Low manganese and high iron levels in amniotic uid correlate with fetal chromosomal abnormalities in pregnant women

This study aimed to check the association of essential and toxic elements in amniotic uid (AF) with chromosomal abnormalities. A total of 156 pregnant Polish white Caucasian women between the age of 20 and 43 years participated in the study. AF samples were collected during routine diagnostic and treatment procedures in pregnant women. Inductively coupled plasma mass spectrometry (ICP-MS) was used to determine the levels of various elements in AF. Genomic hybridization and cytogenic karyotype analysis were used. The results of the karyotype analysis indicated chromosomal abnormalities in 19 fetuses (over 12% of the total population) and it was mainly trisomy 21 (N=11), trisomy 18 (N=2), triploidy (N=2) and other chromosomal aberrations. It was found that a low concentration of manganese in AF was associated with chromosomal abnormalities in the foetus. High levels of iron and advanced age of the mother increased the risk of aneuploidy in the fetus. Principal component analysis (PCA) and Spearman correlation showed a strong correlation between essential and toxic elements in AF, especially in groups with chromosomal abnormalities. The results of this exploratory study indicate that the levels of essential and toxic elements in AF are associated with chromosomal abnormalities in the human fetus.


Introduction
Prenatal diagnostic procedures use amniotic uid (AF) as material in search of chromosomal abnormalities, fetus anomalies and diseases. From the second trimester of pregnancy, fetal urine appears in AF. During early development, AF is in contact with fetal skin, gastrointestinal tract and lungs. It can therefore be assumed that the composition of AF re ects the composition of fetal plasma 1 . The composition of AF depends on the mother's nutrition, including essential nutrients, and also on the mother's exposure to toxic substances 2 .
So far, few studies have described the relationship of mineral nutrition and the concentration of toxic elements in the mother's body, including amniotic uid, with the risk of genetic and chromosomal abnormalities in the fetus and newborns. Because one of the most common chromosomal abnormalities in humans is Down's Syndrome (DS) which is genetically trisomy 21, most of the results indicate the association between the concentration of elements and that disease. A meta-analysis showed that calcium, selenium and zinc levels in blood were lower in people with DS, while higher concentrations of zinc and copper were found in red cells in DS compared with the healthy group 3 . Altered iron metabolism and neuroin ammation was observed in DS and also in Alzheimer's disease 4 . Moreover, it was found that the prevalence of anaemia is higher in DS than in the general population 5 . In chromosomal abnormalities such as trisomy 18 and 21, clinical symptoms include neurological disorders 6,7 . Some studies have suggested that overexpression of amyloid in the brain is the key element in the development of cognitive decline in people. It was found that extracellular accumulation of zinc and copper in amyloid and intracellular excess iron in neurons is associated with DS in humans, whereas in an animal study, in a mouse model of DS that calcium ion channel activity was observed to be higher than in the wild-type mouse 8 . Some environmental factors may increase the risk of chromosomal aberration in pregnancy. It is known that metals such as Cd, Al, As, Pb are neurotoxic and early exposure to these metals may cause neuronal developmental disorders 9 .
In our previous study, we indicated that some essential and toxic elements were associated with fetal congenital anatomical defects and growth disorders. We found that low concentrations of magnesium and vanadium in amniotic uid and high concentrations of aluminium, cadmium and lead increased the risk of non-genetic birth defects 10 .
The in uence of essential and toxic elements on maternal health and fetal development has not yet been fully investigated and the mechanisms are poorly understood. Recent research results indicate that mineral elements may be more important than assumed in assessing the risk of chromosomal abnormalities in the human fetus. Therefore, this exploratory analysis aimed to determine the association of selected essential and toxic elements in amniotic uid with chromosomal disorders detected in fetuses of the population of white Caucasian mothers.

Study design
This was a retrospective cross-sectional study aiming to determine the levels of elements in the amniotic uid of pregnant women classi ed as high risk for chromosomal abnormalities on the basis of rsttrimester screening tests. The AF was collected during a routine prenatal invasive procedure (amniocentesis) at 15-22 weeks of gestation performed between 2015 and 2020.
The study was conducted on 156 pregnant Polish women (white/Caucasian) who were under routine obstetric care. The women voluntarily participated in the study after signing an informed consent form.
They consented to the secondary use of their samples. The study was conducted on patients quali ed to participate based on inclusion and exclusion criteria.
The inclusion criteria were single viable intrauterine pregnancy in the second trimester and informed consent. The exclusion criteria were: multiple pregnancies, lack/withdrawal of the consent, too small amniotic uid sample, excessive blood contamination in the amniotic uid sample, maternal or intrauterine infection, fetal anatomic anomalies, maternal chronic illnesses, multidrug therapy.
A routine medical interview was conducted to collect data on maternity and pregnancy history of illnesses and operations, drugs and medications used. The week of gestation was determined by rsttrimester obstetric ultrasonography. Samples of amniotic uid were collected from all women and genomic hybridization and cytogenic karyotype analysis were performed. According to the results of chromosome analysis, women were divided into two groups: the control group (normal karyotype) and the study group (with chromosomal abnormalities).
The study protocol was approved by the Bioethics Commission at Poznan University of Medical Sciences (approval no. 30/15) in Poland. This study was conducted in accordance with the Declaration of Helsinki. The study has been registered with ClinicalTrials.gov, ID: NCT03598361.

Collection of biospecimens
Amniotic uid (AF) samples were obtained (5 mL) for genetic and element analysis following routine diagnostic amniocentesis at 15-22 weeks of gestation. Samples were collected by transabdominal aspiration using sterile 22-25G spinal and diagnostic puncture needles and plastic syringes free of organic and non-biological contamination. Each sample of fresh amniotic uid was centrifuged for 5 min at 1300 rpm. Next, 2 ml of supernatant was transferred to a new 2 ml tube and frozen at -80°C for analysis.

Element measurements
Measurement of essential and toxic elements in AF has been described by Markiewicz et al. 11 and Kocylowski et al. 10 . This method involves the use of inductively coupled plasma mass spectrometry (ICP-MS) technique to assess the levels of Li, Mg, Al, Ca, Cr, Mn, Fe, Co, Ni, Cu, As, Se, Sr and Cd in AF. Samples were mineralized in a high-pressure, closed microwave digestion system (Ethos One, Milestone). An Elan DRC II ICP-MS (PerkinElmer SCIEX, Ontario, Canada) was used to determine the content of elements. Calibration based on a weighted least squares calibration curve was employed for all elements. Linearity -calculated as R 2 -was acceptable for all analyzed elements (R 2 > 0.999). The trueness of the analytical method was assessed by analyzing the certi ed reference material (CRM) -Seronorm ™ Trace Elements Serum L-2. The values of recovery were within an acceptable range for all analytes, which demonstrates that the described analytical procedure is t for the intended purpose.

Genomic hybridization and cytogenic karyotype analysis DNA Isolation
Genomic DNA for CMA (chromosomal microarray analysis) was extracted from fresh amniotic uid (AF) using the Sherlock kit (A&A Biotechnology, Poland) according to the manufacturer's recommendations.

Array design
Array Comparative Genomic Hybridization (aCGH) was performed using 60K microarrays 8x60K from Oxford Gene Technology (CytoSure ISCA, v3). The array contains 51,317-mer oligonucleotide probes covering the whole genome with an average spatial resolution of 60 Kb.

Array Comparative Genomic Hybridization analysis
Procedures for DNA denaturation, labelling, hybridization and washing were performed according to the manufacturer's instructions. Genomic DNA was labelled using a CytoSure Labelling Kit (Oxford Gene Technology), with no enzyme digestion. Hybridization was performed between 24 and 48 hours at 65°C in a rotator oven (Agilent). Arrays were washed using Agilent wash buffer 1 and 2, scanned using Agilent Technologies microarray scanner and signal intensities were measured using Feature Extraction software (Agilent Technologies). Agilent Feature Extraction software (V10.0) was used to quantify all scanned images.
CytoSure Interpret Software (Oxford Gene Technology) was used to perform data analysis based on the reference genome (NCBI37/hg19).

Cytogenetic karyotype
In most cases, part of the amniotic uid sent for aCGH was also used for cytogenetic karyotype analysis.
Amniocytes were cultured using in situ vessels. Colony growth and mitotic activity were daily controlled after the 8th day of culture. Metaphase chromosomes were prepared according to standard procedures of GTG banding.

Statistical analysis
All statistical analyzes were performed with Statistica 13 for Windows and RStudio software (R version 3.4.0) R Core Team (2017). Data were tested for normal distribution using the Shapiro-Wilk test. Mann-Whitney test was used to compare the differences between groups. The probability function of chromosomal abnormalities depending on the values of elements in AF was analyzed. The symbol p0 indicates the probability of chromosomal abnormalities for various values of the analyzed elements. Principal component analysis (PCA) was used to analyze correlations in the data set. The correlations between elements in the group with abnormalities and in the healthy group were analyzed using Spearman's rank-order correlation. The level of statistical signi cance was set at p < 0.05.

Results
Characteristics of the participants were shown in Table 1. The study population consisted of a total of 156 pregnant women with a median age of 32 years (range, 20-43 years) and pregnancy week 17 (15-22 weeks). The results of the karyotype analysis indicated 49% male and 51% female fetuses.
Chromosomal abnormalities were found in 19 fetuses (over 12% of the total population) and it was mainly trisomy 21 (N = 11), trisomy 18 (N = 2), triploidy (N = 2) and other chromosomal aberrations. In Table 2, the content of essential and toxic elements in amniotic uid in the total population was shown. Following genetic analysis, the samples were divided into two groups: control group (C) (N = 137) with normal karyotype and study group (A) (N = 19) with chromosomal abnormalities. Women in the study group were signi cantly older than women in the control group (Table 3). The comparison of analysed elements in AF indicated a signi cant decrease of Mn concentration in the group with chromosomal abnormalities compared to the group with normal karyotype (Figs. 1 and 2).  The probability function of chromosomal anomalies in fetuses depending on the values of elements in AF was analysed (Fig. 3). It was found that the probability of chromosomal abnormalities increased with the age of the pregnant woman. The age over 40 years was associated with more than 20% probability of the occurrence of chromosomal aberrations. A manganese level below 1 µgL − 1 was related with nearly 15% probability of the occurrence of anomalies. Moreover, the probability of abnormal karyotype increased with an increase in iron concentration in AF. An iron level of nearly 1500 µgL − 1 was associated with over 40% probability of the occurrence of chromosomal anomalies. Figure 4 shows principal component analysis chart. It was observed that the rst two principal components explain nearly 23.3% of the total variance and our PCA chart can only be a visualization of the correlation tendencies of the variables under consideration. It should be mentioned that each of the principal components is a linear combination of all the output variables. Principal component analysis (PCA) was based on all the analyzed variables simultaneously.Nevertheless, some trends can be observed.
It was found that Cu and Fe levels in AF in the total population were correlated (arrows are close to one another and point in the same direction). In both groups, these elements were signi cantly correlated (group A, R = 0.84 and group C, R = 0.34) (Figs. 5 and 6). Al and Ni levels in AF are positively correlated (their arrowheads point in the same direction). Figures 5 and 6 show a signi cant correlation between these two toxic elements (in group A, R = 0.63, in group C, R = 0.45). Moreover, a high positive correlation between Ca and Mg and also between Ca and Sr and between Cd and As in AF was found in the total population and in both studied groups ( Fig. 4-6).

Discussion
In this study, we determined the association of essential and toxic elements with chromosomal abnormalities in human amniotic uid in fetuses of the white/Caucasian population in the second trimester of pregnancy. Based on the results, low concentrations of manganese and high concentrations of iron were associated with ahigher occurrence of chromosomal abnormalities (predominantly of trisomy 21, Down's Syndrome). To the best of our knowledge, this is the rst study of Polish pregnant women to demonstrate such a relationship.
This study showed that the presence of chromosomal aberration in fetuses increased with a low level of Mn in AF. Mn plays a crucial role in maternal health and early pregnancy development. Manganese acts by activating certain enzymes and through Mn-dependent metalloenzymes, such as manganese superoxide dismutase, that are needed for proper reproduction and protection against free radicals. In animal and human studies, it was found that a low level of maternal manganese is associated with reduced reproductive function, impaired fetal growth and poor perinatal outcomes. On the other hand, it is known that excess manganese is a potent neurotoxin and may induce adverse neurological, reproductive and respiratory effects. Little is known about the relation between Mn status and chromosomal abnormalities 12 . the mitochondrial oxidative phosphorylation pathway is a major source of free radicals, and MnSOD plays a crucial role in protection against these radicals and maintains the oxidant/antioxidant balance. A decrease in Mn content caused a low activity of MnSOD and disrupted this balance 13 . It was found that oxidative stress increases in amniotic uid in Down's Syndrome 13,14 .
It is also known that excess iron is a major cause of oxidative stress because Fe ions can generate free radicals in the Fenton reaction. Hattori et al. 15 found that the content of catalytic Fe(II) in amniotic uid inabnormal pregnancies (also involving trisomy 21 and trisomy 18) was signi cantly higher than in normal pregnancies. These authors concluded that catalytic iron in AF may be one of the markers of abnormal pregnancy. Our study con rms such a relationship and indicates that high iron levels in AF increased the probability of fetal abnormalities, mainly Down's Syndrome and trisomy 18. We speculate that increased iron in AF may be swallowed by the trisomic fetus and higher Fe concentrations in its body with chromosomal abnormalities accelerate oxidative stress and progressive oxidative damage of its internal organs. We also noticed that elevated concentration of iron in AF was present not only in DS cases. Extremely high levels of iron (over 1600 ug/L) were found in AF with triploidy.
Other trisomies are also associated with increased oxidative stress, in ammation and nervous system disorders. However, most literature reports about the relationship between minerals and chromosomal abnormalities focus on Down's Syndrome. Signi cantly higher ferritin levels and markedly lower transferrin concentrations and TIBC levels were found in adults (18-35 years) with DS compared to healthy individuals 16 .
It seems that increased oxidative stress observed in people with DS may begin in utero and might play a crucial role in phenotypic traits. Moreover, Down Syndrome and many other chromosomal abnormalities are associated with metabolic diseases such as obesity, insulin resistance, lipid disorders with a great role of oxidative stress in etiopathogenesis 7,17 . Therefore, based on the obtained results we can assume that increased oxidative stress linked to high Fe and low Mn concentrations in AF in early development may contribute to increased oxidative stress, brain disorders and other metabolic disorders in the later life of people with aneuploidies. It seems that a treatment strategy involving lowering oxidative stress by regulation of element status in AF may potentially protect against the disorders that are observed in people with chromosomal abnormalities.
The relationship of chromosomal aberration with lower Mn concentration and higher iron level in AF may also be associated with the interaction between Fe and Mn on the transport level. Divalent metal transporter 1 (DMT1) is the primary non-heme iron transporter in the intestine to transport iron but also other elements, including manganese. DMT 1 is found not only in the intestine but also in other tissues, also in the placenta, and the transport of excess iron via this transporter decreases the transport of manganese in tissues 12 .
The concentration of elements in AF may be affected by endogenous and exogenous factors. The correlation between essential elements observed between iron and copper levels as well as between calcium and magnesium levels indicate the transport and metabolic relationships between these elements. It is known that systemic copper de ciency generates cellular iron de ciency. These two essential elements participate in single-electron transfer reactions and owing to this capacity, they also may generate free radicals and increase oxidative status in the body 18 . As mentioned above, an association between iron excess and neurodegenerative diseases including Down's Syndrome was found 15 . In this study, we observed higher iron levels in AF in pregnancy with chromosomal disorders. In a mouse model of Down's Syndrome it was observed that copper accumulation in the brain increased oxidative stress 19 . A change in the homeostasis of copper in the brain is also suggested in Alzheimer's disease 8 . In the human body, magnesium and calcium metabolism are closely related. There are relations between the intestinal absorption and renal excretion of these two elements 20 . Positive correlation between toxic elements, e.g. Al and Ni, observed in our study may be the consequence of exposure to environmental concomitants 21 . Correlation index values are higher for AF with chromosomal abnormalities which may be related to the exposure of pregnant women in this group to different toxic elements during pregnancy. It was found that toxic elements such as Cd, Ni and Al are adversely related to chronic morbidity 21,22 . Results of the study by McLachlan et al. 23 suggest that aluminium accumulation in the brains of patients with Alzheimer's disease and Down's Syndrome may contribute to the neuropathology of those neurological diseases.
It is known that advanced age in women increases the risks of infertility and aneuploidy in the offspring, most predominantly Down's syndrome (trisomy 21), Edwards' syndrome (trisomy 18), and Turner's syndrome (monosomy X),. The postulated mechanism is connected with increased oxidative stress which may alter gene expression. Follicular aging can lead to mitochondrial dysfunction and hence disturb MnSOD activation. Chromosome segregation errors occur in aged oocytes during meiosis, which can result in aneuploidy and poor oocyte quality 24 .
This study has some limitations. First of all, our study group with chromosomal abnormalities was relatively small and rather heterogeneous with different chromosomal abnormalities (nearly 60% were cases with DS). Furthermore, in the present study, we did not determine nutritional and environmental factors that may affect the concentration of elements in AF. We also did not include other maternal parameters such as smoking, body mass, education and socioeconomic status that may in uence the content of elements in AF.

Conclusions
Low manganese concentration and high iron level in amniotic uid in the second trimester of pregnancy are associated with a higher incidence of chromosomal abnormalities in human fetuses.

Declarations
Funding source  Figure 1 The concentration of essential elements in amniotic uid; C-group with normal karyotype, group A-group with chromosomal abnormalities group; p-signi cance level.

Figure 5
Correlation between essential and toxic elements in amniotic uid of group with normal karyotype.

Figure 6
Correlation between essential and toxic elements in amniotic uid of group with chromosomal abnormalities.