Placental expression of FADS1, FADS2, FADS3 desaturases in selected pregnancy pathologies

Background: The period of intrauterine development is a key period in human development. Its progress largely depends on the function of the placenta, which is responsible for the transportation and biosynthesis of fatty acids. Desaturation enzymes play a key role in placental fatty acid metabolism. The expression of genes coding for desaturases may be associated with pregnancy abnormalities. The objective of this study was to determine the transcriptional activity of the placental genes Fatty Acid Desaturase 1 (FADS1), Fatty Acid Desaturase 2 (FADS2) and Fatty Acid Desaturase 3 (FADS3) in women who gave birth to: appropriate for gestational age (AGA), large for gestational age (LGA), small for gestational age (SGA), intrauterine growth restriction (IUGR) and preterm birth (PTB) infants. Method: The study took place at the Tychy Specialist Hospital in Poland. 34 pregnant women aged 21-37 years old took part. The placental samples were taken from a site about 2-3 cm away from the umbilical cord attachment. The collected tissue sections were stored, according to the manufacturer’s protocol, in RNAlater (Sigma-Aldrich, St Louis, MO, USA), until required for molecular analysis. The expression prole of FADS1, FADS2 and FADS3 was determined by RT-qPCR. Results: In terms of the FADS1 and FADS2 genes, there was no difference in the expression between the groups. However, differences in the expression of the FADS3 gene were found. Analysis of the transcriptional activity of the FADS1, FADS2 and FADS3 genes in most of the examined groups showed signicant differences. Conclusions: These ndings suggest that the transcriptional activity of genes changes with the severity intrauterine and is associated with foetal lipid linked to a greater of fat in the foetal


Background
Intrauterine foetal development is a critical period in human development. Subsequent quality of life, health, susceptibility to diseases, intelligence and many other factors are dependent on this foetal stage [1]. The placenta plays a key role during pregnancy. It controls the metabolic processes on the hormonal pathway between the interface of the mother and child organisms -including oxygenation and the supply of building and energy substrates to the foetus [1,2]. Placental abnormalities lead to premature birthswith varying degrees of malnutrition (e.g. Intrauterine Growth Restriction (IUGR)), small for gestational age (SGA) babies, or Large for Gestational Age (LGA) births. Fatty acids (FA) are one of the most important ingredients that determine proper intrauterine development. They are responsible for the structure of the child's nervous system, brain, development of cell membranes, structure and function of the retina as well as ful lling many other metabolic and structural functions [3,4]. The diversity of FA functions results from the high heterogeneity of their structure, determined by the number of carbon atoms and the number and location of unsaturated bonds in the carbon chain. The source of FAs for the developing foetus is the mother's diet [5,6], the release of FA from deposits in the mother's tissue [7], the endogenous biosynthesis of FA by the mother and later -foetal FA synthesis [8]. The placenta is primarily responsible for the maternal-foetal metabolism of FA; this includes both transportation of the FA from mother to foetus, as well as placental biosynthesis and FA modi cation. Placental transfer is determined by numerous factors, such as the mother's health, condition of the foetus, transport e ciency of the placenta and diet during pregnancy [5,9,10]. Some placental disorders can impair FA metabolism and this may lead to intrauterine foetal development disorders and a predisposition to the development of numerous diseases after birth. Changes in the activity of the enzymes responsible for the desaturation of essential fatty acids (EFAs) seem to be particularly important in maternal-foetal homeostasis. The rst of these is Delta-5 Desaturase (D5D) [EC 1.14. 19.44], encoded by the Fatty Acid Desaturase 1 (FADS1) gene. This gene is clustered with family members at 11q12-q13.1 [11]. This desaturase plays one of the key roles in the biosynthesis of long-chain polyunsaturated fatty acids (L-CPUFA) of both the n-3 and n-6 families. D5D introduces a cis double bond at carbon 5 into dihomo-gamma-linoleoate (DGLA) (20: 3n-6) and eicosatetraenoate (ETA) (20: 4n-3) to generate arachidonate (AA) (20: 4n-6) and eicosapentaenoate (EPA) (20: 5n-3), respectively. The second important enzyme in the biosynthesis pathway of L-CPUFA is Delta-6-Desaturase [EC 1.14.19.3]. It introduces a double cis bond at carbon 6 in linoleic acid (LA) (18: 2n-6) and alpha-linolenic acid (ALA) (18: 3n-3). As a result of this reaction, gamma-linoleate (GLA) (18: 3n-6) and stearidonate (18: 4n-3) respectively are formed [12]. The third representative of desaturases is the Delta (13) Desaturase (D13D) enzyme (EC 1.14.19.-) encoded by the FADS3 gene. D13D, in turn, introduces a cis double bond in (11E) -octadecenoate (trans-vaccenoate) at carbon 13 to generate (11E, 13Z) -octadecadienoate, likely participating in the biohydrogenation pathway of LA [13]. Under normal conditions, the activity of these enzymes remains in a delicate dynamic balance, maintaining the biosynthesis of L-CPUFA n-3 and n-6 at the appropriate level. Disturbance of enzymatic activity, which may be caused by altered transcriptional activity of FADS genes, may contribute to the loss of control over the biosynthesis of the membrane phospholipids as well as DHA -key lipids for the development of the foetal nervous system [14], loss of control over the metabolism of in ammatory lipids such as prostaglandin E2 -critical for acute in ammatory response and maintenance of epithelial homeostasis [15] and metabolic disorders, the effects of which may take some time to appear, such as diabetes, lipid disorders, cardiovascular diseases, etc. [16]. The processes controlled by these desaturases are extremely important from the point of view of the intrauterine development of the foetus and, especially, the structure of its nervous system. Therefore, the aim of the study was to analyse the expression (at transcription level) of the FADS1, Fatty Acid Desaturase 2 (FADS2), and Fatty Acid Desaturase 3 (FADS3) genes.

Study population
The research was carried out with the approval of the Ethics Committee in Bielsko-Biala under approval no: 2016/02/11/4. All relevant guidelines and regulations were adhered to and informed consent was obtained from all the participants in writing. The study population consisted of 34 women who gave birth at the Provincial Specialist Hospital No. 1 in Tychy, Poland. The pregnant women were recruited for the study during their rst visit to the hospital. The women were between 21 and 37 years of age. A description of the study population can be found in Table 1. Table 1 Characteristics of the study population.

AGA
LGA Mode of delivery 8n/1cs 1n/8cs 6n 6cs 3n Apgar Score 9/10 8/9 9/10 6/7/8 9/10 n = natural delivery, cs = caesarean section. Apgar Score measured after 1, 3 and 5 minutes To obtain a homogeneous group of women, the following inclusion criteria were applied [2,17]: Polish nationality (excluding naturalised Polish citizens); single pregnancy; pregnancy I-III (consider parity); Stable socioeconomic status; married, secondary level or higher education; living in a highly industrialised urban region, both the women and their husbands having a steady job. Consenting to participate in the study.
The following exclusion criteria were applied [2,17]: Chronic diseases occurring in the women before pregnancy, such as pre-gestational diabetes; pathologies during the course of pregnancy such as infections during pregnancy (any kind of infection in the perinatal period, such as fever, respiratory infections, urinary infections, etc.); miscarriages and/or premature birth resulting in the death of the child or developmental anomalies in the foetus; AIDS and sexually transmitted diseases; Adherence to a vegetarian diet, Mediterranean diet, or any other special diet; Lack of consent by the mother to take part in the research programme or withdrawal of consent during the study.
LGA Women eligible for the study underwent three ultrasound examinations. The rst ultrasound test was performed between the 12 th and 14 th weeks of gestation, the second between the 20 th and 22 nd weeks and the third one between the 32 nd and 33 rd weeks.
Using ultrasound scans, the foetal weight and length were determined primarily according to the gestational age. The 27th gestational week was the crucial week. The dimensions obtained determined the appropriate way to proceed with the foetus. Although the results of Doppler ow are not considered signi cant below the 30 week term, this test was performed on the foetuses at the 27th week. The foetal dimensions, taken every 6-7 days were marked on a growth chart. If the foetus was between the 10th and 3rd percentile and its gestational age was above 27 weeks -further ultrasound scans were carried out on the foetus. The radiological criteria for hypotrophy or intrauterine foetal growth inhibition (IUGR) was an ultrasound assessment of foetal weight and length and the conversion of this data into a growth chart.
The ultrasound was used to measure the bi-parietal diameter (BPD), head circumference (HC), abdomen circumference (AC) and femur length (FL). These are the standard parameters of so-called basic foetal biometry.

Collection of the placentas
The placental samples were taken from a site about 2-3 cm away from the umbilical cord attachment.
For the research, we wanted to standardise the site of collection to the place where, having established the highest metabolic activity and a strong RNA expression, there was the largest blood supply to the placenta. In our opinion, taking samples from different places would mean that the results would not give a full picture of the expression of the genes tested. With our selection criteria for the place of collection, we now have knowledge of how the quantitative process works in the this part of the placenta. Samples were taken immediately after the birth, then transported in ice to the laboratory. The transport time did not exceed one hour. The samples were then weighed and placed in the reagent (immersed in 1 ml RNAlater for 48 hours at 4℃ and then snap frozen). test were also used. The selection of tests was based on the distribution of variables, which was veri ed by the Shapiro-Wilk test.

RT-qPCR
Comparing the β-actin mRNA copy number in the control and study groups, no statistically signi cant differences were found, which indicates that β-actin can be used as an endogenous control in this experiment.

FADS1 and FADS2
After checking the assumptions of normality, a non-parametric Kruskal-Wallis test was performed to compare the median in individual groups. In terms of the FADS1 and FADS2 genes, the groups did not signi cantly differ statistically (p> 0.05). The AGA, LGA, SGA, IUGR and PTB groups were therefore similar in terms of the distribution of FADS1 and FADS2.

FADS3
After checking the assumptions of normality of the distribution, it turned out to be possible to use the parametric analysis of variance ANOVA, to compare the average of the dependent variable in the individual groups. In the FADS3 range, the groups signi cantly differed statistically (p <0.05).
Tukey's post hoc test was performed in order to determine exactly where there were signi cant differences between the groups.
There were signi cant statistical differences (p <0.05) between: AGA -IUGR; In the AGA group, average values of FADS3 M = 9149.67 were recorded, while in the IUGR group M = 4713.50. The AGA group achieved statistically signi cant (p <0.05) higher results than the IUGR group (Fig. 1).
AGA -PTB; In the AGA group, average values of FADS3 M = 9149.67 were recorded, while in the PTB group M = 967.33. The AGA group achieved statistically signi cant (p <0.05) higher results than the PTB group (Fig. 1).
In the next part of the analysis, the transcriptional activity of the FADS1, FADS2 and FADS3 genes was compared within each of the research groups (AGA, LGA, SGA, IUGR, PTB).

AGA Group
Friedman's test showed signi cant differences (p <0.05) statistically between the expression levels of these genes within the AGA group. Bonferroni's post hoc test was conducted in order to determine exactly where the differences between these gene expressions were signi cant,. The test showed that statistically signi cant (p <0.05) differences between them were, for example: FADS1 -FADS3; For FADS1 the average was M = 2640.78 while for FADS3 the average was M = 9149.67; The AGA group had a signi cantly higher level of FADS3 than of FADS1 (Fig. 2). The AGA group had a signi cantly higher level of FADS3 than of FADS2 (Fig. 2).
LGA Group Friedman's test showed statistically signi cant differences (p <0.05) between the expression level of these genes within the LGA group. The Bonferroni test showed that there were statistically signi cant (p <0.05) differences between gene expression -for example: FADS1 -FADS2; For FADS1 the average was M = 2663.20 while for FADS2 the average was M = 8090.90; The LGA group had a signi cantly higher level of FADS2 than of FADS1 (Fig. 3).
FADS1 -FADS3; For FADS1 the average was M = 2663.20, while for FADS3 the average was M = 8487.60; The LGA group had a signi cantly higher level of FADS3 than of FADS1 (Fig. 3).

SGA Group
Friedman's test showed statistically signi cant differences (p <0.05) between the expression levels of these genes within the SGA group. Tukey's post hoc test was performed in order to determine exactly where there were signi cant differences between gene expression. Tukey's test showed that there were statistically signi cant (p <0.05) differences between gene expression -for example: The SGA group had a signi cantly higher level of FADS3 than of FADS1 (Fig. 4).

IUGR Group
Friedman's test showed statistically signi cant differences (p <0.05) between the expression levels of these genes within the IUGR group. Tukey's post hoc test showed that statistically signi cant (p <0.05) differences between gene expression were: The IGUR group had a signi cantly higher level of FADS3 than of FADS1 (Fig. 5).
FADS2 -FADS3; For FADS2 the average was M = 2,123.00 while for FADS3 the average was M = 4,713.50; The IGUR group had a signi cantly higher level of FADS3 than ofFADS2 (Fig. 5).

PTB Group
The test that was performed showed no statistically signi cant differences (p> 0.05) between gene expression in the PTB group. Tukey's post hoc test con rmed there were no differences.

Discussion
Fat is one of the key ingredients necessary for proper foetal development. During pregnancy, a mother's body deposits fat in an amount which corresponds approximately to the baby's weight (3500g) [18].
These processes occur most vigorously in the second and third trimesters (the anabolic period) and will happen even if the mother is malnourished [19][20][21]. The concentration of phospholipids, non-esteri ed FAs and triglycerides increases in the mother's circulation. This mechanism is associated with an insulindependent decrease in lipoprotein lipase activity in adipose tissue and subsequent insulin resistance. As a result of these processes, part of the accumulated fat is transferred to the foetus via the placenta. The third trimester is a catabolic period. Increased lipolysis in the mother's adipose tissue is associated with decreased sensitivity of insulin receptors, which are hormonally controlled by progesterone, cortisol, prolactin and leptin [21,23]. As a result, in comparison to the anabolic period, even greater amounts of fat, including FAs, reach the placenta. The dynamics of changes in fat content in the foetus is different from that found in the mother. First of all, there is no catabolic period, secondly, the anabolic period begins much later than the mother's -between 20 and 22 weeks of pregnancy. Complicated maternal-placentalfoetal fat metabolism, especially of FAs and their derivatives, continues to be controlled by numerous factors, including enzymes whose expression is regulated, inter alia, at the level of transcription. This paper presents the results of testing the expression of three genes which encode strategic desaturases controlling the formation of n-3 and n-6. In the AGA, LGA, SGA, PTB and IUGR groups that were studied, no signi cant differences in transcriptional activity of the FADS1 and FADS2 genes were observed. This may mean that LC-PUFA biosynthesis and pro-in ammatory and anti-in ammatory cytokines are functioning relatively normal. However, this is not absolutely certain because we have not studied the polymorphism of the FADS1 and FADS2 genes as a factor that could have an effect on foetal development. Nevertheless, studies have shown differences in the transcriptional activity of the FADS3 gene in the tests on placenta. Women who gave birth to healthy children on time (AGA) and women who gave birth to children with only minor problems -SGA and LGA had higher FADS3 transcriptional activity than in the PTB and IUGR (higher level of problems) groups. The result is di cult to interpret because FADS3 and the desaturase encoded by it have not been well researched. It is known that the transcriptional activity of the gene in tissues is signi cantly different between males and females (mice, rats). FADS3 encoded desaturase can introduce, like any other desaturase, a double bond into the FA chain but other potential functions should be considered. In the world of living matter, desaturases perform hydroxylation [24,25], acetylenation and epoxidation [26] reactions. Such substances as etherlipid [27], sphingolipid [25] and cholesterol [28] can also be substrates for desaturases. Therefore, its potential physiological role can be broad, especially when the fact that D13D exists in at least three isoforms [24] is taken into account. One concept which could explain the lower transcription activity of FADS3 in the PTB and IUGR groups is the speci c structure of the gene promoter. Regions binding factors NF-κB [29,30], MYCN [31] and p63 protein [24] were identi ed in it, suggesting that FADS3 is the presumed target gene for these factors. It is known that NF-kB, MYCN and p63 are involved in cell pathways associated with proliferation or apoptosis. Studies exist which demonstrate the impact of IUGR on these pathways. For example, IUGR disrupts NF-κB-regulated proangiogenic targets in foetal pulmonary artery endothelial cells, which leads to the abnormal metabolism of extracellular matrix components and, as a result, interferes with pulmonary angiogenesis [32]. In the placenta of pregnancies complicated with IUGR, in which the processes of apoptoptosis are stronger than in a healthy placenta [33], a signi cantly higher NF-κB expression can be observed [34]. It is not known why the rise in NF-κB does not cause an increase in the expression of FADS 3 in the IUGR placenta; in fact, exactly the opposite happens. Higher NF-κB activity is accompanied by lower FADS 3 transcriptional activity. It is likely that NF-κB inhibitors increase during IUGR, or the chromatin is remodelled in such a way that it becomes inaccessible to the FADS 3 promoter. Changes in FADS 3 activity in the course of IUGR may also be related to the functioning of the membrane transport system, which is responsible for maintaining the correct FA ratio in the maternal and foetal circulation. Changes in the F / M ratio were observed in IUGR in SGA and PTB children [2,16]. This work also thoroughly analyses the transcriptional activity of FADS genes in the AGA, LGA, SGA, PTB, and IUGR groups. Except for the PTB group which had the lowest number of samples, no differences in FADS gene expression were observed. In the other groups there were differences in expression between all genes, only in the LGA group there were no differences between FADS2 and FADS3. Comparing the average FADS between the AGAs and LGAs, it can be assumed that the lack of differences in the LGA group was due to increased FADS2 activity and slightly reduced FADS3 activity.
LGAs are a group of children who, in addition to increased body weight (> 90 percentile), have increased body fat. Both of these, of course, are involved in fat metabolism, so with the increased fat mass of the child, changes in D6D and D13D activity are highly likely, although surprising, for example, in the case of the FADS3 gene product. It has been previously shown that the increased expression of FADS3 in adipose tissue is characteristic of hyperlipidemia [35]. Our research shows that this is the opposite for the placenta. So, perhaps the reduced placental expression of FADS3 in LGA children is a type of compensatory mechanism that regulates foetal fat metabolism. From a clinical point of view, however, it is more interesting to see a signi cant increase in the transcriptional activity of the FADS2 gene encoding desaturase 6, catalysing the reactions of the main biosynthesis pathway n-3 and n-6. One of the factors (although not studied in this research) that could affect FADS2 expression is the pregnant woman's diet. With an ample supply of plant oils, such as sun ower-seed oil, sa ower oil or corn oil, which contain large amounts of LA, then less DHA is produced from ALA as a result of n-6 desaturase inhibition leading to decreased EPA biosynthesis. The n-6 pool then increases, which could be a risk factor for the development of LGA [1,36]. Furthermore, it may have an in uence on the metabolism of medium-chain fatty acids (MCFA), especially miristic acid (C14: 0) and lauric acid (C12: 0), which have a signi cant impact on the conversion of EPA to DHA [37]. This, in turn, in addition to placental biosynthesis LC-PUFA, can disturb the speci ed hierarchy DHA> AA> LA> ALA de ning the order of transport of the acids across the placental barrier [38,39].

Conclusions
The placenta ful ls a hormonal, nutritional and metabolic role. Its task is to control the development of the foetus, although the hormonal-metabolic mechanisms occurring in the placenta also affect, to a large extent, the body of a pregnant woman. Fatty acids play a key role in these mechanisms. Some of them are transported through the placental barrier, others undergo biosynthesis in the placenta. Often, placental biosynthesis involves the elongation and desaturation processes of pre-existing acids with shorter carbon chains and which either lack or have fewer double bonds. Desaturases are involved in these processes. This is an important group of enzymes because they maintain the balance of levels of n-3 and n-6 FAs, have a signi cant role in the development of the nervous system and cell membranes, and affect general maternal-placental-foetal homeostasis. Our studies have shown that the transcriptional activity of the FADS1 and FADS2 genes remain at similar levels in the groups we examined. It was only in the FADS3 gene that differences were discovered. Its lowest activity was observed in the placenta of women who gave birth to premature babies. In this group, no differences were observed in the transcriptional activity of the tested FADS1, FADS2 and FADS3. However, in the SGA group, differences were revealed only between the FADS3 and FADS1 genes. The AGA, LGA and IUGR groups had a similar expression pro le.
The FADS3 gene dominated and the FADS1 gene had the lowest activity, although the LGA group did not show differences between the FADS3 and FADS2 genes. The IUGR group had the lowest transcriptional activity of all genes, while maintaining statistical differences between them. The largest number of differences in gene activity were observed in the placenta of women who gave birth to children with a mild degree of disorder -i.e. PTB and SGA children.

Consent to publish
Not applicable.
Availability of data and materials