Relationship Between Elevated Hair Mercury Levels, Essential

Element Status, and Metabolic Profile in Overweight

and Obese Adults

Anatoly V. Skalny

1,2,3

& Jung-Su Chang

& Igor P. Bobrovnitsky

& Philippe Yu Kopylov

& Margarita G. Skalnaya

Shih-Yi Huang

& Monica Maria Bastos Paoliello

7,8

& Ekaterina S. Ivanova

& Weu Wang

& Alexey A. Tinkov

1,2

Received: 12 August 2020 / Accepted: 8 October 2020

Springer Science+Business Media, LLC, part of Springer Nature 2020

Abstract

The objective of the present study was to evaluate hair essential and trace element levels and metabolic risk markers in

overweight and obese subjects in relation to body mercury burden. According to 2 × 2 factorial design a total of 440 adults were

distributed to four groups: (i) low-Hg normal-weight subjects (n = 114); (ii) high-Hg normal weight subjects (n = 113); (iii) low-

Hg overweight (BMI > 25) subjects (n = 110); (iv) high-Hg overweight (BMI > 25) subjects (n = 110). Hg-exposed groups

consisted of subjects characterized by frequent seafood consumption (> 4 times/week) subsequently evaluated by hair analysis

(> 0.58 μg/g). Dietary-exposed subjects were characterized by a more than 3-fold higher hair Hg content irrespectively of body

weight values. Both low-Hg and high-Hg overweight subjects were characterized by significantly higher ALT activity, as well as

elevated serum glucose, LDL, and triglyceride levels as compared to the respective groups of normal weight subjects. High Hg

body burden had a more significant effect on metabolic parameters in overweight and obese adults. Particularly, high-Hg

overweight subjects were characterized by significantly higher serum creatinine and uric acid levels, as well as increased GGT

and CK activity as compared to low-Hg overweight counterparts. In addition, hair Mg, Mn, and Sr content in high-Hg overweight

subjects was significantly lower than that in low-Hg normal weight and overweight examinees. In turn, high Hg levels in

overweight subjects were associated with significantly higher hair Se and Zn levels when compared to unexposed overweight

adults. Generally, the obtained data demonstrate that increased hair Hg levels in overweight and obese subjects is associated with

adverse metabolic profile. It is proposed that observed metabolic alterations may be at least partially mediated by Hg-associated

disturbances in essential trace element and mineral metabolism.

Keywords Mercury

Metabolic risk

Magnesium

Zinc

Gamma-glutamyl transferase

Introduction

Obesity is considered as a worldwide epidemic that raised

drastically during the last 50 years reaching the mean preva-

lence of up to 19.5% worldwide in 2015 [1]. Although posi-

tive caloric balance due t o increased food intake and low

energy expenditure is the primary cause of obesity, certain

other factors including the environmental ones may also con-

tribute to increased risk of obesity [2]. In 2002, Baillie-

Hamilton proposed the association between chemical toxin

exposure and obesity epidemic [3]. Since then multiple studies

demonstrated the potential contribution of a variety of organic

* Anatoly V. Skalny

skalny3@microelements.ru

I. M. Sechenov First Moscow State Medical University (Sechenov

University), Moscow 119146, Russia

Yaroslavl State University, Yaroslavl 150000, Russia

ITMO University, St. Petersburg 197101, Russia

Taipei Medical University, Taipei, Taiwan

Centre for Strategic Planning, Russian Ministry of Health,

119435 Moscow, Russia

Micronutrients Ltd., Moscow, Russia

Department of Molecular Pharmacology, Albert Einstein College of

Medicine, Bronx, NY 10461, USA

Graduate Program in Public Health, Center of Health Sciences, State

University of Londrina, Londrina, PR 86038-350, Brazil

National Medical Research Center for Therapy and Preventive

Medicine of the Ministry of Health of Russia, Moscow, Russia

Biological Trace Element Research

https://doi.org/10.1007/s12011-020-02430-2

compounds, mainly persistent organic pollutants, in the path-

ogenesis of obesity [4]. Despite the observed association be-

tween metal exposure and obesity [5], data on the potential

role of toxic metals as obesogens are scarce [6].

Multiple studies demonstrated a significant association be-

tween mercury (Hg) body burden and increased BMI [7, 8],

adiposity [9], and metabolic syndrome [10], although certain

contradictions exist [11]. Therefore, it is proposed that Hg

may contribute to obesity and metabolic syndrome pathogen-

esis through its impact on adipose tissue, insulin signaling,

energy, lipid, and carbohydrate homeostasis [12, 13].

Correspondingly, a recent study demonstrated that Hg expo-

sure is associated with adverse metabolic profile in obesity

[14]. However, data on the relationship between Hg body

burden in obesity and metabolic risk are insufficient.

In parallel with mercury, essential trace elements and min-

erals are also involved in obesity and metabolic syndrome

through their physiological functions [15]. Particularly, zinc

[16 ], chromium [17], and vanadium [18] are known to be in-

volved in insulin signal transduction. Selenium is also consid-

ered as an important regulator of adipogenesis and adipocyte

physiology [19]. Magnesium possesses antiinflammatory and

antioxidant effect in obesity [20]. Correspondingly, altered es-

sential trace element and mineral metabolism was shown to be

associated with obesity and metabolic syndrome [21–23].

Along with multiple mechanisms of Hg toxicity, it is also

known to interfere with essential metal metabolism through

competitive binding to metalloproteins [24]ordirectinterac-

tion with selenol groups [25]. Correspondingly, experimental

studies demonstrated that Hg exposure may result in alteration

of essential element homeostasis resulting in alteration of their

target functions [26, 27]. However, human data on the poten-

tial role of Hg-associated disturbances in trace element and

mineral metabolism in obesity and metabolic syndrome are

insufficient.

Therefore, the objective of the present study was to evalu-

ate essential and trace element levels in hair, and metabolic

risk markers in overweight and obese subjects in relation to

body mercury burden.

Materials and Methods

The experimental procedures were performed in agreement

with the ethical standards of the Declaration of Helsinki and

its later amendments. All examinees were informed about the

procedures and signed the informed consent form prior the

investigation. The protocol of the present study was approved

by the local ethics committee (Yaroslavl State University,

Yaroslavl, Russia).

A total of 440 adults participants were enrolled in the cur-

rent study. The study involved a 2 × 2 factorial design with

grouping according to BMI values and hair Hg content. Given

that seafood intake is the primary cause of Hg exposure, the

first selection approach to group participants was based on the

individuals’ frequency of seafood consumption (> 4 times/

week). Based on subsequent ICP-MS analysis, only seafood

consumers with hair Hg concentrations > 0.58 μg/g, estimated

as the cutoff of environmental Hg exposure [28], were includ-

ed in the study. The subjects were recruited at the Center for

Biotic Medicine (Moscow, Russia). The groups were defined

as follows:

Group I (n = 114)—low-Hg normal-weight subjects

(controls). Median and IQR for BMI = 21.8 (20.2–23.4)

kg/m

Group II (n = 113)—high-Hg normal weight subjects.

Median and IQR for BMI = 21.8 (20.6–23.1) kg/m

Group III (n = 110)—low-Hg overweight (BMI > 25)

subjects. Median and IQR for BMI = 28.1 (26.3–30.5)

kg/m

Group IV (n = 110)—hig h-Hg overweight (BMI > 25)

subjects. Median and IQR for BMI = 27.5 (26.2–29.7)

kg/m

No significant group difference in age or gender of the

examinees were observed (Table 1). However, in view of

certain age and gender variability, all comparisons were ad-

justed for age and gender. Body mass index (BMI) was eval-

uated using the obtained data on body height (m) and weight

(kg) according to the formula (BMI = body weight (kg)/

height

)).

The examinees were interviewed to obtain information

about dietary habits (specific food habits including frequency

of fish consumption), occupation, smoking status, frequency

of alcohol consumption, city of origin, as well as duration of

living in Moscow. Questions on the use of hair dyes as well as

specific trace element enriched shampoos were also included.

In order to reduce the influence of confounding factors on

the analyzed variables, the following exclusion criteria were

applied: occupational Hg exposure in heavy metal industry;

smoking (current and former smoker status); using trace

element–enrich formulations as hair shampoos or other

haircare products; trace element and mineral supplementation;

metal implants; acute inflammatory diseases.

Venous blood samples were obtained after an overnight

fasting from the cubital vein using anticoagulant-free

“Vacuette” tubes (Greiner Bio-One International AG, 4550

Kremsmünster, Austria). Blood samples were subsequently

subjected to centrifugation at 1600g for 10 min to separate

serum. Only serum samples without hemolysis were used

for analysis.

Blood serum was analyzed for glucose, lipid spectrum, as

well as metabolic risk markers using the respective Randox

(Randox Laboratories Ltd., Crumlin, UK) kits at Tokyo Boeki

(Tokyo Boeki Machinery Ltd., Tokyo, Japan) biochemical

Skalny et al.

analyzer. Lipid spectrum was evaluated through assessment of

serum total cholesterol (TC), low (LDL-C), and high-density

lipoprotein cholesterol (HDL-C), as well as triglyceride (TG)

levels. Serum activity of alanine (ALT) and aspartate amino-

transferases (AST) [29], gamma-glutamyltransferase (GGT)

[30], as well as levels of creatinine [31], urea, and uric acid

(UA) concentration [32] were used as metabolic risk markers.

Hair samples were collected from occipital region in a

quantity of 0.1 g using ethanol-precleaned stainless steel scis-

sors. The collected samples were subjected to washing with

acetone a nd subsequent rinsing distilled deionized water

(18 MΩ cm). After drying on air to a stable weight the samples

were subjected to microwave HNO3-assisted (Sigma-Aldrich

Co., St. Louis, MO, USA) degradation in Berghof SW4 sys-

tem (Berghof Products & Instruments, Eningen, Germany) for

20 min at 170–180 °C.

Hair Hg and essential trace element and mineral levels were

evaluated using inductively-coupled plasma mass-

spectrometry at NexION 300D (PerkinElmer Inc., Shelton,

CT, USA) spectrometer equipped with ESI SC-2 DX4

(Elemental Scientific Inc., Omaha, NE, USA) autosampler.

The system was calibrated using Universal Data Acquisition

Standards Kit (PerkinElmer Inc., S helton, CT, USA).

Yttrium-89 and rhodium-103 standard solutions (10 μg/l) pre-

pared from yttrium (Y) and rhodium (Rh) Pure Single-

Element Standard (PerkinElmer Inc.) were used for internal

online standardization. In addition, the certified reference ma-

terial of human hair GBW09101 (Shanghai Institute of

Nuclear Research, Academia Sinica, Taipei, Taiwan, China)

was used for laboratory quality control with the recovery rates

of 87–111%.

The obtained data were processed using Statistica 10.0

software (StatSoft, Tulsa, OK, USA). Shapiro–Wilk test re-

vealed skewed distribution of the laboratory data. Therefore,

medians and interquartile range (IQR) boundaries were used

as descriptive statistics. In view of high variability of demo-

graphic parameters, group comparisons were performed using

analysis of covariance (ANCOVA) with adjustment for age

and gender of examinees with subsequent Bonferroni adjust-

ment. In order to reveal the association between Hg-associated

differences in essential element status and metabolic parame-

ters, multiple linear regression analysis was performed.

Metabolic parameters characterized by significant group dif-

ference (GGT, creatinine, CK, UA) were used as dependent

variables, whereas hair Hg, as well as essential trace element

and mineral levels were considered as independent predictors

in age, gender, and BMI-adjusted models. The level of signif-

icance was set as p < 0.05 for all statistical tests.

Results

Hair analysis using inductively-coupled plasma demonstrated

that dietary-exposed subjects were characterized by more than

3-fold higher hair Hg content as compared to the respective

body weight groups, in both normal and overweight groups

(Fig. 1).

Table 1 Metabolic risk markers in lean and overweight subjects in relation to hair Hg levels

Parameter Normal weight Overweight and obese

I. Low Hg II. High Hg III. Low Hg IV. High Hg

Age, years 43.99 ± 11.85 43.88 ± 10.53 44.95 ± 12.74 45.85 ± 8.55

Gender (M/F), % 29%/71 25%/75% 34%/66% 36%/64%

BMI, kg/m

21.80 ± 1.84 21.81 ± 1.74 28.79 ± 3.40 28.47 ± 3.05

ALT, U/l 17.97 (15.05–24.02) 18.17 (13.91–25.29) 24.67 (17.76–31.05)

1,2

26.53 (20.11–36.31)

1,2

AST, U/l 21.1 (18.26–25.79) 20.18 (17.16–25.3) 21.87 (18.83–25.48) 23.41 (18.19–27.76)

TC, mmol/l 5.14 (4.49–5.65) 5.19 (4.41–6.02) 5.45 (4.67–6.17) 5.11 (4.62–5.88)

Creatinine, μmol/l 81.36 (74.38–89.26) 80.67 (73.22–91.18) 87.11 (74.86–98.23)

90.79 (80.25–100.6)

1,2,3

CK, U/l 85.48 (68.81–119.75) 90.25 (65.03–127.38) 98.14 (63.78–148.45) 113.55 (79.64–172.2)

1,2,3

GGT, U/l 12.92 (8.48–20.6) 14.32 (10.46–22.65)

20.78 (12.65–34.41)

1,2

25.39 (15.5–37.52)

1,2,3

Glucose, mmol/l 5.16 (4.88–5.44) 5.1 (4.81–5.64) 5.65 (5.14–6.02)

1,2

5.47 (5.14–5.94)

1,2

HDL-C, mmol/l 1.56 (1.33–1.83) 1.62 (1.4–1.91) 1.42 (1.19–1.72)

1,2

1.34 (1.19–1.56)

1,2

LDL-C, mmol/l 2.59 (2.16–3.18) 2.8 (2.1–3.48) 3.24 (2.68–3.95)

1,2

3.26 (2.62–3.72)

1,2

TG, mmol/l 0.78 (0.64–1.08) 0.73 (0.58–0.97) 1.09 (0.78–1.58)

1,2

1.04 (0.81–1.36)

1,2

UA, μmol/l 189.3 (150.1–229.2) 190.4 (154.9–243.3) 259.1 (198.1–311.5)

1,2

277.7 (215.8–313.2)

1,2,3

Urea, mmol/l 4.66 (3.49–6.06) 4.58 (3.95–5.95) 5.17 (4.33–6.28)

1,2

5.4 (4.67–6.65)

1,2,3

Data expressed as median and IQR;

1,2,3

significant group difference in comparison to groups I, II, and III at p < 0.05 according to ANCOVA Bonferroni

post hoc. ALT alanine aminotransferase; AST aspartate aminotransferases; BMI body mass index; CK creatine kinase; GGT gamma-glutamyltransferase;

HDL-C high-density lipoprotein cholesterol; LDL-C low-density lipoprotein cholesterol; TG triglycerides; TC total cholesterol; UA uric acid

Relationship Between Elevated Hair Mercury Levels, Essential Element Status, and Metabolic Profile in...

Differences in both hair Hg content and BMI had a signif-

icant impact on metaboli c parame ters of blood ser um

(Table 1). The obtained data demonstrate that increased hair

Hg levels in normal weight subjects is associated with 11%

higher serum GGT activity as compared to the control sub-

jects. No significant group difference in other metabolic

markers was revealed between high-Hg and low-Hg normal

weight adults.

In contrast to normal weight subjects, high Hg body burden

had a more significant effect on metabolic parameters in over-

weight and obese adults. Particularly, serum creatinine and

uric acid levels in high-Hg overweight examinees exceeded

those in low-Hg overweight counterparts by 4% and 7%, re-

spectively. At the same time, serum creatine kinase and GGT

activity in high-Hg overweight group was 16% and 22%

higher when compared to the respective weight group with

low hair Hg content.

Both low-Hg and high-Hg overweight subjects were char-

acterized by significantly higher ALT activity, as well as ele-

vated serum glucose, LDL-C, and triglyceride levels as com-

pared to the respective groups of normal weight subjects.

High Hg body burden was also associated with significant

differences in hair essential trace element levels (Table 2). In

particular, normal weight high-Hg subjects were characterized

by significantly elevated hair Co, Cu, Se, Si, V, and Zn as

compared to the contro l group by 56%, 11%, 11%, 9%,

67%, and 9%, respectively. In overweight subjects with low

hair Hg content the levels of Ca, Mg, and Zn were found to be

30%, 37%, and 11% lower than those in low-Hg normal

weight controls. More profound differences were observed

in high-Hg overweight group. Specifically, hair Mg levels in

high-Hg overweight subjects were significantly lower than

that in controls and high-Hg normal weight examinees by

49% and 54%, respectively. Hair Mn and Sr concentrations

in high-Hg overweight group compared to low- and high-Hg

lean, and low-Hg obese groups, were lower by 25%, 39%,

22%, and 42%, 43%, 30%, respectively. The highest hair Se

levels were also observed in high-Hg overweight subjects,

being higher than those in groups I, II, and III by 24%, 13%,

and 20%, respectively. It is also notable that the observed 4%

higher Zn levels in high-Hg overweight subjects were signif-

icantly different from low-Hg overweight examinees.

In order to reveal the relative contribution of hair Hg and

essential element levels into the observed difference in GGT,

creatinine, and uric acid, multiple regression analysis was per-

formed (Table 3). After adjustment for age, gender, and BMI

(in all models) hair Hg was considered as a significant predic-

tor only for GGT activity. Hair Cu and Sr levels demonstrated

positive association with serum GGT, whereas hair Mn and

Zn levels were considered as negative predictors of serum

GGT activity. Serum creatinine concentration was inversely

associated with hair Co and Cr content. In turn, hair Fe levels

were positively interrelated with this parameter. Creatine ki-

nase activity was positively and negatively associated with

hair Se and Si levels, respectively. In an adjusted regression

model, hair Cu levels were positively associated with serum

uric acid concentration, whereas hair Si and Zn content was

considered as negative predictor of this parameter. The overall

models accounted for 25%, 33%, 9%, and 40% of GGT, cre-

atinine, CK, and UA variability, respectively.

Discussion

The obtained data demonstrate that Hg-exposed overweight

and obese subjects are characterized by a tendency for higher

metabolic risk as well as altered essential trace element status.

Moreover, Hg-associated differences in essential element sta-

tus were tightly associated with metabolic parameters.

Generally, observed increase in serum glucose, TG, LDL,

and decreased HDL-C levels in overweight low-Hg patients

correspond to the clearly demonstrated association between

obesity, insulin resistance, and atherogenic dyslipidemia. In

turn, elevated GGT and creatine kinase activity, as well as

creatinine and uric acid levels are considered as markers of

metabolic syndrome. Increased ALT activity [33] and urea

[34] concentrations may be indicative of higher risk of nonal-

coholic fatty liver disease in patients with obesity.

At the same time, patients with high hair Hg concentrations

were characterized by more profound increase in serum creat-

inine and uric acid levels, and higher GGT and creatine kinase

activities, thus being indicative of higher metabolic risk.

Recent findings corroborate the earlier data. Particularly, a

significant association between blood Hg levels and increased

GGT activity was demonstrated [35]. Our previous findings

also revealed a positive correlation between hair Hg content

and serum GGT activity [36]. It is also notable that higher Hg

Fig. 1 Hair Hg levels in relation to body weight index and dietary Hg

exposure. Data expressed as median (line), IQR (box), and nonoutlier

range (whiskers). *Significant group difference at p < 0.05 according to

ANCOVA Bonferroni post hoc

Skalny et al.

levels in overweight subjects was associated with elevated

creatinine concertation and creatine kinase levels. Given the

role of these parameters as markers of renal damage [37],

these findings corroborate the existing data on the role of Hg

as a nephrotoxic agent [38], and may be indicative of higher

risk of chronic kidney disease in obesity and metabolic syn-

drome [39]. Serum uric acid levels were also associated with

blood Hg levels in Korean general population [40]. Generally,

Table 3 Multiple regression analysis on the association between metabolic risk markers characterized by significant group differences (dependent

variable) and hair element content (independent variables)

Parameter GGT Creatinine CK UA

β p β p β p β p

Hg 0.122 0.008 * 0.019 0.658 0.025 0.630 0.029 0.484

Ca 0.055 0.413 0.065 0.306 − 0.059 0.424 − 0.056 0.349

Co − 0.033 0.534 − 0.110 0.027 * − 0.017 0.773 − 0.001 0.980

Cr − 0.045 0.344 − 0.135 0.003 * − 0.058 0.267 − 0.007 0.864

Cu 0.131 0.006 * 0.055 0.219 − 0.024 0.642 0.085 0.044 *

Fe 0.040 0.452 0.109 0.030 * 0.034 0.557 − 0.049 0.296

Mg 0.032 0.665 0.033 0.635 − 0.002 0.981 0.119 0.076

Mn − 0.130 0.019 * − 0.008 0.875 0.063 0.297 − 0.090 0.068

Se − 0.057 0.196 0.019 0.653 0.102 0.039 * 0.025 0.530

Si 0.009 0.830 − 0.070 0.083 − 0.124 0.009 * − 0.084 0.028 *

Sr 0.186 0.002 * − 0.020 0.723 − 0.032 0.622 − 0.004 0.946

V − 0.023 0.644 0.087 0.064 0.010 0.858 0.047 0.291

Zn − 0.217 < 0.001 * 0.010 0.835 0.004 0.948 − 0.099 0.026 *

Age 0.312 < 0.001 * 0.555 < 0.001 * 0.263 < 0.001 * 0.478 < 0.001 *

Gender (M/F) 0.146 0.001 * 0.080 0.060 − 0.007 0.882 0.095 0.018 *

BMI 0.231 < 0.001 * 0.062 0.168 − 0.019 0.720 0.260 < 0.001 *

Multiple R 0.527 0.594 0.344 0.650

Multiple R

0.277 0.353 0.119 0.423

Adjusted R

0.250 0.328 0.085 0.401

p for a model < 0.001 * < 0.001 * < 0.001 * < 0.001 *

Data expressed as regression coefficients (β) and the respective p values; *The association is significant at p <0.05

Table 2 Hair essential trace element and mineral levels in normal weight and overweight adults in relation to hair Hg concentrations (μg/g)

Parameter Normal weight Overweight and obese

I. Low Hg II. High Hg III. Low Hg IV. High Hg

Ca 657.4 (401.9–1414.4) 681.9 (473.8–1314.1) 463.1 (286.3–848.0)

1,2

471.9 (335.1–675.2)

1,2

Co 0.009 (0.005–0.018) 0.014 (0.008–0.034)

0.009 (0.005–0.015)

0.009 (0.006–0.019)

Cr 0.138 (0.072–0.241) 0.176 (0.101–0.295) 0.156 (0.094–0.287) 0.183 (0.097–0.274)

Cu 12.96 (10.94–18.44) 14.43 (11.16–25.84)

12.27 (9.95–15.69)

13.24 (10.6–17.95)

Fe 11.62 (8.77–17.89) 12.06 (8.89–17.95) 11.88 (8.26–15.8) 10.97 (8.34–16.19)

Mg 83.42 (41.62–158.08) 92.44 (45.78–166.08) 52.85 (25.27–127.33)

1,2

42.57 (25.22–69.35)

1,2,3

Mn 0.388 (0.256–0.807) 0.411 (0.247–0.861) 0.373 (0.215–0.68) 0.291 (0.194–0.476)

1,2,3

Se 0.358 (0.281–0.487) 0.392 (0.317–0.499)

0.370 (0.269–0.493) 0.444 (0.308–0.574)

1,2,3

Si 16.04 (10.61–20.75) 17.55 (12.58–24.24)

17.17 (12–23.9) 15.33 (11.98–20.83)

Sr 2.182 (1.026–4.696) 2.195 (0.962–4.83) 1.802 (0.726–5.701) 1.270 (0.687–2.59)

1,2,3

V 0.009 (0.006–0.019) 0.015 (0.009–0.034)

0.012 (0.007–0.020)

0.010 (0.007–0.019)

Zn 194.0 (163.2–228.3) 210.7 (181.7–246.1)

171.9 (137.0–205.2)

1,2

178.5 (152.8–215.3)

2,3

Data expressed as median and IQR;

1,2,3

Significant group difference in comparison to groups 1, 2, and 3 at p < 0.05 according to ANCOVA Bonferroni

post hoc

Relationship Between Elevated Hair Mercury Levels, Essential Element Status, and Metabolic Profile in...

these obtained data corroborate findings by Lee [14] who

demonstrated adverse metabolic profile and tendency for met-

abolic unhealthy obesity with increased blood Hg levels.

In the present study, analysis of hair essential trace element

and mineral content demonstrated that overweight and obesity

were associated with significantly reduced Ca, Mg, and Zn

levels when compared to normal weight controls.

The observed levels of Ca in hair of overweight subjects

generally correspond to the earlier finding of lower Ca intake

in obesity [40, 41]. Experimental data also demonstrated that

low Ca intake is associated with higher susceptibility to diet-

induced obesity [42] and associated metabolic disturbances

[43]. Generally, the relationship between obesity and Ca me-

tabolism may be mediated by the role of the latter in regulation

of adipogenesis, lipid metabolism, thermogenesis, and gut mi-

crobiota [44]. It is also notable that the relationship between

Ca and obesity may be mediated by vitamin D [45].

Recent data demonstrate significantly reduced intracellular

Mg levels in subjects with obesity and diabetes [46], being in

agreement with our findings. Correspondingly, Mg intake is

known to be inversely associated with obesity and metabolic

syndrome [47]. In addition, low Mg levels is known to be

associated with adverse metabolic profile in obesity [48].

Antiinflammatory, antioxidant, hypolipidemic, as well as

insulin-sensitizing effects of Mg may underlie the observed

differences [20, 49].

Similarly, Zn deficiency was shown to be a frequent find-

ing in overweight and obese subjects [23], being related to

obesity-associated metabolic disturbances including dyslipid-

emia, insulin resistance, and hypertension [50]. In turn, Zn

supplementation was reported to improve metabolic status in

obese subjects [51]. A close association between zinc, obesity,

and metabolic syndrome is mediated by the role of Zn in

insulin secretion and signaling, adipogenesis and adipocyte

functioning, lipi d metabolism, as well as antioxid ant and

antiinflammatory effect [16].

Along with more profound alterations of metabolic param-

eters in overweight subjects, Hg exposure was associated with

aberrant essential trace element profile in hair. Particularly, Hg

exposure aggravated overweight-associated decrease in hair

Mg content. Experimental data demonstrated that methylmer-

cury exposure significantly reduced kidney and spleen Mg

levels in rats [26]. These effects may result from competitive

binding of Hg

to Mg-containing enzymes [24].

Although no further decrease in hair Zn levels were ob-

served in Hg-exposed overweight subjects, h igh hair Hg

levels were associated with elevation of hair Zn that may be

also indicative of Hg-induced alterations in Zn metabolism.

Correspondingly, inorganic Hg exposure was shown to in-

crease kidney Zn levels in vivo [26]. At the same time, zinc

treatment may counteract Hg-induced metabolic disturbances

including altered ALT activity, as well as urea, creatinine, and

glucose levels [52], whereas Zn-induced metallothionein

production may reduce tissue Hg deposition thus preventing

its adverse effects [53].

It is also notable that dietary Hg exposure in the overweight

group led to increased hair Cr levels. Despite a significant

increase in hair Cr levels in high-Hg overweight subjects,

these findings may be indicative of lower serum Cr levels as

demonstrated in the study of paired hair-serum-urine samples

in obese examinees [54]. However, the particular mechanisms

of Hg-induced alterations of Cr homeostasis were not estimat-

ed. Given the role of Cr as insulin signaling and adipose tissue

physiology [17], one can propose that altered Cr metabolism

may significantly contribute to obesity-associated metabolic

disorders, especially in high-Hg group.

TheobserveddecreaseinhairMnlevelsinoverweightHg-

exposed subjects generally corresponds to the results of an

earlier in vivo study demonstrating a significant reduction in

hepatic Mn levels in methylmercury-exposed rats due to Hg-

induced arginase I inhibition and Mn

release [55]. In turn,

Mn was also shown to be involved in metabolic syndrome

pathogenesis through its interference with redox homeostasis,

lipid and carbohydrate metabolism [56].

Increased hair Hg content was associated with elevated hair

Se levels both in normal weight and overweight subjects. Se is

known to be Hg antagonist acting through a variety of mecha-

nisms including direct chemical interaction with the formation

of Hg-Se compounds that are excreted from the organism [25]

that may underlie a direct relationship between these elements

in hair. In addition, given the primary role of seafood intake in

elevated hair Hg content, this dietary pattern may also underlie

higher hair Se levels due to the role of sea fish as a dietary

source of Se [57]. At the same time, Hg-associated increase in

hair Se content was more profound in overweight group, being

in agreement with our observation of higher body Se burden in

obese subjects [54]. Se was also shown to modify the associa-

tionbetweenmercuryandmetabolic syndrome [58].

The obtained data correspond well to the role of seafood

consumption as the source of mercury [13]. At the same time,

seafood intake also contributes significantly to exposure to

polychlorinated biphenyls and organochlorine pesticides

[59] that are considered as potent obesogens and promotors

of metabolic syndrome due to endocrine disruption [60].

Therefore, the observation of adverse metabolic profile in sea-

food consumers may be at least partially mediated by organic

obesogens.

Generally, the obtained data demonstrate that increased

hair Hg levels in overweight and obese subjects is associated

with adverse metabolic profile. It is proposed that observed

metabolic alterations may be at least partially mediated by Hg-

associated disturbances in essential trace element and mineral

metabolism. However, further laboratory and epidemiological

studies are required to unravel the relationship between Hg

exposure and trace element and mineral metabolism in obesity

and its impact on metabolic syndrome pathogenesis.

Skalny et al.

Funding The study was supported by grants No. 20-515-S52003 from

the Russian Foundation for Basic Research (Russia) and MOST 109-

2923-B-038-001-MY3 from the Ministry of Science and Technology

(Taiwan).

Compliance with Ethical Standards

Conflict of Interest The authors declare that they haave no conflict of

interest.

References

1. Organisation for Economic Co-operation and Development.

Obesity update 2017. OECD https://www.oecd.org/els/health-

systems/Obesity-Update-2017.pdf (2017)

2. Blüher M (2019) Obesity: global epidemiology and pathogenesis.

Nat Rev Endocrinol 15(5):288–298. https://doi.org/10.1038/

s41574-019-0176-8

3. Baillie-Hamilton PF (2002) Chemical toxins: a hypothesis to ex-

plain the global obesity epidemic. J Altern Complement Med 8(2):

185–192. https://doi.org/10.1089/107555302317371479

4. Nappi F, Barrea L, Di Somma C, Savanelli MC, Muscogiuri G,

Orio F, Savastano S (2016) Endocrine aspects of environmental

“Obesogen” pollutants. Int J Environ Res Public Health 13(8):

765. https://doi.org/10.3390/ijerph13080765

5. Wang X, Mukherjee B, Park SK (2018) Associations of cumulative

exposure to heavy metal mixtures with obesity and its comorbidities

among U.S. adults in NHANES 2003-2014. Environ Int 121(Pt 1):

683–694. https://doi.org/10.1016/j.envint.2018.09.035

6. Heindel JJ (2019) History of the Obesogen field: looking back to

look forward. Front Endocrinol (Lausanne) 10:14. https://doi.org/

10.3389/fendo.2019.00014

7. Skalnaya MG, Tinkov AA, Demidov VA, Serebryansky EP,

Nikonorov AA, Skalny AV (2014) Hair toxic element content in

adult men and women in relation to body mass index. Biol Trace

Elem Res 161(1 ):13–19. https://doi.org/10 .1007/s12011 -014-

0082-9

8. Shin YY, Ryu IK, Park MJ, Kim SH (2018) The association of total

blood mercury levels and overweight among Korean adolescents:

analysis of the Korean National Health and Nutrition Examination

Survey (KNHANES) 2010-2013. Korean J Pediatr 61(4):121–128.

https://doi.org/10.3345/kjp.2018.61.4.121

9. Park JS, Ha KH, He K, Kim DJ (2017) Association between blood

mercury level and visceral adiposity in adults. Diabetes Metab J

41(2):113–120. https://doi.org/10.4093/dmj.2017.41.2.113

10. Eom SY, Choi SH, Ahn SJ, Kim DK, Kim DW, Lim JA, Choi BS,

Shin HJ, Yun SW, Yoon HJ, Kim YM, Park K, Seo E (2016)

Association between toenail mercury and metabolic syndrome is

modified by selenium. Nutrients 8(7):424. https://doi.org/10.3390/

nu8070424

11. Rothenberg SE, Korrick SA, Fayad R (2015) The influence of

obesity on blood mercury levels for U.S. non-pregnant adults and

children: NHANES 2007-2010. Environ Res 138:173–180. https://

doi.org/10.1016/j.envres.2015.01.018

12. Tinkov AA, Ajsuvakova OP, Skalnaya MG, Popova EV, Sinitskii

AI, Nemereshina ON, Gatiatulina ER, Nikonorov AA, Skalny AV

(2015) Mercury and metabolic syndrome: a review of experimental

and clinical observations. Biometals 28(2):231–254. https://doi.org/

10.1007/s10534-015-9823-2

13. Roy C, Tremblay PY, Ayotte P (2017) Is mercury exposure causing

diabetes, metabolic syndrome and insulin resistance? A systematic

review of the literature. Environ Res 156:747–760. https://doi.org/

10.1016/j.envres.2017.04.038

14. Lee K (2018) Blood mercury concentration in relation to metabolic

and weight phenotypes using the KNHANES 2011-2013 data. Int

Arch Occup Environ Health 91(2):185–193. https://doi.org/10.

1007/s00420-017-1269-0

15. Wiernsperger N, Rapin J (2010) Trace elements in glucometabolic

disorders: an update. Diabetol Metab Syndr 2:70. https://doi.org/10.

1186/1758-5996-2-70

16. Olechnowicz J, Tinkov A, Skalny A, Suliburska J (2018) Zinc

status is associated with inflammation, oxidative stress, lipid, and

glucose metabolism. J Physiol Sci 68(1):19–31. https://doi.org/10.

1007/s12576-017-0571-7

17. Vincent JB (2019) Effects of chromium supplementation on body

composition, human and animal health, and insulin and glucose

metabolism. Curr Opin Clin Nutr Metab Care 22(6):483–489.

https://doi.org/10.1097/MCO.0000000000000604

18. Treviño S, Diaz A (2020) Vanadium and insulin: partners in meta-

bolic regulation. J Inorg Biochem 208:111094. https://doi.org/10.

1016/j.jinorgbio.2020.111094

19. Tinkov AA, Ajsuvakova OP, Filippini T, Zhou JC, Lei XG,

Gatiatulina ER, Michalke B, Skalnaya MG, Vinceti M, Aschner

M, Skalny AV (2020) Selenium and selenoproteins in adipose tis-

sue physiology and obesity. Biomolecules 10(4):658. https://doi.

org/10.3390/biom10040658

20. Nielsen FH (2018) Magnesium deficiency and increased inflamma-

tion: current perspectives. J Inflamm Res 11:25–34. https://doi.org/

10.2147/JIR.S136742

21. Astrup A, Bügel S (2010) Micronutrient deficiency in the aetiology

of obesity. Int J Obes 34(6):947–948. https://doi.org/10.1038/ijo.

2010.81

22. Via M (2012) The malnutrition of obesity: micronutrient deficien-

cies that promote diabetes. ISRN Endocrinol 2012:103472–

103478. https://doi.org/10.5402/2012/103472

23. Gu K, Xiang W, Zhang Y, Sun K, Jiang X (2019) The association

between serum zinc level and overweight/obesity: a meta-analysis.

Eur J Nutr 58(8):2971–2982. https://doi.org/10.1007/s00394-018-

1876-x

24. Ynalvez R, Gutierrez J, Gonzalez-Cantu H (2016) Mini-review:

toxicity of mercury a s a consequence o f enz yme alterati on.

Biometals. 29(5):781–788. https://doi.org/10.1007/s10534-016-

9967-8

25. Bjørklund G, Aaseth J, Ajsuvakova OP, Nikonorov AA, Skalny

AV, Skalnaya MG, Tinkov AA (2017) Molecular interaction be-

tween mercury and selenium in neurotoxicity. Coord Chem Rev

332:30–37. https://doi.org/10.1016/j.ccr.2016.10.009

26. Bogden JD, Kemp FW, Troia no RA, Jortner BS, Timpone C,

Giuliani D (1980) Effect of mercuric chloride and methylmercury

chloride exposure on tissue concentrations of six essential minerals.

Environ R es 21(2):350–359. h ttps://doi.org/10.10 16/0013-

9351(80)90037-7

27. Feng W, Wang M, Li B, Liu J, Chai Z, Zhao J, Deng G (2004)

Mercury and trace element distribution in organic tissues and re-

gional brain of fetal rat after in utero and weaning exposure to low

dose of inorganic mercury. Toxicol Lett 152(3):223–234. https://

doi.org/10.1016/j.toxlet.2004.05.001

28. Trasande L, DiGangi J, Evers DC, Petrlik J, Buck DG, Šamánek J,

Beeler B, Turnquist MA, Regan K (2016) Economic implications

of mercury exposure in the context of the global mercury treaty: hair

mercury levels and estimated lost economic productivity in selected

developing countries. J Environ Manag 183:229–235. https://doi.

org/10.1016/j.jenvman.2016.08.058

29. Kim HR, Han MA (2018) Association between serum liver en-

zymes and metabolic syndrome in Korean adults. Int J Environ

Res Public Health 15(8):1658. https://doi.org/10.3390/

ijerph15081658

30. Lee DS, Evans JC, Robins SJ, Wilson PW, Albano I, Fox CS,

Wang TJ, Benjamin EJ, D' Agostino RB, Vasan RS (2007)

Relationship Between Elevated Hair Mercury Levels, Essential Element Status, and Metabolic Profile in...

Gamma glutamyl transferase and metabolic syndrome, cardiovas-

cular disease, and mortality risk: the Framingham Heart Study.

Arterioscler Thromb Vasc Biol 27(1):127–133. https://doi.org/10.

1161/01.ATV.0000251993.20372.40

31. Wang J, Li X, Han X, Yang K, Liu B, Li Y, Wu P, Liu X, Yu K, Dai

X, Yuan J, Yao P, Zhang X, Guo H, Wang Y, Chen W, Wei S,

Miao X, Min X, Liang Y, Yang H, Hu FB, Wu T, He M (2015)

Serum creatinine levels and risk of metabolic syndrome in a middle-

aged and older Chinese population. Clin Chim Acta 440:177–182.

https://doi.org/10.1016/j.cca.2014.11.025

32. Yang T, Chu CH, Bai CH, You SL, Chou YC, Chou WY, Chien

KL, Hwang LC, Su TC, Tseng CH, Sun CA (2012 Feb) Uric acid

level as a risk marker for metabolic syndrome: a Chinese cohort

study. Atherosclerosis. 220(2):525–531. https://doi.org/10.1016/j.

atherosclerosis.2011.11.014

33. Chen ZW, Chen LY, Dai HL, Chen JH, Fang LZ ( 2008)

Relationship between alanine aminotransferase levels and metabol-

ic syndrome in nonalcoholic fatty liver disease. J Zhejiang Univ Sci

B9(8):616–622. https://doi.org/10.1631/jzus.B0720016

34. Liu X, Zhang H, Liang J (2013) Blood urea nitrogen is elevated in

patients with non-alcoholic fatty liver disease.

Hepatogastroenterology 60(122):343–345

35. Seo MS, Lee HR, Shim JY, Kang HT, Lee YJ (2014) Relationship

between blood mercury concentrations and serum γ -

glutamyltranspeptidase level in Korean adults using data from the

2010 Korean National Health and Nutrition Examination Survey.

Clin Chim Acta 430:160–163. https://doi.org/10.1016/j.cca.2014.01.

042

36. Tinkov AA, Skalnaya MG, Demido v VA, Serebryansky EP,

Nikonorov AA, Skalny AV (2014) Hair mercury association with

selenium, serum lipid spectrum, and gamma-glutamyl transferase

activity in adults. Biol Trace Elem Res 161(3):255–262. https://doi.

org/10.1007/s12011-014-0124-3

37. Gowda S, Desai PB, Kulkarni SS, Hull VV, Math AA, Vernekar

SN (2010) Markers of renal function tests. N Am J Med Sci 2(4):

170–173

38. Bridges CC, Zalups RK (2017) The aging kidney and the nephro-

toxic effects of mercury. J Toxicol Environ Health B Crit Rev

20(2):55–80. https://doi.org/10.1080/10937404.2016.1243501

39. Chen J, Gu D, Chen CS, Wu X, Hamm LL, Muntner P, Batuman V,

Lee CH, Whelton PK, He J (2007) Association between the meta-

bolic syndrome and chronic kidney disease in Chinese adults.

Nephrol Dial Transplant 22(4):1100–1106. https://doi.org/10.

1093/ndt/gfl759

40. Schrager S (2005) Dietary calcium intake and obesity. J Am Board

Fam Pract 18(3):205–210. https://doi.org/10.3122/jabfm.18.3.205

41. Skowrońska-Jóźw iak E, Jawo rski M, Lo renc R, K arbownik -

Lewińska M, Lewiński A (2017) Low dairy calcium intake is as-

sociated with overweight and elevated blood pressure in Polish

adults, notably in p remenopausal women. Public Health Nutr

20(4):630–637. https://doi.org/10.1017/S1368980016002706

42. He YH, Li ST, Wang YY, Wang G, He Y, Liao XL, Sun CH, Li Y

(2012) Postweaning low-calcium diet promotes later-life obesity

induced by a high-fat diet. J Nutr Biochem 23(10):1238–1244.

https://doi.org/10.1016/j.jnutbio.2011.07.004

43. Marotte C, Bryk G, Gonzales Chaves MM, Lifshitz F, de Portela

ML, Zeni SN (2014) Low dietary calcium and obesity: a compara-

tive study in genetically obese and normal rats during early growth.

Eur J Nutr 53(3):769–778. https://doi.org/10.1007/s00394-013-

0581-z

44. Zhang F, Ye J, Zhu X, Wang L, Gao P, Shu G, Jiang Q, Wang S

(2019) Anti-obesity effects of dietary calcium: the evidence and

possible mechanisms. Int J Mol Sci 20(12):3072. https://doi.org/

10.3390/ijms20123072

45. Pannu PK, Calton EK, Soares MJ (2016) Calcium and vitamin D in

obesity and related chronic disease. Adv Food Nutr Res 77:57–100.

https://doi.org/10.1016/bs.afnr.2015.11.001

46. Takaya J, Yamato F, Kuroyanagi Y, Higashino H, Kaneko K

(2010) Intracellular magnesium of obese and type 2 diabetes

mellitus children. Diabetes Ther 1(1):25–31. https://doi.org/10.

1007/s13300-010-0003-7

47 . Sarrafzadegan N, Khosravi-Boroujeni H , Lotfizadeh M,

Pourmogaddas A, Salehi-Abargouei A (2016) Magnesium status and

the metabolic syndrome: a systematic review and meta-analysis.

Nutrition 32(4):409–417. https://doi.org/10.1016/j.nut.2015.09.014

48. Guerrero-Romero F, Rodriguez-Moran M (2013) Serum magne-

sium in the metabolically-obese normal-weight and healthy-obese

subjects. Eur J Intern Med 24(7):639–643. https://doi.org/10.1016/

j.ejim.2013.02.014

49. Volpe SL (2013) Magnesium in disease prevention and overall

health. Adv Nutr 4(3):378S–383S. https://doi.org/10.3945/an.112.

003483

50. Freitas EP, Cunha AT, Aquino SL, Pedrosa LF, Lima SC, Lima JG,

Almeida MG, Sena-Evangelista KC (2017) Zinc status biomarkers

and cardiometabolic risk factors in metabolic syndrome: a case

control study. Nutrients 9(2):175. https://doi.org/10.3390/

nu9020175

51. Kim J, Ahn J (2014) Effect of zinc supplementation on inflamma-

tory markers and adipokines in young obese women. Biol Trace

Elem Res 157(2):101–106. https://doi.org/ 10.1007/s12011-013-

9885-3

52. Franciscato C, Moraes-Silva L, Duarte FA, Oliveira CS, Ineu RP,

Flores EM, Dressler VL, Peixoto NC, Pereira ME (2011) Delayed

biochemical changes induced by mercury intoxication are

prevented by zinc pre-exposure. Ecotoxicol Environ Saf 74(3):

480–486. https://doi.org/10.1016/j.ecoenv.2010.11.011

53. Peixoto NC, Serafim MA, Flores EM, Bebianno MJ, Pereira ME

(2007) Metallothionein, zinc, and mercury levels in tissues of

young rats exposed to zinc and subsequently to mercury. Life Sci

81(16):1264–1271. https://doi.org/10.1016/j.lfs.2007.08.038

54. Tinkov AA, Skalna ya MG, Ajsuvakova OP, Serebryansky EP,

Chao JC, Aschner M, Skalny AV (2020) Selenium, zinc, chromi-

um, and vanadium levels in serum, hair, and urine samples of obese

adults assessed by inductively coupled plasma mass spectrometry.

Biol Trace Elem Res 1-8. https://doi.org/10.1007/s12011-020-

02177-w.

55. Kanda H, Sumi D, Endo A, Toyama T, Chen CL, Kikushima M,

Kumagai Y (2008) Reduction of arginase I activity and manganese

levels in the liver during exposure of rats to methylmercury: a

possible mechanism. Arch Toxicol 82(11):803–808. https://doi.

org/10.1007/s00204-008-0307-9

56. Li L, Yang X (2018) The essential element manganese, oxidative

stress, and metabolic diseases: links and interactions. Oxidative

Med Cell Longev 2018:7580707–7580711. https://doi.or g/10.

1155/2018/7580707

57. Yamashita Y, Yamashita M, Iida H (2013) Selenium content in

seafood in Japan. Nutrients 5(2):388–395. https://doi.org/10.3390/

nu5020388

58. Park K, Seo E (2016) Association between toenail mercury and

metabolic syndrome is modified by selenium. Nutrients 8(7):424

59. Qian Z, Luo F, Wu C, Zhao R, Cheng X, Qin W (2017) Indicator

polychlorinated biphenyls (PCBs) and organochlorine pesticides

(OCPs) in seafood from Xiamen (China): levels, distributions, and

risk assessment. Environ Sci Pollut Res 24:10443–10453

60. Heindel JJ, Blumberg B (2019) Environmental obesogens: mecha-

nisms and controversies. Annu Rev Pharmacol Toxicol 59:89–106

Publisher’sNoteSpringer Nature remains neutral with regard to jurisdic-

tional claims in published maps and institutional affiliations.

Skalny et al.