Apolipoprotein-E Gene Polymorphism and Lipid Composition among IUGR and AGA Neonates

Abstract

Objective  The objective of this study is to study the profile of apolipoprotein E ( APOE ) gene polymorphism and lipid profile among intrauterine growth restriction (IUGR) and appropriate for gestational age (AGA) neonates. This is an observational study. This study was done at the neonatal unit of a teaching hospital in South India. All consecutively born IUGR neonates (cases) of more than 32 weeks' gestational age and AGA neonates (controls) were enrolled for the study. Genomic DNA extraction was done from a total of 102 peripheral venous blood samples. Genotyping of the APOE rs429358 and rs7412 defining the ε2, ε3, and ε4 alleles was done by polymerase chain reaction–restriction fragment length polymorphism method. Prefeed venous blood was collected and analyzed for lipid profile estimation. The allelic frequencies of cases versus control were ε2—9 (8.7%) versus 3 (2.9%); ε3—88 (84.6%) versus 81 (79.4%); and ε4–7 (6.7%) versus 18 (17.6%). The frequency of ε4 isoform allele, associated with adult onset of metabolic diseases was less among the IUGR group. The mean total cholesterol (TC), Low-Density Lipoprotein (LDL), High-Density Lipoprotein, and triglyceride (TG) were 107.59 ± 35.99, 51.69 ± 24.68, 21.75 ± 9.58, and 151.22 ± 61.84 mg/dL, respectively, in the IUGR group. The mean TC and LDL levels in IUGR group were marginally higher than AGA neonates (107 ± 35.99 vs. 100.37 ± 22.69 mg/dL and 51.69 ± 24.68 versus 46.9 ± 19.51 mg/dL, p  > 0.05). In both groups, the mean TC and TGL levels were elevated in the ε4 isoform subgroup ( p  > 0.05). In our study, the ε2 allele was the second most predominant APOE isoform and the ε4 allele of the APOE gene associated with adult-onset diseases was not increased among IUGR neonates. Neonates with ε4 allele showed an abnormal lipid profile in both study groups suggesting a possible association.

Introduction

The term intrauterine growth restriction (IUGR) has been defined as the subnormal rate of fetal growth with respect to the growth potentials of a specific fetus in accordance to the gender and race. These neonates are born with clinical features of in utero malnutrition and may or may not have a birth weight less than 10th centile for the particular gestational age. ¹ Approximately 78% of all IUGR infants delivered globally are seen in Asian continent with highest incidence seen in India. ² The IUGR infants are prone for short-term and long-term complications making them high-risk neonates. Few of the short complications include perinatal asphyxia, persistent pulmonary hypertension, hypothermia, hypoglycemia, necrotizing enterocolitis, and so on. ³ The in utero insult to the developing fetus is postulated to have an association with adult diseases, termed as Developmental Origin of Health and Disease (DOHaD). The hypothesis of “thrifty phenotype” by Neel considers the nutritional mismatch in utero and postnatal period as the underlying cause for DOHaD. ⁴

IUGR-induced metabolic syndrome (dyslipidemia and insulin resistance) indirectly leads to atherosclerosis, but studies has shown in IUGR animal models that primary vascular remodeling could happen due to in utero fetal perfusion abnormalities. ^{5 6}

Apolipoprotein E (APOE) is a receptor-binding ligand glycoprotein consisting of 299 amino acids which enhances the clearance of very low-density lipoprotein (LDL) remnants and chylomicrons from plasma and thus mediates the metabolism of triglyceride (TG) and cholesterol. ^{7 8} The APOE gene has the common alleles, designated as ε2, ε3, and ε4, and these alleles form six different genotypes: three homozygous (ε2/ε2, ε3/ε3, and ε4/ε4) and three heterozygous (ε2/ε3, ε3/ε4, and ε2/ε4). APOE gene polymorphism has shown significant modulatory effects on plasma lipid parameters and thus influencing coronary artery disease (CAD) and cerebrovascular diseases. ⁹ Various studies in the past have shown that through platelet aggregation, allele-specific antioxidant and immune activity, and cholesterol effect in the macrophages, the APOE genes might influence on the development of CAD. ^{10 11 12} Dallongeville et al in their meta-analysis reported that subjects with ε2 and ε4 alleles had lower and higher plasma cholesterol values than subjects with ε3/ε3 genotypes, suggesting that APOE ε4 allele may increase the risk of CAD. ¹³

The pattern of distribution of APOE genotypes and its alleles in IUGR neonates is less studied and there exists a paucity of data in Indian context. In the present study, an attempt was made to study the profile of APOE gene polymorphism among IUGR and appropriate for gestational age (AGA) neonates and their association with the lipid composition.

Materials and Methods

Study Subjects

This cross-sectional study was conducted over a period of 2 years (June 2015–May 2017) in the Department of Pediatrics, PSG Institute of Medical Sciences & Research, a tertiary care teaching hospital in South India. The Institutional Human Ethics Committee approved the study (IHEC approval No. 14/328) and all subjects were recruited after obtaining informed consent from one of the parents. A convenient sample of 50 in each group (i.e., IUGR group and AGA control group) was enrolled for the study after obtaining informed consent. All consecutively born IUGR neonates of more than 32 weeks' gestational age were enrolled for the study. A neonate is defined to be IUGR if the ponderal index is less than centile for the gestational age and any two of the following antenatal ultrasonographic evidence of IUGR such as fetal abdominal circumference < 10th centile for the corresponding gestational age or no increase in fetal abdominal circumference noted in two consecutive antenatal scans at least 2 weeks apart or increase in Doppler umbilical artery indices such as absent/reversed end diastolic flow or if beyond 34 weeks, decrease in middle cerebral artery Doppler index with oligohydramnios, that is, amniotic fluid Index < 5. Neonates with birth weight AGA born on the same day as the IUGR newborns were recruited as controls. Infants born with major congenital malformations, clinical phenotype fitting into a specific chromosomal disorder, infants with cholestatic jaundice, and those who did not consent were excluded from the study. Clinical details on maternal age, parity, family history of CAD, cerebrovascular disease, diabetes mellitus in adults younger than 45 years, and gestational medical disorders data were recorded on specified data collection form. After delivery, the neonates were weighed with an electronic weighing scale, and length was measured using an infantometer.

APOE Genotyping

Venous blood samples in EDTA containing 10-mL Vacutainer tubes (BD Biosciences, United States) were collected, aliquoted, and stored at −80°C. Genomic DNA was extracted from 0.2 mL of each blood sample using High Pure PCR Template Preparation Kit (Roche, United States) and stored at −20°C. Genotyping of the APOE rs429358 and rs7412 defining the ε2, ε3, and ε4 alleles was done by polymerase chain reaction (PCR)–restriction fragment length polymorphism method. PCR amplifying the fragment of APOE gene was optimized in a 20-µL reaction containing 100 ng of template DNA, 0.5 µM of APOE-forward and APOE-reverse primers (APOEF: 5′TCCAAGGAGCTGCAGGCGGCGCA3′; APOER: 5′GCCCCGGCCTGGTACACTGCCA3′), 1X Taq DNA Polymerase Master Mix RED (Ampliqon, Germany) and nuclease-free water was performed. The cycling parameters included 94°C for 2 minutes, 35 cycles of 94°C for 30 seconds, 68°C for 45 seconds, 72°C for 45 seconds, and a final extension at 72°C for 5 minutes to yield a 218-bp product. The PCR product was individually digested with Afl III and Hae II enzymes (New England Biolabs, United States). The Afl III digestion mixture contained 2 µL PCR products and 5U of Afl III in the buffer supplied by the manufacturer (CutSmart buffer). The Hae II digestion mixture contained 2 µL PCR products mixed with 10U Hae II in the buffer supplied by the manufacturer (CutSmart buffer). The two reactions were allowed to proceed for 3 hours at 37°C. The digested product was run on 15% polyacrylamide gel electrophoresis. Table 1 indicates the digestion pattern for the respective allele.

Table 1. Baseline characteristics of the study population.

Parameters	IUGR group ( n  = 51) (Mean ± SD)	AGA control group ( n  = 51) (Mean ± SD)	p -Value
Gestational age (wk)	35.67 ± 1.883	38.04 ± 1.371	<0.001
Birth weight (g)	1,852.75 ± 415.06	2,979.80 ± 363.59	<0.001
Length (cm)	42.78 ± 3.38	49.69 ± 2.29	<0.001
Head circumference (cm)	30.56 ± 1.76	34.47 ± 3.10	<0.001
Ponderal index	2.37 ± 0.43	2.43 ± 0.20	0.061
Maternal age (y)	24.75 ± 4.92	23.92 ± 3.14	0.381
Gender (male/female)	29/22	27/24	0.932
No. of primigravida mothers, n (%)	29 (56.8)	18 (35.2)	0.093

Abbreviations: AGA, appropriate for gestational age; IUGR, intrauterine growth restriction; SD, standard deviation.

Lipid Profile Analysis

For estimation of serum lipids, 2 mL of (prefeed) venous blood was collected from neonates of both the groups once they have reached full feeds, that is, feed volume more than 100 mL/kg/d and off total parenteral nutrition for at least 5 days for IUGR group and on Days 3 to 4 in AGA controls. Plasma total cholesterol (TC), TG, LDL, and high-density lipoprotein (HDL) were estimated using automated clinical chemistry analyzer and enzyme-based colorimetric lists supplied by Roche Diagnostics, Germany.

Statistical Analysis

All statistical analyses were performed with IBM SPSS v24 (IBM Corporation; Armonk, New York, United States). All data were expressed as mean ± standard deviation (SD). Kolmogorov–Smirnov's test was used for testing normality of the data. Normally distributed continuous variables were compared using t -test. To evaluate the mean difference between three groups, one-way analysis of variance (ANOVA) test was applied. Categorical data were evaluated by chi-square test. Allele and genotype difference between groups and deviations from Hardy–Weinberg equilibrium were tested by chi-square tests.

Results

A total of 102 neonates were included in the study, of which 51 were IUGR cases and 51 were AGA controls. The baseline characteristics of the study population are summarized in Table 1 . The mean ± SD of the IUGR group was 2.37 ± 0.43 and the maternal age, parity, and gender were similar in two groups.

The lipid profile of IUGR cases and controls are presented in Table 2 . The mean ± SD of TC and LDL among cases (107.59 ± 35.99 and 51.69 ± 24.67 mg/dL) was slightly elevated than the controls (100.37 ± 22.27 and 46.90 ± 19.512 mg/dL) ( p -value = 0.226). There was no difference in the HDL levels between the groups (21.75 ± 9.57 versus 21.96 ± 15.15 mg/dL) ( p -value = 0.932) and serum TGL levels were slightly high in the AGA control group (158.65 ± 38.34 vs. 151.22 ± 61.84). Overall, there was no statistically significant difference in the lipid profile parameters among cases and controls.

Table 2. Lipid composition of IUGR cases and AGA controls.

Parameters	IUGR group ( n  = 51) (Mean ± SD)	AGA control group ( n  = 51) (Mean ± SD)	p -Value
Total cholesterol (mg/dL)	107.59 ± 35.99	100.37 ± 22.69	0.226
Triglyceride (mg/dL)	151.22 ± 61.84	158.65 ± 38.35	0.468
HDL (mg/dL)	21.75 ± 9.58	21.96 ± 15.15	0.932
LDL (mg/dL)	51.69 ± 24.68	46.90 ± 19.51	0.280

Abbreviations: AGA, appropriate for gestational age; HDL, high-density lipoprotein; IUGR, intrauterine growth restriction; LDL, low-density lipoprotein; SD, standard deviation.

The frequency of APOE genotypes and alleles in IUGR and AGA neonates are summarized in Tables 3 and 4 . In our study, the APOE genotypes showed variations in both groups and their genotypic variations were in Hardy–Weinberg equilibrium. The ε3/ε3 genotype was more common in both cases and controls (76.4 and 64.7%). The homozygous ε4/ε4 and ε2/ε2 was not seen in the IUGR cases group and AGA controls group, respectively. The homozygous ε4/ε4 and heterozygous ε3/ε4 genotypes were found in increased frequency among the controls than cases ( p  < 0.05).

Table 3. Apolipoprotein E genotypic frequencies in IUGR cases and AGA controls.

Genotype	IUGR group		AGA control group
	N	Frequency%	N	Frequency%
ε2/ε2	2	3.9	13	25.5
ε3/ε2	3	5.9	2	3.9
ε3/ε3	39	76.4	33	64.7
ε4/ ε3	6	11.8	13	25.5
ε4/ε4	–	–	2	3.9
ε4/ ε2	1	1.9	–	–

Abbreviations: AGA, appropriate for gestational age; IUGR, intrauterine growth restriction.

Table 4. Lipid compositions allelic frequency in IUGR cases based on allele groups.

Carriers of APOE allele ( n )	Allelic frequency (%)	TC (mg/dL)	TG (mg/dL)	HDL (mg/dL)	LDL (mg/dL)
ε2 (6)	8 (7.8)	115.40 ± 38.79	145.60 ± 56.81	30.80 ± 9.36	64.60 ± 16.95 a
ε3 (39)	87 (85.3)	102.82 ± 30.68	147.79 ± 55.45	21.18 ± 8.97	49.51 ± 25.15
ε4 (6)	7 (6.7)	123 ± 59.03 a	161.67 ± 100.46 a	18.67 ± 11.62 b	50.83 ± 26.86
p -Value c		0.370	0.870	0.074	0.445

Abbreviations: HDL, high-density lipoprotein; IUGR, intrauterine growth restriction; LDL, low-density lipoprotein; TC, total cholesterol; TG, triglyceride.

^aHigher than other groups.

^bLower than other groups.

^cOne-way analysis of variance.

The frequency of ε3 allele was predominant in both cases and controls (85.3 and 79.4%). The ε4 allele frequency was significantly lesser among the cases (6.7 vs. 18.6%), odds ratio (OR) (95% confidence interval [CI]) = 0.343 (0.137–0.858), p < 0.05. On the contrary, ε2 allele frequency was higher among cases (7.8 vs. 1.96%), OR (95% CI) = 3.724 (0.768–18.058), p  = 0.107.

The lipid composition of IUGR cases with respect to APOE alleles is presented in Table 4 . IUGR neonates in the APOE ε3 subgroup had 102.82 ± 30.68, 147.79 ± 55.44, 21.18 ± 8.97, and 49.51 ± 25.15 mg/dL of TC, TGL, HDL, and LDL, respectively. IUGR neonates in ε4 subgroup comparatively had significantly higher TC, TGL, and LDL, with lowest levels of HDL than other groups, whereas those in APOE ε2 subgroup had higher HDL (30.8 ± 9.365 mg/dL) levels. On one-way ANOVA, there was no significant difference in the means of the allele subgroups for the lipid compositions ( p  > 0.05).

Discussion

To our knowledge, the present study is the first of its kind to analyze the pattern of APOE gene polymorphism in IUGR neonates born in India, in comparison with AGA controls and their lipid profiles. Our results showed a higher frequency of APOE ε2, a presumed cardioprotective allele and predominance of ε2/ε3 and ε2/ε2, genotypes in IUGR neonates with comparable lipid levels in both groups. This observation is contrary to the study by Infante-Rivard et al who showed a significantly less transmission of this allele among IUGR neonates. ¹⁴

Barker hypothesized that undernutrition during fetal life compromises the differentiation of organs and its growth causing persistent maladaptation, which could result in secondary diseases in adulthood. ^{15 16 17 18} Many epidemiological evidence have strongly correlated with early life environment, low birth weight, and developmental of cardiovascular diseases later in life. ^{19 20} Adverse intrauterine environment could directly affect the expression of genes regulating vascular integrity and myocardial structure of the developing fetus. Crispi et al have demonstrated decreased stroke volume and increased heart rate in children with in utero growth restriction, secondary to fetal growth restriction induced primary cardiovascular changes. ²¹ It indirectly affects the cardiovascular system by impairing insulin sensitivity and resulting metabolic syndrome.

APOE is a receptor-binding ligand glycoprotein that is involved in transport and metabolism of TG and cholesterol. Of its three isoforms (ε2, ε3, and ε4), there exists considerable variation in the cholesterol absorption and postprandial remnant clearance based on their affinity toward the APOE receptor. ²² Throughout the world, studies have shown substantial variations in the allelic expression of the APOE locus with ranges 60 to 90% for ε3, 0 to 20% for ε2, and 10 to 20% for ε4. ^{23 24} Singh et al have investigated its allelic frequencies among Asian Indians and demonstrated its ranges as 0.031 to 0.094 for ε2, 0.803 to 0.968 for ε3, and 0.000 to 0.133 for ε4. ²⁵ In our study, the frequency of ε2 isoform was significantly higher in IUGR group than the control group (7.8 vs. 1.96%). The ε2 isoform has lower affinity for APOE receptors than the other isoforms and causes upregulation of hepatic LDL receptors and thus lowers serum cholesterol levels. ²⁶ This isoform is associated with lower risk of cardiovascular disease and found to be less transmitted in IUGR infants as demonstrated by Infante-Rivard et al in their family-based study design to evaluate the APOE genotypes in very preterm neonates with IUGR. ¹⁴ Contrary to this observation, the variation in the distribution of APOE gene isoforms among IUGR neonates in our study could have been influenced by the race, sampling bias or size of the study population and needs further studies for confirmation.

Only few studies have investigated the distribution of APOE genotype in IUGR/SGA infants. Akisu et al found no difference between the control groups and IUGR neonates with respect to the frequency of APOE isoforms. ²⁷ Norda et al also found no association between the APOE genotype and birth weight percentile with their cohort of preterm infants. ²⁸ We suggest further research to confirm the increased frequency of ε2 isoforms among IUGR neonates in our study.

The APOE ε4 isoform has been strongly associated with increased risk of type 2 diabetes mellitus, CAD, and dyslipidemia. ^{29 30 31 32} Chaudhary et al observed that ε4 genotype showed a lower HDL and higher very LDL and TGL suggesting that ε4 isoform had influence on the lipid profile and might be associated with development of type 2 diabetes and CAD. ³³ In our study, though the IUGR group and AGA controls had no difference in the plasma lipid concentration, on subgrouping the IUGR group based on APOE alleles, the ε4 alleles had higher TC, TGL, and lowest HDL levels when compared with other allele groups ( Table 4 ). The APOE ε4 allele is considered a “thrifty gene” which enhances insulin and cholesterol production, and thus, decreases the serum HDL and increases the serum LDL levels in high fat intake population. Our observation suggests that the IUGR neonates with ε4 allele might have an increased risk for developing dyslipidemia in adulthood. If this observation is reconfirmed in a larger population of IUGR infants, then early identification of their APOE allelic pattern and modification of their diet and lifestyle might avert the possible occurrence of dyslipidemia, type 2 diabetes mellitus, etc., in these infants during their adulthood.

The strength of this study is that the IUGR population has been clearly defined based on fetal growth faltering based on antenatal scans and anthropometric measures at birth, unlike the previous studies where IUGR was defined based only on birth weight centile. The small sample size is a limitation, and hence, our findings need to be confirmed on a larger population.

Conclusion

To conclude, our study showed that ε3 as the predominant isoform and ε4 allele, often attributed for adult-onset metabolic diseases was seen less frequently among IUGR neonates. There was no difference in the lipid profile composition between the IUGR and AGA groups, but the IUGR neonates with the ε4 allele had elevated serum levels of cholesterol, TG, and low HDL than other alleles. We endorse further research on a larger sample size to confirm our observations.

What Is Already Known on This Topic?

• IUGR neonates are at risk for developing metabolic diseases in adulthood.

• APOE gene polymorphisms are shown to modify serum lipid levels and thus influence vascular modeling.

What This Study Adds?

• ε2 allele was the second most predominant APOE isoform, seen more frequently than cardioadverse ε4 allele.

• The lipid levels among IUGR and AGA neonates were comparable, but neonates with ε4 allele showed higher lipid levels than other isoforms.

Funding Statement

Funding This study was funded by PSG PRIME, 11/32 (to R.N.T.).