**1. Preeclampsia—Background**

Preeclampsia (PE) is a serious, pregnancy-specific, multi-organ disease process of compound aetiology. It affects 3–6% of expecting mothers worldwide and it persists as a leading cause of maternal and foetal morbidity and mortality [1]. The diagnosis of PE is clinical. The diagnostic criteria were revised in 2013 and 2014: it is defined as new onset hypertension developing after 20 weeks of gestation and the coexistence of a minimum of one of the following new onset conditions: proteinuria, maternal end- organ dysfunction (including renal, hepatic, haematological, or neurological complications), or uteroplacental dysfunction reflected in foetal growth restriction (FGR) [2,3]. The disease can be further clinically classified as PE with or without severe features, as well as early-onset syndrome (presenting before 34 weeks of gestation, versus late-onset after completed 34 weeks), preterm PE (occurring from 34 + 1 but before 37 + 0 weeks), and term PE (after completed 37 weeks) [4]. The current managemen<sup>t</sup> of the disease mainly depends on gestational age and assessment of PE severity, focusing on blood pressure control, maternal and foetal surveillance, and it aims to deliver the baby in optimal condition prolongating the pregnancy without worsening state of the mother [3]. It requires individualized calculations of risks and benefits, but unfortunately delivery still remains the only definitive treatment.

In recent times, a huge progress has been made in understanding the disease, getting scientist and doctors closer to explain biological mechanisms underlying the development of PE that possibly can be used to create new therapeutic strategies targeting them.

Within the last decade, subsequent studies confirmed the hypothesis of Roberts and colleagues from 1989, who suggested that PE clinical manifestations might be due to maternal endothelium dysfunction [5]. In fact, the hallmark features of PE result from vessel involvement and demonstrate maternal endothelium as a target tissue. However, the placenta, as the interface between mother and fetus, is also regarded a key and causative player in pathogenesis of PE. Growing evidence suggests that chronic placental hypoperfusion triggers the production and release of certain agents that are responsible for endothelial activation and injury. (Figure 1)


**Figure 1.** Physiological roles of NO pathway in pregnancy and their possible influence on preeclampsia (PE) development in two-stage model of disease. NO PATHWAY ROLE: Nitric oxide (NO) pathway role; STAGES OF PE: Stages of preeclampsia (PE).

The development of PE involves a two-stage process [6]. The first, crucially important step is asymptomatic and it takes place during placental invasion and differentiation. While, during normal placentation, the embryo-derived cytotrophoblast properly invades the uterine wall, including the myometrium and spiral arterioles and it leads to transformation of maternal spiral arteries into large capacitance and low resistance vessels; this process is defective in preeclampsia [7–9]. The invasion of cytotrophoblast is incomplete, restricted to superficial layers of decidua that provides inadequate access to maternal oxygen and nutrients for the placenta and growing foetus. Poor placental invasion leads to diminished uteroplacental perfusion pressure and ischemia.

Abnormalities of placental invasion anticipate maternal disorder. Clinical manifestations that define PE represent the second stage of disease. Chronic placental hypoperfusion triggers abnormal production and the release of numerous bioactive factors into the maternal circulation. These circulating substances target endothelial cells resulting in widespread endotheliosis, endothelial dysfunction, generalized multi-system vasospasm, reduced plasma volume, oxidative stress, and hyperinflammatory state. Excessive expression of antiangiogenic proteins, like soluble fms-like tyrosine kinase 1 (sFlt-1) and soluble Endoglin (sEng), which catch circulating, decreased proangiogenic substances, like vascular endothelial growth factor (VEGF), placental growth factor (PlGF), and transforming growth factor β (TGFβ) result in an understanding of PE as an antiangiogenic state [10–13].

Defective trophoblast invasion is an early event in preeclampsia development. However, it has not been resolved whether it is the reason or result of another underlying problem. It remains unclear why trophoblast invasion is interrupted, but an altered immunological response at maternal-foetal interphase, genetics, and environmental factors are believed to contribute, although their role may vary between patients [14,15]. Furthermore, it is suggested that maternal susceptibility and response to placental derangements determines the onset, severity, clinical manifestations, and progression of the disease [16]. The most recent theory identifies PE as a complex disease with two distinct clinical presentations. The first, placental phenotype is associated with shallow trophoblastic invasion and restricted foetal growth, as opposed to PE associated with maternal metabolic syndrome. The second phenotype is associated with normal fetal growth and maternal low-grade inflammation, mainly due to placental oxidative stress, placental villi overcrowding, and decidual lesions [17,18].

Nitric oxide (NO) is one of the key players in the regulation of placental blood flow. It is actively engaged in cytotrophoblast endovascular invasion and development of the placenta, through its unique angiogenic and vasculogenic properties [19]. Current evidence supports altered NO production in the feto-placental unit in preeclampsia, which, by reduced bioavailability, may contribute to vasoconstriction of the placental bed, abnormal placental perfusion, and its maternal consequences, like increased blood pressure, systemic vascular resistance, and sensitivity to the pressors [20–25].

Asymmetric dimethylarginine (ADMA), which is an endogenous inhibitor of NO synthase (NOS), has also been associated with impaired endothelial function and with uterine artery flow disturbances that are characteristic for preeclampsia [24,26,27].

Homocysteine (Hcy) elevated concentrations in preeclamptic women lead to elevated ADMA levels, since Hcy has an inhibitory e ffect on ADMA metabolism. Hyperhomocysteinemia (HHcy) is also associated with endothelial cells lesions due to vascular fibrosis, which results in alterations in the coagulation system, enhanced platelet activation, and thrombogenesis—changes that are noted in preeclampsia [28–30].

In this review, we will present the latest findings on the role of nitric oxide, ADMA, and homocysteine in the etiopathogenesis of preeclampsia and their possible clinical implications.

#### **2. Metabolism and Biological Role of NO, ADMA, and Homocysteine**

Furchgott initially described nitric oxide as an endothelium-derived relaxant factor (EDRF) in 1980 after attributing a vasodilatory e ffect on vascular smooth muscle by stimulation of cholinergic nerves to the endothelium [31]. The identification of EDRF as NO was reported seven years later and was awarded a Nobel Prize in Physiology or Medicine for Furchgott, Ignarro, and Murad in 1998 for their discoveries concerning nitric oxide as key transmitter in the cardiovascular system. NO is produced through L-arginine-NO synthase pathway by converting L-arginine to L-citrulline in the presence of oxygen and the cofactor tetrahydrobiopterin or alternative enzymatic and non-enzymatic nitrate-nitrite-NO pathways [32,33].

Nitric oxide synthase (NOS) possess three di fferent isoforms, namely neuronal NOS (nNOS) or type 1, inducible NOS (iNOS) or type 2 and endothelial NOS (eNOS) or type 3 [32]. NOS1 and NOS3 are considered as constitutive NOS. Endothelial NOS is stored in plasma membrane caveolae and its distribution and activity are regulated by numerous mechanisms [34]. Being released from endothelial cells, NO is quickly transported to the closest vascular smooth muscle cells, where it exerts its role by inducing the production of cyclic guanosine monophosphate (cGMP) as a second messenger. It may be neutralized by reactive oxygen species on the way to its target cells [35].

NO is the key transmitter for the endothelium-dependent regulation of the vascular tone that is controlled by humoral, metabolic and mechanical factors, for example, in response to increased blood flow [36]. Furthermore, NO inhibits the adhesion and activation of platelet aggregation, abolishes the toxic activity of superoxide ions, and acts as an anticoagulant and antiatherogenic substance [22,32].

It is also considered to have major e ffects on the gestational endothelial function as well as to play a supportive role in promoting embryo survival, tissue remodelling, immunosuppression, and vasoregulation critical for placental nutrient transport [37–39]. The human foeto-placental vasculature lacks autonomic innervation and, therefore, NO confers autocrine and/or paracrine e ffects, influencing di fferent aspects of physiological pregnancy. In particular, NO is the main vasodilator that is involved in foeto-placental vascular reactivity regulation, placental bed vascular resistance, trophoblast invasion and apoptosis, and platelet adhesion and aggregation in the intervillous space [40].

Further, the role of NO is also established in vasculogenesis, which results from the de novo formation of vessels derived from pluripotent precursor cells and angiogenesis, the formation of functional capillaries from pre-existing vasculature. Vascular endothelial growth factor (VEGF) is a key particle in these processes. Its expression is mediated by NO release and was required for initiation of vasculogenesis [41]. NO is also a critical downstream mediator of other than VEGF potent angiogenic substances, like basic fibroblast growth factor (FGF), and angiopoietin-1 [42]. The critical role of NO in angiogenesis has been shown in eNOS knockout mice [43]. NOS inhibition is accompanied by defective angiogenesis, as exemplified by deficient vascular sprouting. Interestingly, NO may also act upstream of angiogenic growth factors, because hypoxia-inducible factor-1 (HIF-1) perhaps mediated the e ffect of NO on VEGF production [44].

Asymmetric dimethylarginine (ADMA), which is an analogue of L-arginine, constitutes a natural metabolite that is found in human plasma. Dimethylarginines are formed as a result of the degradation of methylated arginine residues in proteins [45]. Approximately 80% of ADMA undergoes enzymatic transformation by two dimethylarginine dimethylaminohydrolases (DDAH-1 and -2) to L-citrulline and dimethylamine, whereas kidneys excrete the rest. ADMA is endogenous competitive inhibitor of L-arginine for all three isoforms of NOS. Elevated levels of ADMA block NO synthesis and limit the cellular uptake of L-arginine, thereby contributing to oxidative stress and disrupting further NO biogenesis. In this way, ADMA impairs the endothelial function and thus promotes atherosclerosis. Therefore, it is recognized as a biomarker of endothelial disorders. The ADMA levels are found to be elevated in patients with various cardiovascular and metabolic conditions, such as hypercholesterolemia, atherosclerosis, hypertension, chronic heart or renal failure, diabetes mellitus, stroke, and hyperhomocysteinemia [29,45–47]. ADMA has been shown to increase systemic vascular resistance in humans, as an endogenous inhibitor of NOS [45].

Homocysteine (Hcy) is a sulfur-containing amino acid that is produced during the conversion of essential amino acid methionine (Met) to cysteine (Cys) [48]. Its synthesis occurs in the transsulfuration of dietary methionine, which is abundant in animal protein, but it can also occur in demethylation that is related to fasting conditions. Hcy is metabolized by one of the two following pathways: remethylation to methionine, which requires the addition of a methyl group from 5-methyltetrahydrofolate (5-methyl THF) and the cofactor vitamin B12 (or betaine in an alternative reaction, restricted to the liver and independent of vitamin B12); and, transsulfuration to cystathionine; and finally, to cysteine, which requires vitamin B6 as a cofactor [49]. Methionine derivative, S-adenosyl methionine, is a cofactor that serves as a most important methyl donor of the body, whereas cysteine is used for glutathione synthesis or it is metabolised into taurine.

5-methyltetrahydrofolate (5-MTHF), which is the predominant circulating form of folate is the result of a reduction of 5,10-methylenetetrahydrofolate (5,10-MTHF) catalysed by the MTHFR (methylenetetrahydrofolate reductase) enzyme, coded by MTHFR gene, whose locus is on chromosome 1 at the end of the short arm (1p36.6) [50,51]. Polymorphisms of the MTHFR gene play a significant role in the pathogenesis of hyperhomocysteinemia.

The definition of hyperhomocysteinemia (HHcy), generally understood as increased homocysteine in the blood, di ffers between authors [52]. The total fasting concentration of Hcy in plasma of healthy patients is low and its level is 5.0–12.0 μmol/l when the immunoassay methods are used or between 5.0 and 15.0 μmol/L when assessed with the use of HPLC (high-performance liquid chromatography) [53]. Modarate HHcy is diagnosed if the levels are within the range of 16 to 30 μmol/L, 31–100 μmol/L is considered to be intermediate and a value above 100 μmol/L is classified as severe hyperhomocysteinemia [54].

The main causes of elevations in homocysteine levels are vitamin deficiency (B6,B12,folate), aforementioned genetic defects in enzymes that are involved in its metabolism (cystathionineβ-synthase deficiency and MTHFR), and disease conditions that interfere in the metabolism of cofactor levels, disturbing the transsulphuration and remethylation processes. In the general population, higher values of Hcy are observed in men then in women, although the discrepancy diminishes with age and in postmenopausal patients who tend to have higher Hcy levels [55].

In general, we can divide HHcys in two types: severe, but rare forms due to major genetic defects (individuals with the rare homocystinuria typically have levels of >100 μmol/L) and more common, moderately elevated homocysteine levels that are related to a pathogenesis, such as genetic and environmental factors, which is observed in up to 5% to 12% of the general population [52,56].

The most common cause of severe hyperhomocysteinemia and classic homocystinuria (congenital homocystinuria) is considered to be the homozygous deficiency of CβS (cystathionine-β-synthase). This defect is responsible for an increase as much as up to 40-fold in fasting total homocysteine. Other not often observed genetically conditioned states of HHcy are the homozygous deficiency of MTHFR, deficiency of methionine synthase, and impaired activity of methionine synthase due to impaired vitamin B12 metabolism [54].

However, the most common genetic deficiency, which occurs at large rates in various populations, is single nucleotide polymorphism of MTHFR that has been associated with mild and moderate (25–60 μmol/L) hyperhomocysteinemia [57]. A point mutation C-to-Tsubstitution at nucleotide 677 (677C →T) in the gene for MTHFR causes a thermolabile variant of the enzyme and has half-reduced activity, whereas in people who are homozygousfor MTHFR C677T, there is only 30% of normal enzyme function [54]. Another point mutation, called MTHFR A1298C, leads to 60% of normal enzyme function. Double heterozygous (1 abnormal MTHFR C677T gene plus 1 abnormal MTHFR A1298C gene) results in decreased reductase activity as those homozygous for the C677Tpolymorphism [58].

However, the leading cause of HHcy is folate, vitamin B12, and less commonly, B6 deficiency due to low supply, malabsorption, and treatment with substances, such as cyclosporin, methotrexate, fibrates, Levodopa (L-DOPA), and carbamazepine that interfere with the metabolic paths of these vitamins [59,60]. High Hcy levels have been also associated with impaired renal function, high plasma creatinine, smoking, co ffee consumption, and alcoholism [52].

HHcy is generally recognized as an independent risk factor for coronary, cerebral, and peripheral atherosclerosis, which was first reported by McCully in 1969 and later confirmed in a meta-analysis of numerous additional studies [61–63]. An extend meta-analysis suggested that an increment of homocysteine of 5 mmol/L is comparable to the increase in the risk of coronary artery disease caused by cholesterol elevation of 0.5 mmol/L [62]. An association between HHcys and cardiovascular disease, as well as some age-related pathologies, like stroke, Alzheimer's disease, Parkinson's disease, chronic renal failure, and osteoporosis is widely described [61–72]. There are ongoing e fforts to understand if HHcy observed in vascular diseases is a causative factor or a consequence of endothelial activation [73].

#### **3. NO, ADMA and Homocysteine in Pregnancy**

During uncomplicated pregnancy, increased NOS activity in human uterine artery leads to higher NO levels [74]. NOS3 expression raises primarily in the syncytiotrophoblasts and NOS2 activity grows throughout pregnancy, with a peak around mid-gestation [38,39,75,76]. Physiological reduction of blood pressure during pregnancy may greatly rely on the vasodilatory action of NO. NO contributes to the vasodilatation of blood vessels and the decrease in vascular resistance observed during early pregnancy, when maternal blood volume expands, while systemic vascular resistance and systemic blood pressure both decline [22,25,75–77].

In normal pregnancy, also levels of cGMP, a second messenger of NO signalling is particularly increased during the first trimester in plasma and urine [78]. Furthermore, a NO-cGMP pathway is present in the human uterus and it may be responsible for maintaining its relaxation. Spontaneous contractility in vitro was enhanced by the NOS inhibitor L-NAME (nitro-L-arginine methyl ester) and

decreased by NO. Thus, uterine reduced the responsiveness to nitric oxide at term may play a role in the initiation of labour [79].

Different studies on total NO in pregnancy gave conflicting results. The measurement of its relatively stable metabolites, nitrate, and nitrite (NOx) is often employed as an indicator of NO production and as a marker of NOS enzyme activity because NO is highly labile molecule [80]. Still, the plasma level is influenced, not only by the production, but also by the clearance of NO derivatives [79]. Some studies found that NO production increases with gestational age during normal pregnancy, especially in the second trimester, and it peaks in the third trimester [81–83]. However, contrary results were also published reporting that maternal circulating nitrite level decreased with advancing gestation [84], or even that there were no changes in NO production when compared to the nonpregnan<sup>t</sup> state [85,86].

Likewise, studies investigating the circulating levels of NO in preeclampsia have also reported conflicting results [87]. These observations sugges<sup>t</sup> that the status of NO biosynthesis in women during normal pregnancy and preeclampsia remains to be defined.
