Skip to main content
Download PDF

Sexual dimorphism in metabolomic and phenotypic spectra of UGT deficiency: findings from the Canadian Longitudinal Study on Aging

Biology of Sex Differences volume 16, Article number: 26 (2025) Cite this article

Abstract

Background

Two of the most frequently deleted genes in the human genome are the UDP-glycosyltransferases UGT2B17 and UGT2B28. They encode metabolic enzymes of the glucuronidation pathway that plays a pivotal role in the maintenance of cellular homeostasis for a variety of small molecule metabolites. These deletions may impact health, yet their effects remain poorly understood. We evaluated the impact of UGT deficiency on the plasma metabolome and examined the association between altered metabolites and health outcomes.

Methods

The metabolomic profiles of 4262 proficient gene carriers were compared with those of 352 UGT2B17-deficient, 97 UGT2B28-deficient, and 20 double-gene-deficient individuals from the Canadian Longitudinal Study on Aging. Significant metabolites found in these comparisons were analyzed for their associations with common diseases.

Results

The unexpectedly broad molecular divergence found in UGT-deficient metabolomes, which affected > 10% of metabolites, implies their significant influence across various metabolite classes—particularly lipids and amino acids — extending beyond their known substrates. The metabolic profiles of UGT2B17-deficient men and UGT2B28-deficient women were most impacted, with UGT2B17 deficiency affecting various metabolites linked to metabolic diseases, arthritis, and osteoporosis. Metabolites impacted by a UGT2B28 deficiency such as amino acids, were linked to metabolic disorders in women.

Conclusion

The findings significantly advance our understanding of the metabolic landscape associated with these frequently deleted genes in the human genome, which may influence susceptibility to various diseases in a sex-specific manner, laying the groundwork for determining their pathological mechanisms and impact on human health.

Highlights

Divergent metabolomes of deficient glycosyltransferase individuals (UGT KO).

Highest impact of UGT deficiency on lipid profiles.

Unique metabolic signatures characterize UGT2B17 KO men and UGT2B28 KO women.

UGT KO metabolic signatures are linked to a variety of diseases.

Plain language summary

The human body constantly produces a variety of small molecules, including steroids, bile acids, fatty acids, and hormones. UDP-glycosyltransferases (UGTs) are a key family of enzymes that help balance these molecules and remove them from the body through a process called glucuronidation. Interestingly, two UGT genes, UGT2B17 and UGT2B28, are often naturally deleted, meaning some people are born without these genes. This gene absence results in a complete deficiency of the associated UGT proteins, potentially disrupting the balance of certain metabolites in the body. To understand how missing UGT2B17 or UGT2B28 affects the body’s full profile of metabolites (known as the circulating metabolome), we compared individuals without one or both of these genes (deletants) to those with both gene copies (references). We found distinct, sex-specific changes in metabolite levels: men lacking UGT2B17 had higher levels of steroids and phospholipids, while women lacking UGT2B28 had lower levels of fatty acids. Additionally, these disrupted metabolite levels were linked to different diseases in a sex-specific way. These insights help clarify how these commonly deleted genes influence metabolism and may lead to disease, opening avenues for future health research.

Background

The blood metabolome mirrors the biochemical activities of different organs and exhibits significant variation between sexes [1]. When combined with genomics, metabolomics has the potential to uncover gene functions and their influence on different aspects of health and disease [2]. Although single nucleotide polymorphisms account for the majority of base-pair variations in human genomes, a small subset of genes undergoes complete germline deletion (herein referred to as knockout, KO), with largely unclear metabolomic and functional implications. Individuals with natural gene KOs offer a unique opportunity to investigate overall gene functions, metabolomic characteristics, and their associations with human health and disease [3,4,5].

Relevant to metabolomic changes, metabolic genes have been identified to harbor adaptive structural variants including complete gene deletions, the frequency of which varies among different ethnic populations [3,4,5], possibly because of differences in diet [6]. The UDP-glycosyltransferase genes UGT2B17 and UGT2B28 stand out because they frequently undergo germline whole-gene deletion, leading to a complete deficiency [3, 5]. In fact, they represent two of the most frequently deleted genes in the human genome [5]. The frequency of UGT2B17 KO varies greatly from 9% for Caucasians to > 70% for Asians [3, 5]; for UGT2B28 KO, the frequency ranges from 2 to 3% for Caucasians and Asians to 10% for Africans [4, 7]. Individuals with either germline KO do not express the corresponding UGT2B17 or UGT2B28 proteins in any tissue, including metabolic-intensive organs such as the liver [8,9,10]. Conversely, gene-proficient individuals express UGT2B17 and UGT2B28 in various tissues, including the metabolically active liver, kidneys, gastrointestinal tissues, bladder and a variety of other organs and cells [11].

UGTs encode enzymes of the glucuronidation pathway that utilize sugar nucleotides as co-substrates. This pathway plays a pivotal role in preserving cellular homeostasis and protecting against drugs and other foreign substances. It regulates the cellular abundance (and thus bioactivity) of endogenous metabolites such as bilirubin and sex steroids [12,13,14,15,16], which is fundamental to cellular metabolic pathways. Moreover, UGTs participate in pharmacokinetics, contributing to the inactivation of more than 50% of commonly prescribed medications and consequently influencing target-site exposure pharmacological effects [12, 17]. A deficiency in a UGT enzyme, such as the bilirubin-conjugating UGT1A1, can be detrimental in humans, leading to the accumulation of unconjugated neurotoxic bilirubin, which is characterized by liver dysfunction [18]. For the UGT2B17 and UGT2B28 enzymes, the absence of their genes is not lethal; however, an increasing number of studies have found that differences in their expression patterns and/or deletions are associated with osteoporosis, autoimmune diseases and cancers [19,20,21,22,23,24,25,26,27,28,29]. However, the underlying mechanisms and the potentially affected metabolites in these conditions remain unknown. There are few known metabolite substrates for these enzymes, including androgenic steroids, with UGT2B28 being particularly understudied [30,31,32,33]. Thus, the detailed spectrum of metabolites that serve as substrates or influenced by the UGT2B17 and UGT2B28 metabolic pathways remains unknown.

Our goal was to enhance our understanding of the metabolic processes associated with the UGT2B17 and UGT2B28 pathways. We analyzed population-based genomic and metabolomic data from the Canadian Longitudinal Study on Aging (CLSA) [34]. This large cohort includes numerous UGT KO individuals, enabling an examination of genetically influenced systemic metabolomic profiles and evaluation of whether the metabolites affected by these genes are linked to common age-related conditions. We hypothesized that the absence of the UGT2B17/UGT2B28 metabolic pathway could lead to alterations in metabolite levels, resulting in significant systemic changes. We suggest that these metabolic changes may affect disease susceptibility. Findings provide evidence for divergent influences of UGT deletions on the plasma metabolome of males and females, with unexpected metabolic changes, which were particularly notable for UGT2B17 KO males and UGT2B28 KO females. The impacted metabolites were linked to prevalent health conditions and chronic diseases, namely obesity, diabetes, hypertension, osteoporosis and arthritis.

Methods

Sex as a biological variable

The study included an equal number of male and female participants. Both overall and sex-based analyses were conducted, and we report the sex-related influence of UGT KOs on the circulating metabolome.

Study population

The CLSA is a longitudinal study established to study the genetic and environmental contributions to human health and diseases and determinants of aging by collecting information on the changing biological, medical, psychological, social, lifestyle and economic aspects of participants’ lives. It comprises 51,338 community-dwelling Canadians aged 45–85 years at recruitment that are prospectively followed every 3 years for at least 20 years or until death [34]. Detailed descriptions of the study population, protocols and data collection have been reported and may be accessed online [35]. The current study focused on baseline genetics and metabolomic data derived from blood samples of the comprehensive CLSA cohort (30,097 participants) collected between 2010 and 2015. The CLSA participants were originally randomly selected from within 25–50 km of 11 data-collection sites in seven Canadian provinces.

Genotyping

Blood samples were collected as previously described [36]. Whole blood buffy coats were isolated from peripheral blood drawn into EDTA vacutainers, following centrifugation at 2000 × g for 10 minutes and removal of the plasma layer. Samples were immediately moved to -80°C storage. The time from blood collection to -80°C storage was under two hours for all participants. Genomic DNA was extracted from buffy coat samples using the purification protocol ”Chemagic DNA Buffy Coat Kit special 200µl prefilling VD151007” on the Chemagic MSM I instrument (Perkin-Elmer article No. CMG-533). The detailed procedure may be accessed online [37]. A total of 26,622 individuals from the CLSA comprehensive cohort were genotyped across 794,409 genetic markers using the Affymetrix UK Biobank Axiom as well as 308 million genetic variants imputed from the Trans-Omics for Precision Medicine program (TOPMed) reference panel as previously described [36]. Data were available in the binary PLINK format (Affymetrix UK Biobank Axiom Array) and BGEN V.1.2 format (TOPMed). PLINK software was used for SNP extraction and assessment of allele frequencies. The occurrence of UGT2B17 and UGT2B28 deletions was established based on the tag SNPs rs2708666 and rs11249532, which are in linkage disequilibrium with the respective deletions [5] and imputed from the genomic data. The UGT2B17 KO group comprised participants with the rs2708666A/A homozygous genotype who were homozygous carriers of UGT2B28 (tagged by rs11249532A/A). The UGT2B28 KO group comprised those with the rs11249532T/T homozygous genotype (e.g., absence of both UGT2B28 copies) who were homozygous carriers of UGT2B17 (rs2708666G/G). Participants with both rs2708666A/A and rs11249532T/T comprised the double KO group. The reference group comprised participants with homozygous rs2708666G/G and rs11249532A/A genotypes, thus carrying both copies of UGT2B17 and UGT2B28.

Metabolomics

Metabolomics analyses were conducted using plasma collected by centrifugation of whole blood collected in EDTA vacutainers and stored at − 80 °C within 2 h of blood collection as described above. Untargeted global metabolomic profiling was conducted for 9992 CLSA participants selected from the Comprehensive cohort to be representative of each data collection site, age groups and sex. Briefly, 1458 metabolites were quantified using the ultra-HPLC–tandem mass spectroscopy HD4 platform of Metabolon Inc. Detailed descriptions for sample preparation, metabolomics data acquisition and processing may be accessed online [38]. Metabolite levels of each sample were normalized using a pool of well-characterized EDTA plasma quality control (QC) samples used throughout the analysis of the dataset (QC-normalization). Missing values were imputed with the minimum value. Classification of metabolites within superpathways, subpathways and subclasses were as assigned by Metabolon Inc. Xenobiotics (n = 218) and metabolites measured in fewer than 50% of individuals (n = 147) were excluded from further analysis, leaving a total of 1093 metabolites for subsequent analysis, as detailed in Supplementary Fig. 1.

Lifestyle, physical measures and chronic diseases

Baseline sociodemographic data of CLSA participants included in this study were the age at enrolment, sex, smoking and alcohol intake. Smoking status was classified as current (daily smoker), occasional (former daily smoker/never a daily smoker or has smoked less than 100 cigarettes in a lifetime), former (former daily and former occasional smoker) and never smoker. Alcohol intake was classified as current (almost every day, 4–5 times per week, 2–3 times per week, once per week, 2–3 times per month and about once a month), occasional (less than once a month) and non-drinkers (never).

Baseline physical measures were collected by trained CLSA staff following standard operating procedures. BMI (kg/m2) was calculated in terms of body weight (kg) and height (m). The list of diseases was taken from the self-reported physician-diagnosed chronic diseases included in the CLSA questionnaires that existed at the time of reporting. Some of the reported diseases were combined following recommendations [39] and ICD-10 codes, which groups diseases according to sufficient pathophysiological similarity. Disease categories are described in Supplementary Table 1.

Statistics

Levels of measured metabolites were natural log-transformed prior to statistical analysis. Fold-change values were determined based on the mean of each group compared to the reference group. An additional analysis was performed exclusively on postmenopausal women, as the number of premenopausal women in the study was too limited. Relevant details are provided in the figure legends where applicable. A one-way analysis of variance F-test was used to compare the means of the different groups. Statistical significance was determined with unadjusted and Tukey-adjusted P-values, and both are reported in Supplementary Tables 2 and 2.1. Pathway enrichment analysis of affected metabolites was carried out according to the classification of 785 metabolites (Supplementary Fig. 1). Enrichment scores were calculated as follows [7]:

Enrichment Score = (k/m)/((n-k)/(N-m)).

m = number of metabolites in the pathway.

k = number of significant metabolites in the pathway.

n = total number of significant metabolites.

N = total number of metabolites.

For categorical phenotype metabolic data, the Chi-square test was performed, with P-values either not adjusted or adjusted for false-discovery rate. For continuous phenotype data, the one-way analysis of variance F-test with unadjusted and Tukey-adjusted P-values was performed. Univariate and multivariate logistic regression analyses were conducted to estimate odds ratios and 95% confidence intervals for the association between chronic conditions as dependent categorical variables and significantly altered metabolites as independent variables. Covariables used in the multivariate model included age, smoking status, alcohol consumption and sex in the overall analyses. Additionally, a secondary model included menopausal status as well as ever-use of hormone replacement therapy (HRT) as covariables. Details are provided in the figure legends. A two-way ANOVA was used to examine the effect of UGT KOs and sex on metabolite levels. The associations between significantly changed metabolites (as independent variables) and blood biomarkers (as dependent variables) were evaluated using linear regressions. Multiple linear regressions, incorporating the same covariates as in the logistic regression analyses (with categorical variables encoded using one-hot encoding), were then used to adjust these associations. The lm function in R was used for these analyses. Analyses were conducted using R (version 4.4.0).

Results

Study cohort

Among the 9992 CLSA participants with available metabolomics data (Fig. 1A), 469 individuals had homozygous deletion (i.e., KO) of UGT2B17 (n = 352) or UGT2B28 (n = 97) or both genes (n = 20); these individuals were compared to a reference group of 4262 subjects who were homozygous for both genes (Fig. 1A). The prevalence of these deletions in Caucasians (∼9% for UGT2B17, ∼3% for UGT2B28) aligns with previous reports [3,4,5]. The frequency of simultaneous deletion of both genes was ∼0.2%. Table 1 lts the main characteristics of the study groups, with nearly equal numbers of men and women with similar body mass index (BMI) ranging from 26.9 to 28.2 kg/m² and an average age of 62.8 years at the time of blood collection. Most participants were Caucasian (98%). As anticipated given the allele frequencies [3], participants of Asian and other ethnic backgrounds were represented in the KO groups. Additional details are provided in Supplementary Table 3, including the prevalence of various diseases among reference and KO groups.

Fig. 1
figure 1

UGT deletions significantly impact the metabolic profile of plasma. (A) The study cohort consisted of homozygous reference gene carriers or UGT knockout (KO) individuals (n = 4731) from the genotyped comprehensive CLSA cohort for which metabolomics profiles (n = 1458 metabolites) were assessed. (B) Number of significantly altered metabolites in each UGT KO group relative to the gene-proficient reference group (P < 0.05). Total number of metabolites of lower or higher abundance is given for each UGT KO group. A detailed report on the metabolites affected in each group is available in Supplementary Table 2. C–E. Venn diagrams depicting the number of altered metabolites in the overall and sex-based analyses; (C) UGT2B17 KO, (D) UGT2B28 KO and (E) double KO. F–H. Distribution of altered metabolites within superpathways for each UGT KO group in the overall and sex-based analyses. Results were consistent when limited to postmenopausal women (Supplementary Table 2.1). The numbers above each bar represent the percentage of altered metabolites relative to the total number of metabolites measured in each superpathway

Full size image
Table 1 Characteristics of the study cohort
Full size table

Divergent metabolomes of subjects with UGT KOs and disparities between sexes

Of the 1093 metabolites that were consistently measured among individuals within each experimental group, 202 (18%) differed significantly for subjects with UGT2B17 KO, 157 (14%) for UGT2B28 KO, and 86 (8%) for the double KO group relative to the reference group (Fig. 1B–E). In the UGT2B17 KO group, the abundance of most changed metabolites was increased compared to the reference group, whereas in UGT2B28 KO and double KO groups, abundance decreased for most of the affected metabolites (Fig. 1B). In each KO group, only small numbers of metabolites (< 7%) were commonly affected in both males and females, indicating sexual dimorphism with respect to UGT KOs (Fig. 1C–E). In the UGT2B17 KO group, the number of affected metabolites varied more for males than females (109 metabolites compared with 63 metabolites), whereas in the UGT2B28 KO group, metabolite abundance was affected considerably more for females than males (108 metabolites vs. 39 metabolites) (Fig. 1C–E). A distinct example of sex-divergent effects in UGT KOs involved glucuronide metabolites, which are produced by UGT enzymes [40]. The abundance of glucuronide derivatives of deoxycholate, a bile acid, and of catechol, a tyrosine metabolite, changed substantially in each KO group, with a particularly pronounced reduction of deoxycholate glucuronide in UGT2B17 KO males, and an increase of catechol glucuronide in UGT2B28 KO females (Padj<0.05; Supplementary Table 4). However, no sex difference was apparent in most tissues with respect to UGT2B17 or UGT2B28 expression apart from UGT2B17 in the liver (higher in males than females) and in the intestine and colon (higher in females than males) (Supplementary Fig. 2) [8, 9]. Metabolites belonging to the lipid, amino acid and carbohydrate superpathways were the most affected in each UGT KO group compared to the reference group (Fig. 1F–H; Supplementary Tables 2 and 4).

Sex-divergent effect of UGT KOs on certain lipid classes

Metabolites of many lipid subclasses including cholesterol-derived steroids and bile acids, sphingolipids, fatty acids, acyl carnitines and phospholipids were higher in UGT2B17 KO males and lower in UGT2B28 KO females (Fig. 2, Supplementary Table 4). For UGT2B17 KO males, the abundance of 15% of measured lipids was altered compared with 6% for UGT2B17 KO females (Fig. 1F, Supplementary Fig. 3A). For UGT2B28 KO subjects, males had far fewer differences in lipid abundance (3%) than females (13%) (Fig. 1G, Supplementary Fig. 3B).

Fig. 2
figure 2

Distinct lipid subclasses are affected by each UGT KO. Enrichment analysis was conducted for significantly (P < 0.05) altered lipids in the overall (O), male (M), and female (F) analyses. Results were consistent when limited to postmenopausal women (Supplementary Table 2.1). Bubble sizes are proportional to the enrichment score. BCAA: branched-chain amino acids. Number of metabolites measured in each lipid subclass is provided

Full size image

Sex steroids, and more specifically several androgens, were significantly elevated in UGT2B17 KO males, consistent with the reported enzymatic specificity of UGT2B17 (Padj<0.05; Fig. 2, Supplementary Tables 2 and 4). Among the 40 sex steroids we measured, 50% were altered, with all but one more abundant, resulting in greater androgenic exposure in the UGT2B17 KO male group (Fig. 3A, Supplementary Tables 2 and 4) [33]. Some metabolites from this lipid sub-class were significantly affected by UGT2B17 KO status and male sex in the interaction analysis including androsterone sulfate (P = 0.004), 5α-androstan-3α,17β-diol monosulfate (P = 0.005) and pregnanolone/allopregnanolone sulfate (P = 0.028). By contrast, the levels of several sphingolipids, particularly sphingomyelins, were lower compared to the reference group, whereas two precursors in the ceramide metabolic pathway, namely sphinganine and sphingosine, were elevated (Fig. 3B, Supplementary Table 2); this implied that these signaling molecules had accumulated in UGT2B17 KO males. Several phospholipids, and particularly those belonging to the glycerophospholipid and plasmalogen classes, were also elevated in UGT2B17 KO males (Fig. 3C and Supplementary Table 2). The limited number of altered lipids in the female UGT2B17 KO mainly featured higher levels of sulfated androgens, sulfated bile acids, and hexosylceramides but lower levels of glucuronidated androgens, non-sulfated bile acids and tetrahydrocortisol (Padj<0.05; Supplementary Table 2). An 84% reduction in the levels of the primary bile acid cholic acid glucuronide and of the secondary bile acid deoxycholic acid glucuronide (a known substrate of UGT2B17) constituted the most drastic changes in UGT2B17 KO individuals, irrespective of sex (Padj<0.05; Supplementary Table 4, Supplementary Fig. 4A, B). In the UGT2B28 KO group, females (but not males) exhibited significantly lower levels of several fatty acid subtypes, particularly saturated, monounsaturated, and polyunsaturated long-chain fatty acids as well as acyl carnitines (Figs. 2 and 3D-E), from which saturated and polyunsaturated were significant at adjusted p-values (Supplementary Table 2). These included multiple omega-3 and omega-6 fatty acids such as arachidonic acid and docosahexanoic acid (Supplementary Table 2). The abundance of bile acids was also altered only in the female UGT2B28 KO group, with higher levels of glucuronide derivatives of deoxycholic acid and glycochenodeoxycholic acid (Supplementary Table 2). Among steroids, only progestins were changed, and more abundant, including pregnanediol-3 glucuronide (Fig. 3A). The build-up of glucuronide derivatives in UGT2B28-deficient individuals would be counterintuitive if UGT2B28 is involved in the glucuronidation of bile acids and progestins, suggesting that UGT2B28 impedes their glucuronidation by other UGTs and/or the presence of another mechanism that mediates their accumulation in plasma. Our analysis support an interaction between UGT2B28 KO status and female sex for a subset of metabolites including those that belong to the category of fatty acids (dodecadienoate (12:2) (P = 0.043), 3-hydroxyoleate (P = 0.038), behenate (22:0) (P = 0.020) and (2 or 3)-decenoate (10:1n7 or n8) (P = 0.011)), as well as the acyl carnitine 3-hydroxydecanoylcarnitine (P = 0.015), the bile acid lycochenodeoxycholate glucuronide (P = 0.015) and a metabolite part of the N-acylethanolamine family (endocannabinoid stearoyl ethanolamide (P = 0.031)).

The lipids altered in UGT2B28 KO males were primarily of the phospholipid subclasses, with levels of the lysophospholipid and glycerophosphoinositol classes being particularly lower, unlike in UGT2B17 KO males (Fig. 3C). N-palmitoyl-sphinganine and dihydrosphingomyelin were reduced in UGT2B28 KO males, similar to observations in UGT2B17 KO males, supporting a role for these two UGTs in regulating sphingolipid signaling, particularly in men (Supplementary Table 2).

The double KO group exhibited fewer significantly altered lipids, potentially reflecting the smaller sample size of this group and hence the limited ability to detect significant changes (Fig. 1H). Males were more affected than females, like the UGT2B17 KO group, but both males and females exhibited mostly lower lipid levels, similar to observations for the UGT2B28 KO group (Supplementary Fig. 3C). Of note, glucuronide derivatives of cholic and deoxycholic acids were less abundant, like UGT2B17 KO, without any sex disparity (Padj<0.05); Supplementary Fig. 4A, B). The testosterone sulfated derivative 5α-androstan-3α,17α-diol-sulfate was significantly higher in the double KO overall group (Fig. 3A).

Fig. 3
figure 3

Lipid subclasses most affected by UGT KO. A. Schematic pathway of cholesterol-derived progestogens and androgens. Steroids that differ significantly in at least one UGT KO group are shown as tile arrays depicting sense of fold change relative to the reference group (blue, lower; red, higher). Most steroids that were measured were sulfate (-S) or glucuronide (-G) conjugates. Each asterisk denotes that the steroid has two sulfate groups. O: overall; M: males; F: females. B. Simplified sphingolipid metabolic pathway and sphingolipids significantly altered in UGT2B17 KO overall analysis. The global sense of the fold change for each sphingolipid subclass is indicated by color (blue, lower; light red, higher). The numbers of measured (center) and altered sphingolipids (parts of the donut) are given for each subclass. DCER, dihydroceramides; DhSM, dihydrosphingomyelins; CER, ceramides; HCER, hexosylceramides; LCER, lactosylceramides; SM, sphingomyelins. C. Phospholipid profiles in UGT2B17 KO and UGT2B28 KO groups relative to reference. LPC, lysophosphatidylcholines; LPE, lysophosphatidylethanolamines; LPI, lysophosphatidylinositol; LPG, lysophosphatidylglycerol; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PI, phosphatidylinositol. *P < 0.05. (D) Fatty acid and (E) acylcarnitine profiles for UGT2B28 KO females relative to the reference group. Fold changes are shown relative to the reference group level set at 1.0 (dashed line). The x axis indicates carbon chain length and includes fatty-acid dicarboxylates (DC), monohydroxy fatty acids or hydroxy acyl carnitines (2OH, 3OH, 16OH), and branched-chain fatty acids (CH3). 3-CMPFP, 3-carboxy-4-methyl-5-pentyl-2-furanpropionate; CAR, carnitine. The terms 22:3* and 22:5* denote omega-6 polyunsaturated fatty acids

Full size image

Notable changes in amino acids, carbohydrates and certain uncharacterized metabolites in UGT2B17 KO and UGT2B28 KO individuals

Amino acid and carbohydrate metabolites constituted the other superpathways most affected in UGT KO individuals (Fig. 4A, Supplementary Table 2). The “glycine, serine and threonine”, “urea cycle and arginine and proline”, “glycolysis” and “fructose” pathways were particularly altered (Fig. 4A). Changes unique to one sex were commonly observed, with UGT2B17 KO males characterized by higher levels of several metabolites related to the degradation of vitamin E and lower levels of bilirubin and by-products of its degradation (Fig. 4B). Furthermore, the levels of pyrimidine metabolite cytidine were significantly affected by the interaction between UGT2B17 KO status and female sex (P = 0.002).

UGT2B28 KO females exhibited higher levels of fructose as well as N-lactoyl-phenylalanine, an amino acid derivative, indicative of mitochondrial dysfunction that is prevalent in plasma of individuals experiencing septic shock or with certain health conditions such as diabetes [41, 42] (Padj<0.05; Fig. 4C; Supplementary Tables 2, 4.1). The effect of UGT2B28 KO and female sex was significant on the levels of tryptophan (P = 0.002), and UGT2B28 KO and male sex on the carbohydrate arabinose (P = 0.015).

Fig. 4
figure 4

Non-lipid metabolic pathways that are altered by UGT KO. A. Enrichment analysis of each subclass of metabolites for the overall (O), male (M) and female (F) analyses. Bubble size represents the enrichment score for each UGT KO group. B–D. Relative abundance of significantly altered representative metabolites in each KO group. Results were consistent when limited to postmenopausal women (Supplementary Table 2.1). Boxes represent interquartile range, the median (horizontal bar), and mean (+). Whiskers depict minimum and maximum values. Ref: reference group. *P < 0.05, **P < 0.01, ***P < 0.001

Full size image

The double KO individuals had distinct metabolic alterations within the same subclasses, with lower levels of several branched-chain amino acid and tryptophan metabolites such as indolelactate in males and kynurenine in females (Fig. 4D, Supplementary Table 2). Additionally, xylose was one of the most elevated metabolites (Fig. 4D). Females also displayed lower levels of pyrimidine nucleotides such as cytidine and of ascorbic acid (vitamin C) and its metabolites (Supplementary Table 2).

A substantial number of uncharacterized metabolites also exhibited distinct altered abundance in males and females of each KO group. This included 60% lower levels of X-19141, representing a highly discriminating metabolite in UGT2B17 KO and double KO cases, irrespective of sex (Padj<0.05; Supplementary Tables 2, 4.2; Supplementary Fig. 4C). These metabolites were significantly influenced by the double KO status and sex, with males showing changes in X-21607 (P < 0.05) and females in X-21834 (P = 0.007).

Metabolites impacted by UGT KOs are associated with common health conditions

We examined the associations of significantly altered metabolites with prevalent metabolic diseases and chronic health conditions seen in the CLSA cohort (Supplementary Tables 1, 2). Obesity, diabetes and hypertension were associated with the largest numbers of altered metabolites (Fig. 5A). In particular, 10 metabolites belonging to a variety of metabolic classes were consistently associated with these conditions (Fig. 5B). They encompassed two androgen glucuronide conjugates, 3-methylglutaryl carnitine (a branched-chain catabolite of leucine), the monocarboxylic acid amide urea, and three unidentified metabolites, with elevated levels linked to increased disease risk. Conversely, elevated levels of the plasmalogen 1-(1-enyl-palmitoyl)-2-linoleoyl-glycerophosphocholine (P-16:0/18:2), glycerate, and the vitamin C derivative oxalate, were linked to a reduced risk of obesity, diabetes and hypertension (Fig. 5B). One of these core metabolites, namely the inactive androgen metabolite 5α-androstan-3α,17β-diol-17-glucuronide, is a direct product of the reaction catalyzed by UGT2B17 [43], whereas metabolites of other classes are not substrates of UGT2B17. Notably, certain metabolites were consistently associated with biomarkers of these diseases, such as BMI, waist circumference, blood pressure and blood levels of hemoglobin A1c, further lending strong support to these metabolite-disease associations (Fig. 5C).

Fig. 5
figure 5

Association of UGT2B17 KO divergent metabolites with the most common metabolic diseases and chronic conditions. A. Number of metabolites per subpathway significantly associated with diseases. CVD, cardiovascular diseases. B. Odds ratio (OR) and 95% confidence interval (CI) for the core metabolites commonly associated with obesity, hypertension and diabetes. 3α-diol-G, 5α-androstan-3α,17β-diol-17-glucuronide; 11β-ADT-G, 11β-hydroxyandrosterone glucuronide; PC (P-16:0/18:2), 1-(1-enyl-palmitoyl)-2-linoleoyl-glycerophosphocholine (P-16:0/18:2); 3MG carnitine, 3-methylglutarylcarnitine. Associations were significant (Padj<0.05) in a model adjusted for age, smoking and alcohol consumption. The overall analysis was also adjusted for sex. C. Significant associations (Padj<0.05) between the core metabolites and biomarkers related to obesity, diabetes and hypertension. BMI, body mass index; WC, waist circumference; A1C, hemoglobin A1c; HDL, high-density lipoprotein; TGL, triglycerides; DBP, diastolic blood pressure; SBP, systolic blood pressure

Full size image

A deeper examination of altered metabolites linked to the UGT2B17 KO status revealed interconnectivity among other seemingly unrelated diseases (Supplementary Fig. 5). Higher levels of androgenic and pregnenolone steroids were associated with a lower risk of osteoporosis and arthritis, alongside obesity (Supplementary Fig. 5A, F, G). This observation is consistent with previous reports suggesting that UGT2B17 contributes to the pathogenesis of osteoporosis [21]. Similar to the elevated levels of sex steroids in UGT2B17 KO, higher levels of phosphatidylcholine derivatives consistently correlated with a reduced risk of obesity, diabetes and heart failure (Supplementary Fig. 5A, B,E), whereas higher levels of sphingolipids were associated with an increased risk of obesity and arthritis (Supplementary Fig. 5A, G). These observations suggested a reduced risk of developing obesity for UGT2B17 KO individuals, which aligns with studies reporting lower BMI for these individuals [44, 45]. Another example pertains to higher levels of glycine derivatives and lower tocopherol/vitamin E metabolites associated with a higher risk of obesity, diabetes and hypertension (Supplementary Fig. 5A–C).

Similar results were observed among female individuals lacking UGT2B17, with significant associations with obesity, hypertension and diabetes (Fig. 5A). In addition, metabolites that were elevated only in UGT2B17 KO females were significantly associated with certain phenotypes; these included hexosylceramide glycosyl-N-tricosanoyl-sphingadienine (d18:2/23:0) linked to a reduced risk of obesity, diabetes and hypertension, the pyrimidine nucleotide N4-acetylcytidine linked to an increased risk of obesity, diabetes and hypertension, tetrahydrocortisol-glucuronide linked to a higher risk of obesity and diabetes, and higher levels of the uncharacterized metabolite X-12283 associated with a lower hypertension and increased risk of osteoporosis (Supplementary Fig. 6A-E). These relationships remained significant after adjustment for menopausal status and ever-use of HRT (Supplementary Fig. 6A-E).

A comprehensive analysis of the altered metabolites in UGT2B28 KO individuals revealed fewer significant associations compared with UGT2B17 deficiency (Fig. 6A). Metabolites belonging to the amino acid superpathway were particularly associated with obesity, diabetes, hypertension and cardiovascular diseases; these included higher levels of taurine associated with a reduced risk of obesity and BMI and of N-lactoyl derivatives of phenylalanine, leucine and tyrosine associated with an increased risk of diabetes and hypertension (Fig. 6B–D). These associations remained significant after adjustment for menopausal status and ever-use of HRT. For females, elevated levels of the glutathione derivative cysteine-glutathione disulfide, the sterol 3β-cholestenoate, and the unknown metabolite X-25790 were all linked to a reduced risk of obesity and inversely associated with BMI and waist circumference. After adjusting for menopausal status and ever use of HRT, most of the mentioned metabolites remained significant.

Fig. 6
figure 6

Metabolomic profile of UGT2B28 KO is associated with chronic diseases. A. Number of metabolites per subpathway significantly associated with each disease. B–F. Odds ratio (OR) and 95% confidence interval (CI) for metabolites associated with obesity, hypertension, diabetes, cardiovascular diseases and arthritis. 3β-cholestenoate, 3β-hydroxy-5-cholestenoate; Cys-glutathione-SS, cysteine-glutathione disulfide; 2-HPAA, 2-hydroxyphenylacetate; N-lactoyl-Leu, N-lactoyl-leucine; N-lactoyl-Phe, N-lactoyl-phenylalanine; N-lactoyl-Tyr, N-lactoyl-tyrosine. Associations were significant (Padj<0.05) in a model adjusted for age, smoking and alcohol consumption, and for sex for the overall analysis. In UGT2B28 KO females, after adjusting for these covariates as well as menopausal status and history of HRT use, all metabolites associated with obesity, diabetes, and X-23,644 in arthritis remained significant (Padj<0.05)

Full size image

For UGT2B28 KO females, several free fatty acids and the acyl carnitines octanoylcarnitine and decanoylcarnitine that were less abundant only in females were all linked to a reduced risk of cardiovascular diseases, whereas 3-hydroxy derivatives of decanoylcarnitine were associated with an increased risk of arthritis (Fig. 6E, F). Higher levels of fructose, which was one of the metabolites most enriched in plasma samples from UGT2B28 KO females, were also associated with an increased risk of cardiovascular diseases (Fig. 6E). These relationships remained significant after accounting for menopausal status and ever-use of HRT. For UGT2B28 KO males, characterized by globally lower levels of phospholipids, higher levels of 1-stearoyl-glycerophosphoethanolamine and 1-palmitoyl-2-stearoyl-glycerophosphatidylcholine were strongly linked to a lower risk of obesity and hypertension, whereas taurine, which was lower in males, was associated with a reduced risk of arthritis (Supplementary Fig. 6F).

Discussion

Our results provide a comprehensive metabolomic profile of individuals with deletions of the metabolic UGT2B17 and UGT2B28 pathways, encoded by two of the most commonly deleted genes in the genome [3,4,5]. Despite the prevalence of these gene KOs in the population and the increasing number of diseases that have been linked to the complete deficiency of either enzyme in several small studies [21,22,23,24, 26, 27, 29, 46], research to date has predominantly focused on the association between UGT2B17 and prostate cancer [19, 20, 24], with limited reports on UGT2B28 [24]. Our study of the large CLSA cohort reveals the marked impact of each KO on many classes of metabolites, especially on specific lipid classes such as steroids and sphingolipids as well as amino acids, carbohydrates and uncharacterized metabolites, with important sex disparities. Systemic metabolic changes linked to each UGT KO were significantly associated, positively and negatively, with various health conditions including obesity, diabetes, hypertension, osteoporosis and arthritis, extending the potential impact of UGT deletions on health and disease. These findings underscore the potential importance of UGTs in maintaining metabolic balance and suggest their involvement in susceptibility to healthy aging and common health issues.

Our exploratory study included an unprecedentedly large number of individuals in each UGT KO group, enabling us to garner deep metabolic insights into the KO individuals. It also enabled an exploration of double KO, which revealed fewer significant alterations, likely due to its smaller sample size that limited the ability to detect changes. The age and sex distributions of the KO groups were consistent with those of the broader aging Canadian population, as reflected in the CLSA cohort [34]. One important hallmark of UGT KO individuals was their markedly altered blood lipid profiles and the clear unique impact on UGT2B17 KO males and UGT2B28 KO females. Probable contributors include the relatively high expression of UGT2B17 in the liver of males [8, 9] and lower expression in gastrointestinal tissue [47] relative to females, alongside UGT expression in sex-specific organs (Supplementary Fig. 2). Steroid hormones, several of which serve as substrates for UGT2B17 and UGT2B28, may contribute to the observed sexual dimorphism [32, 33], a hypothesis supported by our data. For instance, UGT2B17 KO males had reduced levels of androgen glucuronide conjugates that are known catalytic products of UGT2B17. A compensatory increase in androgen sulfate conjugates was also observed, as recently reported for UGT2B17 KO male patients with prostate cancer [7]. The overall increased androgenic exposure observed in UGT2B17 KO individuals underscores the pivotal role of the UGT2B17 pathway in regulating androgenic activity and androgen signaling. This does not imply that UGT2B17 KO has no effect on the female steroid profile, as females had reduced levels of etiocholanolone-glucuronide. Profiling the blood metabolome—and more specifically two bile-acid glucuronides—also offers a method for monitoring individuals with UGT2B17 KO, regardless of their sex. These metabolites, namely cholic acid and deoxycholic acid glucuronides, are end-products of the UGT2B17 enzymatic pathway [7] and were found to be linked to UGT2B17 variants in recent genomic-metabolomic studies, as were two unknown metabolites in our study (X-24947 and X-19141) [48, 49]. For UGT2B28, despite substantial (>80%) amino acid sequence similarity with UGT2B17, the metabolomic profiles of deficient individuals were inversely impacted compared with those of UGT2B17 KO, suggesting a distinct function for UGT2B28. UGT2B28 was initially reported to catalyze the conjugation of selected sex steroids but with much less efficiency compared with UGT2B17 [32], and UGT2B28 was found to be associated with a substantial accumulation of glucuronide derivatives in the CLSA cohort. This observation hints towards an inhibitory impact of UGT2B28 on glucuronidation that could be mediated by functional interaction with other UGTs, as observed for other UGT proteins [50, 51].

The two KOs we included in this study were defined using SNPs that are closely associated with deletion of UGT2B28 or UGT2B17. SNP markers perfectly matched the UGT2B28 deletion, whereas for UGT2B17, the genetic status relied on a SNP with strong but imperfect linkage disequilibrium (R2 < 0.8) [5]. Despite this limitation, some of our findings are supported by previous research reinforcing the validity of our findings. For example, the higher androgen abundance observed in UGT2B17 KO individuals highlights the critical role of this pathway in regulating androgen bioavailability and androgen receptor signaling, especially in men, supported by its association with the progression of prostate cancer [52]. Thus, by associating UGT KOs with particular metabolic pathways and metabolites, the blood metabolome profiling provides molecular insights into the processes influencing disease pathogenesis, potentially linking UGT KOs to specific diseases. We further broadened our study to include associations with obesity, heart failure, osteoporosis and arthritis, in keeping with previous small size studies linking UGT2B17 SNPs to body weight [44, 45] and as a susceptibility gene for osteoporosis [21]. Our findings for UGT2B17 KO males suggest favorable metabolic health and possibly reduced prevalence of obesity, diabetes and hypertension. As for UGT2B28 deficiency, its impact on sex steroid levels was limited to progestins, which were elevated only in UGT2B28 KO females. Although the association between progestins and osteoporosis, a female-predominant disease, did not reach significance for UGT2B28 KO females, progestins were significantly associated vitamin D, a biochemical marker of osteoporosis, linked to a UGT2B28 genetic variant in a recent genome-wide association study [53]. These observations support the physiological relevance of the metabolic changes caused by UGT2B28 KO, particularly in females. The unexpectedly broad molecular divergence found among the UGT KO metabolomes implies their involvement in various pathways well beyond those specific to steroid hormones. It extends to metabolite subclasses that are not known UGT substrates, such as amino acids, fatty acids and carbohydrates. The mechanisms by which metabolites are affected remain to be demonstrated, potentially through direct enzymatic conjugation or other mechanisms that have yet to be characterized.

This study significantly broadened the range of metabolites impacted by UGT2B17 deficiency and identified several lipid subclasses (sphingolipids, phospholipids, carnitines) and amino acid derivatives that could be grouped into a core signature associated with metabolic diseases, namely hypertension, diabetes and obesity. Several of these metabolites —especially sphingolipids and phospholipids— were previously reported to be associated with risk of obesity, diabetes and hypertension [54,55,56]. In addition, recent genome-wide association studies identified UGT2B17 as an effector gene for circulating levels of total cholesterol, apolipoprotein B [53], low density lipoprotein-cholesterol [52] and triglycerides [57] as well as associations between androgen levels and metabolic diseases [58]. In the case of UGT2B28-deficient individuals, the highly impacted fatty acids and amino-acid derivatives in women were closely associated with several metabolic disorders, an unexpected finding that opens a novel avenue of investigation regarding UGT2B28 function. For example, the positive association between elevated N-lactoyl amino acid blood levels and the risk of diabetes and hypertension is consistent with recent studies linking these amino acid derivatives to diabetic retinopathy and diabetes [42, 59]. Our findings suggest that both UGT2B17 and UGT2B28 are significant determinants of complex aging traits, while the potential compensatory roles of other genes warrant further investigations. Despite the unprecedented number of gene KO individuals included in the study, mostly Caucasians, it remains exploratory in nature, and further confirmatory research is required to establish broader clinical implications. A limitation of our study is the inability to comprehensively adjust for therapeutic interventions, which emphasizes the need for future studies to include detailed treatment information and investigate how therapies influence the effects of UGT2B17 and UGT2B28 deletions on the metabolome. Another limitation is the lack of diversity in age and menopausal status among the participants, as the majority were postmenopausal women. This may limit the applicability of our findings, particularly since the influence of UGT2B17 and UGT2B28 genetic status could differ in younger populations or premenopausal women.

Perspectives and significance

This study reveals how natural and common germline UGT deletions —specifically UGT2B17 and UGT2B28—disrupt metabolic balance and may contribute to sex-specific disease risks. These UGT enzymes are known to regulate steroids [32, 33]. By comparing individuals deficient in these UGTs to those who are proficient, we identified significant and distinct metabolic shifts associated with sex-specific disease susceptibilities, suggesting a broader role for these enzymatic pathways in shaping health outcomes. The variable frequencies of these gene deletions, which vary considerably based on ethnic backgrounds [3,4,5], underscore the need for further replication studies to validate our findings across diverse human populations, as our study was limited in terms of ethnic diversity. Existing literature supports a causal role, particularly for UGT2B17, in regulating androgen levels and influencing hormone-sensitive diseases such as prostate cancer progression [19, 20, 24]. Future studies investigating how UGT gene deletions interact with therapeutic interventions to influence the metabolome and disease outcomes could provide valuable insights for developing personalized medicine approaches. Our findings highlight the need for further studies to validate these associations, assess causal links, and explore potential applications in personalized medicine.

Conclusion

By analyzing extensive population-based genomic and metabolomic data, our research emphasizes the crucial roles of UGT2B17 and UGT2B28 in shaping the systemic metabolome. Although these enzymes are primarily known for their role in steroid metabolism, our findings demonstrate that their influence extends to a wider array of metabolites, many of which are associated with both favorable and unfavorable effects on common health conditions and chronic diseases in a sex-divergent manner. This underscores a significant gap in our current understanding of their functions in both men and women across human populations, as well as the health implications related to common deficiencies in these genes.

Data availability

Data are available from the Canadian Longitudinal Study on Aging (www.clsa-elcv.ca) for researchers who meet the criteria for access to de-identified CLSA data.

References

  1. Costanzo M, Caterino M, Sotgiu G, Ruoppolo M, Franconi F, Campesi I. Sex differences in the human metabolome. Biol Sex Differ. 2022;13(1):30.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Long T, Hicks M, Yu HC, Biggs WH, Kirkness EF, Menni C, et al. Whole-genome sequencing identifies common-to-rare variants associated with human blood metabolites. Nat Genet. 2017;49(4):568–78.

    Article  CAS  PubMed  Google Scholar 

  3. Xue Y, Sun D, Daly A, Yang F, Zhou X, Zhao M, et al. Adaptive evolution of UGT2B17 copy-number variation. Am J Hum Genet. 2008;83(3):337–46.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Menard V, Eap O, Harvey M, Guillemette C, Levesque E. Copy-number variations (CNVs) of the human sex steroid metabolizing genes UGT2B17 and UGT2B28 and their associations with a UGT2B15 functional polymorphism. Hum Mutat. 2009;30(9):1310–9.

    Article  CAS  PubMed  Google Scholar 

  5. McCarroll SA, Hadnott TN, Perry GH, Sabeti PC, Zody MC, Barrett JC, et al. Common deletion polymorphisms in the human genome. Nat Genet. 2006;38(1):86–92.

    Article  CAS  PubMed  Google Scholar 

  6. Hurles ME, Dermitzakis ET, Tyler-Smith C. The functional impact of structural variation in humans. Trends Genet. 2008;24(5):238–45.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Rouleau M, Van Nguyen F, Turcotte V, Caron P, Lacombe L, Aprikian A, et al. Extensive metabolic consequences of human glycosyltransferase gene knockouts in prostate cancer. Br J Cancer. 2023;128(2):285–96.

    Article  CAS  PubMed  Google Scholar 

  8. Gallagher CJ, Balliet RM, Sun D, Chen G, Lazarus P. Sex differences in UDP-glucuronosyltransferase 2B17 expression and activity. Drug Metab Dispos. 2010;38(12):2204–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Bhatt DK, Basit A, Zhang H, Gaedigk A, Lee SB, Claw KG, et al. Hepatic abundance and activity of Androgen- and Drug-Metabolizing enzyme UGT2B17 are associated with genotype, age, and sex. Drug Metab Dispos. 2018;46(6):888–96.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Emond JP, Labriet A, Desjardins S, Rouleau M, Villeneuve L, Hovington H, et al. Factors affecting interindividual variability of hepatic UGT2B17 protein expression examined using a novel specific monoclonal antibody. Drug Metab Dispos. 2019;47(5):444–52.

    Article  CAS  PubMed  Google Scholar 

  11. Zhou L, Montalvo AD, Collins JM, Wang D. Quantitative analysis of the UDP-glucuronosyltransferase transcriptome in human tissues. Pharmacol Res Perspect. 2023;11(6):e01154.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Guillemette C, Levesque E, Rouleau M. Pharmacogenomics of human uridine diphospho-glucuronosyltransferases and clinical implications. Clin Pharmacol Ther. 2014;96(3):324–39.

    Article  CAS  PubMed  Google Scholar 

  13. Beutler E, Gelbart T, Demina A. Racial variability in the UDP-glucuronosyltransferase 1 (UGT1A1) promoter: a balanced polymorphism for regulation of bilirubin metabolism? Proc Natl Acad Sci U S A. 1998;95(14):8170–4.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Little JM, Kurkela M, Sonka J, Jantti S, Ketola R, Bratton S, et al. Glucuronidation of oxidized fatty acids and prostaglandins B1 and E2 by human hepatic and Recombinant UDP-glucuronosyltransferases. J Lipid Res. 2004;45(9):1694–703.

    Article  CAS  PubMed  Google Scholar 

  15. Kato Y, Ikushiro S, Emi Y, Tamaki S, Suzuki H, Sakaki T, et al. Hepatic UDP-glucuronosyltransferases responsible for glucuronidation of thyroxine in humans. Drug Metab Dispos. 2008;36(1):51–5.

    Article  CAS  PubMed  Google Scholar 

  16. Barbier O, Trottier J, Kaeding J, Caron P, Verreault M. Lipid-activated transcription factors control bile acid glucuronidation. Mol Cell Biochem. 2009;326(1–2):3–8.

    Article  CAS  PubMed  Google Scholar 

  17. Leong IUS, Cabrera CP, Cipriani V, Ross PJ, Turner RM, Stuckey A et al. Large-Scale pharmacogenomics analysis of patients with cancer within the 100,000 genomes project combining Whole-Genome sequencing and medical records to inform clinical practice. J Clin Oncol. 2025;43(6):682–93.

  18. Bosma PJ, Chowdhury NR, Goldhoorn BG, Hofker MH, Oude Elferink RP, Jansen PL, et al. Sequence of exons and the flanking regions of human bilirubin-UDP-glucuronosyltransferase gene complex and identification of a genetic mutation in a patient with Crigler-Najjar syndrome, type I. Hepatology. 1992;15(5):941–7.

    Article  CAS  PubMed  Google Scholar 

  19. Gallagher CJ, Kadlubar FF, Muscat JE, Ambrosone CB, Lang NP, Lazarus P. The UGT2B17 gene deletion polymorphism and risk of prostate cancer. A case-control study in Caucasians. Cancer Detect Prev. 2007;31(4):310–5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Karypidis AH, Olsson M, Andersson SO, Rane A, Ekstrom L. Deletion polymorphism of the UGT2B17 gene is associated with increased risk for prostate cancer and correlated to gene expression in the prostate. Pharmacogenomics J. 2008;8(2):147–51.

    Article  CAS  PubMed  Google Scholar 

  21. Yang TL, Chen XD, Guo Y, Lei SF, Wang JT, Zhou Q, et al. Genome-wide copy-number-variation study identified a susceptibility gene, UGT2B17, for osteoporosis. Am J Hum Genet. 2008;83(6):663–74.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. McCarroll SA, Bradner JE, Turpeinen H, Volin L, Martin PJ, Chilewski SD, et al. Donor-recipient mismatch for common gene deletion polymorphisms in graft-versus-host disease. Nat Genet. 2009;41(12):1341–4.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Bronstad I, Wolff AS, Lovas K, Knappskog PM, Husebye ES. Genome-wide copy number variation (CNV) in patients with autoimmune Addison’s disease. BMC Med Genet. 2011;12:111.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Nadeau G, Bellemare J, Audet-Walsh E, Flageole C, Huang SP, Bao BY, et al. Deletions of the androgen-metabolizing UGT2B genes have an effect on Circulating steroid levels and biochemical recurrence after radical prostatectomy in localized prostate cancer. J Clin Endocrinol Metab. 2011;96(9):E1550–7.

    Article  CAS  PubMed  Google Scholar 

  25. Gruber M, Bellemare J, Hoermann G, Gleiss A, Porpaczy E, Bilban M, et al. Overexpression of uridine diphospho glucuronosyltransferase 2B17 in high-risk chronic lymphocytic leukemia. Blood. 2013;121(7):1175–83.

    Article  CAS  PubMed  Google Scholar 

  26. Angstadt AY, Berg A, Zhu J, Miller P, Hartman TJ, Lesko SM, et al. The effect of copy number variation in the phase II detoxification genes UGT2B17 and UGT2B28 on colorectal cancer risk. Cancer. 2013;119(13):2477–85.

    Article  CAS  PubMed  Google Scholar 

  27. Hu L, Wu Y, Guan X, Liang Y, Yao X, Tan D, et al. Germline copy number loss of UGT2B28 and gain of PLEC contribute to increased human esophageal squamous cell carcinoma risk in Southwest China. Am J Cancer Res. 2015;5(10):3056–71.

    CAS  PubMed  PubMed Central  Google Scholar 

  28. Belledant A, Hovington H, Garcia L, Caron P, Brisson H, Villeneuve L, et al. The UGT2B28 Sex-steroid inactivation pathway is a regulator of steroidogenesis and modifies the risk of prostate cancer progression. Eur Urol. 2016;69(4):601–9.

    Article  CAS  PubMed  Google Scholar 

  29. Shen Q, Lee K, Han SK, Ahn HJ, Kim S, Lee JH. Variants at potential loci associated with Sjogren’s syndrome in Koreans: A genetic association study. Clin Immunol. 2019;207:79–86.

    Article  CAS  PubMed  Google Scholar 

  30. Turgeon D, Chouinard S, Belanger P, Picard S, Labbe JF, Borgeat P, et al. Glucuronidation of arachidonic and Linoleic acid metabolites by human UDP-glucuronosyltransferases. J Lipid Res. 2003;44(6):1182–91.

    Article  CAS  PubMed  Google Scholar 

  31. Allain EP, Rouleau M, Le T, Vanura K, Villeneuve L, Caron P, et al. Inactivation of prostaglandin E(2) as a mechanism for UGT2B17-Mediated adverse effects in chronic lymphocytic leukemia. Front Oncol. 2019;9:606.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Levesque E, Turgeon D, Carrier JS, Montminy V, Beaulieu M, Belanger A. Isolation and characterization of the UGT2B28 cDNA encoding a novel human steroid conjugating UDP-glucuronosyltransferase. Biochemistry. 2001;40(13):3869–81.

    Article  CAS  PubMed  Google Scholar 

  33. Beaulieu M, Levesque E, Hum DW, Belanger A. Isolation and characterization of a novel cDNA encoding a human UDP-glucuronosyltransferase active on C19 steroids. J Biol Chem. 1996;271(37):22855–62.

    Article  CAS  PubMed  Google Scholar 

  34. Raina P, Wolfson C, Kirkland S, Griffith LE, Balion C, Cossette B, et al. Cohort profile: the Canadian longitudinal study on aging (CLSA). Int J Epidemiol. 2019;48(6):1752–j3.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Canadian Longitudinal Study on Aging. Researchers 2024 [Available from: https://www.clsa-elcv.ca/researchers/. [Accessed 22 January 2024].

  36. Forgetta V, Li R, Darmond-Zwaig C, Belisle A, Balion C, Roshandel D, et al. Cohort profile: genomic data for 26 622 individuals from the Canadian longitudinal study on aging (CLSA). BMJ Open. 2022;12(3):e059021.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Canadian Longitudinal Study on Aging. Genome-wide Genetic Data Release (version 3) 2023 [Available from: https://www.clsa-elcv.ca/wp-content/uploads/2023/06/clsa_gwas_v3.pdf.. [Accessed 11 Novembre 2023].

  38. Canadian Longitudinal Study on Aging. Metabolomic Profiling on 9,992 Participants using Ultra-performance Liquid Chromatography and Mass Spectrometer Data Support Document 2023 [Available from: https://www.clsa-elcv.ca/wp-content/uploads/2024/01/CLSA_DataSupportDoc_Metabolomics_v2.0_2023Aug03.pdf.. [Accessed 16 September 2023].

  39. Fortin M, Almirall J, Nicholson K. Development of a research tool to document self-reported chronic conditions in primary care. J Comorb. 2017;7(1):117–23.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Allain EP, Rouleau M, Levesque E, Guillemette C. Emerging roles for UDP-glucuronosyltransferases in drug resistance and cancer progression. Br J Cancer. 2020;122(9):1277–87.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Rogers RS, Sharma R, Shah HB, Skinner OS, Guo XA, Panda A, et al. Circulating N-lactoyl-amino acids and N-formyl-methionine reflect mitochondrial dysfunction and predict mortality in septic shock. Metabolomics. 2024;20(2):36.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Fernandes Silva L, Hokkanen J, Vangipurapu J, Oravilahti A, Laakso M. Metabolites as risk factors for diabetic retinopathy in patients with type 2 diabetes: A 12-Year Follow-up study. J Clin Endocrinol Metab. 2023;109(1):100–6.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Sten T, Kurkela M, Kuuranne T, Leinonen A, Finel M. UDP-glucuronosyltransferases in conjugation of 5alpha- and 5beta-androstane steroids. Drug Metab Dispos. 2009;37(11):2221–7.

    Article  CAS  PubMed  Google Scholar 

  44. Zhu AZ, Cox LS, Ahluwalia JS, Renner CC, Hatsukami DK, Benowitz NL, et al. Genetic and phenotypic variation in UGT2B17, a testosterone-metabolizing enzyme, is associated with BMI in males. Pharmacogenet Genomics. 2015;25(5):263–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Swanson C, Mellstrom D, Lorentzon M, Vandenput L, Jakobsson J, Rane A, et al. The uridine diphosphate glucuronosyltransferase 2B15 D85Y and 2B17 deletion polymorphisms predict the glucuronidation pattern of androgens and fat mass in men. J Clin Endocrinol Metab. 2007;92(12):4878–82.

    Article  CAS  PubMed  Google Scholar 

  46. Albertsen HM, Matalliotaki C, Matalliotakis M, Zervou MI, Matalliotakis I, Spandidos DA, et al. Whole exome sequencing identifies hemizygous deletions in the UGT2B28 and USP17L2 genes in a threegeneration family with endometriosis. Mol Med Rep. 2019;19(3):1716–20.

    CAS  PubMed  PubMed Central  Google Scholar 

  47. Hartman RJG, Mokry M, Pasterkamp G, den Ruijter HM. Sex-dependent gene co-expression in the human body. Sci Rep. 2021;11(1):18758.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Chen Y, Lu T, Pettersson-Kymmer U, Stewart ID, Butler-Laporte G, Nakanishi T, et al. Genomic atlas of the plasma metabolome prioritizes metabolites implicated in human diseases. Nat Genet. 2023;55(1):44–53.

    Article  PubMed  PubMed Central  Google Scholar 

  49. Hysi PG, Mangino M, Christofidou P, Falchi M, Karoly ED, Nihr Bioresource I et al. Metabolome Genome-Wide association study identifies 74 novel genomic regions influencing plasma metabolites levels. Metabolites. 2022;12(1).

  50. Rouleau M, Audet-Delage Y, Desjardins S, Rouleau M, Girard-Bock C, Guillemette C. Endogenous protein interactome of human UDP-Glucuronosyltransferases exposed by untargeted proteomics. Front Pharmacol. 2017;8:23.

    Article  PubMed  PubMed Central  Google Scholar 

  51. Bellemare J, Rouleau M, Harvey M, Tetu B, Guillemette C. Alternative-splicing forms of the major phase II conjugating UGT1A gene negatively regulate glucuronidation in human carcinoma cell lines. Pharmacogenomics J. 2010;10(5):431–41.

    Article  CAS  PubMed  Google Scholar 

  52. Kanoni S, Graham SE, Wang Y, Surakka I, Ramdas S, Zhu X, et al. Implicating genes, pleiotropy, and sexual dimorphism at blood lipid loci through multi-ancestry meta-analysis. Genome Biol. 2022;23(1):268.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Weeks EM, Ulirsch JC, Cheng NY, Trippe BL, Fine RS, Miao J, et al. Leveraging polygenic enrichments of gene features to predict genes underlying complex traits and diseases. Nat Genet. 2023;55(8):1267–76.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Yousri NA, Suhre K, Yassin E, Al-Shakaki A, Robay A, Elshafei M, et al. Metabolic and Metabo-Clinical signatures of type 2 diabetes, obesity, retinopathy, and dyslipidemia. Diabetes. 2022;71(2):184–205.

    Article  CAS  PubMed  Google Scholar 

  55. Lind L, Sundstrom J, Elmstahl S, Dekkers KF, Smith JG, Engstrom G, et al. The metabolomic profile associated with clustering of cardiovascular risk factors-A multi-sample evaluation. PLoS ONE. 2022;17(9):e0274701.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Beyene HB, Giles C, Huynh K, Wang T, Cinel M, Mellett NA, et al. Metabolic phenotyping of BMI to characterize cardiometabolic risk: evidence from large population-based cohorts. Nat Commun. 2023;14(1):6280.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Forgetta V, Jiang L, Vulpescu NA, Hogan MS, Chen S, Morris JA, et al. An effector index to predict target genes at GWAS loci. Hum Genet. 2022;141(8):1431–47.

    Article  CAS  PubMed  Google Scholar 

  58. Pietzner M, Stewart ID, Raffler J, Khaw KT, Michelotti GA, Kastenmuller G, et al. Plasma metabolites to profile pathways in noncommunicable disease Multimorbidity. Nat Med. 2021;27(3):471–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  59. Xia JG, Li B, Zhang H, Li QX, Lam SM, Yin CL, et al. Precise metabolomics defines systemic metabolic dysregulation distinct to acute myocardial infarction associated with diabetes. Arterioscler Thromb Vasc Biol. 2023;43(4):581–96.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

The authors thank Professors Éric Lévesque (Faculty of Medicine, Université Laval) and Danielle Laurin (Faculty of Pharmacy, Université Laval) for helpful discussions.

Funding

This study was funded by Canadian Institutes of Health Research (CIHR) to CG (FRN-167269 and FRN-187247). The project was also made possible with the support of the Canada Foundation for Innovation to CG (John R. Evans Leaders Funds #34272). This research was made possible using the data/biospecimens collected by the CLSA. Funding for the CLSA is provided by the Government of Canada through the CIHR under grant reference LSA 94473 and the Canada Foundation for Innovation as well as the following provinces: Newfoundland, Nova Scotia, Quebec, Ontario, Manitoba, Alberta, and British Columbia. This research was conducted using the CLSA dataset [CLSA Metabolomics data (v1), Baseline Comprehensive Dataset (v7) and genomics data (v3)] under Application Number 2109011. The CLSA is led by Drs. Parminder Raina, Christina Wolfson and Susan Kirkland. CG holds the Canada Research Chair in Pharmacogenomics (Tier I CRC-2020-000067). ALRH was supported by a research and innovation studentship from the Fondation CHU de Québec- Université Laval and Desjardins.

Author information

Authors and Affiliations

  1. Faculty of Pharmacy, Université Laval, Québec, QC, G1V 0A6, Canada

    Ana Lucia Rivera-Herrera, Michèle Rouleau & Chantal Guillemette

  2. Université Laval Cancer Research Center, Québec, QC, G1R 3S3, Canada

    Ana Lucia Rivera-Herrera, Michèle Rouleau, Tania Cuppens, Julien Prunier, Arnaud Droit & Chantal Guillemette

  3. Centre Hospitalier Universitaire (CHU) de Québec Research Center - Université Laval, R4701.5, 2705 Blvd Laurier Québec, Québec, QC, G1V 4G2, Canada

    Ana Lucia Rivera-Herrera, Michèle Rouleau, Mahukpe Narcisse Ulrich Singbo, Tania Cuppens, Julien Prunier, Arnaud Droit, David Simonyan & Chantal Guillemette

  4. Faculty of Medicine, Department of Molecular Medicine, Université Laval, Québec, QC, G1V 0A6, Canada

    Tania Cuppens, Julien Prunier & Arnaud Droit

Authors
  1. Ana Lucia Rivera-Herrera
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Michèle Rouleau
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Mahukpe Narcisse Ulrich Singbo
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Tania Cuppens
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Julien Prunier
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Arnaud Droit
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. David Simonyan
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Chantal Guillemette
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

ALRH contributed to the analysis of data, interpretation of data, prepared the figures, and writing and edition of the manuscript. MR contributed to study supervision, analysis of data, interpretation of data, and writing and edition of the manuscript. TC and JP retrieved the genotyping data. MNUS and DS analyzed the data. CG contributed to the study concept, study supervision, analysis of data, interpretation of data, and writing and edition of the manuscript. Critical revision of the manuscript for intellectual content: All authors.

Corresponding author

Correspondence to Chantal Guillemette.

Ethics declarations

Ethics approval and consent to participate

This study was approved by the local ethics committee (CHU de Québec – UL #2012 − 362). All participants provided written informed consent upon enrollment into the CLSA.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Disclaimer

The opinions expressed in this manuscript are those of the authors and do not reflect the views of the Canadian Longitudinal Study on Aging.

Additional files

File name: Additional file 1

File format. Pdf.

Title of data: Supplementary Figures of Sexual Dimorphism in Metabolomic and Phenotypic Spectra of UGT deficiency: Findings from the Canadian Longitudinal Study on Aging.

Description of data:

Supplementary Fig. 1

Metabolites quantified in this study and number of metabolites in each super-pathway that were available for analysis.

Supplementary Fig. 2

Gene expression by sex for (A) UGT2B17 and (B) UGT2B28 in human tissues.

Supplementary Fig. 3

Number of significantly altered metabolites in each UGT KO group relative to the gene-proficient reference group (P < 0.05) by subpathway.

Supplementary Fig. 4

Relative abundance of significantly dysregulated metabolites in each KO group.

Supplementary Fig. 5

Associations between significantly altered metabolites in UGT2B17 KO and diseases.

Supplementary Fig. 6

Associations between significantly altered metabolites in UGT2B17 KO females, UGT2B28 KO males, and diseases.

File name: Additional file 2

File format. Excel.xlsx.

Title of data: Supplementary tables of Sexual Dimorphism in Metabolomic and Phenotypic Spectra of UGT deficiency: Findings from the Canadian Longitudinal Study on Aging.

Description of data:

Supplementary Table 1

Detail of the chronic diseases / categories of chronic diseases used in this study.

Supplementary Table 2

Fold change of the plasma metabolites measured in KO and reference groups.

Supplementary Table 2.1

Significant changes in plasma metabolites for postmenopausal women.

Supplementary Table 3

Characteristics of the study groups.

Supplementary Table 4

Top 5 most changed metabolites in UGT2B17 KO, UGT2B28 KO and Double KO for overall and sex-based analyses.

Supplementary Table 4.1

Top 5 most changed metabolites including unknown and partially characterized metabolites in UGT2B17 KO, UGT2B28 KO and Double KO for overall and sex-based analyses.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Rivera-Herrera, A.L., Rouleau, M., Singbo, M. et al. Sexual dimorphism in metabolomic and phenotypic spectra of UGT deficiency: findings from the Canadian Longitudinal Study on Aging. Biol Sex Differ 16, 26 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13293-025-00708-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13293-025-00708-5

Keywords

Biology of Sex Differences

ISSN: 2042-6410

Contact us