The Causal Association Between Obesity And Early Pubertal Timing In Children: A Bidirectional Mendelian Randomization Analysis Based On The Longitudinal Study

ObjectiveBased on the longitudinal study design,this bidirectional Mendelian randomization（MR）study aimed to explore the potential bidirectional causal relationship between childhood obesity and early pubertal timing.The genetic risk scores（GRS）,constructed by multiple single nucleotide polymorphisms（SNP）,were applied as instrumental variables for childhood obesity and early pubertal timing,respectively.Methods1.A total of 1199 boys and 1033 girls from grade 1 to 4 were selected from the Guangzhou primary and secondary school students cohort of body composition and health promotion.At baseline and follow-up,physical examination,body composition measurement,pubertal development assessment and questionnaire were performed.Body weight,percentage body fat（PBF）,and muscle mass were measured by the Bioelectrical Impedance Analysis.Due to the age and gender specificity of children’s body composition,PBF was standardized by gender and age（Z-score was calculated）.P₈₅ of gender-age specific PBF was used as the cut-off value to define obesity.The testicular volume of boys and the breast development of girls were measured by palpation and visual diagnosis according to the Tanner stages.The early pubertal timing was defined according to the median age of Tanner II-V.The basic characteristics,living behavior and other information of the subjects were collected through questionnaires.In traditional observational study,the generalized linear mixed-model was used to investigate the observed effect of childhood obesity（independent variable）on early pubertal timing（dependent variable）and the observed effect of early pubertal timing（independent variable）on childhood obesity（dependent variable）.2.DNA was extracted from blood samples of subjects during baseline survey.Childhood obesity related SNPs and puberty related SNPs were screened from previous genome-wide association studies.Genotyping was performed by real-time quantitative polymerase chain reaction（PCR）.Unweighted,weighted and order GRS of childhood obesity or early pubertal timing were constructed as instrumental variables,respectively.In mendelian randomization analysis,the Wald ratio method was used to calculate the causal effects based on the generalized linear mixed-model,and the bootstrap method was used to calculate the 95%Confidence intervals（CI）of the causal effects.Pleiotropic diagnosis and sensitivity analysis were performed using the MR-PRESSO method.Results1.Traditional observational analysis:the generalized linear mixed-model of obesity-related indicators（independent variables）on the early pubertal timing（dependent variables）showed a negative association between Z-scores of PBF and the risk of early pubertal timing（OR=0.73,95%CI=0.65-0.82,P<0.001）.When compared with boys with non-obesity,obesity boys had a 44%lower risk of early pubertal timing（OR=0.56,95%CI=0.40-0.78,P<0.001）.In girls,conversely,Z-scores of PBF were positively associated with the risk of early puberty（OR=1.79,95%CI=1.58-2.03,P<0.001）.As compared with girls with non-obesity,obesity girls had a 1.41-fold increased risk of early puberty（OR=2.41,95%CI=1.82-3.20,P<0.001）.The generalized linear mixed-model of early pubertal timing（independent variable）on childhood obesity（dependent variable）showed that there was negative association between early pubertal timing and increased Z-score of PBF（β=-0.12,95%CI=-0.18--0.06,P<0.001）,and non-significant association between early pubertal timing and obesity in boys（P>0.05）.In girls,early pubertal timing was significantly associated with increased Z-scores of PBF（）and increased risk of obesity（Z-score of PBF:β=0.10,95%CI=0.04-0.16,P<0.001;obesity:OR=1.78,95%CI=1.16-2.72,P=0.008）.2.Mendelian randomization analysis:when analyzed the influence of childhood obesity on early pubertal timing,no significant association was observed between obesity-related indicators and early puberty in boys.In girls,when used weighted GRS as instrumental variable,the risk of early pubertal timing increased by 2.01-fold（OR=3.01,95%CI=2.21-8.53,P=0.008）for each unit increase in Z-scores of PBF.As compared to girls with non-obesity,obesity girls had a 51%increased risk of early puberty（OR=1.51,95%CI=1.27-2.09,P<0.001）.Similar results were seen between obesity and girls’pubertal timing when applied order GRS as the instrumental variable.In the sensitivity analysis,there was no horizontal pleiotropy for the candidate SNPs of obesity by the global test.MR-PRESSO analysis suggested that the risk of early puberty increased by 3.77-fold（OR=4.77,95%CI=2.91-7.81,P<0.001）for each unit increase in Z-score of PBF.When compared to girls with non-obesity,obesity girls had an 54%increased risk of early pubertal timing（OR=1.54,95%CI=1.28-1.86,P<0.001）.When explore the influence of early pubertal timing on childhood obesity,using weighted GRS as instrumental variable,boys with early puberty had 0.08units higher Z-score of PBF（β=0.08,95%CI=0.04-0.10,P<0.001）,as compared with boys with normal puberty.Similar results were seen between boys’pubertal timing and obesity when used unweighted GRS and order GRS as instrumental variables.No significant associations of early pubertal timing were observed with obesity in boys,and Z-score of PBF and obesity in girls（P>0.05）.In the sensitivity analysis,there was no horizontal pleiotropy for the candidate SNPs of puberty by the global test.MR-PRESSO suggested that compared with boys with normal pubertal timing,those with early puberty had0.19 units higher Z-score of PBF（β=0.19,95%CI=0.04-0.34,P=0.026）.ConclusionThis bidirectional mendelian randomization study suggested a sexual dimorphisms in the relationship between childhood obesity and early pubertal timing.In girls,increased body fat or obesity may raise the risk of early puberty,but no such relationship was found in boys.Early puberty is associated with an increase in body fat percentage in boys,but it may not affect body fat percentage or obesity risk in girls.