Elevated Plasma Levels of Arginines During Labor Among Women with Spontaneous Preterm Birth: A Prospective Cohort Study

Preterm birth (PTB) is a leading cause of infant mortality and morbidity. The pathogenesis of PTB is complex and involves many factors, including socioeconomy, inflammation and infection. Asymmetric dimethylarginine, ADMA and symmetric dimethylarginine, SDMA are involved in labor as inhibitors of nitric oxide, a known relaxant of the uterine smooth muscles. Arginines are scarcely studied in relation to PTB and we aimed to investigate arginines (ADMA, SDMA and L‐arginine) in women with spontaneous PTB and term birth.


Introduction
Preterm birth (PTB), birth before gestational week 37, is a leading cause of infant mortality and morbidity [1].In addition to causing one million deaths globally each year, PTBs are associated with long-term complications, including impaired neurodevelopment, lung disease and adverse cardiovascular effects [1].The frequency of PTB varies globally but is estimated to be approximately 11% worldwide [2].These numbers are on the rise because of an increase in induction or caesarean delivery of preterm pregnancy, as well as PTB connected to multiple pregnancies following assisted reproduction [1,3].A total of 65%-75% of all PTB occur after spontaneous rupture of the fetal membranes or premature contractions of the uterus, which is called spontaneous PTB (sPTB) [4].The molecular mechanism behind sPTB is not fully understood, but it is believed to result from mechanisms that prematurely stimulate the physiological process which involves the activation of pro-inflammatory pathways within the uterus [5][6][7].
Nitric oxide (NO) is synthesized from amino acid L-arginine by the enzyme nitric oxide synthase (NOS).Nitric oxide promotes vascular protection [8], and relaxes the uterus muscle which is important until the start of labor.L-Arginine, when methylated, can also give rise to both asymmetric dimethylarginine (ADMA) and symmetric dimethylarginine (SDMA) [9].ADMA acts as an endogenous inhibitor to the synthesis of NO.In normal pregnancy, ADMA levels are low in the first trimester, thus facilitating the NO effect, and gradually increase in the second and third trimester [10].Higher levels of ADMA have been shown in cardiovascular events [11], in women with preeclampsia [12], gestational diabetes [13] and isolated intrauterine growth restriction [14].Despite these findings, there is much to be learned about the role of ADMA in pregnancy, pregnancy-related complications, and during labor.Systemic inflammation markers like C-reactive protein (CRP), tumor necrosis factor (TNF), along with its receptors TNF-R1 and TNF-R2, and growth differentiation factor (GDF)−15, have been studied extensively for their roles and levels at initiation and maintenance of labor, including preterm labor (PTL), with varying and inconclusive results [7,[15][16][17][18][19][20][21].
We recently showed that levels of ADMA and SDMA were higher in term vaginal birth when compared to birth by elective caesarean section [22].However, to this date, there are no studies that have measured plasma levels of arginines (L-arginine, ADMA and SDMA) in sPTB.Our objective was to examine if arginines were higher during labor among women with sPTB compared to term birth.As a secondary aim, we wanted to assess common inflammation markers CRP, TNF-R1, TNF-R2 and GDF-15, to study possible correlations between the levels of these markers and arginines.

2
Materials and Methods

Study Design
This study was undertaken as a sub-study based on the population-based, prospective cohort BASIC (Biology, Affect, Stress, Imaging and Cognition; https://www.basicstudie.se)study at the Department of Obstetrics and Gynecology, Uppsala University Hospital, Sweden.The method of recruitment has been extensively described in a previous study of the cohort profile of the BASIC population [23].In summary, all pregnant women in the county of Uppsala are referred to Uppsala University Hospital for a routine ultrasound examination at gestational weeks 16-18, whereby they were offered written information and invited to the BASIC study.Participants gave their written consent separately for every modality they wished to take part in, such as blood sampling.Exclusion criteria were: age below 18 years, insufficient ability to read and understand Swedish, protected identity, known blood borne infections and/or non-viable fetus.Between September 2009 and November 2018, 31 687 pregnant women had been invited to the study of which 6478 (20.4%) chose to participate.Participants were followed up at multiple points from baseline at gestational week 16-18 through 1-year postpartum.Biological samples were collected at several of these points and at childbirth.Participants in the study also gave consent to reviewing of their medical records.

Population
After review of the BASIC database, 34 women with vaginal sPTB (gestational week 22 + 0 to 36 + 6) and 45 women with vaginal spontaneous term birth (from gestational week 37 + 0) were included after inclusion and exclusion criteria were met.The women with term birth have been described in one of our previous publications [22].Both nulliparous and parous women were included.Exclusion criteria were multifetal pregnancy, polyhydramnios, chorioamnionitis, hypertensive disorders of pregnancy and labor dystocia.Information about sociodemographic factors, medical history including obstetrics/gynecology, current medication, pregnancy complications, labor characteristics (e.g., mode of onset, induction methods, cervical dilatation at blood sampling, gestational length at birth, duration of labor, expulsion time and postpartum bleeding) and neonatal information (e.g., gender, birth weight and Apgar scores) were gathered prospectively.

Blood Sample Collection and Processing
Peripheral blood samples had been collected upon admission to the delivery ward at labor in the scope of BASIC.The process of collection, preparation and storage of plasma samples were performed as previously reported [23].The samples were centrifuged, within 2 h, at 1500 × g for 10 min and stored at −70 • C until analysis.
L-Arginine, ADMA and SDMA were quantitatively analyzed with Ultra-High Performance Liquid Chromatography-tandem quadrupole mass spectrometry (UHPLC-MS/MS) after addition of isotope-labeled internal standards and protein precipitation using the same method as in one of our previous publications [22], (Waters ACQUITY, Waters Inc., Milford, MA, USA) coupled to tandem quadrupole mass spectrometry (XEVO TQ-S, Waters Inc.).Chromatographic separation was achieved on a Premier HSS T3 column (150 mm, 3.0 mm, 1.7 µm, Waters, Milford, MA, USA) at 40 • C. Mobile phase A consisted of 20 mM ammonium formate in water and mobile phase B was methanol.Isocratic elution was performed with 10% of mobile phase B. The mass spectrometric detection was performed using positive ionization electrospray ionization.Quantification was carried out using a selected reaction monitoring method with the following transitions: for ADMA m/z 203.1 > 116.1 (collision energy: 18 eV), ADMA-d6 m/z 209.2 > 52.

Ethical Approval
The original application was made in the context of the BASIC project and was approved by the Regional Ethical Review Board in Uppsala (Reference number 2009/171).A supplementary application has been approved later by the Swedish Ethical Review Authority regarding the analysis of blood samples in this study (Reference number 2020-00633).

Statistical Analysis
Statistical analysis was performed using the SPSS version 27.0 (SPSS Inc.PASW statistics), for the Windows software package.We chose to use non-parametric tests because the data was not normally distributed.Results are presented as medians (interquartile ranges) for continuous variables and as numbers (percentages) for categorical variables.For comparison between the two groups, the Mann-Whitney U test was used for continuous data and chi-square for categorical data.Differences between groups were adjusted for gestational length at birth, BMI at the early second trimester, mode of onset of labor, cervical dilatation at sampling and fetal birth weight using Univariate General Linear Model.Spearman's rank correlation test was used to measure the potential correlations between demographic data and clinical characteristics versus arginine levels to identify possible covariates, and to explore the correlations between arginine levels and inflammation markers.The level of significance was set at p < 0.05.

Demographics and Clinical Characteristics
A comparison of the demographic and clinical characteristics of the study population is summarized in Table 1.No significant differences were seen between groups regarding demographic data.However, there were differences regarding some of the clinical characteristics: in the preterm group, the mode of onset of labor was more commonly spontaneous rupture of membranes (p < 0.001), the cervix was less dilatated at sampling (p = 0.02), gestational length at birth was lower (p < 0.001), fetal birth weight was lower (p < 0.001) and neonatal intensive care unit care was more common (p < 0.001).There was no reported use of alcohol, smoking or chewing tobacco during pregnancy in either group.

Plasma Levels of Arginines and Inflammation Markers
Plasma levels of arginines and inflammation markers are presented in Table 2. Levels of ADMA (p < 0.001) and L-arginine (p = 0.027) were higher at labor in the sPTB group than in the term birth group, Table 2 and Figure 1.No differences were seen in levels of SDMA and L-Arginine/ADMA ratio and L-Arginine/SDMA ratio, Table 2 and Figure 1.Regarding inflammation markers, no differences were found between the groups except for TNF-R1 (p = 0.013), Table 2. Differences between the groups are adjusted for gestational length, BMI, mode of onset of labor, cervical dilatation and fetal birth weight at sampling.In a subanalysis, in women with sPTB, we found no differences in levels of arginines or inflammation markers regarding parity and mode of onset of labor (data not presented as a table).

Correlation Analysis of Arginines to Inflammation Markers
No correlations could be observed between demographic (maternal age or BMI at the early second trimester) or clinical characteristics (gestational length at birth, cervical dilation at sampling, duration of labor, time of expulsion or fetal birth weight) in relation to arginines and inflammation markers (results not shown in table format).Further, no correlations were found when comparing arginines with inflammation markers in women with sPTB, except between ADMA and CRP (rs 0.37 and p = 0.03), Table 3.

Main Findings
The purpose of this study was to examine the role of arginines in sPTB while controlling for potential correlation with already established inflammation markers.Levels of ADMA and L-arginine were significantly higher in sPTB compared to uncomplicated spontaneous term birth.Among inflammation markers, only TNF-R1 was significantly higher in sPTB compared to term birth.However, there were no significant correlations between arginines and inflammation markers, except between ADMA and CRP.The women with term birth have been described in one of our previous publications [22].

TABLE 3
Correlations between arginines and inflammation markers in women with spontaneous preterm birth.

Interpretation of Results
The elevated levels of ADMA and L-arginine in the preterm group may suggest a possible role of arginines in the pathophysiology of sPTB.At the same time, it remains unclear if these elevated levels of arginines should be considered as a cause or an effect of the initiation of PTL.There are no previous studies on arginine levels in human PTL for comparison.However, we recently showed that plasma levels of arginines were significantly higher in women at term labor, than in women at term but not in labor [22], which is in line with the findings of the present study and might supports the theory that higher arginine levels lead to PTL.Additionally, previous findings of gradual increase in ADMA from first to third trimester [10] and significant lower levels of NOS in women during PTL, compared to women with preterm and term pregnancies that were not in labor, also supports this hypothesis.Furthermore, inhibition of NO synthesis by ADMA may contribute to vascular dysfunction and vasoconstriction which could lead to many other diseases or complications in pregnancy, like preeclampsia [12,24], gestational diabetes [13,25] and isolated intrauterine growth restriction [14].Higher levels of L-arginine in the preterm group might indicate a compensatory mechanism in response to elevated concentrations of ADMA, to facilitate NO synthesis, since it also requires L-arginine as a substrate [11].Moreover, the absence of differences in L-arginine/ADMA and L-arginine/SDMA ratios could suggest a relatively stable level of L-arginine compared to dimethylarginines.As there were no correlations between arginines and inflammation markers in sPTB, we conclude that the significantly higher levels of arginines in women with sPTB are not a result of inflammation and that arginines were elevated independently.There are some studies about dimethylarginines in uncomplicated pregnancy with a gradual increase of dimethylarginines from the first trimester to third trimester [10] and from gestational week 36 to about 3 days postpartum [26].To this date, there are no studies regarding plasma levels of arginines at labor in human PTB.We recently showed that levels of dimethylarginines were higher at birth in women with term vaginal birth when compared to birth by elective caesarean section [22].There is a theory suggesting that the lower level of dimethylarginines in pregnancy compared to non-pregnant women, facilitate NO and maintain uterine quiescence [10].Thus, it can be hypothesized that the initiation of labor could be associated to downregulation of NO through upregulation of dimethylarginines.It has been shown that NOS expression in rat myocytes fell significantly from mid-gestation to term and was barely detectable in term labor [27,28].In a similar study, however, in human myocytes, Bansal et al.
showed that the expression of NOS was highest in myometrium of preterm non-laboring patients, which decreased by 75% at term, and was barely detectable in specimens from preterm in-labor or term in-labor patients [27].
Our findings of no difference in CRP in sPTB compared to term birth, were not consistent with the results of previous studies [29,30].This might be explained by the fact that in our study PTBs with infection were excluded.The significantly elevated inflammatory marker TNF-R1 in the sPTB group supports previously reported associations between PTB and inflammation markers [21].Similarly, our results regarding no differences in levels of TNF-R2 between sPTB and term birth are in line with previous reports [18,21].According to a few studies, levels of GDF-15 were higher in obese women [31], women with hyperemesis gravidarum [32] and with abnormal glucose metabolism [33].
However, studies about GDF-15 in PTB are very scarce.

Strengths and Limitations
This is the first study, to our knowledge, that explores the potential role of arginines in human PTB and provides novel findings regarding arginines in sPTB.Similarly, data on inflammation markers allowed for the analysis of potential correlations with arginines to be considered.Risk pregnancies that could be associated to higher levels of arginines were excluded which made the results applicable to a low-risk population.Finally, both groups were relatively homogeneous with respect to their sociodemographic characteristics, reducing confounding factors.Recall bias was minimized as the information on maternal and neonatal health care was collected prospectively.
The major limitation was that, because of the cross-sectional nature of the study, it was not possible to conclude the causality between the studied biomarkers and sPTB.Future studies that measure arginines and inflammation markers in PTB patients, for example at several time points in pregnancy, at labor and postpartum, would provide more insight into the potential causality of arginines in PTB.

Conclusion
In conclusion, this study presents novel evidence for the potential involvement of arginines in the pathophysiology of PTL.Further, it seems that arginine levels at PTL are varying independently of several inflammatory markers.These findings contribute to a better understanding of the molecular mechanisms behind sPTB.
Our findings may open up new avenues for research, with a potential of ADMA as therapeutic targets in the prevention and management of sPTB, by preventing its inhibitory effect on NOsynthesis and thereby minimizing the disruptive effect of uterine quiescence.

FIGURE 1
FIGURE 1 Boxplot analysis to show the differences in (A) ADMA, asymmetric dimethylarginine; (B) SDMA, symmetric dimethylarginine; (C) L-arginine; (D) L-arginine/ADMA and (E) L-arginine/SDMA in women with spontaneous preterm birth and spontaneous term birth.The top and the bottom of the boxes represent the third and the first quartiles.The horizontal lines within the boxes represent the median values.The bars on the side of the boxes represent the highest and the lowest values.Red and blue filled circles represent extreme values.a p < 0.001.The women with term birth have been described in one of our previous publications [22].

TABLE 1
Comparison of demographic and clinical characteristics between spontaneous preterm and term birth groups.
[22]s: Continuous variables are presented as median (interquartile ranges) and number (percentages) for categorical variables.pvalueswere calculated with the Mann-Whitney U test for continuous data while chi-square was used for categorical data.The women with term birth have been described in one of our previous publications[22].Abbreviations: Apgar, appearance, pulse, grimace, activity, respiration; BMI, body mass index; g, gram; kg, kilograms; m, meter; mL, milliliter; mm, millimeter; NICU, neonatal intensive care unit; VE, vacuum extraction.

TABLE 2
Plasma levels of arginines and inflammation markers at labor in women with spontaneous preterm and term birth.

levels Preterm birth (n = 34) Term birth (n = 45) Adjusted p-value
[22]s: Values are presented as median with interquartile ranges.Differences between groups are analyzed by the Mann-Whitney U test and are adjusted for gestational length, BMI, mode of onset labor, cervical dilatation and fetal birth weight at sampling, by Univariate General Linear Model.The women with term birth have been described in one of our previous publications[22].Abbreviations: µM/L, micromole/liter; ADMA, asymmetric dimethylarginine; CRP, C-reactive protein; GDF, growth differentiation factor; mg/L, milligram/liter; pg/mL, picogram/liter; SDMA, symmetric dimethylarginine; TNF, tumor necrosis factor.