Concentrations of Persistent Organic Pollutants in Women’s Serum in the European Arctic Russia

Persistent organic pollutants (POPs) are heterogeneous carbon-based compounds that can seriously affect human health. The aim of this study was to measure serum concentrations of POPs in women residing in the Euro-Arctic Region of Russia. A total of 204 women from seven rural settlements of the Nenets Autonomous Okrug (NAO) took part in the study. We measured serum concentrations of 11 polychlorinated biphenyls (PCBs) and 17 organochlorine pesticides (OCPs) across the study sites and among Nenets and non-Nenets residents. Measurement of POPs was performed using an Agilent 7890A gas chromatograph equipped with an Agilent 7000 series MS/MS triple quadrupole system. The concentrations of all POPs were low and similar to findings from other Arctic countries. However, significant geographic differences between the settlements were observed with exceptionally high concentrations of PCBs in Varnek located on Vaygach Island. Both ΣDDT (p = 0.011) and ΣPCB (p = 0.038) concentrations were significantly lower in Nenets. Our main findings suggest that the serum concentrations of the legacy POPs in women in the Euro-Arctic Region of Russia are low and similar to those in other Arctic countries. Significant variations between settlements, and between Nenets and non-Nenets residents, were found. Arctic biomonitoring research in Russia should include studies on the associations between nutrition and concentrations of POPs.


Introduction
Persistent organic pollutants (POPs) are heterogeneous carbon-based compounds of natural or anthropogenic origin [1]. According to the Stockholm Convention on Persistent 1). The data from Nelmin-Nos and Krasnoe were merged with that of Pechora due to similarities between these two settlements situated along the Pechora river. All adult residents of the abovementioned settlements were informed about the study and invited to participate by the local health care units. Among those who participated, we selected only those who lived at least half of their lives in the index settlement. After applying exclusion criteria [25] and excluding those with acute illnesses and pregnant and lactating women, the final sample consisted of 204 women. The participation rate ranged from 5.2% of the total female population in Pechora to 22% in Varnek. All participants filled out a questionnaire with the assistance of trained local health care workers. Fasting blood samples were taken the same day [25]. By ethnicity, the participants were divided into Nenets (n = 113) representing the indigenous population and non-Nenets (n = 91) consisting of mostly Russians permanently living in NAO. Body mass index (BMI) was calculated for all participants. Overweight and obesity were defined as 25.0 < BMI < 29.9 kg/m 2 and BMI > 30.0 kg/m 2 , respectively. More details on the settings and recruitment of the participants, in addition to blood and serum sample collection, transportation, and storage are presented elsewhere [25].
The supernatant of the n-hexane layer was transferred to a second 15 mL centrifuge tube. Then, 2 mL of n-hexane was added to the first tube containing H 2 SO 4 and plasma, mixed and centrifuged as described above; this step was repeated twice.
To the resulting 7 mL supernatant volume, 2 mL of conc. H 2 SO 4 was added before mixing and centrifugation, as described above. The final supernatant was transferred to a glass conical bottom centrifuge tube and evaporated to dryness in a vacuum rotary evaporator EV400 LabTech (Italy) at 53 • C. Then, the sample was transferred to the micro vials with two portions of acetone in 50 µL each. If extraction emulsions formed at any stage, they were eliminated by adding 10-15 drops of MilliQ water before centrifugation [26].

Instrumental Analysis
An Agilent 7890A gas chromatograph (GC) with an Agilent 7000 series MS/MS triple quadrupole system (Santa Clara, CA, USA) worked in electron-ionization (EI) mode (70 eV) was used as the main equipment. An Agilent ultra-inert GC column (HP-5MS UI, 30 m × 0.25 mm × 0.25 µm) with a Restek Guard column (Rtx-5M) was used for analysis.
A multimode injector was operated in splitless mode at a temperature of 250 • C. The portions of extracts were 1 µL. The ultra-inlet liner was with a glass wool frit 5190-2293 from Agilent (Santa Clara, CA. USA).
GC MS/MS settings were the same as in the article [24].

Quality Assurance, Quality Control and Method Validation
Three replicate test portions of the 5 serum samples were weighed into 15 mL tubes. For validation, a blank matrix and matrix spiked at 0.2 and 1 ng/mL were used ( Table 1). Blank matrix was obtained by mixing together equal parts of different blood and was used in matrix-matched (MM) calibration at 0, 2, 5, 20, 40, and 80 ng/mL levels. Daily validation included a reagent blank analysis. In real sample analysis, each batch (10 samples) included the hexane blank and reagent sample blank. Limit of detection (LOD) was calculated as three times the standard deviation of the procedural blank levels. Limit of quantification (LOQ) was calculated as ten times the standard deviation of the procedural blank levels. If the signal was not detected, then serum with a minimal addition of analytes was used to determine LOD and LOQ (Table 1). If values were below the LOD, we assigned the values of 1 2 LOD. Blank and calibration samples were prepared daily. Linear regression coefficients were ≥0.99 and all relative standard deviations (RSD) were below 20%. For assessing the accuracy of obtained results, a spiked mixture of serum blood with known content of analytes (0.5 ng/mL) was analyzed in each batch. Average batch accuracy is presented Table 1.

Determination of Total Serum Lipids
All POP concentrations were adjusted for serum lipids. The total lipid content (mg/dL) was calculated from the concentrations of total cholesterol and triglycerides according to the following equation [27,28]: Total lipids = 2.27 * total cholesterol + triglycerides + 62.3 Total cholesterol and triglyceride were assessed by enzymatic methods using an automatic biochemical analyzer Random Access A-15 (Biosystems, Barcelona, Spain) at the Northern State Medical University (Arkhangelsk).

Statistical Analysis
To ensure comparability of our results with the findings of other studies we used means, medians, and geometric means (GM) as measures of central tendency and standard deviations and range as measures of variability. In addition, 95% confidence intervals (CI) were calculated for GM. Distributions of continuous variables were assessed using Shapiro-Wilk tests. Given that most of the variables were skewed to the right, non-parametric statistical tests were applied for all analyses. Kruskal-Wallis tests were used for analyses across the settings, and differences between Nenets and non-Nenets were studied by Mann-Whitney tests. Categorical variables were compared using Pearson's chi-squared tests. IBM SPSS software, version 23.0 (IBM Corp., Armunk, NY, USA) was used for all calculations.

Ethical Considerations
The study was approved by the Ethical Committee at the Northern Medical State University, Arkhangelsk, Russia (protocol no. 06/09-17 of 27 September 2017). All study participants signed an informed consent.

Basic Characteristics of the Study Participants
The age of the women ranged from 19 to 87 years. Nenets constituted 55.4% of the sample. Every fourth woman was a smoker with no differences across the settlements (p = 0.253) or ethnic background (p = 0.935). The participants' age, BMI, total cholesterol, triglycerides, total serum lipids and proportion of smokers across study sites and ethnic backgrounds is presented in Table 2. Only age and ethnic background varied significantly between the settlements Nenets women were significantly younger (p < 0.001) and tended to have lower concentrations of triglycerides (p = 0.053) and total lipids (p = 0.058). As many as 66.0% of the women were overweight (40.4%) or obese (25.6%) with no significant ethnic differences.

Distribution of POPs across the Study Settlements
The concentrations of POPs across the settlements are presented in Table 3.   Among the organochlorine pesticides (OCPs), p,p -DDE showed the highest concentrations among the participants from all settlements with a median of 68.8 ng/g lipid. Regarding PCBs, significant variations in concentrations of OCPs between the settlements were observed.
ΣDDT was the dominant POP in all settlements, except Varnek and Indiga, where ΣPCB and HCB, respectively, were the dominant compounds. The main congener of the ΣDDT group was p,p -DDE. The group "others" included β-HCH, aldrin, and mirex ( Figure 2).

Distribution of POPs between Nenets and Non-Nenets
Nenets women had higher concentration of PCB 183 (p = 0.013), but lower concentration of PCB 118 (p = 0.001). Concentration of o,p -DDE was greater in Nenets (p < 0.001), whereas the concentration of p,p -DDE was greater among non-Nenets (p = 0.006). Both ΣDDT (p = 0.011) and ΣPCB (p = 0.038) concentrations were significantly lower in Nenets. However, ΣDDT comprised 34% of the measured POPs in the serum of Nenets but 53% of non-Nenets, whereas ΣPCB comprised 36% of POPs in Nenets but 17% in non-Nenets (Figure 3).
Among the organochlorine pesticides (OCPs), p,p'-DDE showed the highest concentrations among the participants from all settlements with a median of 68.8 ng/g lipid. Regarding PCBs, significant variations in concentrations of OCPs between the settlements were observed.
ΣDDT was the dominant POP in all settlements, except Varnek and Indiga, where ΣPCB and HCB, respectively, were the dominant compounds. The main congener of the ΣDDT group was p, p'-DDE. The group "others" included β-HCH, aldrin, and mirex (Figure 2).

Distribution of POPs between Nenets and Non-Nenets
Nenets women had higher concentration of PCB 183 (p = 0.013), but lower concentration of PCB 118 (p = 0.001). Concentration of o,p'-DDE was greater in Nenets (p < 0.001), whereas the concentration of p,p'-DDE was greater among non-Nenets (p = 0.006). Both ΣDDT (p = 0.011) and ΣPCB (p = 0.038) concentrations were significantly lower in Nenets. However, ΣDDT comprised 34% of the measured POPs in the serum of Nenets but 53% of non-Nenets, whereas ΣPCB comprised 36% of POPs in Nenets but 17% in non-Nenets ( Figure 3).

Discussion
To the authors' knowledge this is the largest study on the concentration of POPs in women's serum conducted in the European part of the Russian Arctic. Most previous studies were performed in a limited number of settlements. Studies from Greenland and Northern Canada are exceptions because POP biomonitoring is regulated by the state [19,[29][30][31].

Discussion
To the authors' knowledge this is the largest study on the concentration of POPs in women's serum conducted in the European part of the Russian Arctic. Most previous studies were performed in a limited number of settlements. Studies from Greenland and Northern Canada are exceptions because POP biomonitoring is regulated by the state [19,[29][30][31].
Geometric means for ∑PCB 11 and ∑DDT were 58.5 ng/g lipid and 136 ng/g lipid, respectively, which is similar to that reported in international literature [22,[32][33][34]. Geometric means of marker polychlorinated biphenyls (∑PCB 6 ) were 37.9 ng/g lipid. Increased concentrations of PCB 153 were observed in Varnek, located on Vaygach Island, p,p -DDE in the west (Shoina), PCB 138 in the east (Amderma), and HCB in the central part of NAO (Indiga, Nelmin-Nos, Krasnoe). Concentrations of POPs in Arctic residents have been shown to be associated with traditional food items rich in fat [16,17,22,29]. The differences on serum concentrations of POPs observed in this study may be partly attributed to the differences between the settlements in food patterns [35], although the historic use of DDT as an insecticide in the studied settlements may also be an important factor. Further research is needed to elucidate the reasons behind the observed differences.
Concentrations of most POPs in human serum have decreased in recent decades [19,22].  The downward trend in concentrations of POPs has also been reported in other countries (Table 4). For example, the concentration of p,p'-DDE and PCB153 in the serum of women in Greenland halved, the concentration of HCB decreased six times, and the concentration of β-HCH decreased 10 times in the period from 2005 to 2015 [29][30][31]. The most likely explanation for this is prohibition of the use of POPs.  The downward trend in concentrations of POPs has also been reported in other countries (Table 4). For example, the concentration of p,p -DDE and PCB153 in the serum of women in Greenland halved, the concentration of HCB decreased six times, and the concentration of β-HCH decreased 10 times in the period from 2005 to 2015 [29][30][31]. The most likely explanation for this is prohibition of the use of POPs. Higher average concentrations and maximum values for PCB 153 (2413 ng/g lipid) and p,p -DDE (300 ng/g lipid) concentrations in Varnek may be associated with the increased frequency in dietary consumption of marine mammals, which accumulate these substances. Associations between location-specific food patterns and concentrations of POPs and other environmental pollutants in residents of the European Arctic Russia are a subject for our future research.
Significant variations in concentrations of selected POPs between the settlements suggest that national or regional data can mask local geographical differences, warranting more detailed biomonitoring activities.
Although our study presents unique findings of different POPs in women's serum in European Arctic Russia, the results should be interpreted with caution taking into account potential limitations of the study. In spite the fact that all residents of the selected settlements had an opportunity to take part in the study, one may suspect that those with poorer health were more likely to contact healthcare facilities for participation. If we hypothesize that health is inversely associated with exposure to POPs, this situation may have led to overestimation of the average values of POPs in this study. This concern is partly supported by the observed suboptimal average levels of cholesterol and triglycerides, although the prevalence of obesity is comparable with the national average [25].
Although we collected the data from seven rural settlements, they may not be representative of NAO, where 73% of inhabitants live in towns. However, they were selected from east, west, and central NAO, and the islands, for better representativeness of the rural part of NAO. Nevertheless, given that the majority of the population live in urban areas and the regional capital Naryan-Mar, which is home to more than a half of the population of NAO, these residents should also be included in biomonitoring activities.
We emphasize that biomonitoring studies should be carried out regularly in selected locations to obtain the most accurate picture of the concentrations of POPs in humans and their trends. Because the ban on the use of POPs is regulated by the Stockholm Convention, the main remaining sources of legacy POPs in the Arctic appears to be local food. Since 2017, the Arctic biomonitoring laboratory at NArFU (Arkhangelsk) has been collecting food samples among the population of NAO.

Conclusions
Our findings suggest that the concentrations of most POPs in women's serum in NAO of European Arctic Russia are low and similar to those in other Arctic countries. Significant geographical variations between settlements, and between Nenets and non-Nenets in the study area were found. Arctic biomonitoring research in Russia should include studies of associations between nutrition and concentrations of POPs.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.