Advertisement
American Journal of Kidney Diseases

Joint Associations of Maternal-Fetal APOL1 Genotypes and Maternal Country of Origin With Preeclampsia Risk

  • Author Footnotes
    ∗ X.H. and A.Z.R. contributed equally to this work.
    Xiumei Hong
    Footnotes
    ∗ X.H. and A.Z.R. contributed equally to this work.
    Affiliations
    Department of Population, Family and Reproductive Health, Center for the Early Life Origins of Disease, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD
    Search for articles by this author
  • Author Footnotes
    ∗ X.H. and A.Z.R. contributed equally to this work.
    Avi Z. Rosenberg
    Footnotes
    ∗ X.H. and A.Z.R. contributed equally to this work.
    Affiliations
    Department of Pathology, Johns Hopkins University, Baltimore, MD
    Search for articles by this author
  • Boyang Zhang
    Affiliations
    Department of Biostatistics, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD
    Search for articles by this author
  • Elizabeth Binns-Roemer
    Affiliations
    Molecular Genetic Epidemiology Section, Frederick National Laboratory for Cancer Research in the Basic Research Laboratory, National Cancer Institute, Frederick, MD
    Search for articles by this author
  • Victor David
    Affiliations
    Basic Research Laboratory, Center for Cancer Research, National Cancer Institute, Frederick, MD
    Search for articles by this author
  • Yiming Lv
    Affiliations
    Department of Population, Family and Reproductive Health, Center for the Early Life Origins of Disease, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD
    Search for articles by this author
  • Rebecca C. Hjorten
    Affiliations
    Department of Medicine, Division of Nephrology, University of Washington, Seattle, WA
    Search for articles by this author
  • Kimberly J. Reidy
    Affiliations
    Department of Pediatrics, Division of Pediatric Nephrology, Children’s Hospital at Montefiore, Albert Einstein College of Medicine, Bronx, NY
    Search for articles by this author
  • Teresa K. Chen
    Affiliations
    Department of Medicine, Division of Nephrology and Welch Center for Prevention, Epidemiology, and Clinical Research, Johns Hopkins Medical Institutions, Baltimore, MD
    Search for articles by this author
  • Guoying Wang
    Affiliations
    Department of Population, Family and Reproductive Health, Center for the Early Life Origins of Disease, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD
    Search for articles by this author
  • Yuelong Ji
    Affiliations
    Department of Population, Family and Reproductive Health, Center for the Early Life Origins of Disease, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD
    Search for articles by this author
  • Claire L. Simpson
    Affiliations
    Department of Genetics, Genomics, and Informatics, University of Tennessee Health Science Center, Memphis, TN
    Search for articles by this author
  • Robert L. Davis
    Affiliations
    Center for Biomedical Informatics, Department of Pediatrics, University of Tennessee Health Science Center, Memphis, TN
    Search for articles by this author
  • Jeffrey B. Kopp
    Affiliations
    Kidney Diseases Section, National Institute of Diabetes and Digestive and Kidney Disease, National Institutes of Health, Bethesda, MD
    Search for articles by this author
  • Xiaobin Wang
    Correspondence
    Xiaobin Wang, MD, MPH, ScD, 615 N. Wolfe Street, E4132, Baltimore, MD 21205
    Affiliations
    Department of Population, Family and Reproductive Health, Center for the Early Life Origins of Disease, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD

    Division of General Pediatrics and Adolescent Medicine, Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, MD
    Search for articles by this author
  • Cheryl A. Winkler
    Correspondence
    Address for Correspondence: Cheryl A. Winkler, PhD, Frederick National Laboratory for Cancer Research, 8560 Progress Drive, Frederick, MD 21702
    Affiliations
    Molecular Genetic Epidemiology Section, Frederick National Laboratory for Cancer Research in the Basic Research Laboratory, National Cancer Institute, Frederick, MD
    Search for articles by this author
  • Author Footnotes
    ∗ X.H. and A.Z.R. contributed equally to this work.
Published:December 22, 2020DOI:https://doi.org/10.1053/j.ajkd.2020.10.020

      Rationale & Objectives

      Preeclampsia, which disproportionately affects Black women, is a leading cause of preterm delivery and risk for future hypertension and chronic kidney disease (CKD). Apolipoprotein L1 (APOL1) kidney risk alleles, common among Black individuals, contribute substantially to CKD disparities. Given the strong link between preeclampsia and CKD, we investigated whether maternal and fetal APOL1 risk alleles can jointly influence preeclampsia risk, and explored potential modifiers of the association between APOL1 and preeclampsia.

      Study Design

      Nested case-control study.

      Setting & Participants

      426 Black mother-infant pairs (275 African Americans and 151 Haitians) from the Boston Birth Cohort.

      Exposure

      Maternal and fetal APOL1 risk alleles.

      Outcomes

      Preeclampsia.

      Analytical Approach

      Logistic regression models with adjustment for demographic characteristics were applied to analyze associations between fetal and maternal APOL1 risk alleles and risk of preeclampsia and to investigate the effects of modification by maternal country of origin.

      Results

      Fetal APOL1 risk alleles tended to be associated with an increased risk of preeclampsia, which was not statistically significant in the total genotyped population. However, this association was modified by maternal country of origin (P < 0.05 for interaction tests): fetal APOL1 risk alleles were significantly associated with an increased risk of preeclampsia among African Americans under recessive (odds ratio [OR], 3.6 [95% CI, 1.3-9.7]; P = 0.01) and additive (OR, 1.7 [95% CI, 1.1-2.6]; P = 0.01) genetic models but not in Haitian Americans. Also, maternal-fetal genotype discordance at the APOL1 locus was associated with a 2.6-fold higher risk of preeclampsia (P < 0.001) in African Americans.

      Limitations

      Limited sample size in stratified analyses; self-reported maternal country of origin; pre-pregnancy estimated glomerular filtration rate (eGFR) and proteinuria data in mothers were not collected; unmeasured confounding social and/or environmental factors; no replication study.

      Conclusions

      This study supports the hypothesis that fetal APOL1 kidney risk alleles are associated with increased risk for preeclampsia in a recessive mode of inheritance in African Americans and suggests that maternal-fetal genotype discordance is also associated with this risk. These conclusions underscore the need to better understand maternal-fetal interaction and their genetic and environmental factors as contributors to ethnic disparities in preeclampsia.

      Graphical abstract

      Index Words

      Preeclampsia, characterized by increased blood pressure after 20 weeks of pregnancy, as well as other abnormalities (eg, protein in the urine), is dangerous to mothers and their infants. Previous studies found that individuals with African ancestry may carry APOL1 genetic variants that increase risk for chronic kidney disease. This study found that fetal high-risk APOL1 genotypes and maternal-fetal APOL1 genotype discordance independently contribute to preeclampsia risk in African American mothers. This association was not observed in Haitian mother-infant pairs possibly because of different environmental exposures and cultural milieu. Additional studies are required to understand why APOL1 associations with preeclampsia differ by maternal country of origin and to improve management of mothers at risk for preeclampsia.
      Editorial, p. 863
      Preeclampsia, principally characterized by the onset of hypertension after the 20th week of pregnancy, is a leading cause of medically indicated preterm delivery, future risk of cardiovascular diseases,
      • Bergen N.E.
      • Schalekamp-Timmermans S.
      • et al.
      Hypertensive disorders of pregnancy and subsequent maternal cardiovascular health.
      • Veerbeek J.H.
      • Hermes W.
      • Breimer A.Y.
      • et al.
      Cardiovascular disease risk factors after early-onset preeclampsia, late-onset preeclampsia, and pregnancy-induced hypertension.
      • Hammad I.A.
      • Meeks H.
      • Fraser A.
      • et al.
      Risks of cause-specific mortality in offspring of pregnancies complicated by hypertensive disease of pregnancy.
      and chronic kidney disease (CKD)
      • Kristensen J.H.
      • Basit S.
      • Wohlfahrt J.
      • Damholt M.B.
      • Boyd H.A.
      Pre-eclampsia and risk of later kidney disease nationwide cohort study.
      • Khashan A.S.
      • Evans M.
      • Kublickas M.
      • et al.
      Preeclampsia and risk of end stage kidney disease: A Swedish nationwide cohort study.
      • Barrett P.M.
      • McCarthy F.P.
      • Kublickiene K.
      • et al.
      Adverse pregnancy outcomes and long-term maternal kidney disease: a systematic review and meta-analysis.
      in mothers and offspring.
      • Bokslag A.
      • van Weissenbruch M.
      • Mol B.W.
      • de Groot C.J.
      Preeclampsia: short and long-term consequences for mother and neonate.
      Preeclampsia disproportionately affects Black women worldwide.
      • Urquia M.L.
      • Glazier R.H.
      • Gagnon A.J.
      • et al.
      Disparities in pre-eclampsia and eclampsia among immigrant women giving birth in six industrialised countries.
      Despite decades of research, the cause of preeclampsia remains unclear, and there are no established genetic or environmental factors that account for the higher risk of preeclampsia in Black women. Genetic variants in the apolipoprotein L1 (APOL1) gene account for much of the excess risk of kidney disease in Black individuals.
      • Reidy K.J.
      • Hjorten R.
      • Parekh R.S.
      Genetic risk of APOL1 and kidney disease in children and young adults of African ancestry.
      • Genovese G.
      • Friedman D.J.
      • Ross Maryland
      • et al.
      Association of trypanolytic ApoL1 variants with kidney disease in African Americans.
      • Parsa A.
      • Kao W.H.
      • Xie D.
      • et al.
      APOL1 risk variants, race, and progression of chronic kidney disease.
      • Williams W.W.
      • Pollak M.R.
      Health disparities in kidney disease---emerging data from the human genome.
      It is estimated that 12%-14% of African Americans carry 2 APOL1 kidney risk alleles. Given that pre-existing CKD increases the risk of developing preeclampsia
      • Masuyama H.
      • Nobumoto E.
      • Okimoto N.
      • Inoue S.
      • Segawa T.
      • Hiramatsu Y.
      Superimposed preeclampsia in women with chronic kidney disease.
      ,
      • Bramham K.
      • Briley A.L.
      • Seed P.T.
      • Poston L.
      • Shennan A.H.
      • Chappell L.C.
      Pregnancy outcome in women with chronic kidney disease: a prospective cohort study.
      and preeclampsia increases the risk for subsequent CKD,
      • Kristensen J.H.
      • Basit S.
      • Wohlfahrt J.
      • Damholt M.B.
      • Boyd H.A.
      Pre-eclampsia and risk of later kidney disease nationwide cohort study.
      • Khashan A.S.
      • Evans M.
      • Kublickas M.
      • et al.
      Preeclampsia and risk of end stage kidney disease: A Swedish nationwide cohort study.
      • Barrett P.M.
      • McCarthy F.P.
      • Kublickiene K.
      • et al.
      Adverse pregnancy outcomes and long-term maternal kidney disease: a systematic review and meta-analysis.
      it is likely that APOL1 variants play a role in preeclampsia susceptibility. Bruggeman et al
      • Bruggeman L.A.
      • Wu Z.
      • Luo L.
      • et al.
      APOL1-G0 or APOL1-G2 transgenic models develop preeclampsia but not kidney disease.
      reported that pregnant transgenic mice expressing the APOL1 variant G0 or G2 allele in the placenta developed a preeclampsia-like phenotype. Although the preeclampsia phenotype was observed in mice, whether they were transgenic for APOL1 G0 or G2, the latter exhibited more severe disease. Preeclampsia was dependent on the APOL1 genotype of the pup rather than the dam, suggesting transgenic APOL1 expressed by the fetus or placenta was responsible for this phenotype. Consistent with this, Reidy et al
      • Reidy K.J.
      • Hjorten R.C.
      • Simpson C.L.
      • et al.
      Fetal-not maternal-APOL1 genotype associated with risk for preeclampsia in those with African ancestry.
      found that fetal but not maternal APOL1 high-risk genotypes were associated with a 2-fold increased risk of preeclampsia in 2 African American cohorts.
      Previous studies have indicated that both maternal and fetal genetic backgrounds influence preeclampsia risk.
      • Gray K.J.
      • Saxena R.
      • Karumanchi S.A.
      Genetic predisposition to preeclampsia is conferred by fetal DNA variants near FLT1, a gene involved in the regulation of angiogenesis.
      Discordance between maternal and fetal genotypes may lead to altered risk of preeclampsia through the maternal immune system.
      • Parimi N.
      • Tromp G.
      • Kuivaniemi H.
      • et al.
      Analytical approaches to detect maternal/fetal genotype incompatibilities that increase risk of pre-eclampsia.
      For example, a study of family triads showed that particular combinations of mother-child HLA-G genotypes influenced the risk for preeclampsia.
      • Hylenius S.
      • Andersen A.M.
      • Melbye M.
      • Hviid T.V.
      Association between HLA-G genotype and risk of pre-eclampsia: a case-control study using family triads.
      We therefore planned to test the novel hypothesis that maternal-fetal genotype combinations at the APOL1 locus influence risk of preeclampsia.
      Within the US Black population, there is a significant variation in country of origin (eg, maternal self-identified country of identity) and nativity (birthplace), which may affect disease profiles through factors such as different proportions of African descent, distinct within-group cultural norms, and health behaviors.
      • Wartko P.D.
      • Wong E.Y.
      • Enquobahrie D.A.
      Maternal birthplace is associated with low birth weight within racial/ethnic groups.
      • Hammond W.P.
      • Mohottige D.
      • Chantala K.
      • Hastings J.F.
      • Neighbors H.W.
      • Snowden L.
      Determinants of usual source of care disparities among African American and Caribbean black men: findings from the National Survey of American Life.
      • Huffman F.G.
      • De La Cera M.
      • Vaccaro J.A.
      • et al.
      Healthy eating index and alternate healthy eating index among Haitian Americans and African Americans with and without type 2 diabetes.
      • Messiah S.E.
      • Atem F.
      • Lebron C.
      • et al.
      Comparison of early life obesity-related risk and protective factors in non-Hispanic black subgroups.
      Given that the frequencies of APOL1 risk alleles may vary among distinct ethnic populations in Africa and among the African diaspora,
      • Nadkarni G.N.
      • Gignoux C.R.
      • Sorokin E.P.
      • et al.
      Worldwide frequencies of APOL1 renal risk variants.
      ,
      • Limou S.
      • Nelson G.W.
      • Kopp J.B.
      • Winkler C.A.
      APOL1 kidney risk alleles: population genetics and disease associations.
      further studies are needed to investigate whether country of origin and nativity may modify the associations between APOL1 genotypes and risk of preeclampsia.
      We genotyped APOL1 genetic variants in 426 Black mother-infant pairs enrolled in the Boston Birth Cohort (BBC), a predominantly urban, low-income population at high risk for preeclampsia. We tested maternal and fetal APOL1 kidney risk allele-preeclampsia associations separately and explored whether these associations were affected by 1) maternal country of origin (African Americans versus Haitians); and 2) maternal birthplace (US-born vs non–US-born). We further tested whether maternal-fetal genotype discordance (or mismatch) of the APOL1 genotypes affected preeclampsia risk.

      Methods

      Study Design, Setting, and Participants

      Using a nested case-control study design, we studied 213 Black mother-infant pairs with preeclampsia (cases) and 213 Black pairs without preeclampsia (controls) from the BBC. The BBC study began in 1998 at the Boston Medical Center, as previously reported elsewhere.
      • Wang G.
      • Divall S.
      • Radovick S.
      • et al.
      Preterm birth and random plasma insulin levels at birth and in early childhood.
      ,
      • Wang X.
      • Zuckerman B.
      • Pearson C.
      • et al.
      Maternal cigarette smoking, metabolic gene polymorphism, and infant birth weight.
      The BBC study population mirrors the patient population of the Boston Medical Center and contains a relatively high prevalence of preterm births. Mothers who delivered singleton live births were invited to participate in the BBC study within 24-72 hours after delivery. Pregnancies that involved multiple gestations, fetal chromosomal abnormalities, or major birth defects or were the result of in vitro fertilization were excluded. After each mother provided written informed consent, she was interviewed using a standardized questionnaire to gather sociodemographic and other epidemiologic data. Their electronic medical records were abstracted at delivery. The study protocol was approved by the Institutional Review Boards of Boston University Medical Center and Johns Hopkins Bloomberg School of Public Health.
      Selection of cases and controls for the present study is presented in Figure 1. Briefly, we included all available Black mother-infant pairs with preeclampsia (cases, n = 213), and we sought to enroll the same number of Black controls. For control selection (Figure 1), we first performed individual matching on 3 variables including maternal age (+/- 5 years), parity (defined as the number of times that the mother has given birth to a fetus with a gestational age of 24 weeks or more) and infant’s sex. For those cases without matching controls, we selected controls that matched with the cases on either 2 of the 3 variables. After selection, cases and controls were largely comparable as to the 3 matching variables (Table 1).
      Figure thumbnail gr1
      Figure 1Flow chart of the study participants.
      Table 1Characteristics of 426 Black Mother-Infant Pairs With and Without Preeclampsia
      VariablesEnrolled Pairs
      The difference of population characteristics between preeclampsia cases and controls was tested based on univariate logistic regression models.
      Eligible Controls in the BBC
      The differences in population characteristics between the enrolled 213 mother-infant pairs without preeclampsia and all eligible Black mother-infant pairs without preeclampsia in the Boston Birth Cohort by t-test for continuous variables and χ2 test for categorical variables.
      CasesControls
      No. of pairs2132131,691
      Maternal age, y29.6 ± 7.029.2 ± 6.928.2 ± 6.6
      Marital status
       Married75 (35.2%)70 (32.9%)584 (34.5%)
       Others136 (63.9%)137 (64.3%)1,076 (63.6%)
       Missing2 (0.9%)6 (2.8%)31 (1.8%)
      Maternal prepregnancy BMI
       18.5-24.9 kg/m263 (29.6%)106 (49.8%)783 (46.3%)
       25.0-29.9 kg/m259 (27.7%)
      P < 0.01.
      49 (23.0%)455 (26.9%)
       ≥ 30 kg/m280 (37.6%)
      P < 0.001.
      40 (18.8%)351 (20.8%)
       Missing11 (5.2%)18 (8.5%)102 (6.0%)
       Haitian maternal country-of-origin75 (35.2%)76 (35.7%)538 (31.8%)
      Maternal nativity or birthplace
       Non-US born118 (56.2%)138 (64.8%)1,008 (59.6%)
       US born92 (43.8%)
      P < 0.05.
      75 (35.2%)674 (39.9%)
       Missing3 (1.4%)09 (0.5%)
      Years stayed in the United States
       <5.0 years40 (18.8%)53 (24.9%)390 (23.1%)
       5.0-9.9 years29 (13.6%)31 (14.6%)245 (14.5%)
       ≥10.0 years34 (16.0%)36 (16.9%)225 (13.3%)
       Born in the United States92 (43.2%)75 (35.2%)674 (39.9%)
       Missing18 (8.4%)18 (8.4%)157 (9.3%)
      Maternal highest education level
       ≤High school137 (64.3%)127 (59.6%)1,005 (59.5%)
       College or above75 (35.2%)83 (39.0%)662 (39.1%)
       Missing1 (0.5%)3 (1.4%)24 (1.4%)
       Nulliparity112 (52.6%)105 (49.3%)723 (42.8%)
      Maternal smoking during pregnancy
       Never178 (83.5%)187 (87.8%)1,402(82.9%)
       Quitter20 (9.4%)12 (5.6%)127 (7.5%)
       Current smoker14 (6.6%)10 (4.7%)144 (8.5%)
       Missing1 (0.5%)4 (1.9%)18 (1.1%)
      Alcohol drinking during pregnancy
       No180 (84.5%)192 (90.1%)1,488 (88.0%)
       Yes22 (10.3%)13 (6.1%)126 (7.5%)
       Missing11 (5.2%)8 (3.8%)77 (4.5%)
      Chronic hypertension
       No153 (71.8%)209 (98.1%)1,630 (96.4%)
       Yes59 (27.7%)
      P < 0.001.
      4 (1.9%)55 (3.3%)
       Missing1 (0.5%)06 (0.4%)
      Preterm birth116 (54.6%)
      P < 0.001.
      00
      Cesarean delivery116 (54.6%)
      P < 0.001.
      47 (22.1%)487 (28.9%)
      The differences in population characteristics between the enrolled 213 mother-infant pairs without preeclampsia and all eligible Black mother-infant pairs without preeclampsia in the Boston Birth Cohort by t-test for continuous variables and χ2 test for categorical variables.
      P < 0.05.
      Female sex of newborn115 (54.0%)109 (51.2%)841 (49.7%)
      Small for gestational age43 (20.2%)
      P < 0.001.
      18 (8.5%)193 (11.4%)
      Values are mean ± SD for maternal age and count (%) for other variables. Black mother-infant pairs with preeclampsia are cases; and matched mother-infant pairs without pre-eclampsia are controls. Eligible controls are all eligible Black mother-infant pairs without preeclampsia in the parent Boston Birth Cohort.
      Abbreviations: BBC, Boston Birth Cohort; BMI, body mass index.
      a The difference of population characteristics between preeclampsia cases and controls was tested based on univariate logistic regression models.
      b The differences in population characteristics between the enrolled 213 mother-infant pairs without preeclampsia and all eligible Black mother-infant pairs without preeclampsia in the Boston Birth Cohort by t-test for continuous variables and χ2 test for categorical variables.
      c P < 0.01.
      d P < 0.001.
      e P < 0.05.

      Definition of Outcome

      Physician diagnoses of preeclampsia were extracted from the maternal electronic medical records.
      • Bustamante Helfrich B.
      • Chilukuri N.
      • He H.
      • et al.
      Maternal vascular malperfusion of the placental bed associated with hypertensive disorders in the Boston Birth Cohort.
      Preeclampsia was defined according to the report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy, as systolic blood pressure of ≥140 mm Hg or diastolic blood pressure of ≥90 mm Hg on at least 2 occasions, and proteinuria ≥1+ after 20 weeks of gestation.
      • Thakoordeen-Reddy S.
      • Winkler C.
      • Moodley J.
      • et al.
      Maternal variants within the apolipoprotein L1 gene are associated with preeclampsia in a South African cohort of African ancestry.
      For women with preexisting hypertension, evidence of worsening hypertension (a systolic blood pressure ≥160 mm Hg or a diastolic blood pressure of ≥110 mm Hg) was required for a diagnosis of preeclampsia. Hemolysis, elevated liver enzymes, and low platelets developing during pregnancy (HELLP) syndrome was considered present when a physician made this diagnosis contemporaneously; the 213 cases included 210 mothers with preeclampsia and 3 with HELLP.

      Data Measurement

      Using a standard questionnaire interview, maternal epidemiological factors were collected, including self-reported ethnicity defined by maternal country of origin (African Americans vs Haitian), birthplace (US-born vs non–US-born), age at delivery, highest education level, parity, smoking, and alcohol consumption during pregnancy. Non–US-born mothers were asked about how long they had resided in the United States, which was converted into a categorical variable (<5.0, 5.0-9.9, ≥10.0 years). Maternal pre-pregnancy body mass index, calculated as self-reported weight in kilograms divided by height in meters squared, was divided into 4 groups: normal (<25.0 kg/m2), overweight (25-29.9 kg/m2), obese (≥30 kg/m2), and unknown. Clinical complications before and during pregnancy, including chronic hypertension and pre-existing or gestational diabetes, were extracted from electronic medical records. The missing rate of each covariate is presented in Table 1. In data analyses, we replaced missing data with the most frequent values (for categorical variables with missingness in ≤5 participants) or by using a missing indicator/category (for categorical variables with missingness in >5 participants).

      APOL1 Genotyping, and Genetic Ancestry Proportion

      Maternal DNA was isolated from blood within 1-3 days of delivery, and fetal DNA was isolated from umbilical cord blood. Three APOL1 variants (G1 allele [rs73885319 A>G and rs60910145 T>G] and G2 allele [rs71785313 6-base pair deletion]) were genotyped using customized TaqMan assays (Thermo Fisher Scientific) in maternal and fetal DNA samples.
      Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy.
      The APOL1 alleles, including variants G1, G2, and G0 (the last being neither the G1 nor the G2 allele), were coded accordingly. The number of APOL1 kidney risk alleles carried by each participant was coded as 0 for the G0/G0 genotype; 1 for the G0G1 or G0G2 genotype; or 2 for the G1G1, G1G2, or G2G2 genotypes. The APOL1 high-risk genotype was defined as carriage of any combination of 2 G1 and/or G2 risk alleles and the low-risk genotype was defined as carriage of 0 or 1 risk alleles. The APOL1 risk alleles were analyzed for associations with preeclampsia under 3 genetic modes of inheritance: recessive (2 vs 1 or 0 risk alleles; or high-risk vs low-risk genotype); dominant (1 or 2 risk alleles vs 0 risk alleles); and additive (coded as 0, 1, or 2 risk alleles as a continuous variable).
      A random subset of 161 mother-infant pairs (82 with preeclampsia and 79 without preeclampsia) were genotyped using the Infinium Quality Control array (Infinium QC Array-24), containing 15,949 ancestry-informative markers for ancestry estimation. A principal component analysis (“EIGENSTRAT” function in the R “AssocTests” application; R Foundation software) was performed to calculate eigenvectors for each participant. The first 2 eigenvectors, PC1 and PC2, representing the estimated genetic ancestry for each participant, were then plotted by maternal country of origin and birthplace to indicate whether genetic ancestry varied by these 2 variables.

      Statistical Analyses

      Maternal and infant case characteristics were compared with control characteristics using univariate logistic regression models. Maternal and fetal APOL1 genotypes in the controls were examined separately for agreement with Hardy-Weinberg equilibrium (HWE) expectations using χ2 tests. Logistic regression models were applied to analyze the association between each maternal or fetal APOL1 risk allele and risk of preeclampsia, adjusting for conventional factors associated with preeclampsia (pre-pregnancy body mass index, smoking status, alcohol consumption, chronic hypertension), and country of origin, and factors varied between cases and controls in our cohort (years the mothers resided in the United States). Point estimates and 95% confidence intervals (CI) of odds ratios (OR) were calculated. Recessive, additive, and dominant modes of inheritance were tested. Interaction effects between APOL1 genotypes and maternal country of origin (African Americans vs Haitians) and maternal birthplace (US-born vs non–US-born) were tested by adding those 2 variables and their interaction term into the same logistic model. Stratified analyses by maternal country of origin and birthplace were also performed.
      To investigate whether maternal and fetal APOL1 genotype discordance affected the risk of preeclampsia, we generated a binary variable (“MFG_index” application) to represent the APOL1 maternal-fetal genotype discordance state, which was coded as “yes” if there were any differences between maternal and fetal APOL1 genotypes, or “no” if maternal and fetal APOL1 genotypes were the same. The association between the APOL1 maternal-fetal genotype discordance and the risk of preeclampsia was analyzed using the logistic regression model, with adjustment of covariates as described above, as well as with additional adjustment for the main effect of maternal and/or fetal APOL1 genotypes.

      Results

      Participant Characteristics

      As shown in Table 1, the prevalence of pre-pregnancy overweight/obesity, chronic hypertension, and US-born nativity was higher in mothers with preeclampsia (cases, n = 213) than in mothers without preeclampsia (controls, n = 213); all are well-established risk factors for preeclampsia. Preeclamptic mothers were more likely to undergo cesarean delivery and have a small-for-gestational age infant for the preeclamptic pregnancy (all P < 0.05). Participant characteristics among the 213 controls enrolled in this study and all the available Black controls (n = 1,691) in the BBC (Fig 1) are shown in Table 1. These variables were comparable between those 2 control groups, except that cesarean delivery was less frequent in the 213 controls (Table 1).

      Distribution of APOL1 Kidney Risk Alleles and Associations with Preeclampsia

      In mothers without preeclampsia (controls), the minor allele frequencies for rs73885319, rs60910145, and rs71785313 were approximately 19%, 19%, and 9%, respectively, which were comparable to minor allele frequencies in their infants (Table S1). Genotypic distribution of rs71785313 obeyed HWE expectations in both mothers and infants. Genotypic distribution of rs73885319 and rs60910145 (in nearly complete linkage disequilibrium with each other, R2 ~0.99) in the mothers whose conditions varied slightly from HWE expectations (P < 0.05), with an excess of the rs73885319 GG variant genotype and the rs60910145 GG variant genotype (Table S1). This variation was no longer significant when the HWE test was performed separately in Haitian and African American mothers.
      Of the 421 successfully genotyped mothers, 214 (51%), 155 (37%), and 52 (12%) carried 0, 1, or 2 APOL1 risk alleles, respectively. Among the 417 genotyped infants, 201 (48%), 169 (41%), and 47 (11%) carried 0, 1, and 2 risk alleles, respectively. Among all genotyped participants, neither fetal (Table 2) nor maternal (Table 3) APOL1 risk alleles, under the 3 genetic models, had significant associations with risk of preeclampsia, although fetal APOL1 high-risk genotypes showed a trend toward an increased risk of preeclampsia.
      Table 2Association of Fetal APOL1 Risk Alleles With Risk of Preeclampsia
      APOL1 Risk AlleleControlsCasesOR (95% CI)
      Adjusted for pre-pregnancy BMI category, smoking status during pregnancy, alcohol drinking during pregnancy, chronic hypertension, country-of-origin (for the analyses in the total sample only), and years the mothers stayed in the United States.
      P
      Adjusted for pre-pregnancy BMI category, smoking status during pregnancy, alcohol drinking during pregnancy, chronic hypertension, country-of-origin (for the analyses in the total sample only), and years the mothers stayed in the United States.
      Total sample
      0110 (52.9%)91 (43.5%)1.00 (reference)
      179 (38.0%)90 (43.1%)1.17 (0.73-1.87)0.5
      2
      The 2-degree of freedom test was performed for APOL1 risk allele, with bP=0.3 in the total sample, cP=0.02 in African Americans, and dP=0.6 in Haitians.
      19 (9.1%)28 (13.4%)1.79 (0.87-3.70)0.1
      Recessive19 (9.1%)28 (13.4%)1.67 (0.83-3.34)0.2
      Dominant98 (47.1%)118 (56.5%)1.28 (0.83-1.99)0.3
      Additive1.28 (0.93-1.77)0.1
      APOL1 ×maternal country-of-origin interaction
      In the interaction test, maternal country-of-origin was treated as a binary variable with the African American category as the reference group. APOL1 risk allele was considered in a recessive, dominant, or additive model as shown.
      Recessive model0.04
      Dominant model0.2
      Additive model0.05
      Stratified by maternal country-of-origin
      African Americans
       074 (55.6%)54 (40.3%)1.00 (reference)
       152 (39.1%)59 (44.0%)1.35 (0.75-2.44)0.3
       2
      The 2-degree of freedom test was performed for APOL1 risk allele, with bP=0.3 in the total sample, cP=0.02 in African Americans, and dP=0.6 in Haitians.
      7 (5.3%)21 (15.7%)4.06 (1.43-11.6)0.009
       Recessive7 (5.3%)21 (15.7%)3.55 (1.29-9.74)0.01
       Dominant59 (44.4%)80 (59.7%)1.64 (0.94-2.87)0.08
       Additive1.72 (1.11-2.64)0.01
      Haitian
       036 (48.0%)37 (49.3%)1.00 (reference)
       127 (36.0%)31 (41.3%)0.93 (0.42-2.07)0.9
       2
      The 2-degree of freedom test was performed for APOL1 risk allele, with bP=0.3 in the total sample, cP=0.02 in African Americans, and dP=0.6 in Haitians.
      12 (16.0%)7 (9.3%)0.55 (0.17-1.78)0.3
       Recessive12 (16.0%)7 (9.3%)0.57 (0.18-1.74)0.3
       Dominant39 (52.0%)38 (50.6%)0.81 (0.39-1.71)0.6
       Additive0.79 (0.46-1.35)0.4
      Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio.
      a Adjusted for pre-pregnancy BMI category, smoking status during pregnancy, alcohol drinking during pregnancy, chronic hypertension, country-of-origin (for the analyses in the total sample only), and years the mothers stayed in the United States.
      b,c,d The 2-degree of freedom test was performed for APOL1 risk allele, with bP = 0.3 in the total sample, cP = 0.02 in African Americans, and dP = 0.6 in Haitians.
      e In the interaction test, maternal country-of-origin was treated as a binary variable with the African American category as the reference group. APOL1 risk allele was considered in a recessive, dominant, or additive model as shown.
      Table 3Associations of Maternal APOL1 Risk Alleles With Preeclampsia
      APOL1 Risk AlleleControlsCasesOR (95% CI)
      Adjusted for pre-pregnancy BMI category, smoking status during pregnancy, Alcohol drinking during pregnancy, chronic hypertension, country-of-origin (for the analyses in the total sample only), and years the mothers stayed in the United States.
      P
      Adjusted for pre-pregnancy BMI category, smoking status during pregnancy, Alcohol drinking during pregnancy, chronic hypertension, country-of-origin (for the analyses in the total sample only), and years the mothers stayed in the United States.
      Total Sample
      0115 (55.0%)98 (46.4%)1.00 (reference)
      169 (33.0%)86 (40.8%)1.50 (0.93-2.40)0.09
      2
      The 2-degree of freedom test was performed for maternal APOL1 risk allele, with bP=0.2in the total sample, cP=0.1 in African Americans, and dP=0.1in Haitian subset.
      25 (12.0%)27 (12.8%)1.11 (0.55-2.23)0.8
      Recessive25 (12.0%)27 (12.8%)0.95 (0.49-1.86)0.9
      Dominant94 (45.0%)113 (53.6%)1.39 (0.90-2.14)0.1
      Additive1.17 (0.86-1.60)0.3
      APOL1 ×maternal country-of-origin interaction
      In the interaction test, maternal country-of-origin is treated as a binary variable with the African American category as the reference group. APOL1 risk allele was considered in a recessive, dominant, or additive model as shown.
      Recessive model0.02
      Dominant model0.1
      Additive model0.03
      Stratified by maternal country-of-origin
      African-Americans
       075 (56.4%)56 (40.9%)1.00 (reference)
       145 (33.8%)60 (43.8%)1.63 (0.89-2.98)0.1
       2
      The 2-degree of freedom test was performed for maternal APOL1 risk allele, with bP=0.2in the total sample, cP=0.1 in African Americans, and dP=0.1in Haitian subset.
      13 (9.8%)21 (15.3%)2.20 (0.91-5.34)0.08
       Recessive13 (9.8%)21 (15.3%)1.81 (0.77-4.23)0.2
       Dominant58 (43.6%)81 (59.1%)1.76 (1.01-3.06)0.05
       Additive1.53 (1.02-2.28)0.04
      Haitian
       040 (52.6%)42 (56.8%)1.00 (reference)
       124 (31.6%)26 (35.1%)1.15 (0.51-2.56)0.7
       2
      The 2-degree of freedom test was performed for maternal APOL1 risk allele, with bP=0.2in the total sample, cP=0.1 in African Americans, and dP=0.1in Haitian subset.
      12 (15.8%)6 (8.1%)0.31 (0.08-1.17)0.08
       Recessive12 (15.8%)6 (8.1%)0.30 (0.08-1.09)0.07
       Dominant36 (47.4%)32 (43.2%)0.83 (0.40-1.74)0.6
       Additive0.72 (0.42-1.23)0.2
      a Adjusted for pre-pregnancy BMI category, smoking status during pregnancy, Alcohol drinking during pregnancy, chronic hypertension, country-of-origin (for the analyses in the total sample only), and years the mothers stayed in the United States.
      b,c,d The 2-degree of freedom test was performed for maternal APOL1 risk allele, with bP = 0.2 in the total sample, cP = 0.1 in African Americans, and dP = 0.1 in Haitian subset.
      e In the interaction test, maternal country-of-origin is treated as a binary variable with the African American category as the reference group. APOL1 risk allele was considered in a recessive, dominant, or additive model as shown.

      Modification Effects by Maternal Country-of-Origin and Birthplace

      To test whether the associations between APOL1 genotypes and risk of preeclampsia varied by maternal country of origin (Haitian vs African Americans), we performed interaction tests and found a borderline interaction between fetal APOL1 risk alleles and maternal country of origin on risk of preeclampsia, with P < 0.05 for the interaction term when APOL1 alleles were analyzed using the recessive or additive mode of inheritance (Table 2). In stratified analyses, we found that, among the African American subset, fetal APOL1 risk alleles were significantly associated with an increased risk of preeclampsia under recessive (OR, 3.6 [95% CI, 1.3-9.7]; P = 0.01) and additive (OR, 1.7 [95% CI, 1.1-2.6]; P = 0.01) modes of inheritance. No such associations were observed in Haitians (Table 2). In comparison, no significant interactions were observed between fetal APOL1 risk alleles and maternal birthplace, although fetal APOL1 risk alleles were significantly associated with a higher risk of preeclampsia under an additive (OR, 2.2 [95% CI, 1.2-4.0]; P = 0.01) and a dominant (OR, 2.1 [95% CI, 1.0-4.4]; P = 0.04) model in US-born but not in non–US-born subsets (Table S2).
      Similar interactions were also observed between maternal APOL1 risk alleles and country of origin on risk of preeclampsia, in recessive (P = 0.02) and additive (P = 0.03) modes of inheritance, respectively (Table 3). With stratified analyses and in an additive mode of inheritance, maternal APOL1 risk alleles were borderline significantly associated with a higher risk of preeclampsia (OR, 1.5 [95% CI, 1.0-2.3]; P = 0.04) only in the African American subset, and not the Haitian subset. A similar borderline association was observed when maternal APOL1 risk alleles were analyzed in a dominant mode (Table 3). No significant effect modification of maternal birthplace was observed for the relationships between maternal APOL1 risk alleles and risk of preeclampsia.
      In a subset of the mother-infant pairs, we performed ancestry principal component analyses, which did not reveal significant differences in population substructure by maternal country-of-origin or birthplace (Fig S1).

      APOL1 Mother-Fetal Genotype Discordance and Risk of Preeclampsia

      Among the 411 pairs with available genotypic data in both mothers and infants, 198 pairs had a discordance between maternal and fetal APOL1 genotypes. Preeclampsia risk was significantly higher in pairs with APOL1 maternal-fetal genotype discordance compared with those without such discordance (P = 0.04). Again, we found that this association was significantly modified by maternal country-of-origin (with P = 0.006 for interaction) (Table 4): APOL1 maternal-fetal genotype discordance was associated with ~2.6-fold higher risk of preeclampsia (95% CI, 1.5-4.6; P < 0.001) in African Americans but not in Haitians. These associations remained largely unchanged after adjusting for the main effect of maternal and/or fetal APOL1 risk alleles (Table 4).
      Table 4Associations Between APOL1 MFG Discordance and Risk of Preeclampsia
      ControlsCasesModel 1
      Model 1: associations were tested using the logistic regression model, adjusted for pre-pregnancy BMI category, smoking status during pregnancy, alcohol drinking during pregnancy, chronic hypertension, country of origin (for the analyses in the total sample only), and years mothers stayed in the United States.
      Model 2
      Model 2=Model 1+maternal APOL1 allele status.
      Model 3
      Model 3=Model 1+maternal APOL1 allele status+fetal APOL1 allele status.
      OR (95% CI)POR(95% CI)POR(95% CI)P
      Total Sample
      MFG discordance (N)
      Maternal-fetal genotype (MFG) discordance was coded as “Y” if there were any differences between maternal and fetal APOL1 genotypes or “N” if there were no differences between maternal and fetal APOL1 genotypes.
      113 (55.4%)100 (48.3%)1.00 (reference)1.00 (reference)1.00 (reference)
      MFG discordance (Y)
      Maternal-fetal genotype (MFG) discordance was coded as “Y” if there were any differences between maternal and fetal APOL1 genotypes or “N” if there were no differences between maternal and fetal APOL1 genotypes.
      91 (44.6%)107 (51.7%)1.58 (1.02-2.44)0.041.53 (0.95-2.45)0.081.46 (0.89-2.38)0.1
      MFG discordance ×maternal country-of-origin interaction0.0060.0060.004
      African American mother-infant pairs
      MFG discordance (N)
      Maternal-fetal genotype (MFG) discordance was coded as “Y” if there were any differences between maternal and fetal APOL1 genotypes or “N” if there were no differences between maternal and fetal APOL1 genotypes.
      80 (62.0%)59 (44.4%)1.00 (reference)1.00 (reference)1.00 (reference)
      MFG discordance (Y)
      Maternal-fetal genotype (MFG) discordance was coded as “Y” if there were any differences between maternal and fetal APOL1 genotypes or “N” if there were no differences between maternal and fetal APOL1 genotypes.
      49 (38.0%)74 (55.6%)2.61 (1.48-4.60)<0.0012.34 (1.29-4.26)0.0052.19 (1.19-4.03)0.01
      Haitian mother-infant pairs
      MFG discordance (N)
      Maternal-fetal genotype (MFG) discordance was coded as “Y” if there were any differences between maternal and fetal APOL1 genotypes or “N” if there were no differences between maternal and fetal APOL1 genotypes.
      33 (44.0%)41 (55.4%)1.00 (reference)1.00 (reference)1.00 (reference)
      MFG discordance (Y)
      Maternal-fetal genotype (MFG) discordance was coded as “Y” if there were any differences between maternal and fetal APOL1 genotypes or “N” if there were no differences between maternal and fetal APOL1 genotypes.
      42 (56.0%)33 (44.6%)0.60 (0.29-1.27)0.20.68 (0.30-1.57)0.40.68 (0.27-1.68)0.4
      Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio.
      a Maternal-fetal genotype (MFG) discordance was coded as “Y” if there were any differences between maternal and fetal APOL1 genotypes or “N” if there were no differences between maternal and fetal APOL1 genotypes.
      b Model 1: associations were tested using the logistic regression model, adjusted for pre-pregnancy BMI category, smoking status during pregnancy, alcohol drinking during pregnancy, chronic hypertension, country of origin (for the analyses in the total sample only), and years mothers stayed in the United States.
      c Model 2 = Model 1 + maternal APOL1 allele status.
      d Model 3 = Model 1+ maternal APOL1 allele status + fetal APOL1 allele status.

      Discussion

      This study confirmed the adverse impact of fetal APOL1 risk alleles on preeclampsia risk among African Americans but not among Haitians, which suggests that other environmental or genetic factors may modify APOL1 risk allele penetrance for preeclampsia. Importantly, we found that APOL1 maternal-fetal genotype discordance was associated with a 2.6-fold higher risk of preeclampsia in African Americans, which was not confounded by maternal or fetal APOL1 genotypes. This finding suggests that both maternal and fetal APOL1 high-risk genotypes contribute to the pathogenesis of preeclampsia.
      Although the pathophysiological mechanisms by which APOL1 risk variants damage the kidney are not yet fully defined, current evidence suggests multiple mechanisms may contribute, including plasma membrane cation channel activity, endosomal trafficking, mitochondrial function, pro-inflammatory cytokine expression, activation of inflammasomes and protein kinase R, and interactions with APOL3.
      • David V.A.
      • Binns-Roemer E.A.
      • Winkler C.A.
      Taqman assay for genotyping CKD-associated APOL1 SNP rs60910145: a cautionary note.
      • Ma L.
      • Chou J.W.
      • Snipes J.A.
      • et al.
      APOL1 renal-risk variants induce mitochondrial dysfunction.
      • Shah S.S.
      • Lannon H.
      • Dias L.
      • et al.
      APOL1 kidney risk variants induce cell death through mitochondrial translocation and opening of the mitochondrial permeability transition pore.
      • Jha A.
      • Kumar V.
      • Haque S.
      • et al.
      Alterations in plasma membrane ion channel structures stimulate NLRP3 inflammasome activation in APOL1 risk milieu.
      • O’Toole J.F.
      • Schilling W.
      • Kunze D.
      • et al.
      ApoL1 overexpression drives variant-independent cytotoxicity.
      In the placenta, APOL1 is expressed in trophoblasts, Hoffbauer cells, and endothelial cells. Previously, we showed very low levels of placental growth factor expression that, together with high APOL1 expression in trophoblasts, raised the possibility that an APOL1-mediated defect in placentation contributed to preeclampsia. Consistently, circulating antibodies against APOL1 are found in women with preeclampsia, and serum APOL1 levels are higher in women with preeclampsia.
      • Okamoto K.
      • Rausch J.W.
      • Wakashin H.
      • et al.
      APOL1 risk allele RNA contributes to renal toxicity by activating protein kinase R.
      ,
      • Wen Q.
      • Liu L.Y.
      • Yang T.
      • et al.
      Peptidomic identification of serum peptides diagnosing preeclampsia.
      Very few studies have investigated the role of APOL1 risk alleles in preeclampsia.
      • Reidy K.J.
      • Hjorten R.C.
      • Simpson C.L.
      • et al.
      Fetal-not maternal-APOL1 genotype associated with risk for preeclampsia in those with African ancestry.
      ,
      • Elliott S.E.
      • Parchim N.F.
      • Liu C.
      • et al.
      Characterization of antibody specificities associated with preeclampsia.
      In 2 US study cohorts, Reidy et al
      • Reidy K.J.
      • Hjorten R.C.
      • Simpson C.L.
      • et al.
      Fetal-not maternal-APOL1 genotype associated with risk for preeclampsia in those with African ancestry.
      reported that fetal but not maternal APOL1 high-risk genotype was associated with an approximately 2-fold higher risk of preeclampsia. Miller et al
      • Elliott S.E.
      • Parchim N.F.
      • Liu C.
      • et al.
      Characterization of antibody specificities associated with preeclampsia.
      , in 395 preeclamptic cases and 282 controls, confirmed that infant APOL1 genotype was associated with preeclampsia but in a dominant inheritance model (Fig S2).
      • Elliott S.E.
      • Parchim N.F.
      • Liu C.
      • et al.
      Characterization of antibody specificities associated with preeclampsia.
      Neither study investigated potential factors that might modify APOL1-preeclampsia associations, nor did they test the impact of maternal-fetal genotype discordance. Robertson et al
      • Miller A.K.
      • Azhibekov T.
      • O’Toole J.F.
      • et al.
      Association of preeclampsia with infant APOL1 genotype in African Americans.
      analyzed data from 2 genome-wide association studies of prematurity, including 960 mothers from our BBC study and 519 mothers and 867 infants from GENEVA (Gene Environment Association Studies). The study found no significant associations between APOL1 high-risk genotypes and preeclampsia. The 2 genome-wide association studies were originally designed for genetic studies of prematurity, which might have led to participant selection bias when preeclampsia was analyzed as the outcome of interest, because preeclampsia is a cause of prematurity. Also, fetal genome-wide association studies were not available from the BBC, and determination of the APOL1 G2 variant in the mothers was based on imputation. Finally, in GENEVA, preeclampsia and chronic hypertension were assessed as a pooled binary trait. Using chronic hypertension to define whether the mother had preeclampsia could lead to phenotype misclassification. Also, only ~50% of the study population had data for this binary trait, limiting statistical power. In the African American subset from the present study, we replicated the association between fetal APOL1 risk alleles and preeclampsia under the recessive mode of inheritance as reported by Reidy et al
      • Reidy K.J.
      • Hjorten R.C.
      • Simpson C.L.
      • et al.
      Fetal-not maternal-APOL1 genotype associated with risk for preeclampsia in those with African ancestry.
      in Black individuals from New York City, New York and from Memphis, Tennessee. Although the present results indicated that this association remained significant in an additive mode of inheritance, we note that the effect size for carrying 1 risk allele (OR, 1.4) was modest compared to the effect size of carrying 2 risk alleles (OR, 4.1). Therefore, we propose that the impact of fetal APOL1 genotypes on the risk of preeclampsia is best explained by the recessive model, but this awaits further validation in a larger cohort.
      The present study suggests that maternal country-of-origin modifies the relationship between fetal APOL1 high-risk genotypes and risk of preeclampsia. One possible explanation is that there are differences in regional African or European genetic contributions between Haitians and African Americans. Although our admixture analyses indicate that African American and Haitian subsets had comparable genetic ancestry, the ancestry-informative markers genotyped in this study were designed to differentiate between European and African ancestry and may not be able to distinguish regional African or European contributions that may differ between Haitians and African Americans. Another possibility is that there are distinct environmental/social exposures or other unmeasured genetic variants between these 2 subsets that modify the APOL1-preeclampsia associations.
      We observed a borderline-significant positive relationship between maternal APOL1 risk allele and preeclampsia risk in African Americans under both the additive and dominant modes of inheritance compared to a null association as reported by Reidy et al.
      • Reidy K.J.
      • Hjorten R.C.
      • Simpson C.L.
      • et al.
      Fetal-not maternal-APOL1 genotype associated with risk for preeclampsia in those with African ancestry.
      In comparison, the study by Thakoordeen-Reddy et al
      • Robertson C.C.
      • Gillies C.E.
      • Putler R.K.B.
      • et al.
      An investigation of APOL1 risk genotypes and preterm birth in African American population cohorts.
      reported that the maternal APOL1 G1 risk allele was associated with 2.2 times higher risk of having early-onset preeclampsia in 428 South African women of African ancestry. Maternal APOL1 risk alleles may affect the risk of preeclampsia directly or through transmission of the risk allele to the infant, because mothers would be obligate carriers of at least 1 risk allele for a fetus carrying 2 risk alleles and have a 50% chance of being the source of the 1 risk allele in the heterozygous fetus. We further demonstrated that APOL1 maternal-fetal genotype discordance was a risk factor for preeclampsia in African Americans. This finding, if validated in other studies, suggests a potential joint mother-fetal genetic contribution to preeclampsia risk, highlighting the need to evaluate combined maternal-fetal genotypes when assessing pregnancy complications as the outcomes. Although the underlying biological mechanism for the impact of maternal-fetal genotype combination is unknown, a possible explanation is suggested by the immunological maladaptation theory, which proposes that the maternal immune system does not adapt fully to the semi-allogenic fetus. Although the mother is usually tolerant of discordant fetal antigens, there are exceptions. These include paternal antigens that are discordant from those of the mother and are absent from semen. Further studies are needed to validate our findings on the joint effects of mother-fetal APOL1 genotypes.
      This study is to our knowledge the first to systematically investigate the impact of the maternal APOL1 genotype, fetal APOL1 genotype, and their combination effects on preeclampsia in Black women, as well as to explore the modification effects by maternal country-of-origin. Several limitations should be acknowledged. First, statistical power was relatively limited, especially for stratified analyses. Second, maternal country-of-origin was self-reported and may have been subject to reporting bias. Other population characteristics were collected (within 1-3 days after delivery) after the diagnosis of presence or absence of preeclampsia and thus might have led to recall bias. However, none of the participants knew their APOL1 genotypes during their interview, so recall bias, if it existed, will not substantially influence the identified APOL1-preeclampsia associations. Third, pre-pregnancy estimated glomerular filtration rate and proteinuria data in mothers were not collected in this study; therefore, their impact on the APOL1-preeclampsia association cannot be analyzed. Fourth, we acknowledge the possibility that the associations between fetal APOL1 risk allele (or maternal-fetal genotype discordance) and risk of preeclampsia in African Americans may be driven by some unmeasured genetic/environmental variables and that the APOL1 risk allele may actually serve as the surrogate for the unmeasured variable. Fifth, in the analyses for APOL1 maternal-fetal genotype discordance, we applied the most robust coding scheme for the genotypic discordance between the mothers and the infants, as reported.
      • Parimi N.
      • Tromp G.
      • Kuivaniemi H.
      • et al.
      Analytical approaches to detect maternal/fetal genotype incompatibilities that increase risk of pre-eclampsia.
      However, it is likely that other discordance models may be more predictive, and further studies with case-parent triads are needed to validate and extend our findings. Sixth, the data reported here have been collected since 1998, well before the introduction of the 2019 definition of preeclampsia. The new definition, in which proteinuria is no longer necessary for diagnosis, will not significantly change our classification of the case versus control group, and thus, will not significantly change the associations direction as we reported.
      In conclusion, this study suggests that the effects of both maternal and fetal APOL1 risk alleles are risk factors for preeclampsia but APOL1 penetrance may be modified by maternal country of origin, which may reflect differences in as yet undetected environmental and/or genetic exposures. Future research addressing these knowledge gaps would advance understanding of preeclampsia, enable more effective prevention and treatment of preeclampsia and might have implications for the prevention and treatment of APOL1-mediated CKD across generations.

      Article Information

      Authors’ Full Names and Academic Degrees

      Xiumei Hong, MD, PhD, Avi Z. Rosenberg, MD, PhD, Boyang Zhang, BS, Elizabeth Binns-Roemer, BS, Victor David, MS, Yiming Lv, BS, Rebecca C. Hjorten, MD, Kimberly J. Reidy, MD, Teresa K. Chen, MD, MHS, Guoying Wang, MD, PhD, Yuelong Ji, PhD, Claire L. Simpson, PhD, Robert L. Davis, MD, MPH, Jeffrey B. Kopp, MD, Xiaobin Wang, MD, MPH, ScD, and Cheryl A. Winkler, PhD.

      Authors’ Contributions

      Contributed to the study design: AZR, XH, XW, CAW, JBK; oversaw the genotyping performed in this study and genetic analyses: CAW, VD; performed data cleaning and data analyses: XH, BZ, YL; provided critical inputs on study design, data analyses, interpretation of the study findings: all authors; founded and principal investigator of the Boston Birth Cohort (the parent study): XW; oversaw participant recruitment and data collection for the parent study: XW; contributed to participant recruitment and data collection for the parent study: XH, GW, YJ. Each author contributed important intellectual content during manuscript drafting or revision and agrees to be personally accountable for the individual’s own contributions and to ensure that questions pertaining to the accuracy or integrity of any portion of the work, even one in which the author was not directly involved, are appropriately investigated and resolved, including with documentation in the literature if appropriate.

      Support

      The Boston Birth Cohort (the parent study) was supported by National Insitutes of Health ( NIH) grants R03HD096136 , R21HD085556 , 2R01HD041702 , R01HD086013 , R01HD098232 , and R21AI154233 . This project was supported by National Institute of Diabetes and Digestive and Kidney Diseases ( NIDDK) Intramural Research Program grant ZO1 DK-043308 (to JBK), an NIH Bench to Bedside grant supplementing ZO1 DK-043308 and 3R01HD098232-02S1 , and National Cancer Institute of NIH Intramural Research Program (to Dr Winkler) and under contract HHSN26120080001E. Dr Hong is supported by the Johns Hopkins Population Center (Eunice Kennedy Shriver National Institute of Child Health and Human Development grant R24HD042854) . Dr Chen was supported by the Extramural Grant Program by Satellite Healthcare, a not-for-profit renal care provider. Dr Reidy receives support from a catalytic seed grant (NIH Clinical and Translational Science Award 1 UL1 TR001073 [Einstein/Montefiore]), NIDDK, and the Preeclampsia Foundation. The funders were not involved in study design, data collection, analysis, reporting, or the decision to submit for publication.

      Financial Disclosure

      The authors declare that they have no relevant financial interests.

      Acknowledgements

      The authors thank study participants in the Boston Birth Cohort study for their support. The authors also thank for the dedication and hard work of the field team at the Department of Pediatrics, Boston University School of Medicine, and for the support of the obstetric nursing staff at Boston Medical Center. The authors also thank Linda Rosen, Boston University Clinical Data Warehouse, for assistance in obtaining relevant clinical information. The Clinical Data Warehouse service is supported by Boston University Clinical and Translational Institute and the NIH Clinical and Translational Science Award U54-TR001012.

      Disclaimer

      Research reported in this publication was supported through the NIH Bench-to-Bedside award made possible by Office of Research on Women’s Health (ORWH). The content of this publication does not necessarily reflect the view or policy of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the government.

      Peer Review

      Received April 23, 2020. Evaluated by 2 external peer reviewers, with direct editorial input from a Statistics/Methods Editor, an Associate Editor, and the Editor-in-Chief. Accepted in revised form October 24, 2020.

      Supplementary Material

      References

        • Bergen N.E.
        • Schalekamp-Timmermans S.
        • et al.
        Hypertensive disorders of pregnancy and subsequent maternal cardiovascular health.
        Eur J Epidemiol. 2018; 33: 763-771
        • Veerbeek J.H.
        • Hermes W.
        • Breimer A.Y.
        • et al.
        Cardiovascular disease risk factors after early-onset preeclampsia, late-onset preeclampsia, and pregnancy-induced hypertension.
        Hypertension. 2015; 65: 600-606
        • Hammad I.A.
        • Meeks H.
        • Fraser A.
        • et al.
        Risks of cause-specific mortality in offspring of pregnancies complicated by hypertensive disease of pregnancy.
        Am J Obstet Gynecol. 2020; 222: 75 e71-75 e79
        • Kristensen J.H.
        • Basit S.
        • Wohlfahrt J.
        • Damholt M.B.
        • Boyd H.A.
        Pre-eclampsia and risk of later kidney disease nationwide cohort study.
        BMJ. 2019; 365: l1516
        • Khashan A.S.
        • Evans M.
        • Kublickas M.
        • et al.
        Preeclampsia and risk of end stage kidney disease: A Swedish nationwide cohort study.
        PLoS Med. 2019; 16e1002875
        • Barrett P.M.
        • McCarthy F.P.
        • Kublickiene K.
        • et al.
        Adverse pregnancy outcomes and long-term maternal kidney disease: a systematic review and meta-analysis.
        JAMA Netw Open. 2020; 3e1920964
        • Bokslag A.
        • van Weissenbruch M.
        • Mol B.W.
        • de Groot C.J.
        Preeclampsia: short and long-term consequences for mother and neonate.
        Early Hum Dev. 2016; 102: 47-50
        • Urquia M.L.
        • Glazier R.H.
        • Gagnon A.J.
        • et al.
        Disparities in pre-eclampsia and eclampsia among immigrant women giving birth in six industrialised countries.
        BJOG. 2014; 121: 1492-1500
        • Reidy K.J.
        • Hjorten R.
        • Parekh R.S.
        Genetic risk of APOL1 and kidney disease in children and young adults of African ancestry.
        Curr Opin Pediatr. 2018; 30: 252-259
        • Genovese G.
        • Friedman D.J.
        • Ross Maryland
        • et al.
        Association of trypanolytic ApoL1 variants with kidney disease in African Americans.
        Science. 2010; 329: 841-845
        • Parsa A.
        • Kao W.H.
        • Xie D.
        • et al.
        APOL1 risk variants, race, and progression of chronic kidney disease.
        N Engl J Med. 2013; 369: 2183-2196
        • Williams W.W.
        • Pollak M.R.
        Health disparities in kidney disease---emerging data from the human genome.
        N Engl J Med. 2013; 369: 2260-2261
        • Masuyama H.
        • Nobumoto E.
        • Okimoto N.
        • Inoue S.
        • Segawa T.
        • Hiramatsu Y.
        Superimposed preeclampsia in women with chronic kidney disease.
        Gynecol Obstet Invest. 2012; 74: 274-281
        • Bramham K.
        • Briley A.L.
        • Seed P.T.
        • Poston L.
        • Shennan A.H.
        • Chappell L.C.
        Pregnancy outcome in women with chronic kidney disease: a prospective cohort study.
        Reprod Sci. 2011; 18: 623-630
        • Bruggeman L.A.
        • Wu Z.
        • Luo L.
        • et al.
        APOL1-G0 or APOL1-G2 transgenic models develop preeclampsia but not kidney disease.
        J Am Soc Nephrol. 2016; 27: 3600-3610
        • Reidy K.J.
        • Hjorten R.C.
        • Simpson C.L.
        • et al.
        Fetal-not maternal-APOL1 genotype associated with risk for preeclampsia in those with African ancestry.
        Am J Hum Genet. 2018; 103: 367-376
        • Gray K.J.
        • Saxena R.
        • Karumanchi S.A.
        Genetic predisposition to preeclampsia is conferred by fetal DNA variants near FLT1, a gene involved in the regulation of angiogenesis.
        Am J Obstet Gynecol. 2018; 218: 211-218
        • Parimi N.
        • Tromp G.
        • Kuivaniemi H.
        • et al.
        Analytical approaches to detect maternal/fetal genotype incompatibilities that increase risk of pre-eclampsia.
        BMC Med Genet. 2008; 9: 60
        • Hylenius S.
        • Andersen A.M.
        • Melbye M.
        • Hviid T.V.
        Association between HLA-G genotype and risk of pre-eclampsia: a case-control study using family triads.
        Mol Hum Reprod. 2004; 10: 237-246
        • Wartko P.D.
        • Wong E.Y.
        • Enquobahrie D.A.
        Maternal birthplace is associated with low birth weight within racial/ethnic groups.
        Matern Child Health J. 2017; 21: 1358-1366
        • Hammond W.P.
        • Mohottige D.
        • Chantala K.
        • Hastings J.F.
        • Neighbors H.W.
        • Snowden L.
        Determinants of usual source of care disparities among African American and Caribbean black men: findings from the National Survey of American Life.
        J Health Care Poor Underserved. 2011; 22: 157-175
        • Huffman F.G.
        • De La Cera M.
        • Vaccaro J.A.
        • et al.
        Healthy eating index and alternate healthy eating index among Haitian Americans and African Americans with and without type 2 diabetes.
        J Nutr Metab. 2011; 2011: 398324
        • Messiah S.E.
        • Atem F.
        • Lebron C.
        • et al.
        Comparison of early life obesity-related risk and protective factors in non-Hispanic black subgroups.
        Matern Child Health J. 2020; 24: 1130-1137
        • Nadkarni G.N.
        • Gignoux C.R.
        • Sorokin E.P.
        • et al.
        Worldwide frequencies of APOL1 renal risk variants.
        N Engl J Med. 2018; 379: 2571-2572
        • Limou S.
        • Nelson G.W.
        • Kopp J.B.
        • Winkler C.A.
        APOL1 kidney risk alleles: population genetics and disease associations.
        Adv Chronic Kidney Dis. 2014; 21: 426-433
        • Wang G.
        • Divall S.
        • Radovick S.
        • et al.
        Preterm birth and random plasma insulin levels at birth and in early childhood.
        JAMA. 2014; 311: 587-596
        • Wang X.
        • Zuckerman B.
        • Pearson C.
        • et al.
        Maternal cigarette smoking, metabolic gene polymorphism, and infant birth weight.
        JAMA. 2002; 287: 195-202
        • Bustamante Helfrich B.
        • Chilukuri N.
        • He H.
        • et al.
        Maternal vascular malperfusion of the placental bed associated with hypertensive disorders in the Boston Birth Cohort.
        Placenta. 2017; 52: 106-113
      1. Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy.
        Am J Obstet Gynecol. 2000; 183: S1-S22
        • David V.A.
        • Binns-Roemer E.A.
        • Winkler C.A.
        Taqman assay for genotyping CKD-associated APOL1 SNP rs60910145: a cautionary note.
        Kidney Int Rep. 2019; 4: 184-185
        • Ma L.
        • Chou J.W.
        • Snipes J.A.
        • et al.
        APOL1 renal-risk variants induce mitochondrial dysfunction.
        J Am Soc Nephrol. 2017; 28: 1093-1105
        • Shah S.S.
        • Lannon H.
        • Dias L.
        • et al.
        APOL1 kidney risk variants induce cell death through mitochondrial translocation and opening of the mitochondrial permeability transition pore.
        J Am Soc Nephrol. 2019; 30: 2355-2368
        • Jha A.
        • Kumar V.
        • Haque S.
        • et al.
        Alterations in plasma membrane ion channel structures stimulate NLRP3 inflammasome activation in APOL1 risk milieu.
        FEBS J. 2020; 287: 2000-2022
        • O’Toole J.F.
        • Schilling W.
        • Kunze D.
        • et al.
        ApoL1 overexpression drives variant-independent cytotoxicity.
        J Am Soc Nephrol. 2018; 29: 869-879
        • Okamoto K.
        • Rausch J.W.
        • Wakashin H.
        • et al.
        APOL1 risk allele RNA contributes to renal toxicity by activating protein kinase R.
        Commun Biol. 2018; 1: 188
        • Wen Q.
        • Liu L.Y.
        • Yang T.
        • et al.
        Peptidomic identification of serum peptides diagnosing preeclampsia.
        PLoS One. 2013; 8e65571
        • Elliott S.E.
        • Parchim N.F.
        • Liu C.
        • et al.
        Characterization of antibody specificities associated with preeclampsia.
        Hypertension. 2014; 63: 1086-1093
        • Miller A.K.
        • Azhibekov T.
        • O’Toole J.F.
        • et al.
        Association of preeclampsia with infant APOL1 genotype in African Americans.
        BMC Med Genet. 2020; 21: 110
        • Robertson C.C.
        • Gillies C.E.
        • Putler R.K.B.
        • et al.
        An investigation of APOL1 risk genotypes and preterm birth in African American population cohorts.
        Nephrol Dial Transplant. 2017; 32: 2051-2058
        • Thakoordeen-Reddy S.
        • Winkler C.
        • Moodley J.
        • et al.
        Maternal variants within the apolipoprotein L1 gene are associated with preeclampsia in a South African cohort of African ancestry.
        Eur J Obstet Gynecol Reprod Biol. 2020; 246: 129-133

      Linked Article

      • APOL1 and Preeclampsia: Intriguing Links, Uncertain Causality, Troubling Implications
        American Journal of Kidney DiseasesVol. 77Issue 6
        • Preview
          Preeclampsia is a common, serious, multisystem disorder of pregnancy characterized by hypertension onset after 20 weeks of gestation (or worsening of hypertension in a patient with chronic hypertension) and onset or worsening of proteinuria.1 Data from the National Inpatient Sample showed approximately 177,000 of 3.8 million deliveries (4.7%) have a diagnosis of preeclampsia/eclampsia.2 This disorder is of interest to nephrologists, since women with chronic kidney disease (CKD) are at increased risk to develop preeclampsia and kidney physicians are often consulted to help manage their blood pressure and kidney manifestations.
        • Full-Text
        • PDF