Research ArticleMalaria

Malaria in pregnancy alters l-arginine bioavailability and placental vascular development

See allHide authors and affiliations

Science Translational Medicine  07 Mar 2018:
Vol. 10, Issue 431, eaan6007
DOI: 10.1126/scitranslmed.aan6007
  • Fig. 1 l-Arginine, ADMA, and SDMA change longitudinally over the course of pregnancy.

    Concentrations of l-arginine, ADMA, and SDMA measured by mass spectrometry in n = 603 plasma samples from the study cohort of pregnant women (n = 94) beginning at 16 weeks of gestation. (A to D) Longitudinal assessment of changes during pregnancy in ADMA (P = 0.01) (A), SDMA (P < 0.0001) (B), l-arginine (P > 0.05) (C), and l-arginine/ADMA (P > 0.05) (D) from 16 weeks of gestation to delivery by linear regression.

  • Fig. 2 Malaria at enrollment is associated with altered nitric oxide biosynthesis during pregnancy.

    (A to F) Individual data points colored according to malaria status by microscopy for concentrations of l-arginine (P > 0.95) (A), ADMA (P > 0.65) (B), and SDMA (P < 0.04) (C) or malaria status by PCR for l-arginine (P = 0.031) (D), ADMA (P = 0.006) (E), and SDMA (P = 0.57) (F). The overlaid regression lines are from linear mixed-effects models fitted for a subject with average values (conditional on fixed effects only).

  • Fig. 3 Increased ADMA in pregnancy is associated with SGA birth outcomes.

    Those who went on to have SGA births had higher concentrations of ADMA than those who later had AGA births [χ2(2) = 8.76, P < 0.02]. Individual data points are colored according to SGA or AGA status with overlaid regression lines from linear mixed-effects models, fitted for a subject with average values (conditional on fixed effects only).

  • Fig. 4 Dietary l-arginine supplementation improves fetal outcomes in experimental malaria in pregnancy.

    (A) Fetal weight (in grams) in uninfected vehicle control–treated litters (n = 22), uninfected l-arginine–supplemented litters (n = 37), malaria [Plasmodium berghei ANKA (PbA)]–infected vehicle control–treated litters (n = 26), and malaria (PbA)–infected l-arginine–supplemented litters (n = 36). Box plots depict median, and interquartile range with whiskers depicts maximum and minimum values. (B) Percentage of viable pups per litter in uninfected vehicle control–treated litters (n = 22), uninfected l-arginine–supplemented litters (n = 37), malaria (PbA)–infected vehicle control–treated litters (n = 26), and malaria (PbA)–infected l-arginine–supplemented litters (n = 36). The figure depicts mean ± SD. Results of independent samples t test (fetal weight) and χ2 test (viability); *P < 0.05 and **P < 0.001.

  • Fig. 5 l-Arginine, ADMA, and SDMA concentrations are altered in experimental malaria in pregnancy.

    (A) Malaria infection is associated with decreased l-arginine serum concentrations, as measured by mass spectrometry. (B and C) Malaria-infected dams receiving l-arginine supplementation show reduced serum concentrations of ADMA (B) and SDMA (C). Box plots depict median, and interquartile range with whiskers depicts maximum and minimum values. (D) l-Arginine/ADMA ratio in mice with or without malaria infection and/or l-arginine supplementation. Malaria-infected vehicle control–treated (Veh) (n = 18) and malaria-infected (PbA) l-arginine–treated (l-Arg) dams (n = 17) were compared with uninfected (UI) vehicle control (n = 14) and uninfected l-arginine–treated dams (n = 16). Figures depict mean ± SD. Results of one-way analysis of variance (ANOVA) and Tukey post test; *P < 0.05, **P < 0.01, and ***P < 0.001.

  • Fig. 6 Malaria and l-arginine induce changes in the expression of inflammatory and angiogenic mediators in placental tissue by reverse transcription PCR.

    (A to H) C5 (A), C5ar (B), Icam-1 (C), Tie-2 (D), Ang-1 (E), Ang-2 (F), Vegf-a (G), and Flt-1 (VEGF receptor) (H) in uninfected (UI) vehicle control–treated (Veh) dams (n = 12), uninfected (UI) l-arginine–treated (l-Arg) dams (n = 12), malaria-infected (PbA) vehicle control–treated dams (n = 12), and malaria-infected l-arginine–treated dams (n = 12). Box plots depict median, and interquartile range with whiskers depicts maximum and minimum values. Results of one-way ANOVA and Tukey post test; *P < 0.05, **P < 0.01, and ***P < 0.001.

  • Fig. 7 l-Arginine supplementation increases the number of small vessels in placentas from malaria-infected litters.

    (A and B) Representative micro-computed tomography images of fetoplacental arterial vasculature at gestational day 18 in placentas from malaria-infected vehicle control–treated (A) and l-arginine–treated (B) mice color-coded by vessel diameter. (C) Cumulative distribution of vessel diameters in placentas from uninfected vehicle control–treated (n = 7), uninfected l-arginine–treated (n = 7), malaria-infected vehicle control–treated (n = 8), and malaria infected l-arginine–treated (n = 7) litters. Cumulative vessel segments are depicted as median and SEM of vessels larger than the threshold diameter (0.035 mm) with results of two-way ANOVA and Dunn’s multiple comparison post hoc test; ***P < 0.001.

  • Fig. 8 Dietary l-arginine supplementation improves birth outcomes in malaria-infected dams receiving l-arginine–deficient chow.

    (A) Fetal weight (in grams) of uninfected vehicle control–treated (Veh) litters (n = 22) and uninfected l-arginine–supplemented (l-Arg) litters on regular chow (n = 36), as well as uninfected vehicle control–treated litters (n = 10), uninfected l-arginine–supplemented litters (n = 10), malaria (PbA)–infected vehicle control–treated litters (n = 12), and malaria (PbA)–infected l-arginine–supplemented litters (n = 11) on l-arginine–deficient chow. Box plots depict median, and interquartile range with whiskers depicts maximum and minimum values. (B) Percentages of viable pups per litter in the groups of mice shown in (A). The figure depicts mean ± SD. Results of one-way ANOVA and Tukey post test; *P < 0.05 and **P < 0.01.

  • Table 1 Characteristics of pregnant women enrolled in a prospective cohort study in southern Malawi.

    Data are presented as means (SD) or n (%), unless otherwise indicated. RR, relative risk; CI, confidence interval; BMI, body mass index; MUAC, middle-upper arm circumference; ADMA, asymmetric dimethylarginine; SDMA, symmetric dimethylarginine.

    nPopulation
    (n = 384)
    Longitudinal
    population
    (n = 94)
    Bivariate analysisMultivariate analysis*
    AGA
    (n = 302)
    SGA
    (n = 82)
    PRR (95% CI)P
    Maternal characteristics
      Age (years)38418.0 (1.8)18.0 (1.8)18.0 (1.8)18.1 (1.7)0.8401.05 (0.94–1.16)0.407
      Height (cm)384153 (4.3)154 (4.8)153 (4.4)152 (4.1)0.216
      Weight (kg)38153.4 (6.3)54.1 (7.4)53.8 (6.3)51.8 (6.2)0.011
      BMI (kg/m)38122.8 (2.6)22.8 (3.0)22.9 (2.6)22.2 (2.6)0.0430.93 (0.86–1.01)0.089
      MUAC (cm)23723.5 (1.6)23.8 (1.8)23.5 (1.6)23.4 (1.6)0.680
      Socioeconomic
    status (tertile)
        1381135 (35.4)34 (36.2)110 (36.8)25 (30.5)0.148Reference
        2111 (29.1)20 (21.3)80 (26.8)31 (37.8)1.65 (1.04–2.64)0.035
        3135 (35.4)40 (42.6)109 (36.5)26 (31.7)1.17 (0.70–1.94)0.553
      Education status
    (tertile)
        138168 (17.9)14 (14.9)57 (19.1)11 (13.4)0.471
        2229 (60.1)56 (59.6)176 (58.9)53 (64.6)
        384 (22.1)24 (25.5)66 (22.1)18 (22.0)
      Treatment arm (IPTp),
    n (%)
    382199 (51.8)40 (42.6)158 (52.3)41 (50.0)0.7101.05 (0.71–1.54)0.808
      Hemoglobin (g/dl)37610.3 (1.4)10.0 (1.4)10.3 (1.4)10.1 (1.5)0.272
      Gestational age at
    enrollment (weeks)
    38420.4 (3.1)20.9 (3.4)20.6 (3.0)19.7 (3.2)0.0320.93 (0.86–0.99)0.036
    Malaria at enrollment
      Microscopy, n (%)37991 (24.0)21 (23.1)71 (23.8)20 (24.7)0.8710.96 (0.60–1.55)0.879
      PCR, n (%)381217 (57.0)57 (61.3)171 (57.2)46 (56.1)0.859
      Submicroscopic,
    n (%)
    285137 (48.1)37 (53.6)108 (48.2)29 (47.5)0.926
    Arginine pathway at enrollment
      l-Arginine (μM)37932.3 (19.6)36.8 (17.0)31.1 (19.6)32.9
    (19.6)
    0.750
      ADMA (μM)3790.44 (0.08)0.49 (1.0)0.44 (0.08)0.46
    (0.10)
    0.03021.2 (2.27–197.9)0.007
      SDMA (μM)3790.37 (0.07)0.40 (0.07)0.37 (0.07)0.38
    (0.07)
    0.632
      l-Arginine/ADMA37975.3 (49.4)79.8 (44.0)75.1 (48.3)75.9
    (53.7)
    0.902
    Birth outcomes
      Infant sex (female)384188 (49.0)42 (44.7)145 (48.0)43 (52.4)0.477
      Birth weight (g)3842781 (427)2759 (456)2859 (428)2493
    (272)
    <0.0001
      Gestational age at
    delivery (weeks)
    38438.1 (2.3)38.1 (2.2)37.7 (2.3)39.2 (1.7)<0.0001
    Placental malaria
      Histology356171 (48.0)45 (51.1)124 (44.6)47 (60.3)0.014

    *Log-binomial model.

    †Median (interquartile range). Bivariate analysis by Student’s t test or Wilcoxon rank sum test.

    ‡Submicroscopic malaria is defined as polymerase chain reaction (PCR)–positive and microscopy-negative. Participants who were positive by microscopy were excluded (n = 91, where n = 15 were positive by microscopy but negative by PCR), as were those who were missing microscopy results (n = 5) or missing PCR results (n = 4).

    • Table 2 Association between the nitric oxide biosynthetic pathway and malaria at enrollment.

      Malaria at enrollment is determined in maternal blood at enrollment, data are presented as means (SD) and analyzed using Student’s t test, and log-binomial regression was used to obtain relative risk and corresponding 95% confidence intervals adjusting for maternal age and gestational age at enrollment.

      Malaria-negativeMalaria-positivePUnadjusted model, RR
      (95% CI)
      Adjusted model, RR
      (95% CI)
      Microscopy-defined
      malaria
      n = 285n = 89
        l-Arginine (μM)32.3 (19.2)32.8 (21.2)0.9261.00 (0.99–1.01)1.00 (0.99–1.01)
        ADMA (μM)0.44 (0.09)0.46 (0.08)0.0446.72 (1.05–43.10)6.18 (0.93–40.96)
        SDMA (μM)0.37 (0.07)0.39 (0.06)0.02912.00 (1.24–116.16)24.23 (2.09–280.90)
        l-Arginine/ADMA76.0 (48.4)73.6 (53.5)0.6981.00 (1.00–1.00)1.00 (1.00–1.00)
      PCR-defined malarian = 163n = 214
        l-Arginine (μM)33.9 (21.2)31.1 (18.2)0.1601.00 (0.99–1.00)0.996 (0.987–1.005)
        ADMA (μM)0.42 (0.08)0.46 (0.08)<0.00016.74 (4.13–11.01)6.40 (3.72–11.04)
        SDMA (μM)0.36 (0.07)0.38 (0.07)0.0094.24 (1.49–12.08)3.94 (1.61–9.66)
        l-Arginine/ADMA83.7 (56.0)69.0 (43.0)0.0040.997 (0.994–0.999)0.997 (0.994–0.999)
      Submicroscopic
      malaria*
      n = 147n = 136
        l-Arginine (μM)34.0 (21.7)30.6 (16.1)0.1430.995 (0.988–1.002)0.993 (0.982–1.005)
        ADMA (μM)0.41 (0.08)0.47 (0.09)<0.000111.60 (6.03–22.32)10.29 (5.04–21.02)
        SDMA (μM)0.36 (0.07)0.38 (0.07)0.0376.02 (1.21–29.90)8.44 (1.77–40.29)
        l-Arginine/ADMA84.7 (57.8)66.8 (33.7)0.0020.995 (0.992–0.999)0.995 (0.992–0.999)

      *Submicroscopic malaria is defined as PCR-positive and microscopy-negative. Women were excluded if they were missing microscopy results (n = 5) or PCR results (n = 4) or were microscopy-positive (n = 91).

      • Table 3 Regression analysis examining the nitric oxide biosynthetic pathway and maternal nutritional status.

        Linear regression analysis was used, adjusting for maternal age and gestational age at enrollment.

        BMI (kg/m2)MUAC (cm)Hemoglobin (g/dl)
        Unadjusted
        model, β (SE)
        Adjusted model,
        β (SE)
        Unadjusted
        model, β (SE)
        Adjusted model,
        β (SE)
        Unadjusted
        model, β (SE)
        Adjusted model,
        β (SE)
        l-Arginine (μM)0.005 (0.007),
        P = 0.432
        0.004 (0.007),
        P = 0.523
        0.018 (0.008),
        P = 0.021
        0.024 (0.009),
        P = 0.005
        0.001 (0.004),
        P = 0.818
        0.001 (0.004),
        P = 0.842
        ADMA (μM)−1.23 (1.56),
        P = 0.431
        −1.07 (1.56),
        P = 0.495
        −1.42 (1.20),
        P = 0.241
        −1.09 (1.23),
        P = 0.373
        −6.06 (0.82),
        P < 0.0001
        −6.01 (0.83),
        P < 0.0001
        SDMA (μM)−1.01 (1.90),
        P = 0.597
        −2.63 (1.95),
        P = 0.180
        −0.64 (1.50),
        P = 0.669
        −0.304 (1.57),
        P = 0.845
        −3.80 (1.06),
        P < 0.0001
        −3.80 (1.10),
        P = 0.001
        l-Arginine/ADMA0.003 (0.003),
        P = 0.255
        0.002 (0.003),
        P = 0.357
        0.008 (0.003),
        P = 0.010
        0.010 (0.003),
        P = 0.003
        0.003 (0.001),
        P = 0.060
        0.003 (0.001),
        P = 0.072

      Supplementary Materials

      • www.sciencetranslationalmedicine.org/cgi/content/full/10/431/eaan6007/DC1

        Table S1. Linear mixed-effects modeling of longitudinal changes in ADMA and SGA.

        Table S2. Dams’ peripheral parasitemia at G19 and litter size from all cohorts.

        Table S3. Fetal and placental weight by treatment group.

        Table S4. Fetal viability by treatment group.

        Table S5. RT-PCR primer sequences (5′ to 3′).

      • Supplementary Material for:

        Malaria in pregnancy alters L-arginine bioavailability and placental vascular development

        Chloe R. McDonald, Lindsay S. Cahill, Joel L. Gamble, Robyn Elphinstone, Lisa M. Gazdzinski, Kathleen J. Y. Zhong, Adrienne C. Philson, Mwayiwawo Madanitsa, Linda Kalilani-Phiri, Victor Mwapasa, Feiko O. ter Kuile, John G. Sled, Andrea L. Conroy, Kevin C. Kain*

        *Corresponding author. Email: kevin.kain{at}uhn.ca

        Published 7 March 2018, Sci. Transl. Med. 10, eaan6007 (2018)
        DOI: 10.1126/scitranslmed.aan6007

        This PDF file includes:

        • Table S1. Linear mixed-effects modeling of longitudinal changes in ADMA and SGA.
        • Table S2. Dams’ peripheral parasitemia at G19 and litter size from all cohorts.
        • Table S3. Fetal and placental weight by treatment group.
        • Table S4. Fetal viability by treatment group.
        • Table S5. RT-PCR primer sequences (5′ to 3′).

        [Download PDF]