Serum adropin is unaltered in adolescents with histology-confirmed steatotic liver disease

Abstract

Objectives:

Metabolic dysfunction-associated steatotic liver disease (MASLD) in adolescents is increasing. Adropin is a liver-derived peptide involved in glucose and lipid homeostasis that was shown to be reduced in adults with metabolic disorders and cardiovascular disease (CVD). Serum adropin may also be higher in young men than women. A prior study reported that serum adropin was reduced in adolescents with MASLD, but the relationship between liver histology and CVD risk factors was not reported. We tested the hypotheses that adropin is (1) reduced in adolescents with MASLD compared to adolescents with obesity (Ob) or normal weight (NW) without MASLD, (2) correlated with blood pressure, arterial stiffness, and liver histopathology, and (3) higher in boys than girls.

Methods:

Serum adropin was measured in 47 patients with MASLD, and 27 and 29 control participants with Ob or NW, respectively.

Results:

Adropin was not reduced but was instead 5% and 20% higher (p > 0.42) in the MASLD compared to the Ob and NW groups, respectively. Adropin concentration was not correlated with arterial stiffness or blood pressure. Adropin was 20% higher in boys than girls in the entire study cohort (p = 0.034). This difference was evident in the Ob group (p = 0.018), but not in the NW (p = 0.537) or the MASLD (p = 0.893) groups. Adropin was positively correlated with age within the MASLD group only (r = 0.46, p < 0.001).

Conclusion:

Serum adropin was not reduced in adolescents with Ob or MASLD as reported previously. The positive relationship between age and adropin in adolescents with MASLD requires further examination.

1 |. INTRODUCTION

Much of our understanding of the molecular and metabolic causes of metabolic dysfunction-associated steatotic liver disease (MASLD) comes from studies of adults, rodents, or cell culture models. Despite advances in adults, major gaps remain in defining the pathways and mechanisms unique to pediatric MASLD pathology.^1,2 MASLD is associated with obesity (Ob) and prevalence rates of Ob, MASLD, and the more advanced form of MASLD, metabolic dysfunction associated steatohepatitis (MASH), are rising in adolescents.^3–5

Adropin is a secreted liver peptide encoded by the Energy Homeostasis Associated gene (Enho) that may play a role in MASLD.⁶ Adropin was initially proposed as a regulator of glucose and peripheral lipid homeostasis.⁶ Reduced circulating adropin concentration has been reported in metabolic disorders, such as MASLD and metabolic syndrome.^6,7 Animal models have identified potential mechanisms to explain how adropin regulates metabolic activity.^8–10 Chen et al.⁹ demonstrated that mice with diet-induced Ob, diet-induced MASH, or genetic knockout of Enho had reduced serum adropin and liver pathology consistent with MASLD, whereas treatment with exogenous adropin protected against liver injury. Yang et al.⁸ showed that treadmill exercise reduced liver inflammation in mice on diet-induced models of Ob and MASH, an outcome that was at least partially attributed to an increase in adropin. Two studies using rodents and hepatocyte cultures provided evidence that adropin may play a key role in the relative protection against liver steatosis and injury in females when faced with diet-induced Ob compared to their male counterparts.^10,11 This effect appears to be mediated via the regulatory role of estrogen receptor alpha in the liver.

In studies of adult humans, men under 30 years old were reported to have higher circulating adropin than women, a difference that wasn’t present at older ages.¹⁰ Adults with MASLD were shown to have lower serum adropin compared to healthy control groups.^11,12 Adropin concentration was also reduced in the participants in those studies who had insulin resistance compared to those who did not.^11,12 A study of adults with MASLD who underwent liver biopsy reported an inverse relationship between serum adropin levels and MASLD histological severity.¹³ Adropin was also reduced in adults with type 2 diabetes and appeared to contribute to the increased arterial stiffness in those patients.¹⁴ That finding is in agreement with a study that showed that adropin was reduced in adults with type 2 diabetes and carotid atherosclerosis.¹⁵ Collectively, the studies of adult humans and rodents suggest adropin has a protective role against the development of MASLD and that adropin is reduced when MASLD is present.

There are fewer reports of the role of adropin in children with Ob, MASLD, or related metabolic conditions. A study of adolescents in Turkey with a mean age of 15 years found that those with Ob had 26% lower serum adropin than those who were normal weight (NW).¹⁶ Similarly, in a cohort of younger children (mean age 10 years) in China, serum adropin was lower in those with Ob compared to a NW group, and further reduced in children with Ob and metabolic syndrome.¹⁷ However, another larger study in Spain found no difference between adolescents (mean age 14.6 years) with Ob versus those who were NW.¹⁸ The latter study also showed that in pre-pubertal children (mean age 7.2 years) serum adropin was higher in participants with Ob versus NW, revealing that the relationship between adropin and metabolic status may vary with age and/or developmental stage.¹⁸ In the only prior study of adolescents with a MASLD diagnosis (mean age 13 years), Sayin et al.¹⁹ reported that, compared to a control group with NW, serum adropin was 62% and 68% lower in groups with Ob or Ob plus MASLD, respectively, and inversely correlated with insulin resistance. However, MASLD in that study was diagnosed with liver ultrasound rather than liver biopsy, so it was not possible to test whether adropin varied with MASLD severity. As recognized by Nobili et al.,²⁰ a lack of studies in children with biopsy-proven MASLD has limited our understanding of how unique features of pediatric MASLD influence disease progression.

It remains unclear whether there is a role of adropin in pediatric MASLD. Specifically, it is unknown whether circulating adropin varies with MASLD severity, or if it is associated with cardiovascular outcomes, as has been reported in adults. In the current study, we tested the hypotheses that, (1) adropin would be reduced in adolescent MASLD patients compared to adolescents with Ob or NW, without MASLD, (2) that adropin is correlated to arterial stiffness and varies with the severity of liver histopathology in patients with MASLD, and (3) that there would be a difference in adropin levels between sexes.

2 |. METHODS

2.1 |. Participants

We studied 48 patients with MASLD confirmed by liver biopsy and 2 control groups without MASLD, comprised of 27 adolescents with Ob, and 29 adolescents with NW based on body mass index (BMI). The BMI range for NW in children and adolescents is 5–84th percentile for age and sex, while the obese range is ≥95th percentile.²¹ All patients were between the ages of 10–20 years old. Potential participants were excluded from the study if they had any diagnosed cardiac, metabolic, or untreated thyroid diseases or other endocrine conditions, liver conditions other than MASLD (e.g., Wilson’s disease), cardiovascular or orthopedic conditions that could limit physical activity, or were taking medications that could potentially affect liver function. Use of tobacco, alcohol, or illicit drugs was also cause for exclusion. Each participant and their parents or guardians provided their informed, written consent and/or assent to enroll in the study in accordance with the University of Oklahoma Health Sciences Center’s Institutional Review Board guidelines. Participant testing and blood sample collection occurred from February 2019 to July 2023.

2.2 |. Procedures

Liver biopsies were obtained in patients with MASLD as part of routine care. All biopsy specimens were reviewed by the same certified pathologist, who determined the severity of steatosis and fibrosis, inflammation, and the MASLD Activity Score (MAS), a validated approach to semi-quantitatively determine the presence and severity of MASH.²² All laboratory tests were performed in the morning following an overnight fast. Measurements of height and weight were used to calculate BMI and BMI z-score using growth curves developed by the Centers of Disease Control and Prevention.²¹ Body composition was measured using a whole-body dual-energy X-ray absorptiometry scan (GE/Lunar iDXA; GE-Healthcare). Fasting venous blood was collected, a portion of which was centrifuged, aliquoted into serum and plasma, and stored at −80°C until analysis occurred. Participants rested supine for 30 min in a quiet room maintained at 22–23°C. Afterward, carotid-femoral pulse wave velocity was measured with a SphygmoCor Xcel instrument (Atcor Medical). A blood pressure cuff was placed on the upper thigh to measure the femoral pulse while a tonometer was placed over the carotid artery. The same instrument was used to measure blood pressure with a brachial cuff of appropriate size. Both sets of measurements were performed in triplicate, with the average used for data analysis.

2.3 |. Analysis

Whole blood collected in lithium-heparin tubes was used to measure glucose, total cholesterol, high-density lipoprotein cholesterol (HDL-C), triglycerides, alanine aminotransferase (ALT), and aspartate aminotransferase (AST) using a Piccolo Xpress Chemistry Analyzer (Abbott Point of Care). Low-density lipoprotein cholesterol concentration (LDL-C) was calculated using the formula developed by Sampson et al.²³ Serum insulin was measured with an enzyme linked immunosorbent assay (ELISA) from Alpco (#80-INSHU-CH01). Insulin resistance was calculated as the revised homeostatic model of assessment, iHOMA2-IR, using the glucose and insulin values.²⁴ Serum adropin was measured using an ELISA per protocol from the manufacturer (#EK-032–35; Phoenix Pharmaceuticals Inc.).

2.4 |. Statistical analysis

GraphPad Prism was used to determine the standard curves and for data interpolation of participant samples for both insulin and adropin ELISAs. SPSS (IBM) was used to make group comparisons for all variables using analysis of variance with pairwise comparisons among groups performed using Tukey’s tests adjusted for multiple comparisons. For non-normally distributed variables (serum adropin, triglycerides, ALT, AST, insulin, iHOMA2-IR) a Kruskal Wallis test was used, followed by Dunn’s multiple comparisons test. A Chi-square test was used to measure categorical variables, like the distribution of boys and girls among groups. Pearson’s correlations were calculated to measure the strength of the association between adropin and other biological data. p Values less than 0.05 were considered significant for all tests.

3 |. RESULTS

The clinical and laboratory results for the three groups are summarized in Table 1. The MASLD group had more boys than girls, but the Chi-square test for the distribution of sexes among groups was not significant (p = 0.163). The MASLD group was younger than the Ob group but not the NW group. The groups had similar height, and the Ob and MASLD groups had similar body weight and BMI (kg/m²) to one another, with both of those groups larger than the NW group. However, the MASLD group had a higher BMI z-score than either the Ob or NW groups. Whole body fat percentage was higher in Ob and MASLD than NW and was slightly higher in MASLD than Ob (p = 0.06), while trunk fat percentage differed among all groups, with the highest values in the MASLD group. The MASLD group had higher fasting glucose than either NW or Ob but did not differ from Ob for insulin or insulin resistance. Total and LDL-cholesterol did not differ among groups, but the MASLD group had lower HDL-cholesterol and higher triglycerides, ALT, and AST than either the NW or Ob groups. The MASLD group had higher values for systolic and diastolic BP, PWV, and heart rate than both control groups.

TABLE 1.

Participant characteristics.

	NW control (n = 29)	Ob control (n = 27)	MASLD (n = 47)
Sex (M:F)	17:12	13:14	33:14
Age (years)	15.5 ± 2.4	16.0 ± 2.4	14.3 ± 2.1b
Height (cm)	165.2 ± 10.7	166.5 ± 10.3	165.6 ± 11.0
Weight (kg)	57.2 ±10.6	92.0 ± 20.8a	94.3 ± 20.0a
BMI (kg/m2)	20.8 ± 2.2	32.8 ± 4.8a	34.1 ± 5.2a
BMI z-score	0.20 ± 0.58	2.06 ± 0.32a	2.30 ± 0.32a,b
Body fat (%)	30.8 ±6.8	42.9 ± 7.5a	46.5 ± 6.0a
Trunk fat (%)	29.4 ±7.7	44.4 ± 7.8a	49.9 ± 5.6a,b
Glucose (mg/dL)	81 ±10	81 ± 14	97 ± 20a,b
Insulin (μIU/mL)	10.0 ± 5.6	23.4 ± 19.6a	27.1 ± 20.7a
iHOMA2-IR	1.06 ± 0.60	2.37 ± 2.04a	2.96 ± 2.09a
Total cholesterol (mg/dL)	162 ± 34	154 ± 44	164 ± 40
HDL-C (mg/dL)	57 ± 13	48 ± 13a	39 ± 10a,b
LDL-C (mg/dL)	92 ± 30	92 ± 37	101 ± 36
Triglycerides (mg/dL)	87 ± 41	101 ± 49	197 ± 107a,b
ALT (mU/ml)	19.6 ± 9.6	28.3 ± 16.9	119.4 ± 97.4a,b
AST (mU/ml)	31.0 ± 19.6	30.7 ± 13.3	81.4 ± 52.9a,b
Systolic BP (mmHg)	112 ± 9	118 ± 11a	125 ± 8a,b
Diastolic BP (mmHg)	66 ± 7	67 ± 9	72 ± 6a,b
PWV (m/s)	4.84 ± 0.62	5.19 ± 0.85	5.46 ± 0.60a

Note: Values are shown as mean ± standard deviation. NW, control group with normal weight; Ob, control group with obesity; MASLD, group with obesity and metabolic dysfunction-associated steatotic liver disease.

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; BP, blood pressure; HDL-C, high-density lipoprotein cholesterol; iHOMA2-IR, integrated homeostatic model of assessment of insulin resistance; LDL-C, low-density lipoprotein cholesterol; PWV, pulse wave velocity.

^ap < 0.05 versus the control group with normal weight (NW).

^bp < 0.05 versus the control group with obesity (Ob).

The self-declared race-ethnicity of the three groups did not differ (Chi-square p-value = 0.48). The MASLD group was comprised of 51% Hispanic/Latino, 47%White, 28% Native American, 6% Black, and 6% Asian, with 34% (n = 16) identifying with 1 or more categories. The Ob group was comprised of 41% Hispanic/Latino, 48% White, 33% Native American, 19% Black, and 0% Asian, with 26% (n = 7) identifying with 1 or more categories. The NW group was comprised of 48% Hispanic/Latino, 48% White, 14% Native American, 7% Black, and 14% Asian, with 24% (n = 7) identifying with 1 or more categories.

Liver histology scores for the MASLD group are shown in Supporting Information S2: Table 1. Most of the MASLD group had moderate to advanced steatosis, but lower scores for lobular inflammation and ballooning. Most patients were classified as borderline MASH or MASH based on the MAS. The majority of patients had portal or periportal fibrosis, but 4 patients had no fibrosis, and 6 patients had evidence of bridging fibrosis.

Adropin in the MASLD group was 20% higher than the NW group and 5% higher than the Ob group (Figure 1), but these differences did not reach statistical significance. We explored correlations between adropin and several of the primary physiological and laboratory assessments (Table 2). The only significant finding was a positive correlation between adropin concentration and age within the MASLD group (Figure 2). Within the MASLD group, adropin concentration did not vary across scores for steatosis, inflammation, ballooning, or fibrosis (not shown).

FIGURE 1.

Adropin concentration in each group. Boxes show the median (center line) and interquartile range (IQR) for each group. Whiskers show the range of values within 1.5× of the IQR. Open circles are outlier values outside of 1.5× of the IQR. Between-group comparisons are shown as adjusted p-values from Tukey’s multiple comparison test. MASLD, metabolic dysfunction-associated steatotic liver disease; NW, normal weight; Ob, obesity.

TABLE 2.

Correlations between adropin and selected variables.

	Study cohort	NW group	Ob group	MASLD group
Age	0.044	−0.234	−0.289	0.463*
BMI (kg/m2)	0.122	−0.070	−0.196	0.187
iHOMA2-IR	0.093	0.198	0.034	0.046
Glucose	−0.074	−0.117	−0.059	−0.180
Insulin	0.096	0.164	0.017	0.063
ALT	0.104	−0.244	0.074	0.077
AST	0.111	−0.080	0.048	0.104
PWV	0.036	−0.313	−0.016	0.116
Systolic BP	0.104	−0.051	0.231	−0.061
Diastolic BP	−0.002	−0.194	0.056	−0.055
Heart rate	−0.141	0.128	−0.262	−0.267

Note: Values are for Pearson correlation (r) for the entire cohort (Study cohort) and for each group separately. Study cohort refers to the entire NW, control group with normal weight; Ob, control group with obesity; MASLD, group with obesity and metabolic dysfunction-associated steatotic liver disease.

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; BP, blood pressure; BMI, body mass index; iHOMA2-IR, integrated homeostatic model of assessment of insulin resistance; PWV, pulse wave velocity.

^*p < 0.05.

FIGURE 2.

Correlation between age and adropin within the MASLD group (r = 0.463, p < 0.001). MASLD, metabolic dysfunction-associated steatotic liver disease.

We also found no difference (p = 0.81) in adropin in participants with insulin resistance (defined as iHOMA2-IR ≥ 4.0, n = 36) and those without insulin resistance (iHOM2-IR < 4.0, n = 57). We explored whether there were sex differences in adropin since prior studies of rodents and humans suggested that they may be present.^10,25,26 We found that adropin was 21% higher in males compared to females in the entire study cohort (p = 0.034). This difference was evident within the Ob group (p = 0.018), but not within either the NW (p = 0.537) or the MASLD (p = 0.893) groups when considered separately (Supporting Information S1: Figure 1).

4 |. DISCUSSION

In the present study, we did not find that serum adropin was reduced in adolescents with MASLD as previously reported in studies of adults^11–13 and the one prior study of children and adolescents with MASLD.¹⁹ We also did not find that adropin varied with liver histological features as was reported in one of the prior studies of adults with MASLD.¹³ Likewise, adropin was not associated with central arterial stiffness as was recently shown in adults with type 2 diabetes.¹⁴

One of the strengths of the current study is that it is the first to report serum adropin values in adolescents with biopsy-confirmed MASLD. The only prior study of adolescents with MASLD relied on ultrasound for diagnosis,¹⁹ which can overestimate the presence of MASLD.²⁴ In addition to liver histology, the participants with MASLD in our study also had ALT and AST values that were considerably higher relative to the control groups, compared to the study by Sayin et al.¹⁹ The MASLD group in the present study also had insulin resistance and dyslipidemia (low HDL-C and high TG) consistent with prior studies of adults and children with MASLD that have reported values for adropin.^11–13,19 Thus, we expected to find lower serum adropin values in the Ob and/or MASLD groups in our study, similar to or greater in magnitude than what was previously reported.¹⁹ Our cohort was comprised predominantly of participants who identified as Hispanic/Latino, White, and/or Native American. Since other investigations on adropin in humans have not reported this information, it is unclear how genetics and/or cultural differences among study participants may impact the results. Serum adropin is reported to vary in response to short-term changes in dietary macronutrients in humans and animals.^27,28 We did not perform dietary analyses, nor has dietary information been reported in prior studies of adropin and MASLD, so we cannot determine the role of diet in our findings. At present, there are no obvious explanations for the lack of agreement between our results and what was previously reported in patients with MASLD.

A review published in 2020 noted that variability in adropin results among studies has made it challenging to confirm its role in normal physiology and disease pathology.²⁸ Some of that variability may be due to the use of different assays and the uncertainty about the half-life of adropin in circulation. We used the same assay that was previously used to show that adropin was reduced in adults with MASLD compared to NW controls¹² and to show the association between adropin and carotid-femoral pulse wave velocity in adults with diabetes.¹⁴ The assay performed well and as described by the manufacturer. We also used standardized approaches to collect, aliquot, and store all serum samples so we do not expect that technical error can account for the lack of differences among groups in this study.

We found that adropin was higher in boys than girls, especially in those with Ob. This is consistent with findings that adropin may be regulated by sex hormones. In agreement with our findings, Herrero et al.¹⁸ reported that adropin was higher in boys than girls who were pubertal. However, in pre-pubertal children, adropin was higher in girls. Additionally, Herrero et al.¹⁸ found that adropin was higher in children with Ob, but only in the pre-pubertal category; as we also found, there was no effect of Ob on serum adropin in pubertal adolescents. Adropin may regulated by testosterone, as a study of men with Ob found a positive association between circulating adropin and testosterone.²⁹ Likewise, a study of older adults also reported higher adropin in men than women.¹⁰ However, in rodents Enho was shown to be under the positive regulation of estrogen and the estrogen receptor alpha.^25,26 Whether estrogen-mediated pathways control adropin in humans is less clear.

The cause for the moderate, positive correlation between adropin and age within the MASLD group is not clear. In the study by Herrero et al.¹⁸ adropin concentration was lower in puberal compared to pre-puberal youth, especially in girls. In adults, adropin has been shown to decline with age.¹⁰ Since adropin is increased by high-calorie diets, particularly dietary fructose, both of which increase the risk of developing liver steatosis,²⁸ the finding that an age effect was present only in the MASLD group may reflect a cumulative influence of diet in that group.

One of the potential limitations of the current study is that there were more boys than girls in the MASLD group but not in the control groups. However, there were no differences in serum adropin among groups within either the boys or girls when considered separately. Further, this was a single-site study, and as noted, we cannot yet determine whether differences in the population we studied account for the lack of agreement with some prior reports. We also did not perform a detailed dietary analysis so the role of specific dietary components, which may vary among populations, cannot be determined.

In conclusion, the concentration of serum adropin did not differ in adolescents with MASLD compared to control groups with Ob or NW without MASLD. The moderate positive correlation between age and adropin within the MASLD group is a novel finding that requires additional investigation. The age and sex effects on adropin are not yet fully elucidated and also require further investigation.

What is Known

• Adropin is a secreted liver peptide that protects against inflammation and arterial stiffening in animals and adults.

• Serum adropin was lower in adolescents with metabolic dysfunction-associated steatotic liver disease (MASLD) in one prior study.

What is New

• This study is the first to compare serum adropin in adolescents with biopsy-confirmed MASLD and peers with obesity or normal weight without MASLD.

• Adropin did not differ among groups and did not vary with liver histology or arterial stiffness but was higher in boys than girls.

• The role of adropin in pediatric MASLD is not yet clear.