely related to the increased risk of metabolic disorders in the near and long term.However,previous studies have found that rhGH treatment does not increase the risk of metabolic diseases in SGA children,and can even improve the lipid metabolism,suggesting that there are different regulatory mechanisms between spontaneous catch-up growth and catch-up growth after rhGH treatment in SGA children.The gut microbiota is closely related to the occurrence and development of various metabolic diseases,but it is currently unclear whether rhGH can participate in the regulation of gut microbiota.Thus,this study aimed to investigate the changes in the gut microbiome in SGA children with short stature after receiving recombinant human growth hormone(rhGH)treatment,as well as the correlation between changes in glucose and lipid metabolism and changes in gut microbiome.Furthermore,we aimed to preliminarily explore the role of gut microbiome in the regulation of glucose and lipid metabolism in SGA children with short stature during rhGH treatment.MethodsThis study was a prospective clinical cohort study,which included 34 SGA children with short stature who had not received rhGH treatment previously.The children were not given any intervention for the first 3 months after enrollment as a blank self-control period,and then rhGH treatment was initiated.After treatment,the children were followed up every 3 months until 12 months of rhGH treatment.Clinical data such as height,weight,and laboratory tests before and after treatment were collected.Also,stool samples were collected.16S rRNA sequencing technology was used to analyze the species annotation,diversity analysis,difference analysis between groups,differential species analysis,and functional prediction on the gut microbiome.SPSS 24.0 software was used for statistical analysis,and Spearman correlation was used to analyze the correlation between serum metabolic-related indicators and differential gut microbiota.Results1.Baseline characteristics and rhGH treatment responseA total of 34 SGA children with short stature were included,including 10 girls and 24 boys.The median age was 6.0(4.9-8.1)years old,and the median bone age was 3.8(2.0-5.9)years old.The height of the children increased significantly after rhGH treatment.The standard deviation score(SDS)of the children’s height before treatment(-2.12 ± 0.71)showed statistically significant differences compared to 3 months after treatment(-1.69 ± 0.74),6 months after treatment(-1.56 ± 0.77),9 months after treatment(-1.20 ± 0.80),and 12 months after treatment(-1.16 ± 0.79),with P values of 0.008,0.001,<0.001,and<0.001,respectively.2.The effect of rhGH treatment on glucose and lipid metabolism in SGA children In terms of glucose metabolism,after 12 months of rhGH treatment,the fasting blood glucose(FBG),fasting insulin(FINS),and homeostasis model assessment-insulin resistance(HOMA-IR)of the SGA children were significantly increased compared to those before treatment,but still within the normal range.Among them,FBG increased from 4.79(4.50,5.18)to 5.30(4.90,5.50)mmol/L(P=0.013);FINS increased from 3.90(2.60,6.20)to 9.10(5.65,15.98)μIU/mL(P=0.005);HOMA-IR increased from 0.78(0.52,1.78)to 1.67(1.30,3.91)(P=0.015).In terms of lipid metabolism,there was no significant change in the body mass index(BMI)SDS of the SGA children before and after rhGH treatment.However,after 9 months of treatment,triglycerides(TG)significantly increased compared to that before treatment,from 0.42(0.31,0.50)to 0.68(0.48,0.90)mmol/L(P=0.018),but decreased at 12 months of treatment;After 3 months of treatment,there was a significant decrease in low-density lipoprotein cholesterol(LDL-C).After 6 months of treatment,it decreased from 2.41(2.22,2.89)to 2.26(1.91,2.73)mmol/L(P=0.012)and maintained until 9 months of treatment;After 3 months of treatment,the level of lipoprotein a[Lp(a)]significantly increased compared to that before treatment,rising from 45.00(18.00,232.00)to 55.00(28.50,215.50)mg/L(P=0.007).However,it decreased after 9 months of treatment and was no longer significantly higher than that before treatment.3.The changes in gut microbiota after rhGH treatmentThe number of operational taxonomic units(OTUs),Chao 1,ACE and PD_whole_tree index reflecting the species abundance in Alpha diversity in SGA children during the blank self-control period were similar and had no significant change.However,the above 4 indexes were significantly higher after 12 months of rhGH treatment than those before treatment,suggesting that the species abundance of gut microbiota increased after treatment.In addition,the composition of gut microbiota also changed after rhGH treatment.There was no significant change in the Beta diversity index during the blank self-control period,but at 9 and 12 months of treatment,the Beta diversity index was significantly higher than that before treatment,reflecting significant differences in microbial communities before and after rhGH treatment.Among them,the relative abundance of Lachnospirales,a biomarker of metabolic disorders,was significantly reduced after 12 months of treatment compared to that before treatment.The 18 genera in Proteobacteria,a biomarker of intestinal dysbiosis,were significantly lower than that before treatment.The 3 genera in Bacteroidota,related to the improvement of lipid metabolism,were significantly increased compared with those before treatment,indicating an improvement in gut microbiota status after rhGH treatment.4.The correlation analysis between differential indexes of serum glucose and lipid metabolism and differential gut microbiotaThe MetaStat results showed that after only receiving rhGH treatment for 3 months,the relative abundance of f Peptostreptococcaceae significantly decreased compared to that before treatment and continued until 12 months of treatment;After 9 months of treatment,the relative abundances of f_Butyricicoccaceae and g_Blautia were significantly reduced compared to those before treatment.Besides,the change in f_Peptostreptococcaceae was significantly negatively correlated with the change in TG level(r=-0.554,P=0.005).Also,the changes in f_Butyricicoccaceae and g_Blautia were significantly negatively correlated with the change in FBG level(r=-0.438,P=0.032;r=-0.545,P=0.006).ConclusionsSignificant changes in glucose and lipid metabolism indexes were observed in SGA children with short stature after rhGH treatment.In terms of glucose metabolism,the levels of FBG,FINS,and HOMA-IR were significantly increased.In terms of lipid metabolism,the levels of TG and Lp(a)were significantly increased while the level of LDL-C was significantly decreased.In addition,after 12 months of rhGH treatment,the abundance of gut microbiota significantly increased compared to that before treatment,and the composition of microbial communities also changed significantly.Among them,the changes of f_Peptostreptococcaceae,f_Butyricicoccaceae and g_Blautia were significantly negatively correlated with the changes in FBG and TG levels,indicating that the changes of gut microbiota in SGA children with short stature after rhGH treatment may be involved in the regulation of the glucose and lipid metabolism.ObjectivesThe impact of SGA during the fetal period will permanently change the physiological and metabolic conditions,leading to an increased risk of short-term and long-term metabolic disorders,and such metabolic disorders are closely related to the spontaneous catch-up growth of SGA.Previous studies demonstrated that metabolic disorders during spontaneous catch-up growth of SGA rats were related to changes in gut microbiota status.This study aimed to establish SGA rat models by restricting food intake during pregnancy and administer VSL#3 probiotics,a mixed preparation consisting of eight probiotic strains,by gavage,as well as observe the effect of VSL#3 probiotic intervention on metabolic disorders during spontaneous catch-up growth in SGA rats.Also,this study explored the mechanisms with a focus on the composition and function of gut microbiota,the metabolites of gut microbiota and the related molecules of the signaling pathways,and the expression of liver lipid metabolismrelated genes.MethodsThe modeling method for SGA rats was based on maternal malnutrition during pregnancy.After conception,female Sprague-Dawley rats were randomly divided into a normal-feeding group and a restricted-feeding group.The normal-feeding group fed freely;the restricted-feeding group reduced the food intake by 50%from the first day of conception.Both groups of female rats underwent natural childbirth,with the normal-feeding group serving as the control group(n=15)and the restricted-feeding group SGA rats being randomly divided into two groups(n=7/group)starting from 3 weeks old.The SGA rats were given VSL#3 probiotic solution or PBS solution by gavage daily until the age of 10 weeks.The changes in body length,body weight,fat weight,food intake,glucose and lipid metabolism were observed.The changes in gut microbiota composition,gut microbiota metabolites,leptin,gastrointestinal hormones,inflammatory factors,and the expression of liver lipid metabolism related genes were also detected.Results1.The effects of VSL#3 intervention on the metabolic disorders during spontaneous catch-up growth in SGA ratsThe weight of white adipose tissue(WAT)of SGA rats in the VSL#3 group was significantly lower than that in the PBS group and the control group,at 7.2±1.3g vs.9.3±2.2g and 9.4±1.7g,respectively(P values were 0.033 and 0.021,respectively).The body fat percentage of SGA rats in the VSL#3 group was also significantly lower than that of SGA rats in the PBS group(2.5%vs.3.2%,P=0.030).In addition,the FBG level of SGA rats in the VSL#3 group was significantly lower than that in the PBS group and control group,with values of 4.29±0.59 mmol/L vs.5.77±1.37 and 5.54±1.02 mmol/L(P values of 0.013 and 0.014,respectively).The serum total cholesterol(TC)level of SGA rats in the PBS group was significantly higher than that in the control group(3.71 ±0.37 mmol/L vs.3.24±0.60 mmol/L,P=0.045),but there was no significant difference in the TC level between SGA rats in the VSL#3 group and the control group.Similarly,the serum LDL-C level of SGA rats in the PBS group was significantly higher than that in the control group(1.57±0.16 mmol/L vs.1.37±0.26 mmol/L,P=0.040),but there was no significant difference in the LDL-C level between SGA rats in the VSL#3 group and the control group.2.The effects of VSL#3 intervention on the gut microbiota in SGA ratsAfter VSL#3 intervention,there was a significant change in the composition of gut microbiota in the SGA rats,which showed significant differences in the microbial communities compared to the SGA rats in the PBS group and the normal rats in the control group.Among them,as the species with the highest relative abundance at the phylum level,the relative abundance of Firmicutes in the VSL#3 group showed a significant decrease compared to the PBS group(55.86%vs.74.03%,P=0.027),and also showed a decreasing trend compared to the control group.In addition,as the species with the second highest relative abundance of the Phylum level,the relative abundance of Bacteroidota in the VSL#3 group was significantly higher than that in the PBS group(41.89%vs.24.26%,P=0.028),and there was also an upward trend compared to the control group,suggesting that the gut microbiota status of SGA rats in the VSL#3 group was improved.3.The effects of VSL#3 intervention on short chain fatty acids(SCFAs)produced by gut microbiota metabolism in SGA ratsVSL#3 intervention can improve the partial decrease in the SCFAs levels during the spontaneous catch-up growth in SGA rats,specifically manifested by a significantly lower butyric acid level in the PBS group compared to the control group(0.261±0.126μg/mg vs.0.261±0.126 μg/mg,P=0.049),but there was no significant difference in the butyric acid level between the VSL#3 group and the control group.Similarly,the levels of isobutyric acid and isovaleric acid in the PBS group were significantly lower than those in the control group,with values of 0.037±0.017 μg/mg vs.0.057±0.018 μg/mg and 0.029±0.010 μg/mg vs.0.042±0.012 μg/mg(P values of 0.015 and 0.023,respectively),but there was no significant difference in the isobutyric acid and isovaleric acid levels between the VSL#3 group and the control group.In addition,as the target of SCFAs,the serum leptin level of SGA rats in the VSL#3 group was significantly higher than that in the PBS group and the control group,with values of 1.19±0.15 ng/mL vs.0.89±0.12 ng/mL and 1.00±0.10 ng/mL,respectively(P values<0.001 and 0.001,respectively).Meanwhile,the food intake of the SGA rats in the VSL#3 group was significantly reduced compared to the SGA rats in the PBS group and the normal rats in the control group,at 14.7±1.7 g/d vs.16.9±0.8 g/d and 18.7±1.5 g/d(P values of 0.008 and<0.001,respectively).4.The effects of VSL#3 intervention on the expression of lipid metabolism related genes in the liver tissues of SGA ratsThe sterol regulatory element binding protein-1c(SREBP-1c)and its target genes,fatty acid synthase(FAS)and acetyl-CoA carboxylase(ACC),were significantly downregulated in the liver tissues of the SGA rats in the VSL#3 group.Among them,the expression level of SREBP-1c in the liver tissues of the SGA rats in the VSL#3 group decreased to 60.80%of that in the PBS group,FAS expression decreased to 60.10%of that in the PBS group,and ACC expression decreased to 74.08%of that in the PBS group.ConclusionsAfter the spontaneous catch-up growth,SGA rats experienced lipid metabolism disorders,manifested by elevated serum TC and LDL-C levels.VSL#3 intervention can improve the lipid metabolism disorders that occurred during the spontaneous catch-up growth of the SGA rats,and reduce their WAT weight and body fat percentage while not affecting their body length catch-up.The specific mechanisms may include:(1)an improvement in the gut microbiota status of the SGA rats and an increase in the relative abundance of beneficial bacteria;(2)The recovery of metabolites of gut microbiota SCFAs levels,and the increase in the downstream serum leptin level leading to a decrease in the food intake;(3)The expression of SREBP-1c and its target genes FAS and ACC related to fat synthesis in the liver tissues of the SGA rats were downregulated.ObjectivesSmall for gestational age(SGA)is defined as a birth weight below a distribution-based gestational age threshold,usually the 10th percentile.SGA newborns are at increased risk of perinatal morbidity and mortality.Thus,if the condition is identified timely and accurately before delivery,the fetus and the mother can be given closer monitoring and timely delivery,thereby reducing adverse fetal outcomes.Exposures to radiation and pesticides have been associated with an increased risk of delivering SGA newborns.However,there are no tools to predict SGA newborns in pregnant women exposed to radiation or pesticides before pregnancy.We aimed to develop an array of machine learning(ML)models to predict SGA newborns in women exposed to radiation or pesticides before pregnancy.MethodsPatients’ data were obtained from the National Free Preconception Health Examination Project from 2010 to 2012,which was carried out in 220 counties from 31 provinces or municipalities.All singleton live newborns with complete birth records and gestational age of more than 24 weeks were included in the study,and then we selected newborns whose mothers were exposed to radiation or pesticides in their living or working environment before pregnancy.The dataset was divided randomly into the training set(80%)and the testing set(20%)for the development and validation of the prediction models.The traditional logistic regression method and seven mainstream ML algorithms were compared for solving the binary classification of SGA prediction,with the area under the receiver-operating-characteristic curve(AUC)as the main index to measure the performances of the models.The ML algorithms included random forest(RF),gradient boosting decision tree(GBDT),extreme gradient boosting(XGBoost),light gradient boosting machine(LGBM),category boosting(C atBoost),support vector machine(SVM)and multi-layer perceptron(MLP).A post-hoc explainability based on the Shapley Additive Explanation(SHAP)model was used to identify and interpret the most important features that contribute to the prediction outcome.ResultsA total of 455 newborns whose mothers were exposed to radiation before pregnancy were included,with the occurrence of 60 SGA births(13.2%).757 newborns whose mothers were exposed to pesticides before pregnancy were included,with the occurrence of 98 SGA births(12.9%).As for the SGA prediction in women exposed to radiation before pregnancy,the model obtained by XGBoost achieved the highest AUC in the testing set[0.844,95%confidence interval(CI):0.713-0.974].All models showed satisfied AUCs,except for the logistic regression model(AUC:0.561,95%CI:0.355-0.768).After feature selection by recursive feature elimination(RFE),15 features were included in the final prediction model using the XGBoost algorithm,with an AUC of 0.821(95%CI:0.650-0.993).As for the SGA prediction in women exposed to pesticides before pregnancy,similarly,all models showed satisfied AUCs,except for the logistic regression model(AUC:0.691,95%CI:0.554-0.828).The model obtained by CatBoost achieved the highest AUC in the testing set(0.855,95%CI:0.752-0.959).After feature selection by RFE,15 features were included in the final prediction model using the CatBoost algorithm,with an AUC of 0.811(95%CI:0.675-0.947).ConclusionsML algorithms can generate robust models to predict SGA newborns in high-risk pregnant women,which can yield more effective predictions than the conventional logistic regression model.Including only 15 features,the models based on the ML algorithms can achieve effective prediction of SGA,which can be considered as an effective prediction tool for SGA newborns in high-risk pregnant women. |
PDF Full Text Request