Analysis of essential, toxic, rare earth, and noble elements in maternal and umbilical cord blood

Progressive industrialization in recent decades has contributed to the increase of metal levels in the environment, which has a dangerous impact on human health, primarily pregnant women. In this study, we aimed to compare levels of various elements in maternal and umbilical cord (UC) plasma samples collected from 125 healthy pregnant women, conduct correlation analysis among paired plasma samples, and compare our data with other populations worldwide. The study design included the following elements: essential (Mn, Co, Cu, Zn, Se, Mo), non-essential (Be, Al, Ni, As, Rb, Sr, Cd, Sb, Pb, U), rare earth (La, Pr, Ce, Nd, Sm, Eu, Gd, Dy, Ho, Er), and noble metals (Ru, Rh, Re, Pt). Levels of 30 elements were higher in maternal plasma than in UC plasma samples. However, no disparities at the statistically significant level were found for Be, Zn, Rb, Cd, Ce, and Ho. Correlation analysis among paired plasma samples revealed only positive/synergistic correlations of different strengths between most elements. Compared to other countries across the globe, our participants had considerably lower plasma levels of Zn and higher levels of Co, Ni, and As. This study provides not only a new and deeper comprehension, but also the first insight into the levels, correlation, distribution, and potential transplacental transfer of 30 elements.


Introduction
Human biomonitoring (HBM) has become a powerful tool in an algorithm for examining the impact of one or more natural or artificial environmental pollutants on health, particularly in pregnant women and children, as the most vulnerable population groups (Kot et al. 2019;Bocca et al. 2019). HBM studies aim to identify the type of pollutant and exposure route, find specific biomarkers, create normal intervals and establish a scientific database, assess health risks, assist in epidemiological, clinical, and other investigations, etc. (Cerná et al. 2012). A new trend in HBM studies is the examination of solid tissues (e.g., placental, thyroid) instead of body fluids (e.g., serum/plasma, urine) and/or body products (e.g., feces), because each solid tissue has its blood supply and different accumulation capacity for both naturally occurring and anthropogenic origin pollutants . Among the well-identified harmful agents, some elements are recognized as hazardous to human health (Silver et al. 2018). Furthermore, as a result of growing industrialization, the level of these hazardous elements has risen notably in recent decades . Therefore, it is not unexpected that elements' profiles differ among population groups across the globe and that it is necessary to monitor their levels in specific clinical samples (Serafim et al. 2012;Caserta et al. 2013).
The negative effects of elements are pronounced in the pre-and postnatal periods of life (Al-Saleh et al. 2013;Kozikowska et al. 2013). The prenatal period is particularly Responsible Editor: Lotfi Aleya * Aleksandar Stojsavljević aleksandars@chem.bg.ac.rs critical because the fetus undergoes various morphological changes, which are directly dependent on environmental stimuli, even at low levels of exposure (Cerrillos et al. 2019). Due to exposure to specific elements, studies have reported permanent morphological alterations over the first 3 months of fetal development, whereas functional consequences were observed after the end of organogenesis for different degrees of exposure to the elements (Rudge et al. 2009;Al-Saleh et al. 2011). A good example is lead (Pb) since this heavy metal easily crosses the placental barrier and tends to accumulate in the fetus. Moreover, babies and children are subjected to Pb later in life, primarily through toys and dust, while their mothers and the population in general are primarily exposed to Pb via polluted air, contaminated drinking water and food, and so on (Reddy et al. 2014;Iwai-Shimada et al. 2019).
The first aim of this study was to compare levels of 30 elements in maternal and umbilical cord (UC) plasma samples. A further aim of this study was correlation analysis between the same elements in paired maternal and UC plasma samples. The last aim of this study was to compare the levels of investigated elements in maternal and UC plasma samples with other populations worldwide. In accordance with the aims, the following elements were enrolled in the study design: essential (Mn, Co, Cu, Zn, Se, Mo), non-essential (Be, Al, Ni, As, Rb, Sr, Cd, Sb, Pb, U), rare earth (La, Pr, Ce, Nd, Sm, Eu, Gd, Dy, Ho, Er), and noble elements (Ru, Rh, Re, Pt). These elements were selected due to a lack of knowledge about their levels and distribution, as well as their correlation, in maternal and UC plasma samples.

Sample collection
One hundred and twenty-five pregnant women, aged 20-41 years, were included in this study. The course of pregnancy was regular, and each birth was in term. All participants were in excellent health, released from chronic diseases (previous history of malignancies, liver and renal dysfunctions, hypertension, type I or II diabetes, etc.), venous thrombosis, and complications during pregnancy (oligohydramnios, gestational diabetes, preeclampsia, and infection). They were also released from external factors, such as occupational exposure to heavy metals. Furthermore, ineligibility criteria were multi-fetal pregnancies, mothers who gave birth by elective or emergency cesarean section, prolonged labor, etc. Thus, only participants who had spontaneous, vaginal, and uncomplicated childbirth were recruited in this study. The data from participants were collected after the informed consent of each woman.
The investigation was approved by the Institutional Ethical Review Board (No: 05006-2021(No: 05006- -1525. At the beginning of the delivery phase, a sample of the mother's venous blood was taken from the antecubital vein by venipuncture and collected into a lithium-heparin Vacutainer tube (BD Vacutainer Royal Green Cap, Becton Dickinson). A total of 5 mL of whole blood (arterialvenous) was drawn from the umbilical cord up to 5 min after birth, promptly after double tightening of the cord from the maternal umbilical end, but before the placental expulsion. Each sample of umbilical whole blood was collected into a Li-heparin Vacutainer tube from the same manufacturer. After centrifuging whole blood (3000 × g for 10 min), plasma samples from the mother and umbilical cord were separated from the blood cells. All samples were stored at −80 °C until analysis.

Analysis of elements
The procedure for element analysis was adapted from our previous study (Stojsavljević et al. 2018). Briefly, plasma samples were diluted 10-fold with an aqueous solution containing 0.05% nitric acid, 0.1% Triton X-100, and 3% n-butanol. All elements were quantified by inductively coupled plasma quadrupole mass spectrometry (ICP-Q-MS) in collisional (helium) mode and by adding an internal standard solution ( 71 Ga, 115 In, and 159 Tb at concentrations of 10 μg/L), in equal amounts to each solution. Good linearity of the calibration curve (R ≥ 0.994) was recorded for each element in the ranges from 1 to 250 μg/L (Cu, Zn, Se, Rb), 1-25 μg/L (Al, Mn, Co, Ni, As, Sr, Mo, Sb, Cd, Pb), or 0.1-25 μg/L (Be, Ru, Rh, La, Ce, Pr, Nd, Sm, Eu, Dy, Gd, Ho, Er, Re, Pt, U). All elements were successfully (100%) detected in the analyzed plasma samples. The accuracy of the method was tested with certified reference materials (CRMs) Serum Level-1 and Serum Level-2, supplied by Seronorm (Sero AS, Norway). The recoveries for the CRMs spanned from 80.4 to 118% (Table S1a, Suppl. material). Furthermore, the standard addition method was applied to evaluate the accuracy of the analytical technique used. Ten samples of maternal plasma and ten samples of UC plasma were spiked with a standard solution of each element (10 and 20 μg/L). The results are represented in Table S1b (Suppl. material). Based on the recoveries obtained by CRMs and the standard addition method, the following isotopes were chosen: 9 Be, 27 Al, 55

Data analysis
The results were processed using SPSS v.25 statistical software package (SPSS Inc., Chicago, Ill.), while all figures were drawn using GraphPad Prism v.5.0 (CA, USA). The Kolmogorov-Smirnov test was used to check the normality of the data. Differences between elemental levels were analyzed using Chi-square test, Kruskal-Wallis one-way analysis of variance, and Mann-Whitney U test. The Spearman's rho (ρ) correlation test was used to examine the correlation analysis between maternal and UC plasma for each element. In all tests used, P < 0.05 was considered statistically significant.

Results
The most relevant parameters for pregnant women and their newborns are shown in Table 1. It is evident that the enrolled group of participants was notably homogeneous in terms of maternal age and pregnancy duration. Correlation analysis was not conducted for these data since none of the pregnant women smoked or used alcohol and/or drugs. Moreover, none of the participants were professionally exposed to heavy metals. All pregnancies were singleton with a gestational age of 38-42 weeks, while Apgar scores were ≥ 9 in the first minute of postnatal life. Anthropometric parameters of the newborn (weight and height) were within the reference ranges, additionally implying a normal course of pregnancy. Fig. 1 shows the map of Serbia. The shaded fields represent Belgrade and two regions (Šumadija and Podunavlje) from where the samples of participants were acquired. The Chi-square test did not reveal statistically significant differences in the elemental levels in maternal and UC plasma samples between women residing in different regions. This finding could be explained by a similar diet of the examined pregnant women.
The distribution of analyzed elements in the maternal and umbilical cord (UC) plasma samples, based on their values of mean and standard deviation (SD), is given in Figs. 2 and 3. Copper (Cu) was the most dominant element in maternal plasma samples, while it was the second most abundant in UC plasma samples; in these, the first place belonged to zinc (Zn). Holmium (Ho) was the element with the lowest detected levels in both maternal and UC plasma samples. Data from Fig. 2 are further specified in Table 2, together with the P values obtained by the Mann-Whitney U test. Maternal plasma samples had higher levels of each analyzed element compared to UC plasma samples; however, no statistically significant differences were found for beryllium (Be), Zn, rubidium (Rb), cadmium (Cd), cerium (Ce), and Ho. Also, comparing maternal and UC plasma samples, the levels of Rb and Ce were very similar, while maternal/UC levels for other elements were more disparate (Table 2). Table 3 shows the correlation analysis between the same elements in maternal and UC plasma samples. Surprisingly, only positive correlations of varying strength were discovered among the elements. However, comparing maternal and UC plasma samples, levels of aluminum (Al), nickel (Ni), arsenic (As), selenium (Se), strontium (Sr), and platinum (Pt) were not statistically significantly correlated between the two plasma types.

Discussion
Our results showed that the levels of all analyzed elements were higher in maternal than in UC plasma samples. However, not all studied elements had statistically significantly different levels between paired maternal and UC plasma samples. Thus, no disparities were found for Be, Zn, Rb, Cd, Ce, and Ho levels in the two plasma types (P > 0.05). Among these elements, the levels of Zn and Cd have been previously reported while the biochemical functions of Be, Rb, Ce, and Ho remain insufficiently explored. Our data could imply that the fetus withdraws essential Zn from the mother's bloodstream, which could be a reason for the strong positive/synergistic correlation obtained between the maternal and UC plasma samples for this metal (ρ = 0.73). Moreover, this finding should also be interpreted taking into account the previously reported Zn deficiency in our adult population (Jagodić et al. 2021). Therefore, the notably low Zn levels in both our types of plasma samples indicate the need for oral Zn-salt supplementation and/or increased intake of Zn-rich foods. The placenta has been indicated to function as a barrier against Cd toxicity (Al-Saleh et al. 2013). Although inconclusive reports were identified among studies, new research has shown that the placenta has the ability to accumulate Cd and increase the expression of metallothioneins (MTs), small proteins that efficiently bind Cd 2+ and reduce its toxicity (Somsuan et al. 2019;Mazurek et al. 2020). However, MTs also bind Zn 2+ ions, which could explain a lack of Zn if an increased Cd level is present (Iwai-Shimada et al. 2019). Sakamoto et al. (2018) reported statistically higher levels of Cd in maternal plasma samples compared to UC ones. Our data were in agreement with their study, although we did not find a statistically significant value (P = 0.075) between maternal (0.47 ± 0.31 μg/L) and UC plasma samples (0.31 ± 0.15 μg/L).
Other analyzed elements occurred at significantly different levels in maternal and UC blood plasma samples (Table 2). Among them, and to the best of our knowledge, literature data are available for Mn, Co, Ni, Cu, As, Se, Mo, and Pb. Mn is an important antioxidant nutrient and essential trace metal for embryogenesis, myelin development, skeletal formation, etc. (Silver et al. 2018), while elevated Mn levels could be related to complications in fetal nervous system development, gestational hypertension, attenuated effects on birth weight, etc. (Zhou et al. 2019). Our Mn levels in maternal plasma samples were slightly higher than reported data (Burtis et al. 2012). Also, UC plasma samples (1.61 ± 0.81 μg/L) had slightly lower Mn levels compared to those reported from several Asian countries, including Japan (4.54 ± 1.21 μg/L) (Osada et al. 2002), Anhui Province, China (5.44 μg/L) (Liang et al. 2019), and Zhejiang province, China (4.20 μg/L) (Silver et al. 2018), but were in good agreement with Mn in UC plasma samples collected from Valencia, Spain (2.92 μg/L) (Bermúdez et al. 2015).
The notably high Co levels recorded in both types of plasma samples can be clarified by the use of supplements containing vitamin B12, since 85% of the human body's Co content is in this form (Bocca et al. 2019). As a result, no comparison with research from other nations has been conducted, considering pregnant women in Serbia frequently take B12 supplements from different manufacturers in varying doses.
The considerably high Ni levels in maternal and UC plasma samples agreed with our previously reported high Ni levels in placental samples from the same participants . Since Ni-contaminated food and drinking water are the main sources of this metal, along with Ni-containing products (cosmetics, jewelry, etc.) (Bocca et al. 2020), it is difficult to ascertain the exact routes of Ni exposure. However, notably high Ni levels deserve further analysis, as studies have shown an association between high Ni levels and some disturbances in anthropometric parameters, gestational hypertension, and impaired neurodevelopment (Kot et al. 2019).
Cu is necessary for proper fetus growth and many lifesupporting biochemical pathways. Our data for Cu in maternal plasma samples (2127 ± 523 μg/L) were consistent with previously published findings (Butler Walker et al. 2006a, b;Bermúdez et al. 2015), as estrogen is known to increase Cu levels through enhanced ceruloplasmin expression. Also, the mean Cu level in UC plasma samples (276 ± 114 μg/L) was consistent with other studies worldwide, including 54 participants from Valencia, Spain (248 μg/L) (Bermúdez et al. 2015), 357 participants from Zhejiang Province, China (243 μg/L) (Silver et al. 2018), and 2382 participants from Anhui Province, China (300 μg/L) (Liang et al. 2019).
Curiously, compared to other data worldwide, we found notably higher As levels in maternal and UC plasma samples (up to 10-fold higher). For example, Liang et al. (2019) reported 1.87 μg/L in UC plasma samples, while McKeating et al. (2020) recorded 0.90 ± 1.83 μg/L in the same sample type. However, Bermúdez et al. (2015) reported 19.9 μg/L in UC plasma, similar to our mean UC plasma level. Since our participants had a normal diet, typical for pregnancy, high As levels should be examined further by applying speciation analysis, since our population is not known for consuming significant amounts of fish or seafood, which are the most dominant sources of organic (harmless) forms of As. In addition to Ni, the findings for As are worrisome, since As easily crosses the placenta and increases the likelihood of preterm birth, fetal loss, and neonatal mortality, while it reduces birth weight and children's thymus size, especially in countries with As-contaminated drinking water, such as Bangladesh and India (Cardenas et al. 2015;Ahmed et al. 2011).  Table 2 Results for LOD, range (min-max), and mean and standard deviation (SD) for the analyzed elements in maternal and umbilical cord (UC) plasma samples (μg/L). Statistically significant differences between the two plasma types are given in  Levels of essential Se and Mo in both our plasma sample types were in very good accordance with the majority of reported studies (Liang et al. 2017, McKeating et al. 2020. Pb is a well-known environmental pollutant without recognized essential functions in human health (Parizi et al. 2021). In fact, even low-level Pb exposure has deleterious effects on children's nervous systems (Rudge et al. 2009). Our data for Pb in maternal and UC plasma samples were consistent with other data across the globe (Butler Walker et al. 2006a, b;Dursun et al. 2016) or were slightly lower than in the study by Liang et al. (2019), indicating low-level exposure to Pb. The low levels of Pb in our UC plasma samples could be the result of passive diffusion since this heavy metal easily crosses the placental barrier (Zhou et al. 2019). It is essential to monitor Pb since long-term exposure could induce premature birth, reduce neonatal anthropometry, and later, can result in impaired neurodevelopment, etc. (Reddy et al. 2014).
To the extent of the authors' knowledge, no similar data for the other elements enrolled in this study (Rb, Sr, Ru, Rh, Sb, La, Pr, Ce, Nd, Sm, Eu, Gd, Di, Ho, Er, Re, Pt, and U) have been previously reported. Thus, this study provides a first look into levels, distribution, and correlation analysis of these elements. We stress that our findings for non-essential, rare earth, and noble elements will assist researchers in further biochemical, toxicological, and clinical studies.

Limitations
This study is limited to the number of samples according to age, lifestyle, place of residence, etc. Furthermore, mercury was not reported due to strong memory effects during measurements on the ICP-MS device.

Conclusion
In summary, we assessed exposure to essential, non-essential, rare earth, and noble elements in paired maternal and umbilical cord plasma samples. Levels of all elements were higher in maternal plasma than in UC plasma; however, no disparities at a statistically significant level were found for Be, Zn, Rb, Cd, Ce, and Ho. Correlation analysis among paired plasma samples revealed only positive/synergistic correlations of different strengths between most elements. Compared to other countries worldwide, our participants had notably higher levels of Co, Ni, and As and lower levels of Zn in their plasma. This research brings not only a new and deeper comprehension, but also the first insight into the levels, distribution, correlation analysis, and potential transplacental transfer of 30 elements. The results presented in this study can be utilized as a database for risk assessment of the pregnancy outcomes and as a starting point for further toxicological, epidemiological, clinical, and other peer-reviewed studies.
Consent to participate All women voluntarily participated in this study, and written informed consent from all study participants was obtained according to the ethical standards defined by the Declaration of Helsinki.