Sage Journals: Discover world-class research

Abstract

Objective:

Cytokines, molecules within the immune system that affect either a pro- or anti-inflammatory response, have previously been shown to influence birth outcomes. The maternal cytokine gene–environment interactions are thought to alter their expression, potentially influencing susceptibility to adverse birth outcomes. The aim of this study was to determine the association between the maternal interleukin-1β (IL-1β) haplotype and expression variation with oxides of nitrogen (NO_x) levels, and thereafter investigate the IL-1β haplotype-specific effects of NO_x exposure levels, IL-1β mRNA expression and other variables on gestational age.

Material and methods:

Using the prospective Mother and Child in the Environment (MACE) birth cohort in Durban, South Africa, 335 participants were genotyped for the IL-1β haplotype. Previous studies showed that three single nucleotide polymorphisms (SNPs), IL-1β-1464G/C, -511C/T and -31C/T, constitute the IL-1β functional haplotype. These SNPs were genotyped using a restriction fragment length polymorphism assay, while IL-1β mRNA expression was measured using a quantitative real-time polymerase chain reaction assay. Individual estimates of NO_x exposure were obtained by land use regression modelling. A multivariate linear regression analysis was employed to test for significant effects on gestational age.

Results:

IL-1β mRNA expression was found to possess a haplotype-dependent effect (p = 0.0001) and its expression levels positively correlated with NO_x levels (r = 0.34; p = 0.006). In the high haplotype model, a unit increase in NO_x exposure level was associated with a decrease in gestational age by 1 week (p = 0.02). Furthermore, gestational age decreased by 0.9 weeks for every unit increase of IL-1β mRNA expression level (p = 0.025). HIV-1 positivity was associated with a 0.2-week decrease in gestational age (p = 0.035) in the intermediate haplotype model and a 0.4-week decrease in the high haplotype model (p = 0.044).

Conclusion:

These data have implications for better understanding the effect of prenatal NO_x exposure on gestational age and demonstrate the role of the IL-1β haplotype in modulating the effects of NO_x exposure.

Keywords

IL-1β haplotype gestational age

Introduction

Ambient air pollution has been linked to adverse birth outcomes, including increased rates of mortality and various diseases in children, the elderly and pregnant women, who are particularly sensitive.^1,2 The risk for neonatal and infantile morbidity and mortality increases with decreasing gestational age³ and is a predictor of mortality and chronic morbid conditions later in childhood and adulthood, namely developmental disabilities and cardiovascular and respiratory diseases. Such conditions are associated with impaired responses to environmental agents, further impacting morbidity.

Oxides of nitrogen (NO_x), that is, nitrogen dioxide (NO₂) and nitric oxide, are a component of industrial and vehicular exhaust emissions. Traffic is reportedly the principal source of NO₂ and the residential distance from roadways impacts levels.^4,5 In Durban (South Africa), the primary source of NO_x includes mobile sources, responsible for emitting 11,500 tons per annum and over 60% of the total emissions, and industry (largely concentrated in the Durban South Industrial Basin (DSIB)), responsible for emitting 5400 tons per annum (29% of the total emissions).⁶ The DSIB comprises communities residing among heavy industry and significant roadways and is notorious for being among the most highly polluted regions in South Africa.⁷ The region is associated with a greater prevalence of adult respiratory conditions (including higher adult asthma rates than those internationally), childhood asthma and marked airway hyperreactivity compared to the northern region in Durban.^6

–9

NO_x, a strong oxidant, generates reactive oxygen species thereby inducing oxidative stress and the resultant pro-inflammatory response in vivo.^10,11 Pregnancy is an inflammatory state characterized by gestational stage-specific immune responses, thus pollutant-induced alterations in response may adversely affect birth outcomes.¹² Interleukin-1β (IL-1β), a mediator of the pro-inflammatory response to infection and oxidative stress, is implicated in the pathophysiology of various chronic diseases through dysregulated or increased synthesis.¹³ Data on the maternal IL-1β response to NO_x exposure, however, are scarce or non-existent. Most studies on air pollution and birth outcomes emanate from high-income countries, with none or few from low- or middle-income countries. In this study, we investigated the effect of prenatal NO_x exposure on the maternal IL-1β response and explored the association between maternal IL-1β expression on gestational age in a predominantly African birth cohort in Durban, South Africa. Since single nucleotide polymorphisms (SNPs) affect transcription in a haplotype-dependent manner,¹⁴ we characterized the maternal IL-1β haplotype distribution and associated expression profile using three major functional IL-1β promoter SNPs (IL-1β-1464, -511 and -31).¹⁵ We also explored the role of such haplotypes in determining the effects of NO_x exposure levels, IL-1β expression and other variables on gestational age, thereby identifying vulnerable subgroups to NO_x-induced and IL-1β haplotype-dependent gestational age outcomes. We hypothesized that chronic elevated NO_x exposure may play a significant role in altering maternal IL-1β expression and negatively influencing gestational age outcomes, while IL-1β haplotype-specific associations between maternal haemoglobin (Hb), body mass index (BMI; trimester 3), NO_x levels, IL-1β mRNA levels, HIV-1, income and education may play a role in influencing gestational age.

Methods

Subjects

Study subjects included a subset of 338 participants from the Mother and Child in the Environment (MACE) prospective birth study that commenced in 2013 in Durban, KwaZulu-Natal, South Africa. MACE participants were recruited from public sector antenatal clinics, during the first trimester. The cohort comprised of individuals from African, Caucasian, Indian and Coloured ethnicity. Informed and signed consent was obtained from each participant. Study exclusion criteria were twin births, pregnancy complications and smoking. The study was approved by the Biomedical Research Ethics Committee of the University of KwaZulu-Natal (reference number: BF263/12).

Genotyping

Maternal DNA was extracted from third trimester whole blood using the FlexiGene DNA Kit (QIAGEN GmbH, Hilden, Germany) as per manufacturer’s protocol. DNA concentration was quantified spectrophotometrically (NanoDrop Spectrophotometer, Biolab, Scoresby, VIC, Australia) and standardized to 10 ng/µl. The IL-1β gene SNPs at the following positions, IL-1β-511 C>T, IL-1β-31 T>C and IL-1β-1464 G>C comprising the IL-1β haplotype, were genotyped by the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method. The PCR primers and cycling conditions used for IL-1β amplification are shown in Table 1. The PCR mixture included 30 ng DNA, 5 μl buffer (5×), 3 μl magnesium chloride (25 mM), 2 μl deoxynucleotide (10 mM), 0.125 µl Taq polymerase (5U/µl; GoTaq^®G2 Flexi DNA Polymerase (Promega, Madison, Wisconsin, USA)), 0.125 μl of each primer (20 pmol/μl) and nuclease-free water (25 µl total volume). To confirm specificity, PCR products were visualized on a 1% agarose gel with a 50-bp molecular marker. PCR products were digested for 16 h at 37°C, IL-1β-511 C>T (10 U AvaI), IL-1β-31 T>C (10U AluI) and IL-1β-1464 G>C (10U StyI). Digested PCR products were visualized on a 2% agarose gel with a 50-bp molecular marker.

Table 1.

Primers used in study. Primers and cycling conditions used are listed in the table below.

Gene	Accession number/reference	Sequence: 5′–3′	Cycling conditions: denaturation, annealing and extension
IL-1β [C/T-511]	rs16944	F: 5′-TGGCATTGATCTGGTTCATC-3′	(95°C, 30 s), (55°C, 30 s) and (72°C, 30 s)
IL-1β [C/T-511]		R: 5′-GTTTAGGAATCTTCCCACTT-3′	(95°C, 30 s), (55°C, 30 s) and (72°C, 30 s)
IL-1β [C/T-31]	rs1143627	F: 5′-AGCTTCCACCAATACTCTTTTCCCCTTTCC-3′	(95°C, 30 s), (57°C, 30 s) and (72°C, 30 s)
IL-1β [C/T-31]		R: 5′-TACACACAAAGAGGCAGAGAGACAGAG-3′	(95°C, 30 s), (57°C, 30 s) and (72°C, 30 s)
IL-1β [G/C-1464]	rs1143623	F: 5′-CCCTCGTGTCTCAAATACTTGC-3′	(95°C, 30 s), (55°C, 30 s) and (72°C, 30 s)
IL-1β [G/C-1464]		R: 5′-GCACAGTGCCTGGCATGTAC-3′	(95°C, 30 s), (55°C, 30 s) and (72°C, 30 s)
GAPDH	NM_002046	F: 5′-AAGGTCGGAGTCAACGGATT-3′	(95°C, 30 s), (65°C, 30 s) and (72°C, 15 s)
		R: 5′-CTCCTGGAAGATGGTGATGG-3′	(95°C, 30 s), (65°C, 30 s) and (72°C, 15 s)
IL-1β	NM_000576.2	F: 5′- ATGATGGCTTATTACAGTGGCAA-3′	(95°C, 30 s), (60°C, 30 s) and (72°C, 30 s)
		R: 5′- GTCGGAGATTCGTAGCTGGA-3′	(95°C, 30 s), (60°C, 30 s) and (72°C, 30 s)

IL-1β: interleukin-1β; GAPDH: glyceraldehyde 3-phosphate dehydrogenase.

RNA isolation and analysis

RNA was isolated from a subset (n = 66) of subjects based on haplotype, HIV status and NO_x exposure. NO_x exposure was categorized into high NO_x (HN) and low NO_x (LN) groups using the median, that is, 36 μg/m³ as the split point. RNA was extracted from 34 HIV-1-positive and 32 HIV-1-negative subjects. RNA was extracted immediately after thawing from 500 μl of whole blood using the TRIzol method (Invitrogen, California, USA) as per manufacturer’s instructions. The total RNA concentration was quantified using a spectrophotometer and all RNA samples were DNase treated. One microgram of total RNA from each sample was reversed transcribed using the iScript cDNA synthesis kit (Bio-Rad, California, USA).

Real-time PCR RNA quantitation

Quantitative real-time PCR was performed for the IL-1β gene on RNA isolated from participants within high and low NO_x exposure groups with comparable haplotypes and HIV-1 infection status. The IL-1β gene was amplified using primers obtained from a public database, PrimerBank, to uniquely amplify the IL-1β gene (NM_000576.2 GenBank accession no.). The primers used were 5′-ATGATGGCTTATTACAGTGGCAA-3′ (forward) and 5′-GTCGGAGATTCGTAGCTGGA-3′ (reverse). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was validated as the most suitable reference gene, based on its PCR efficiency. The GAPDH (NM_002046) primers used were 5′ -AAGGTCGGAGTCAACGGATT-3′ (forward) and 5′-CTCCTGGAAGATGGTGATGG-3′ (reverse).

Each PCR reaction comprised of either 0.5 pmol/μl for IL-1β or 0.25 pmol/μl for GAPDH of each primer, 5 μl SYBR Green I Master Mix (2X) (Bio-Rad), 1 μg cDNA and water to 10 μl. Reactions were run in duplicate on a CFX real-time machine (Bio-Rad) with cycling condition as follows: 1 cycle at 95°C for 10 min, followed by 40 cycles at 95°C for 30 s, at 60°C for 30 s and at 72°C for 30 s. Relative mRNA levels were calculated by generating standard curves from serial dilutions of a known concentration of cDNA sample for IL-1β and GAPDH. Individual samples were then analysed against the standard curve and represented as levels of mRNA per microgram of cDNA. Results are depicted as a ratio of IL-1β to GAPDH per microgram of cDNA. To confirm amplification specificity, the PCR products were subjected to a melting curve analysis and agarose gel electrophoresis.

NO_x exposure data

An estimate of individual exposure levels for NO_x, a widely-utilized surrogate for other traffic-related pollutants including particles, was available. A NO_x passive sampling programme was used to characterize community exposure levels. NO_x monitoring locations were randomly selected for each of the study areas (Durban North and South). Monitoring efforts focused on characterizing the spatial variation in air pollution exposure given the surrounding environmental factors such as land use and topography. Data generated from this monitoring were used in developing land use regression models, which permitted the prediction of exposure at each of the participant’s homes, through geographical positioning data.

For the model, NO_x was measured at selected locations and regressed against site specific a priori defined (i.e. direction of effect) geographic predictors such as road length, land use types, topography and population and housing density in a multivariate regression model. The NO_x measurements were conducted using Ogawa samplers (Ogawa & Co., Inc., V3.98, USA), which were deployed at 40 randomly selected sites (Durban North and South) over a once repeated 2-week periods during midwinter and midsummer,^16,17 thus accounting for seasonal variation. The results from the two sampling periods were adjusted using data from an existing air quality monitoring station consistent with the sampling periods as per the European Study of Cohorts for Air Pollution Effects approach.¹⁸ The annual adjusted averages of NO_x were obtained by adjusting the arithmetic mean of the available measurements (i.e. both sampling periods) per site using the difference between the sampling period and the annual average at the air quality monitoring station. The adjusted NO_x measurements were then used in model development of which the regression coefficients were applied to participants.

Statistical analysis

All statistical analyses were performed using InStat GraphPad Prism version 5 and SPSS Statistics version 25 (SPSS Inc., Chicago, Illinois, USA). The IL-1β haplotypes were constructed using Haploview version 4.2 from IL-1β-1464, -511 and -31 SNPs. The high haplotype (GTC/CTT/CCT), the intermediate haplotype (CTC) and low haplotype (GCT/CCC) were defined as previously published.¹⁵ IL-1β mRNA expression values are expressed as medians. Differences between the medians of comparable haplotype LN and HN groups were determined using the Student’s t-test. High and intermediate haplotypes were combined and compared to the low haplotype for both HIV-1 positive and negative groups. Differences in expression between haplotypes were determined using an analysis of variance. Pearson’s correlation, χ ² or Fisher’s exact test was used to analyse genetic and clinical data. Scatter plots and bar graphs were generated using InStat GraphPad Prism version 5. p Values of <0.05 were considered as statistically significant. SPSS Statistics version 25 was used to conduct a multivariate linear regression analysis for gestational age on all participants for which NO_x data were available (n = 272) with adjusting the following variables chosen a priori: maternal Hb, BMI, NO_x exposure levels, IL-1β/GAPDH, HIV-1 status, education and income.

Results

Cohort characteristics

The multi-ethnic cohort was predominantly African (76%) and had a 34% HIV-1 prevalence (Table 2). The preterm birth (PTB) and low birth weight prevalence were 10% and 15%, respectively. The NO_x exposure range for the cohort was 12.8–56.9 µg/m³ with a median of 36.23 µg/m³. There was no significant difference in the IL-1β SNPs (-1464, -511 and -31) and haplotype distribution between the low and high NO_x exposure groups (p = 0.8, 0.3, 0.5, 0.1, respectively; Table 3). The predominant cohort haplotype is the high producer and the distribution for the African group was 40%, 35% and 25% for the high, intermediate and low haplotypes, respectively.

Table 2.

Maternal clinical, demographic and birth outcome data.^a

	Cohort	High NO_x exposure group (n = 138)	Low NO_x exposure group (n = 134)	p Value
Race (n) %
African	(257) 76	(94) 68	(117) 87	>0.05
Caucasian	(2) 0.6	–	–
Indian	(16) 5	(3) 2	(13) 10	>0.05
Coloured	(63) 19	(41) 30	(4) 3	0.0004
Total	(338) 100
Median maternal age (range)	25 (15–44)	25	25	0.5
Median BMI (kg/m²) trimester 3 (range)	30 (18.64–84)	31.59	28.72	0.01
Median Hb (g/dL) Trimester 1(range)	11.05 (5–16)	11	11.5	0.1
Median birth weight and range (g)	3105 (302–4440)	3200 (1530–4440)	3045 (2066–4260)	0.1
% low birth weight (<2500 g) (n)	15 (38)	17 (31)	9.5 (7)	0.7
Median gestational age and range (weeks)	39 (32–42)	39 (33–41)	39 (32–42)	0.01
% Preterm birth (n)	10 (32)	60 (28)	40 (19)	0.5
HIV positive % (n)	34 (111)	33 (44)	34 (45)	0.1

NO_x: oxides of nitrogen; BMI: body mass index; Hb: haemoglobin.

^aMaternal clinical, demographic and birth outcome data are shown with a comparison by NO_x exposure categories, high (>=36 μg/m²) and low (<36 μg/m²). A p value of <0.05 for the Student’s t-test or χ ² was considered as statistically significant.

Table 3.

IL-1β genotypic and haplotypic distribution in the cohort.^a

	Cohort	High NO_x exposure group	Low NO_x exposure group	p Value
Genotypes	(n) %
IL-1β–1464
GG		–	–
GC	268 (79)	29 (21)	25 (19)	0.8
CC	70 (21)	109 (79)	109 (81)	0.8
Total	338	138	134
IL-1β-511
CC	62 (18.3)	26 (19)	25 (19)	0.3
CT	146 (43.2)	65 (47)	50 (38)
TT	130 (38.5)	47 (34)	58 (43)
Total	338	138	133
IL-1β-31
TT	(41) 12	16 (12)	12 (10)	0.5
TC	(151) 45	62 (48)	52 (42)
CC	(143) 43	51 (40)	60 (48)
Total	335	129	124
Haplotypes
Low	(89) 27	36 (26)	32 (24)	p = 0.1
Intermediate	(107) 32	50 (36)	65 (49)
High	(139) 41	52 (38)	35 (27)
Total	335	138	132

IL-1β: interleukin-1β; NO_x: oxides of nitrogen.

^aMaternal IL-1β genotypic and haplotypic distribution is shown with a comparison by NO_x exposure categories, high (>=36 μg/m²) and low (<36 μg/m²). A p value of <0.05 for the Student’s t-test or χ ² was considered as statistically significant.

IL-1β genotypes associate with mRNA expression levels

IL-1β gene expression was associated with the three major functional IL-1β promoter SNPs (IL-1β-1464, -511 and -31) and the IL-1β haplotypes. High levels of IL-1β expression were significantly associated with IL-1-31 T/T, IL-1-511 C/C and IL-1-1464 G/C (p = 0.0381, 0.038 and 0.046, respectively; Figure 1(a) to (c)). High levels of IL-1β mRNA were associated with GTC when compared to other haplotypes (p = 0.023; Figure 1(d)). Haplotypes CCC, CCT and CTT (Figure 1(e)) were grouped according to expression. Haplotype CCC showed similar expression patterns as the low haplotype, while CCT and CTT had similar expression patterns as the high haplotype. Expression in the high haplotype (GTC/CTT/CCT) was significantly higher than other haplotypes, that is, intermediate (CTC) (p = 0.0017) and low (GCT/CCC; p = 0.0001), and the intermediate haplotype expression was significantly higher than that of the low haplotype (p = 0.0025).

Figure 1.

IL-1β genotypes associate with expression levels. Gene expression analysis was performed on 66 participants for IL-1β-1464, -511 and -31 genotypes and IL-1β haplotypes. Data are depicted as a normalized ratio of IL-1β versus GAPDH (Figure 1(a) to (f)). The horizontal line represents the median. Differences between groups were determined using an unpaired Student’s t-test. A p value of <0.05 was considered as statistically significant. IL-1β: interleukin-1β; GAPDH: glyceraldehyde 3-phosphate dehydrogenase.

Effect of NO_x exposure level on IL-1β expression in the cohort

IL-1β mRNA levels were positively correlated with NO_x levels in the cohort (r = 0.34; p = 0.006; Figure 2). A significant positive correlation was shown independently in both HIV-1 positive and negative groups (p < 0.05; data not shown).

Figure 2.

NO_x exposure level versus IL-1β expression. We assessed whether there was an association between pro-inflammatory cytokine IL-1β mRNA expression and NO_x exposure level in the cohort using the Pearson’s correlation. A p value of <0.05 was considered as significant. IL-1β: interleukin-1β; NO_x: oxides of nitrogen.

Both high and intermediate haplotypes were significantly higher in expression than the low haplotype (Figure 3). In both the HIV-1 positive and negative groups, IL-1β expression in participants with the low haplotype was significantly elevated in the HN exposure group compared to the LN exposure group (p < 0.05; Figure 3). In the HIV-1 negative group, expression in HN was elevated in the high haplotype compared to the low haplotype (p = 0.03; Figure 3).

Figure 3.

IL-1β expression in different NO_x exposure groups. Gene expression analysis was performed on 66 participants by IL-1β haplotype: (high (CCT/CTT/CTC) versus low (CCC/GCT)), HIV-1 status and NO_x exposure category: high (>=36 μg/m²) and low (<36 μg/m²). For this analysis, the IL-1β high and intermediate haplotypes were combined to increase the numbers. Data are depicted as a normalized ratio of IL-1β versus GAPDH (Figure 1(a) to (f)). The horizontal line represents the median. The differences between groups were determined using an unpaired Student’s t-test. A p value of <0.05 was considered as statistically significant. IL-1β: interleukin-1β; NO_x: oxides of nitrogen; GAPDH: glyceraldehyde 3-phosphate dehydrogenase.

Multivariate linear regression model for gestational age

A multivariate linear regression analysis for gestational age was conducted on 272 samples adjusting for maternal Hb (trimester 1), BMI (trimester 3), NO_x exposure levels, IL-1β mRNA expression levels, HIV status, income level and education, with separate models for each haplotype: low (0), intermediate (1) and high (2) (Table 4). In the first model, that is, without stratifying by haplotype and the low haplotype model (haplotype 0), no significant associations were observed. In the high haplotype (haplotype 2), NO_x exposure level, IL-1β mRNA expression level and HIV-1 positivity were negatively associated with gestational age (p < 0.05 for all). In this model (haplotype 2), for each unit increase in NO_x level, a 1-week decrease in gestational age was observed (p = 0.02), while for every unit increase in IL-1β mRNA expression level, a 0.9-week decrease in gestational age was observed (p = 0.025); HIV-1 positivity was associated with a 0.4-week decrease in gestational age (p = 0.044). In the intermediate haplotype model (haplotype 1), HIV-1 positivity was associated with a 0.2-week decrease in gestational age (p = 0.035).

Table 4.

Linear regression model for gestational age.

Dependent variable: gestational age (weeks)	All^a standardized coefficients beta std. Error sig.	Haplotype 0^b standardized coefficients beta std. Error sig.	Haplotype 1^c standardized coefficients beta std. Error sig.	Haplotype 2^d standardized coefficients beta std. Error sig.
HIV-1	0.231 0.410 0.102	–0.481 0.544 0.560	–0.235 0.717 0.035	–0.413 0.714 0.044
Hb (g/dl)	0.161 0.054 0.247	0.381 0.048 0.089	0.450 0.143 0.072	0.669 0.147 0.055
BMI trimester 3 (kg/m²)	0.159 0.016 0.274	0.052 0.030 0.810	0.501 0.022 0.059	0.717 0.029 0.068
NO_x (μg/m³)	–1.093 0.020 0.070	–0.532 0.023 0.280	–0.137 0.008 0.065	–1.024 0.077 0.020
IL-1β/GAPDH	–0.264 0.136 0.083	0.255 0.177 0.245	–0.518 0.264 0.075	–0.949 0.399 0.025
Education level	0.092 0.186 0.511	0.609 0.291 0.064	0.448 0.348 0.146	0.016 0.430 0.949
Income level	0.169 0.180 0.258	0.519 0.226 0.061	–0.643 0.418 0.056	0.531 0.371 0.095

^aFull dataset; ^bHaplotype 0: low; ^cHaplotype 1: intermediate; ^dHaplotype 2: high; NO_x: oxides of nitrogen; IL-1β: interleukin-1β; Hb: haemoglobin; BMI: body mass index; GAPDH: glyceraldehyde 3-phosphate dehydrogenase.

Discussion

In this study of a South African birth cohort, a significant association between NO_x and gestational age was observed, with a strong suggestion of effect modification by the IL-1β haplotype. These findings are evident when adjusting for maternal IL-1β mRNA expression, HIV status, BMI, Hb level, income and education.

We observed a positive association between IL-1β expression and NO_x exposure, and in a multivariate analysis observed further IL-1β haplotype-specific effects of NO_x and IL-1β mRNA expression on gestational age, with an almost twofold decrease in gestational age with the high haplotype when compared to the other haplotypes.

Several epidemiological studies in predominantly Caucasian or Asian cohorts have linked air pollution with adverse birth and health outcomes^2,19,20, including an association between maternal NO_x and NO₂ exposure with either shortened gestational age or PTB.^21

–24 Other studies have shown increased risk for PTB by proximity to traffic or major motorways.^25
–27 Our finding of a negative association between NO_x exposure and gestational age is consistent with other studies. The pathogenetic mechanisms, however, are not completely understood.

Data on the maternal IL-1β response to NO_x exposure are scarce or non-existent. In this study, the observed positive association between NO_x exposure levels and maternal IL-1β expression may indicate an inductive effect of NO_x on oxidative stress,²⁸ which leads to an elevated pro-inflammatory IL-1β response. DNA and cellular component damage²⁹ and tissue damage in target organs, including the placenta, may ensue.³⁰ An excessive pro-inflammatory response may induce further free radical activity, leading to further tissue destruction. In the multivariate models, we observed that NO_x effects on gestational age are haplotype-specific for the high-producer IL-1β haplotype in which a unit increase in NO_x exposure level is associated with a 1-week shortened gestational age. This may suggest that elevated maternal IL-1β plays a role in mechanisms by which NO_x affects gestational age.

During labour and PTB, the level of IL-1β mRNA in peripheral blood was shown to be significantly elevated in comparison to pregnant women not in labour³¹ and in spontaneous PTB.³² Exacerbation of pro-inflammatory responses due to cytokine gene variants and gene-environment interactions at varying stages may threaten the Th2 dominant maternal tolerance of the semiallogeneic foetal-placental unit and increase the risk for decreased gestational age or PTB.¹² Elevated IL-1β contributes to induction of labour by stimulating the expression and production of uterine activation proteins and prostaglandin synthesis, and decreasing tissue progesterone responsiveness.³³ Significant down-regulation of 11β-HSD2 mRNA expression and activity by IL-1β increases the amount of cortisol entering the foetal circulation³⁴ and increases intrauterine cortisol concentrations which drive prostaglandin synthesis potentially contributing to PTB. We observed in the multivariate models that a unit increase in IL-1β mRNA expression level associates with a decrease in gestational age by almost 1 week in the high haplotype only (Table 4), highlighting a vulnerable group to shortened gestational age by elevated IL-1β gene expression.

In a predominantly Black African birth cohort, we report the maternal IL-1β haplotype (-1464, -511, -31) distribution and associated expression, which has been described in African Americans, Caucasians and Koreans^14,15 but not in an African cohort. Similar to the reported distribution in African Americans, the highest transcriptional activity haplotype is most prevalent in Africans at a comparable 40% versus 46.1% in African Americans, while haplotype-associated expression matches previously published data (Figure 1).^14,15,35 The PTB and low birth weight prevalence matches the estimated global prevalence of 9.6% and 16%, respectively.³⁶

We report a positive correlation between NO_x exposure level and IL-1β mRNA expression levels in the cohort. Further study is required to determine the association between elevated IL-1β in response to NO_x and the occurrence and exacerbation of asthma and respiratory morbidity. In immunostimulated pregnant dams, a transient increase in IL-1 among other cytokines in serum and amniotic fluid was shown, and their offspring displayed accelerated development and heightened responsiveness of Th1, Th17 and cytotoxic effector T-cell subsets.^37
–39 Morelli et al. (2015) thus hypothesized that elevated maternal IL-1 among other pro-inflammatory cytokines may influence programming of the foetal immune system towards a pro-inflammatory phenotype with exacerbated responses to immune stimuli that may persist into adulthood.⁴⁰ Interestingly, in a Chinese cohort comprising 2598 children, maternal exposure to NO₂ was significantly associated with increased risk for childhood asthma, allergic rhinitis and eczema.⁴¹

Maternal HIV-1 infection is associated with increased risk for adverse birth outcomes, including lower gestational age, PTB and intrauterine growth retardation; however, the mechanisms underlying these associations are not completely understood.^42
–44 Indeed, in vitro studies show the inductive effect of IL-1β on HIV expression in the latently infected human promonocytic cell line U1.^45,46 In this study, we observe an IL-1β haplotype-specific negative association between HIV-1 positivity and gestational age for the intermediate and high-producer haplotypes with an almost twofold greater effect in the high haplotype, while no association is shown in the low-producer haplotype. We postulate that higher IL-1β expression associated with the high and intermediate haplotypes in comparison to the low haplotype may potentially play a role in mediating or synergizing the association between HIV-1 positivity and lower gestational age outcomes, via a pro-inflammatory state which alters maternal tolerance of the foetal-placental unit and potentially via facilitation of HIV-1 replication. We thus identify vulnerable subgroups of the HIV-1 population to lower gestational age by the IL-1β haplotype. We speculate that significantly elevated IL-1β in response to elevated NO_x exposure may potentially further increase susceptibility of HIV-1-positive individuals to lower gestational age.

Inter-individual variability in HIV-1 infection outcome is an indicator of the role of host genetic and environmental factors in influencing disease progression. The reported positive association between IL-1β expression and elevated NO_x exposure is shown independently in the HIV-1-positive group and (data not shown) is a cause for concern. We speculate that chronic NO_x-induced increase in IL-1β expression may potentially influence HIV-1 infection outcome, disease progression, pathogenesis and transmissibility. It has been shown that increased plasma IL-1β is associated with disease progression in both children and adults,⁴⁷ while in vitro studies show the inductive effect of IL-1β on HIV expression in the latently infected human promonocytic cell line U1.^45,46 IL-1β also plays a key role in pyroptosis, a major mechanism of CD4 T-cell depletion⁴⁸ and IL-1β production is lower in T-cell cultures from women with controlled viral load,^19,49 essential to prevent mother-to-child transmission. However, further work is needed to address whether NO_x exposure levels may impact HIV-1 progression and pathogenesis in a larger characterized cohort matched for IL-1β haplotype and stage of infection.

Conclusion

In conclusion, we report the maternal IL-1β (-1464, -511, -31) haplotype distribution in a predominantly African pregnant cohort and show maternal IL-1β haplotype-specific effects of NO_x levels, IL-1β expression and HIV-1 positivity on gestational age in a birth cohort, which has not been previously reported. These data have implications for better understanding the effect of prenatal NO_x exposure on gestational age outcomes and demonstrate the role of the IL-1β haplotype in modulating the effects of NO_x exposure therein. The study identifies vulnerable subgroups to shortened gestational age, due to IL-1β haplotype and NO_x interactions. The study identifies vulnerable subgroups among HIV-1 infected mothers to lower gestational age outcomes by IL-lβ haplotype and contributes to the understanding of the association between HIV-1 infection and gestational age, which is currently limited. The study provides a basis for further work to explore the association between maternal IL-1β haplotypes and morbid outcomes among children of the cohort.

Footnotes

Acknowledgement

The authors wish to thank the Mother and Child in the Environment cohort study participants and nursing staff at the participating clinics and hospitals.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the National Research Foundation of South Africa (grant number: 90550), the Medical Research Council of South Africa (SIR Grant) and the AstraZeneca Research Trust.

References

Ciliberto

Marx

. Physiological changes associated with pregnancy. Update Anaesth 1998; 9:1–6.

Brunekreef

. Health effects of air pollution observed in cohort studies in Europe. J Expos Sci Environ Epidemiol 2007; 17: S61–S65.

Wilcox

Skjaerven

. Birth weight and perinatal mortality: the effect of gestational age. Am J Public Health 1992; 82: 378–382.

Nakai

Nitta

Maeda

. Respiratory health associated with exposure to automobile exhaust. II. Personal NO2 exposure levels according to distance from the roadside. J Expo Anal Environ Epidemiol 1995; 5: 125–136.

Kodama

Arashidani

Tokui

. Environmental NO2 concentration and exposure in daily life along main roads in Tokyo. Environ Res 2002; 89: 236–244.

Naidoo

Gqaleni

Batterman

. South Durban health study, final project report, multipoint plan: project 4, health study and health risk assessment. Department of Environmental Affairs and Tourism. 2007, p. 1–269.

Matooane

Diab

. Air pollution carrying capacity in the South Durban Industrial Basin. S Afr J Sci 2001; 97: 450–453.

Naidoo

Robins

Batterman

. Ambient pollution and respiratory outcomes among schoolchildren in Durban, South Africa. S Afr J Child Health 2013; 7: 127–134.

Kistnasamy

Robins

Naidoo

. The relationship between asthma and ambient air pollutants among primary school students in Durban, South Africa. Int J Environ Health 2008; 2: 365–385.

10.

Genestra

. Oxyl radicals, redox-sensitive signalling cascades and antioxidants. Cell Signal 2007; 19: 1807–1819.

11.

Kelly

. Oxidative stress: its role in air pollution and adverse health effects. Occup Environ Med 2003; 60: 612–616.

12.

Mor

Cardenas

Abrahams

. Inflammation and pregnancy: the role of the immune system at the implantation site. Ann N Y Acad Sci 2011; 1221: 80–87.

13.

Choy

EHS

Panayi

. Cytokine pathways and joint inflammation in rheumatoid arthritis. N Engl J Med 2001; 344: 907–916.

14.

Chen

Wilkins

Aziz

. Single nucleotide polymorphisms in the human interleukin-1B gene affect transcription according to haplotype context. Hum Mol Genet 2006; 15: 519–529.

15.

Lee

. Functional haplotype frequencies of the interleukin-1b promoter in the Korean population. Genomics Inform 2008; 6: 29–31.

16.

Tyson

Preston-Whyte

. The weather and climate of southern Africa. Oxford and New York: Oxford University Press, 2004.

17.

Tularam

Ramsay

. Synoptic influences on air pollution events in the Durban South Basin, 2006 to 2010. Durban, KwaZulu Natal: University of KwaZulu Natal, 2013.

18.

Beelen

Hoek

. ESCAPE exposure assessment manual. 2010. Available at: http://www.escapeproject.eu/manuals/ESCAPE_Exposure-manualv9.pdf (updated July 2010).

19.

Sram

Binkova

Dejmek

. Ambient air pollution and pregnancy outcomes: a review of the literature. Environ Health Perspect 2005; 113: 375–382.

20.

Kampa

Castanas

. Human health effects of air pollution. Environ Pollut 2008; 151: 362–367.

21.

Ren

Delfino

. Association between local traffic-generated air pollution and preeclampsia and preterm delivery in the south coast air basin of California. Environ Health Perspect 2009; 117: 1773–1779.

22.

Llop

Ballester

Estarlich

. Preterm birth and exposure to air pollutants during pregnancy. Environ Res 2010; 110: 778–785.

23.

Olsson

Ekström

Forsberg

. Temporal variation in air pollution concentrations and preterm birth—a population based epidemiological study. Int J Environ Res Public Health 2012; 9: 272–285.

24.

Olsson

Mogren

Eneroth

. Traffic pollution at the home address and pregnancy outcomes in Stockholm, Sweden. BMJ Open 2015; 5: 5.

25.

Wilhelm

Ritz

. Residential proximity to traffic and adverse birth outcomes in Los Angeles county, California, 1994-1996. Environ Health Perspect 2003; 111: 207–216.

26.

Yang

Chang

Chuang

. Evidence for increased risks of preterm delivery in a population residing near a freeway in Taiwan. Arch Environ Health 2003; 58: 649–654.

27.

Brauer

Lencar

Tamburic

. A cohort study of traffic-related air pollution impacts on birth outcomes. Environ Health Perspect 2008; 116: 680–686.

28.

Mirowsky

Dailey

Devlin

. Differential expression of pro-inflammatory and oxidative stress mediators induced by nitrogen dioxide and ozone in primary human bronchial epithelial cells. Inhal Toxicol 2016; 28: 374–382.

29.

Closa

Folch-Puy

. Oxygen free radicals and the systemic inflammatory response. IUBMB Life 2004; 56: 185–191.

30.

Halliwell

Louie

. Interaction of nitrogen dioxide with human plasma. Antioxidant depletion and oxidative damage. FEBS Lett 1992; 313: 62–66.

31.

Yuan

Jordan

McInnes

. Leukocytes are primed in peripheral blood for activation during term and preterm labour. Mol Hum Reprod 2009; 15: 713–724.

32.

Heng

Pennell

Chua

. Whole blood gene expression profile associated with spontaneous preterm birth in women with threatened preterm labor. PLoS One 2014; 9: e96901.

33.

Christiaens

Zaragoza

Guilbert

. Inflammatory processes in preterm and term parturition. J Reprod Immunol 2008; 79: 50–57.

34.

Chisaka

Johnstone

Premyslova

. Effect of pro-inflammatory cytokines on expression and activity of 11beta-hydroxysteroid dehydrogenase type 2 in cultured human term placental trophoblast and human choriocarcinoma JEG-3 cells. J Soc Gynecol Invest 2005; 12: 303–309.

35.

Lee

Kim

Yook

. TNF/LTA polymorphisms and risk for gastric cancer/duodenal ulcer in the Korean population. Cytokine 2004; 28: 75–82.

36.

Beck

Wojdyla

Say

. The worldwide incidence of preterm birth: a systematic review of maternal mortality and morbidity. Bull World Health Organ 2010; 88: 31–38.

37.

Mandal

Marzouk

Donnelly

. Preferential development of Th17 cells in offspring of immunostimulated pregnant mice. J Reprod Immunol 2010; 87: 97–100.

38.

Mandal

Marzouk

Donnelly

. Maternal immune stimulation during pregnancy affects adaptive immunity in offspring to promote development of TH17 cells. Brain Behav Immun 2011; 25: 863–871.

39.

Mandal

Donnelly

Elkabes

. Maternal immune stimulation during pregnancy shapes the immunological phenotype of offspring. Brain Behav Immun 2013; 33: 33–45.

40.

Morelli

Mandal

Goldsmith

. The maternal immune system during pregnancy and its influence on fetal development. Dovepress 2015; 6: 171–189.

41.

Deng

. Early life exposure to traffic-related air pollution and allergic rhinitis in preschool children. Respir Med 2016; 121: 67–73.

42.

Larsen

Nordly

Rosenfeldt

. Morbidity among HIV-exposed uninfected children compared to children not exposed to HIV – a Danish nationwide study. In: 4th international workshop on HIV & women, Washington, DC, 13–14 January 2014.

43.

Wedi

Kirtley

Hopewell

. Perinatal outcomes associated with maternal HIV infection: a systematic review and meta-analysis. Lancet HIV 2016; 3: e33–e48.

44.

dos Reis

HLB

Araujo

KDS

Ribeiro

. Preterm birth and fetal growth restriction in HIV-infected Brazilian pregnant women. Rev Inst Medicina Trop São Paulo 2015; 57: 111–120.

45.

Poli

Kinter

Fauci

. Interleukin 1 induces expression of the human immunodeficiency virus alone and in synergy with interleukin 6 in chronically infected U1 cells: inhibition of inductive effects by the interleukin 1 receptor antagonist. Proc Natl Acad Sci U S A 1994; 91: 108–112.

46.

Granowitz

Saget

Wang

. Interleukin 1 induces HIV-1 expression in chronically infected U1 cells: blockade by interleukin 1 receptor antagonist and tumor necrosis factor binding protein type 1. Mol Med (Cambridge, Mass) 1995; 1: 667–677.

47.

Ketlinskii

Kalinina

Tsvetkova

. Tumor necrosis factor-alpha and interleukin-1beta in the blood plasma of patients with HIV infection. Vestn Ross Akad Med Nauk 1992; (9–10): 36–41.

48.

Doitsh

Galloway

Geng

. Cell death by pyroptosis drives CD4 T-cell depletion in HIV-1 infection. Nature 2014; 505: 509–514.

49.

Bento

Hygino

Andrade

. IL-10-secreting T cells from HIV-infected pregnant women downregulate HIV-1 replication: effect enhanced by antiretroviral treatment. AIDS (London, England) 2009; 23: 9–18.

IL-1β haplotype influences the effect of NO x exposure on gestational age in the South African MACE birth cohort

Abstract

Objective:

Material and methods:

Results:

Conclusion:

Keywords

Introduction

Methods

Subjects

Genotyping

RNA isolation and analysis

Real-time PCR RNA quantitation

NOx exposure data

Statistical analysis

Results

Cohort characteristics

IL-1β genotypes associate with mRNA expression levels

Effect of NOx exposure level on IL-1β expression in the cohort

Multivariate linear regression model for gestational age

Discussion

Conclusion

Footnotes

Acknowledgement

Declaration of Conflicting Interests

Funding

References

NO_x exposure data

Effect of NO_x exposure level on IL-1β expression in the cohort