A novel murine model of post-implantation malaria-induced preterm birth.

Despite major advances made in malaria treatment and control over recent decades, the development of new models for studying disease pathogenesis remains a vital part of malaria research efforts. The study of malaria infection during pregnancy is particularly reliant on mouse models, as a means of circumventing many challenges and costs associated with pregnancy studies in endemic human populations. Here, we introduce a novel murine model that will further our understanding of how malaria infection affects pregnancy outcome. When C57BL/6J (B6) mice are infected with Plasmodium chabaudi chabaudi AS on either embryonic day (E) 6.5, 8.5, or 10.5, preterm birth occurs in all animals by E16.5, E17.5, or E18.5 respectively, with no evidence of intrauterine growth restriction. Despite having the same outcome, we found that the time to delivery, placental inflammatory and antioxidant transcript upregulation, and the relationships between parasitemia and transcript expression prior to preterm birth differed based on the embryonic day of infection. On the day before preterm delivery, E6.5 infected mice did not experience significant upregulation of the inflammatory or antioxidant gene transcripts examined; however, peripheral and placental parasitemia correlated positively with Il1β, Cox1, Cat, and Hmox1 placental transcript abundance. E8.5 infected mice had elevated transcripts for Ifnγ, Tnf, Il10, Cox1, Cox2, Sod1, Sod2, Cat, and Nrf2, while Sod3 was the only transcript that correlated with parasitemia. Finally, E10.5 infected mice had elevated transcripts for Ifnγ only, with a tendency for Tnf transcripts to correlate with peripheral parasitemia. Tumor necrosis factor deficient (TNF-/-) and TNF receptor 1 deficient (TNFR1-/-) mice infected on E8.5 experienced preterm birth at the same time as B6 controls. Further characterization of this model is necessary to discover the mechanism(s) and/or trigger(s) responsible for malaria-driven preterm birth caused by maternal infection during early pregnancy.

Introduction

Malaria remains a global public health issue despite unprecedented successes in vector control programs and antimalarial treatment. In 2020, the World Health Organization (WHO) reported that in regions with moderate to high Plasmodium falciparum transmission, roughly 11 million pregnancies were exposed to malaria infection [1]. Among those pregnancies, an estimated 819,000 infants were born with low birth weight, a well-documented risk factor for neonatal mortality [1]. Additional maternal-fetal health consequences to infection include maternal anemia and preterm delivery, especially in primigravid women [2]. Notably, malaria infection during pregnancy can manifest as an organ-specific syndrome known as placental malaria (PM), identified by the accumulation of Plasmodium-infected red blood cells (IRBCs), immune cell infiltration, and both malaria pigment (hemozoin) and fibrin deposition in the placenta [3-5]. To protect pregnant women from PM, WHO recommends the use of insecticide-treated mosquito nets (ITNs) and intermittent preventative treatment using antimalarial drugs such as sulfadoxine-pyrimathamine (IPTp-SP). However, in 2018 only 31% of women received the recommended dosages of IPTp-SP during their pregnancy and ITN coverage in sub-Saharan Africa has stalled since 2015 [6].

Despite decades of discovery and billions of dollars invested in malaria research, the underlying mechanisms involved in PM pathogenesis are still incompletely understood. One major reason is that access to the placenta is limited until after delivery, making it difficult to evaluate the impact of infection on the placenta during earlier stages of pregnancy. As a result, most of what is known about human PM comes from studies focused on clinical outcomes throughout gestation, systemic evaluation for biomarkers of disease, and placental biopsies postpartum. Some studies provide evidence for the critical role of inflammation, specifically via the upregulation of tumor necrosis factor (TNF), interleukin-10 (IL-10), interleukin-1 beta (IL-1β), and interferon-gamma (IFN-γ), in driving poor pregnancy outcomes [7-13]. Other studies suggest that oxidative stress, resulting from a cell’s inability to mount an effective antioxidant response to mitigate damage driven by reactive oxygen species (ROS), is a key player in the placental pathology and poor fetal health outcomes associated with PM [14, 15].

Mouse models have been central to furthering our understanding of PM pathogenesis by providing a more accessible and genetically tractable tool for recapitulating malaria infection during pregnancy. Infection with the murine-infective Plasmodium chabaudi chabaudi AS (Pcc) in C57BL/6J (B6) and Swiss Webster mice capture some of the important hallmarks of human PM, such as elevated proinflammatory cytokine production, accumulation of IRBCs in the placenta, increased fibrin deposition, and pregnancy loss [16-20]. In these models, mice infected with Pcc on the first day of gestation (embryonic day 0.5, E0.5) experience a non-lethal infection that is accompanied by severe maternal anemia, high parasite burden, elevated inflammatory responses, hemozoin accumulation, and lipid peroxidation in the placenta, a known sign of oxidative stress. Therapeutic interventions that target the pathogenic contributions of coagulation, TNF, and ROS in B6 mice protect against pregnancy loss at midgestation (E10.5–12.5) [5, 19, 20]. Alternatively, studies in later gestation commonly utilize Plasmodium berghei in either B6 or BALB/c mice; however, the infection must be initiated between E10.5–13.5, due to the maternal lethality of the parasite. In these models, mice experience elevated placental inflammation and fibrin deposition, oxidative stress, IRBC adherence to the placenta, and preterm delivery [21-24]. Moreover, antimalarial and anti-inflammatory drug treatment restore maternal survival, reduce oxidative stress, and improve pregnancy outcomes [22, 25].

While these well-established models are useful for studying PM pathogenesis, our understanding of how the timing of malaria infection determines maternal and fetal health outcomes remains unclear. Studies of women naturally exposed to malaria at various times during gestation report associations between peripheral and placental parasitemia and poor birth outcomes such as low birth weight, preterm delivery, and neonatal mortality [26-29], with some studies that focus on infection early in the first trimester [30-33]. However, due to a lack of histological and pathophysiological information related to PM pathogenesis during the first and second trimesters of pregnancy, it is difficult to envision host-directed therapeutic interventions that mitigate the impact of infection during these times in gestation. This represents a major obstacle in PM prevention and treatment efforts, especially amidst emerging antimalarial drug and insecticide resistance [34, 35].

Here, we describe a mouse model for studying malaria infection during early stages of pregnancy (early post-implantation) using Pcc in B6 mice. Given the developmental similarities between the first trimester human placenta and the last two-third of mouse gestation [36, 37], this model offers an avenue for investigating the pathogenic mechanisms underlying malaria infection during early placentation, when P. falciparum prevalence in pregnant women is highest [2] and fetal health is significantly impacted [27, 30, 33] but the placenta remains inaccessible to study. This work shows that malaria infection during the post-implantation period induces preterm birth and alters the expression of inflammatory and antioxidant-related gene transcripts in the placenta, one day prior to preterm delivery.

Materials and methods

Mice

C57BL/6J (B6), TNFα-knockout (B6;129S-Tnftm1Gkl/J), TNFRp55-deficient (Tnfrsf1atm1Mak/J), and A/J mice were purchased from the Jackson Laboratory (Bar Harbor, ME) and maintained by brother-sister mating for a maximum of ten generations in the University of Georgia Coverdell Vivarium, following guidelines and regulations set forth by the University of Georgia Animal Care and Use Committee. All animals were supplied food (PicoLab® Rodent Diet 20: 5030, St. Louis, MO) and water ad libitum. Mice were adjusted to a 14-hour light/10-hour dark cycle and housed in conditions of 65–75 ˚F and 40–60% humidity. All animal procedures reported in this study were reviewed and approved by the Institutional Animal Care Use Committee (IACUC) at the University of Georgia, protocol number A2018 02-016-Y1-A0. Mice were anesthetized with 2.5% Tribromoethanol before sacrifice and all efforts were made to minimize suffering.

Parasites and infection monitoring

The following reagent was obtained through BEI Resource Repository, NIAID, NIH: Plasmodium chabaudi chabaudi, Strain AS, MR4-741, contributed by David Walliker. Parasites were maintained as frozen stock in accordance with supplier guidelines and passaged in A/J mice for the purposes of infecting experimental B6 mice. Peripheral parasitemia was assessed by flow cytometry with a method adapted from work published by Jimenez-Diaz et al. [38]. A 2μl blood sample was collected by tail clip [39], diluted in 98μl 0.9% NaCl and stained with 0.25μl SYTO-16 Green Fluorescent Nucleic Acid Stain (ThermoFisher Scientific, catalog # S7578) within 4 hours of collection. Stained samples were incubated in the dark for 20 minutes at room temperature, further diluted 1:9 in 0.9% NaCl, then analyzed using a CyAn ADP Flow Cytometer (Beckman Coulter; Brea, CA). 30,000 cells were assessed daily for each mouse; infected red blood cells were distinguished based upon size and fluorescence intensity. An uninfected blood sample was used as an internal negative control. Parasitemia is reported as the percentage of infected red blood cells (IRBCs) to the total number of red blood cells (RBCs).

Experimental design

Female B6 mice aged 8–10 weeks were paired with age-matched males nightly and examined each morning until a vaginal plug was observed, indicating successful mating. The morning a vaginal plug was observed was considered embryonic day 0.5, E0.5. After baseline measurements of weight and hematocrit were recorded, females were left undisturbed until E6.5 to minimize stress and increase the chances of successful blastocyst implantation. Mice were infected intravenously with 1000 P. chabaudi chabaudi AS-iRBCs diluted in 200ul 1X phosphate-buffered saline (PBS) per 20 grams of body weight on E6.5, E8.5, or E10.5 and are termed infected pregnant (IP). Age-matched non-pregnant mice were infected similarly as infection controls (infected non-pregnant, INP). In another control group, uninfected pregnant (UP) mice were sham injected with 200ul PBS per 20 grams body weight on the same gestation days. Immediately prior to infection or sham infection, experimental animals were switched to a high-fat rodent chow (PicoLab Mouse Diet 20 5058; St. Louis, MO) suited for pregnant animals. Weight measurements were recorded on E0.5, E6.5, E8.5, E10.5, and E12.5-E18.5 for all groups, depending on when the infection was initiated, to assess pregnancy progress and allow mice to proceed to spontaneous delivery. Parasitemia and hematocrit (a measure of anemia) were monitored daily in the infected groups beginning five days post-infection to assess the development of infection.

In a second series of experiments, mice infected on E6.5, E8.5, and E10.5, and along with their UP controls, were sacrificed on E15.5, E16.5, and E17.5, respectively. Weight, hematocrit, and parasitemia were recorded as described above until euthanasia. Mice were anesthetized before sacrifice and placentae were collected and preserved for histological and quantitative real-time PCR (RT-qPCR) analysis. Plump, pink, well-vascularized pups that reflexively responded to touch were considered viable and their weights and placenta weights were collected. Pup viability data are summarized in S1 Table.

Histology

Placentae collected at the time of sacrifice were fixed for 24 hours in 10% buffered formalin, processed, and paraffin-embedded. Placental sections 5μm in thickness were mounted to microscope slides and stained with hematoxylin and eosin (H&E) for histological analysis. H&E images were obtained using a Keyence BZ-X710 with BZ Analyzer software and the 100X oil objective. Sections were Giemsa-stained and parasite burden in the placenta was determined by counting at least 1,000 erythrocytes in maternal blood sinusoids of at least two or more different placentae per dam, as previously described [19].

Gene expression by quantitative real-time PCR

Total RNA from mouse placentae collected on E15.5, E16.5, and E17.5 was isolated using Trizol Reagent (Ambion, Ref # 15596026) and a bead shaker (BeadBlaster 24, Benchmark Scientific, SKU: D2400) with a minimum of four placentae were pooled per dam. RNA was DNase-treated (Invitrogen, Ref # AM1906) and then reverse-transcribed with High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Ref # 4368814). Relative transcript abundance for the genes of interest was quantified using PowerSYBR Green PCR Master Mix (Applied Biosystems, Cat # 4367659), the Roche LightCycler 96 Instrument (software version 1.01.01.0050), and the Mic qPCR cycler (Biomolecular systems, firmware version 2.25). Each sample was assayed in duplicate for target and housekeeping genes. Average Ct values of target genes were normalized to average Ct values of Ubc as the reference gene and relative transcript abundance of genes of interest was determined using the ΔΔCt method. Transcript expression in individual mice is presented relative to the mean expression value in UP mice at E6.5. Details of primer sets are summarized in S2 Table.

Statistics

All statistical analyses were performed using GraphPad Prism version 9.2.0 (GraphPad Software; La Jolla, California). All raw clinical data are presented as mean ± SEM. Error bars are not visible if they are shorter than the symbol’s height. The area under the curve (AUC) of percent starting weight, hematocrit, and parasitemia was calculated for each mouse between E0.5 and E18.5, as appropriate. AUC for weight and hematocrit was compared between IP, INP, and UP mice infected on the same day and was analyzed using a Kruskal-Wallis test between IP and INP groups and IP and UP groups. AUC for parasitemia between IP and INP groups was compared using a two-tailed unpaired t-test with Welch’s correction. RT-qPCR data were analyzed using an unpaired t-test with Welch’s correction and presented as a scatterplot with a bar representing the mean. Parasitemia and transcript data for correlation analyses were log-transformed. P values less than or equal to 0.05 were considered statistically significant. Mixed linear models analysis (SAS 9.4) was used to estimate differences in fetal and placental weights in dams sacrificed on E15.5, E16.5, and E17.5, and their controls. Proportional analysis tested by chi-square was used to compare pup viability between IP and UP dams. Univariate and multivariate regression analyses were done in SAS 9.4.

Results

In this observational study, Pcc blood-stage infection was initiated on E6.5, 8.5, or 10.5 in B6 mice. These time points coincide with a period of intense placental growth and vascularization in human pregnancy, which represents the initial point at which placental sequestration of P. falciparum is proposed to occur [40, 41]. Initial measurements of weight and hematocrit were taken on E0.5 and then mice were randomly assigned to one of six groups–E6.5 infected pregnant (IP), E8.5 IP, E10.5 IP, E6.5 uninfected pregnant (UP), E8.5 UP, and E10.5 UP, with the indicated embryonic day representing the day of infection for IP groups (Fig 1A). After the infection was initiated, daily measurements of weight and parasitemia were collected beginning 5 days post-infection (Fig 1B–1G). In each group, IP mice steadily gained weight alongside their UP controls until they experienced a sudden and precipitous decline in weight, suggestive of preterm pregnancy compromise (Fig 1B–1D). Euthanasia of these dams at E19 revealed empty uteri, implicating preterm birth as the likely cause of the observed weight loss. The time to preterm birth was accelerated the later in gestation that infection occurred such that E6.5-, E8.5-, and E10.5-infected dams began losing weight nine, eight, and seven days post-infection, respectively (Fig 1B–1D). UP mice continued gaining weight beyond E18.5, as expected during a normal pregnancy. Parasitemia was greater in IP dams compared to their infected nonpregnant (INP) counterparts (Fig 1E–1G), which was confirmed statistically through area under the curve (AUC) analysis, in all groups except the E8.5 infection group (S1 Fig). Corresponding with higher parasite burdens, AUC for hematocrit was significantly lower in E6.5 infected dams compared to UP controls (S1 Fig), likely due to the prolonged destruction of red blood cells caused by the advanced stage of infection compared to other infection groups [42]. AUC for hematocrit in the other infection groups, E8.5 and E10.5, was not statistically significantly different from UP controls (S1 Fig).

Fig 1

P. chabaudi AS infection in the post-implantation period universally precipitates preterm delivery.

(A) Schematic of experimental design. (B-D) Percent starting weight and (E-G; % IRBCs) parasitemia are presented for infected pregnant (IP), uninfected pregnant (UP), and infected non-pregnant (INP) groups. All IP mice experienced precipitous weight loss during ascending infection. Arrows indicate timepoints chosen for subsequent experiments. (B, E) E6.5 infection group (red): IP, n = 6; UP, n = 6; and INP, n = 4. (C, F) E8.5 infection group (blue): IP, n = 8; UP, n = 6; INP, n = 4. (D, G) E10.5 infection group (green): IP, n = 14; UP, n = 6; INP, n = 7.

To investigate the effect of maternal infection on placental and pup health in this model, dams infected on E6.5, E8.5, or E10.5 were euthanized one day prior to expected preterm parturition on E15.5, E16.5, and E17.5 respectively (Fig 1B–1D, arrows indicate the day of euthanasia when placentae were collected for analysis). Daily weight, parasitemia, and hematocrit measurements were recorded, beginning 5 days post-infection (S2 and S3 Figs). Placental parasitemia in dams of the E8.5 and E10.5 infection groups, but not in those infected at E6.5, was significantly higher than their peripheral parasitemia at sacrifice (Fig 2). To further examine the impact of infection, pup and placental weights and pup viability were measured one day before expected preterm parturition. By mixed linear models analysis, neither pup nor placental weights were significantly impacted by maternal malaria across all infection groups; graphs are shown for visualization purposes only (Fig 3). Pup viability was also not different between IP versus UP dams, regardless of the infection group (S1 Table). When holding infection constant, increased pup viability in mice infected at E10.5 yielded a significant reduction in pup weight (by 0.7546 g, P = 0.0102), consistent with the well-documented observation that increased litter sizes result in reduced fetal weights [43]. Overall, these results indicate that despite having significant parasite burden, including in the placenta, malaria-induced preterm birth in this model occurs independently of changes in placental and fetal weight and viability.

Fig 2

P. chabaudi AS-infected red blood cells accumulate prior to preterm delivery in the placentae of mice infected at E8.5 and E10.5.

Peripheral and placental parasitemia from mice infected on E6.5 (red), E8.5 (blue) and E10.5 (green), sacrificed at E15.5, E16.5 and E17.5, respectively. **P < 0.002, unpaired t-test with Welch’s correction; ns = not significant, P > 0.05.

Fig 3

P. chabaudi AS infection does not significantly impact placenta or pup weight prior to preterm delivery.

(A) Placenta and (B) pup weights depicted for uninfected pregnant (UP) and infected pregnant (IP) mice sacrificed one day before preterm delivery. These data are shown for visualization only; statistical analysis is in the text. E15.5 group: 4 IP dams, 32 placentae; 4 UP dams, 30 placentae. E16.5 group: 4 IP dams, 31 placentae; 4 UP dams, 33 placentae. E17.5 group: 5 IP dams, 40 placentae; 5 UP dams, 41 placentae.

Histology in PM-positive human placenta observed at term is generally characterized by immune cell infiltration, IRBC accumulation and both fibrin and hemozoin deposition in placental tissues [3-5]. To assess the histological condition of the placenta prior to preterm delivery in our model, placentae collected from IP mice and their UP controls were stained with hematoxylin and eosin (H&E) (Fig 4). Representative images of placentae from each group depict IRBCs in the placentae of all infection groups (Fig 4B, 4D and 4F). Giemsa-stained tissue sections further demonstrate the presence of IRBCs in all infection groups (S4 Fig). There was no evidence of immune cell accumulation or excessive fibrin deposition within the labyrinth in any of the placentae examined.

Fig 4

P.chabaudi AS-infected red blood cells in the placental labyrinth region one day prior to preterm delivery.

(A, C, E) micrographs taken with 100x objective lens from uninfected pregnant (UP) controls at E15.5, E16.5, and E17.5. (B, D, F) micrographs from infected pregnant (IP) dams sacrificed one day prior to preterm delivery on E15.5, E16.5, and E17.5, respectively.

The pathogenesis of preterm delivery has long been associated with excessive inflammation, induced by infection or other disorders [44]. To probe the importance of inflammatory mediators in driving preterm delivery in these models, real-time quantitative PCR (RT-qPCR) analysis was performed using RNA isolated from the placental homogenates. The relative abundance of inflammatory gene transcripts, such as tumor necrosis factor (Tnf), interleukin-10 (Il10), interferon-gamma (Ifnγ), interleukin beta (Il1β), and genes important in pregnancy maintenance and parturition, such as cyclooxygenases 1 and 2 (Cox1 and Cox2) were measured (Fig 5). Transcript abundance in placentae collected on E15.5 from mice infected on E6.5 revealed no significant differences between IP and UP dams for all inflammatory targets measured (Fig 5A–5F). However, Il1β transcripts were positively correlated with both peripheral parasitemia at euthanasia (E15.5) and peripheral parasitemia AUC for mice infected on E6.5 (Fig 6A and 6B). Additionally, there was a positive correlation between Cox1 transcripts and peripheral parasitemia at sacrifice (Fig 6F) and a weak tendency toward a positive correlation between Cox1 transcripts and parasitemia AUC for this same group (S5 Fig). In the E8.5 infection group, transcript abundance of all the inflammatory genes targeted, except for Il1β, was significantly elevated in placentae collected on E16.5 (Fig 5A–5F), but no correlative relationships with parasitemia at sacrifice or parasitemia AUC were found. In the E10.5 infection group, only transcripts for Ifnγ were elevated, while all other targets remained unchanged (Fig 5A–5F). There was a tendency for Tnf transcripts to be positively correlated with both peripheral parasitemia at sacrifice and parasitemia AUC in mice infected on E10.5 (S5 Fig).

Fig 5

P. chabaudi infection at E8.5 induces upregulation of inflammatory and parturition-associated gene transcripts in the E16.5 placenta prior to preterm delivery.

(A-F) Mouse Ifnγ, Tnf, Il10, Il1β, Cox1, and Cox2 mRNA abundances normalized to Ubc and quantified by qPCR in placentae taken from infected pregnant (solid color circles) and uninfected pregnant (open black circles) dams. Solid color circles denote pooled placentae from individual infected dams sacrificed on E15.5 (red), E16.5 (blue), and E17.5 (green). Open black circles are pooled placentae from uninfected pregnant controls for each group. Group means and transcript abundance in individual mice are depicted. **P ≤ 0.005, *P < 0.05, unpaired t-test with Welch’s correction; ns = not significant, P > 0.05; # denotes a trending result, P = 0.08.

Fig 6

Inflammatory and antioxidant gene expression in the E15.5 placenta positively correlates with peripheral and placental parasitemia in mice infected on E6.5.

(A-F) Mouse mRNA transcript abundance relative to peripheral parasitemia at the time of sacrifice, peripheral parasitemia AUC, and placental parasitemia one day before expected preterm birth (E15.5).

To determine the parameters most critical in determining inflammatory gene expression, linear regression modeling was performed, considering infection status, day euthanasia was performed, and, as observed at euthanasia, uterus weight, number of embryos and embryo viability. The latter three were considered together in multivariate analysis, and none were found to predict transcript abundance. Infection status significantly influenced all targets (S3 Table). Day of euthanasia also significantly influenced all targets except Cox1; relative to placentae from mice infected at E6.5 (i.e., E15.5), transcripts for Ifnγ, Tnf, Il1β, Il10, and Cox2 in placentae from E8.5 infections (i.e., E16.5) were all elevated. In a multivariate model testing both status and day, membership in the E16.5 group was a significant predictor for Ifnγ, Tnf, Il1β, Il10, and Cox2 abundance (p ≤ 0.0306; S5 Table), while holding infection status constant. Similarly, while holding day constant, infection status maintained a significant influence on Ifnγ, Il1β, Il10, and Cox2 abundance (p ≤ 0.0273), with tendencies for Tnf and Cox1 (p = 0.0569 and p = 0.0645, respectively; S5 Table).

TNF has long been implicated as one of the major contributors to poor pregnancy outcome in malaria-positive primigravid women and in other infections [11, 45, 46]. Moreover, in a related mouse model, in which infection is initiated at E0.5, antibody-mediated ablation of TNF resulted in midgestational rescue from PccAS-induced abortion [19]. To directly evaluate a link between placental inflammation driven by TNF and subsequent precipitation of preterm delivery in the current mouse model, mice deficient in tumor necrosis factor (TNF-/-) and tumor necrosis factor receptor 1 (TNFRI-/-) were infected with 1000 Pcc-IRBCs on E8.5. This time point was chosen based on the multivariate modeling which suggested that Tnf transcripts are elevated prior to preterm delivery (after E16.5) in B6 mice. Contrary to expectation, both TNF-/- and TNFRI-/- IP mice experienced preterm delivery after E16.5, similar to wildtype B6 mice (Fig 7). Although parasitemia appears to be higher in TNF-/- IP compared to B6 mice, AUC analysis revealed no significant differences between the groups for either TNF-/- or TNFRI-/- IP dams (S6 Fig). Taken together, these data demonstrate that TNF and TNF signaling through TNFRI are not required to drive preterm delivery in mice infected on E8.5.

Fig 7

Tumor necrosis factor (TNF-/-) and tumor necrosis factor receptor 1 (TNFRI-/-) deficient mice experience preterm delivery following P. chabaudi infection on E8.5.

(A) Percent starting weight for TNF-/- and (B) TNFRI-/- mice are depicted for infected pregnant (IP) and infected non-pregnant (INP) groups. TNF-/- IP: n = 3, INP n = 3; TNFRI-/- IP: n = 7, INP n = 3; B6 IP: n = 5. All IP mice experience precipitous weight loss, indicating preterm pregnancy compromise.

Oxidative stress, generally described as an imbalance between the formation of free radicals, such as reactive oxygen species (ROS), and antioxidant defense molecules within cells and tissues, leads to cellular damage to lipids, proteins, RNA, and DNA. Recently, oxidative stress has been documented to play a crucial role in pregnancy outcome in some mouse models of PM [20, 23, 24] and is implicated in women experiencing malaria during pregnancy [14, 15] as well as other types of pregnancy complications [44, 47, 48]. To interrogate the role of oxidative stress in this model, placentae were evaluated for alterations in antioxidant gene expression by RT-qPCR (Fig 8). In the placentae of B6 mice infected on E8.5, placental transcript abundance at E16.5 for superoxide dismutases 1 and 2 (Sod1, Sod2), catalase (Cat), and nuclear factor erythroid-2 related factor 2 (Nrf2) were significantly elevated compared to UP controls (Fig 7A, 7B, 7D and 7E). None of these targets were significantly elevated in IP mice infected and euthanized at the other time points (Fig 8). However, when the relationships between antioxidant transcript abundance and parasitemia were considered at E15.5 in the E6.5 infection group, Hmox1 and Cat positively correlated with placental parasitemia. Additionally, Hmox1 positively correlated with peripheral parasitemia AUC (Fig 6C–6E). In the E8.5 infection group, E16.5 placental Sod3 transcripts tended to correlate negatively with both peripheral parasitemia at sacrifice and parasitemia AUC (Fig 9). Significant correlations between antioxidant transcript expression and parasite burden were not found in the placenta at E17.5 in the E10.5 infection group.

Fig 8

P. chabaudi infection at E8.5 induces elevated antioxidant gene transcript expression in the E16.5 placenta prior to preterm delivery.

(A-F) Mouse Sod1, Sod2, Sod3, Cat, Nrf2, and Hmox1 mRNA abundance relative to Ubc and quantified by qPCR in placentae taken from infected pregnant (solid circles) and uninfected pregnant (open black circles) dams. Solid color circles denote pooled placentae from individual infected dams sacrificed on E15.5 (red), E16.5 (blue), and E17.5 (green). Open black circles are pooled placentae from uninfected pregnant controls for each group. Group means and transcript abundance in individual mice are depicted. ***P ≤ 0.005, **P = 0.02; ns = not significant, P > 0.05.

Fig 9

Placental Sod3 gene expression negatively correlates with peripheral parasitemia at sacrifice on E16.5 in mice infected on E8.5.

Mouse Sod3 mRNA transcript abundance measured in placentae collected one day before expected preterm birth (E16.5) relative to (A) peripheral parasitemia at the time of sacrifice and (B) peripheral parasitemia AUC.

Linear regression modeling was applied to assess factors important in driving differences in placental antioxidant gene expression. As with inflammatory markers, this analysis considered infection status, day euthanasia was performed, uterus weight, number of embryos and embryo viability. The latter three were considered together and did not predict antioxidant transcript abundance. Infection status drove significant increases for genes except Sod3 (p ≤ 0.04; S4 Table). Day of euthanasia also significantly influenced all targets except Sod 3 and Hmox1. Transcripts for Nrf2, Sod1, Sod2, and Cat in placentae from E8.5 infections (i.e., E16.5) were all significantly elevated relative to placentae from mice infected at E6.5 (i.e., E15.5; p ≤ 0.01; S4 Table). In a multivariate model testing influence of both status and day on these transcripts, holding day constant revealed a significant impact of infection was retained for Nrf2, Sod1, Sod2, Cat and Hmox1 (p ≤ 0.04; S6 Table). Membership in the E16.5 group independently predicted enhanced expression of Nrf2, Sod1, Sod2, and Cat (p ≤ 0.0254; S6 Table).

Discussion

Mouse models represent an affordable and genetically manipulable tool for investigating the pathogenesis of PM, a syndrome responsible for significant maternal morbidity and poor pregnancy outcomes globally. In this study, a novel murine model for malaria-induced preterm birth is described. This model requires a blood-stage infection in B6 mice using the murine infective parasite, Plasmodium chabaudi chabaudi AS (Pcc) at E6.5, E8.5, or E10.5. Although post-implantation Pcc infection in this model universally resulted in preterm birth, we observed that transcript abundance for certain genes in the placenta prior to delivery differed based on when in gestation the infection occurred. This model is the first to provide the flexibility to study malaria-induced pregnancy compromise driven by early post-implantation infection, without requiring acquired immunity or drug treatment for maternal survival. This is especially relevant to cases where immunity to PM is low or antimalarial treatment is not recommended, such as for primigravid women and those in their first trimester of pregnancy [13, 49].

To varying extents, key features of human PM are recapitulated in this model, including maternal anemia, higher parasitemia in the placenta compared to the periphery, and preterm delivery [4, 13]. Indeed, in the E8.5 and E10.5 infection groups, higher parasitemia in pregnant mice compared to non-pregnant controls corresponds with higher parasitemia in the placenta compared to peripheral parasitemia, corroborating previous studies in both mice and humans demonstrating that pregnancy itself can increase susceptibility to malaria [21, 24, 50] and lead to parasite accumulation in the placenta [3]. Thus, this novel model provides another opportunity for researchers to take a snapshot of the placenta at various times during the course of an acute malaria infection to investigate the relationships between parasite dynamics, placental development, and birth outcome.

Preterm delivery was a universal outcome of infection in the model reported here, with the time to delivery shortening the later in gestation the infection was introduced. When placental weight, pup weight, and pup viability were assessed one day prior to preterm delivery, the infection status of the dam had no impact. Thus, the physiological events that precipitate labor are acute. Analysis of the placenta preceding preterm delivery revealed differential expression of various inflammatory genes, highlighting the possibility that distinct mechanisms or triggers may be involved in driving preterm delivery in a gestational age-dependent manner in this model. Studies of cyclooxygenases, COX1 and COX2, have described their importance in the onset of labor in both mice and humans [51, 52]; however, Cox1 and Cox2 transcripts were elevated only at E16.5 in mice infected at E8.5. Interestingly, whereas Cox1 was positively correlated with parasitemia in the E6.5 infection group, no significant malaria-induced upregulation for either Cox1 or Cox2 relative to uninfected placenta was detected in the E10.5 infection group at E17.5 or in the E6.5 infection at E15.5, both sampled one day prior to expected preterm delivery. Consistent with this, multivariate modeling showed that Cox2 expression was influenced by infection only at E8.5, with a weak tendency for Cox1 to be influenced by infection only.

IFNγ is elevated during acute malaria infection and required for parasite control [53]. Ifnγ transcripts were elevated only in the E8.5 and E10.5 infection groups even though all IP mice were experiencing ascending parasitemia at the time that placental analyses were conducted. Additionally, Tnf and Il10 transcripts were increased in the E8.5 infection group only, and multivariate modeling confirmed that only infection at this day significantly predicted these increases. This was unexpected given the association of these cytokines with human malaria-induced preterm delivery [11, 54]. Likewise, TNF-/- and TNFR1-/- mice infected on E8.5 did not experience improved pregnancy outcomes or any changes in the time to preterm delivery. This was contrary to expectation given that ablation of TNF with antibody treatment successful protected mice from PccAS-induced abortion following infection at E0.5 [19]. These data suggest that although TNF signaling may be an important driver of poor pregnancy outcomes during human malaria infection [8, 19, 54] and early gestational malaria in the B6 mouse, its expression at the placental level is not required within the 24 hours preceding preterm delivery after infection at E8.5. Il1β transcript abundance positively correlated with parasitemia in the E6.5 infection group and multivariate modeling showed that infection at E8.5 increased placental transcription of this factor. This observation is consistent with another model of malaria infection, in which reduction of IL-1β signaling improved pregnancy outcome [22]; thus, IL-1β appears to be a universal driver of poor birth outcomes in malaria, perhaps through inflammasome assembly and/or the initiation of pyroptosis [55].

Oxidative stress has been described in various models of preterm delivery [56] and implicated in both human and mouse PM [14, 15, 20, 23, 24]. To corroborate these observations, antioxidant gene transcripts were measured in placentae collected one day prior to expected preterm delivery. Sod1, Sod2, Cat, and Nrf2 transcripts were elevated in the E8.5 infection group only, with Hmox1 expression being predicted only by infection but not by infection day. Similar results were reported with induction of PccAS infection in B6 mice at E0.5, where Sod1, Sod2, Cat, and Nrf2 transcripts were elevated in conceptuses and antioxidant drug treatment mitigated pregnancy loss [20]. When the relationship between parasite burden and antioxidant gene transcript expression was considered, Cat and Hmox1 transcripts were positively correlated with placental parasitemia at sacrifice and Hmox1 shared this relationship with peripheral parasitemia AUC as well. These results are consistent with the observed strong association between Hmox1 expression and hemozoin deposition in the placentae of PccAS-infected Swiss Webster dams at midgestation [57]. Since E6.5 IP mice exhibited significant anemia and the greatest interval between infection and preterm delivery, infection initiated at this timepoint may present another opportunity to study the impact of Hmox1 regulation of heme in driving negative pregnancy outcomes. Elevated heme levels and altered Hmox1 activity have been implicated in the pathogenesis of both human and murine PM [58-60] and other pregnancy complications [61]. For mice infected on E10.5, none of the antioxidant gene transcripts measured were differentially expressed or correlated with parasite burden. Since this group had an overall lower parasite burden relative to the other groups, these results may collectively indicate that a threshold of infectious burden is required to commensurate certain antioxidant transcript responses in this model.

Our analyses focused on placental inflammation and oxidative stress; however, there are other mechanisms of infection-induced preterm delivery that were not assessed in this model. Some biological functions of clinical relevance in PM that could be explored in future studies include the coagulation pathway [5, 62], the complement system [63], nutrient bioavailability [64] and transport across the placenta [65, 66], angiogenic balance and vascularization [67], placental autophagy [68, 69], and hormonal alterations [65]. Likewise, it is possible that pathological changes leading to preterm birth occurred in tissue types or sites not evaluated here, such as the uteroplacental interface.

Given the many remaining unresolved questions in PM pathogenesis, the model described here may be instrumental for future studies aimed at modeling maternal and/or fetal responses to malaria infection during pregnancy. Although Pcc infection was initiated on three distinct embryonic days in this study, the system might collectively be employed as a single model for malaria-induced pregnancy compromise because the time points assessed (infection and sacrifice days) have been reported to share functional and developmental similarities with 1st and early 2nd trimester human placental development [36, 37]. Malaria infection in the 1st trimester is associated with detrimental effects on pregnancy [2, 32, 70] and represents a period in gestation where the placenta cannot be sampled and preventative antimalarial treatment is not recommended [49]. Detailed reports on the histopathological impact of malaria infection during the 1st and early 2nd trimester of human pregnancy, which correspond to the time points modeled here, are currently unavailable. Thus, this model may provide a unique opportunity for researchers to gain insight into this underreported aspect of PM pathogenesis.

There are some limitations to this model and analysis as reported here. First, mouse gestation remains an imperfect model of human gestation, with a shorter gestational period and the birth of altricial young [37, 71–73]. However, the time points that we evaluate in the mouse reportedly share similarities with first trimester of human placentation [36, 37]. Second, this model did not yield evidence of intervillositis, placental fibrin deposits, intrauterine restriction, or low birth weight, all well-documented features of PM described in postpartum gestational tissues [4, 29, 33, 74], albiet at or close to term. Of note, histological characterization of the early placenta during malaria infection has not yet been reported; thus, our model helps to address this gap in knowledge. Third, it is possible that the timing of our analyses may have contributed to failure to detect critical physiological changes that occurred closer to the time of labor. Dams were euthanized at least 12 hours prior to expected preterm birth (which typically occurred overnight), and triggers for labor could have occurred inside this narrow window. Finally, the ability to successfully detect and characterize physiologically relevant changes by RT-qPCR alone may be limited. Thus, follow-up studies must consider expanding the timing of tissue collection and analyses, types of tissues evaluated, and the breadth of techniques used to identify a more precise cause(s) of preterm birth in this model.

In conclusion, this model provides a new avenue for interrogating the pathological driver(s) of preterm birth following post-implantation malaria infection. This model adds a new dimension to our current arsenal for studying the mechanisms involved in PM pathogenesis and could help improve our understanding of how malaria during pregnancy may galvanize preterm delivery.

Area under the curve analysis for weight, parasitemia, and hematocrit in B6 mice infected with P. chabaudi AS on E6.5, E8.5, and E10.5.

(A-C) Area under the curve (AUC) was calculated for uninfected pregnant (UP), infected pregnant (IP) and infected non-pregnant (INP) mice belonging to the E6.5 infection group (A, D, G; red), E8.5 infection group (B, E, H; blue), or E10.5 infection group (C, F, I; green) in an observational study. No statistical differences are observed in weight change over time between UP and IP animals. (D-F) Parasitemia AUC is higher in some IP animals compared to INP counterparts. (G-I) Hematocrit AUC was statistically different in the E6.5 infection group only. Groups were compared either by using a Kruskal-Wallis test or unpaired t-test with Welch’s correction (for parasitemia AUC). **P ≤ 0.005, *P < 0.05; ns = not significant, P > 0.05.

(TIF)

Click here for additional data file.

Course of P. chabaudi AS infection in mice euthanized one day prior to expected pre-term delivery.

(A-C) Percent starting weight and (D-F) parasitemia (% IRBCs) are presented for infected pregnant (IP), uninfected pregnant (UP), and infected non-pregnant (INP) groups. All mice were sacrificed one day prior to expected preterm delivery (arrows indicate time of sacrifice) and tissues were collected for further studies. (A, D, G) E6.5 infection group, euthanized on E15.5 (red): IP, n = 9; UP, n = 9; and INP, n = 4. (B, E, H) E8.5 infection group, euthanized on E16.5 (blue): IP, n = 15; UP, n = 12; INP, n = 7. (C, F, I) E10.5 infection group, euthanized on E17.5 (green): IP, n = 12; UP, n = 10; INP, n = 6.

(TIF)

Click here for additional data file.

Area under the curve analysis for weight, hematocrit and parasitemia in mice euthanized one day prior to expected pre-term delivery.

Area under the curve (AUC) was calculated for uninfected pregnant (UP), infected pregnant (IP) and infected non-pregnant (INP) mice belonging to the E6.5 infection group (A, D, G; red), E8.5 infection group (B, E, H; blue), or E10.5 infection group (C, F, I; green) for serial sacrifice experiments. (A-C) No statistical differences are observed in weight change over time between UP and IP animals. (D-F) Parasitemia AUC is higher in some IP animals compared to INP counterparts. (G-I) Hematocrit AUC achieved statistical significance between IP and INP mice in the E8.5 infection group only. Groups were compared either by using a Kruskal-Wallis test or unpaired t-test with Welch’s correction (for parasitemia AUC). ****P < 0.0001, **P < 0.005, *P < 0.05; ns = not significant, P > 0.05.

(TIF)

Click here for additional data file.

(A-C) micrographs taken with 100x objective lens of Giemsa-stained placentae from infected pregnant (IP) dams sacrificed one day prior to preterm delivery on E15.5, E16.5, and E17.5, respectively.

(TIF)

Click here for additional data file.

Correlation analyses between inflammation and oxidative stress-associated transcripts and peripheral parasitemia in placentae of mice infected on E6.5 and E10.5.

Mouse mRNA transcript abundance relative to peripheral parasitemia at the time of sacrifice or parasitemia AUC in placenta collected one day before expected preterm birth. (A-B) Tnf transcripts tended to be correlated with parasitemia in the E10.5 infection group. (C-D) Cat and Hmox1 transcripts tended to be positively correlated with parasitemia in the E6.5 infection group and (E) Cox1 transcripts tended to have a positive correlation with parasitemia AUC.

(TIF)

Click here for additional data file.

Parasitemia and area under the curve for parasitemia in TNF-/- and TNFR1-/- mice infected on E8.5.

(A) Parasitemia (% IRBCs) in TNF-/- and (B) TNFRI-/- mice. (C-D) Area under the curve (AUC) analysis do not show a statistically significant increase in parasitemia between IP versus INP groups for both strains. TNF-/- IP: n = 3, INP n = 3; TNFRI-/- IP: n = 7, INP n = 3; ns = not significant, P > 0.05.

(TIF)

Click here for additional data file.

Pup viability from infected and uninfected dams sacrificed one day prior to expected preterm delivery.

No differences were observed in pup viability between infected pregnant (IP) and uninfected pregnant (UP) dams across all infection groups. Statistical significance determined via proportional analysis tested by chi-square.

(DOCX)

Click here for additional data file.

Primer sequences for qPCR targets.

Mouse-specific forward (FP) and reverse (RP) primers used in quantitative real-time PCR for the amplification of mRNA transcripts associated with inflammation, parturition, antioxidant activity, and reference (Ubc) genes.

(DOCX)

Click here for additional data file.

Univariate regression analysis of inflammatory and parturition-associated transcript expression and day of sacrifice.

Analysis performed with proc reg for dichotomous (status) and continuous (parasitemia) and variables and proc glm for categorical variables (sacrifice day). Parasitemia was log10-transformed for the analysis. Dashes indicate that E15.5 is the reference value. Sample sizes for the analysis are as follows: E15.5 IP, n = 4; E15.5 UP, n = 4; E16.5 IP, n = 11; E16.5 UP, n = 4; E17.5 IP, n = 6; E17.5 UP, n = 3.

(DOCX)

Click here for additional data file.

Univariate regression analysis of antioxidant transcript expression and day of sacrifice.

Analysis performed with proc reg for continuous (parasitemia) and dichotomous (status) variables and proc glm for categorical variables (sacrifice day). *Parasitemias were log10-transformed for the analysis. Dashes indicate that E15.5 is the reference value. Sample sizes for the analysis are as follows: E15.5 IP, n = 4; E15.5 UP, n = 4; E16.5 IP, n = 11; E16.5 UP, n = 4; E17.5 IP, n = 6; E17.5 UP, n = 3.

(DOCX)

Click here for additional data file.

Multivariate regression analysis of inflammatory and parturition-associated transcript expression and day of sacrifice.

Analysis performed with proc glm. Dashes indicate that E15.5 is the reference value; dashes and NA indicate that these parameters were not considered in the analysis. Sample sizes for the analysis are as follows: E15.5 IP, n = 4; E16.5 IP, n = 11; E17.5 IP, n = 6.

(DOCX)

Click here for additional data file.

Multivariate regression analysis of antioxidant transcript expression and day of sacrifice.

Analysis performed with proc glm. Dashes indicate that E15.5 is the reference value; dashes and NA indicate that these parameters were not considered in the analysis. Sample sizes for the analysis are as follows: E15.5 IP, n = 4; E16.5 IP, n = 11; E17.5 IP, n = 6.

(DOCX)

Click here for additional data file.

Analysis performed with proc glm. Dashes indicate that E15.5 is the reference value; dashes and NA indicate that these parameters were not considered in the analysis. Sample sizes for the analysis are as follows: E15.5 IP, n = 4; E16.5 IP, n = 11; E17.5 IP, n = 6.

(DOCX)

Click here for additional data file.

Analysis performed with proc glm. Dashes indicate that E15.5 is the reference value; dashes and NA indicate that these parameters were not considered in the analysis. Sample sizes for the analysis are as follows: E15.5 IP, n = 4; E15.5 UP, n = 4; E16.5 IP, n = 11; E16.5 UP, n = 4; E17.5 IP, n = 6; E17.5 UP, n = 3.

(DOCX)

Click here for additional data file.

Analysis performed with proc glm. Dashes indicate that E17.5 is the reference value; dashes and NA indicate that these parameters were not considered in the analysis. Sample sizes for the analysis are as follows: E15.5 IP, n = 2; E16.5 IP, n = 7; E17.5 IP, n = 3.

(DOCX)

Click here for additional data file.

Analysis performed with proc glm. Dashes indicate that E17.5 is the reference value; dashes and NA indicate that these parameters were not considered in the analysis. Sample sizes for the analysis are as follows: E15.5 IP, n = 2; E16.5 IP, n = 7; E17.5 IP, n = 3.

(DOCX)

Click here for additional data file.

(XLSX)

Click here for additional data file.

16 Aug 2021

PONE-D-21-23277

Novel murine models of post-implantation and midgestional malaria-induced preterm birth

PLOS ONE

Dear Dr. Andrew,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

If you can address the specific concerns detailed below, then we would be willing to reconsider a revised version. Before seeing and evaluating such changes, however, we cannot guarantee that your revised article would be accepted for publication. Should you decide to revise the manuscript for further consideration here, your revisions should address all points indicated by reviewers.

Please submit your revised manuscript by within 30 days. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see:  http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at  https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .

We look forward to receiving your revised manuscript.

Kind regards,

Claudio Romero Farias Marinho, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.

pdf

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

3. Thank you for stating the following financial disclosure:

"A.K.A was supported by the 2017-2019 Peach State LSAMP Bridge to the Doctorate Program at the University of Georgia (National Science Foundation, award number 1702361, https://www.nsf.gov). This work was supported by a National Institute of Health grant to J.M.M (award number R01HD46860, www.nih.gov) . The content is solely the responsibility of the authors and does not necessarily represent official views of the Eunice Kennedy Shriver National Institute of Child Health and Human Development, the National Institute of Allergy and Infectious Diseases, or the National Institutes of Health."

Please state what role the funders took in the study.  If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

4. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

5. Please include your full ethics statement in the ‘Methods’ section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this manuscript, the authors describe three murine models of placental malaria, by infectingB6 mice with P. chabaudi, chabaudi at embryonic day 6.5, 8.5, or 10.5. The authors observed no intrauterine growth restriction, in contrast to other models of placental malaria, although they do report pre-term delivery. The authors also examined antioxidant and inflammatory gene transcripts, finding variations which are linked to the day of infection. Lastly, the authors have used mice deficient in TNF and its receptor to demonstrate that these do not drive pre-term delivery as a result of infection at embryonic day 8.5.

There are already a number of murine models of placental malaria, and the one described here provides an additional one which can be used to explore responses which result in pre-term delivery.

The statistical analysis methodology is appropriate considering the nature of the study and other recent work in the field.

There are, however, a few areas, which could be improved in the manuscript, listed below.

1. In lines 98 and 99, the authors state that infection during the second trimester is linked to poor outcomes of pregnancy and provide appropriate citations. They also claim that the mouse midgestation period (E6-12) is analogous to the human second trimester, but neither of the citations provided in that sentence support this. The human placenta is fully formed and exposed to maternal blood by the end of the first trimester, whereas the mouse placenta undergoes significant changes between days E6 and E12. These changes include chorioallantoic fusion around E8, and the fetuses are exposed to the full contents of maternal blood by E11.5-12. It is important that the authors clarify if they mean “the middle third of pregnancy” when calling the mouse midgestation and the human second trimester analogous, or if there are specific functional aspects.

2. Accounting for this series of changes in the mouse placenta during midgestation, it is pertinent for the authors to explain how the 3 models presented differ with respect to the phase of human disease they aim to recapitulate.

3. The authors use precipitous weight loss as an indicator of pre-term delivery (caption for figure 1 and lines 208-210), could this not also be an indicator for resorptions?

4. The histological data used to determine placental parasitemia in Figure 2 would benefit from the inclusion of a representative image showing the difference in maternal and fetal sinusoids in infected mice.

5. The caption for Figure 3 is difficult to follow e.g “…E6.5 infection group 4IP Dams, 32 placentae…”, Whereas the graph has labels corresponding to sacrifice date, and the number of data points do not appear to correspond (E15.5, 9 points).

6. The qPCR method states that pooled samples of minimum 4 placentas were used per dam (lines 167-169), whereas the caption of Figure 4 appears to indicate that each circle represents a placenta (rather than a pool) – this needs to be clarified.

7. For Figure 5, it is not clear why placental parasitemia (measured by histology and averaged, or gene expression in the same samples) was not used instead of, or in addition to maternal parasitemia.

8. In lines 369 to 372, the authors state that these are the first models they know of which do not induce maternal mortality. Models using recrudescence (Sharma et al 2016), specifically in multigravid mice (Marinho et al 2009) have also shown this.

9. The authors do not address the significant physiological differences in the human and mouse placenta within the manuscript, nor do they attempt to explore the potential limitations of their models. Adding this information to the discussion would allow the reader to determine whether the models are appropriate or not.

Reviewer #2: Major comments:

Introduction:

Overall, it was not clear in the introduction the importance of performing comparative studies in plasmodium-infected mice across distinct time points during mid-pregnancy. The authors should emphasize all biological aspects that endorsed the study. Additionally, it was not clear why days E6.5, E8.5 and E10.5 were chosen to create the models, as well as the choice for B6 mice and Plasmodium chabaudi chabaudi AS plasmodium strain. These points should be taken into consideration to provide a more robust background.

Materials and methods

The statistical analysis applied in the study should be summarized, section is very long.

Results

Comparisons across E6.5, E8.5 and E10.5 groups in relation to mRNA expression of pro- and anti-inflammatory cytokine genes should be carried out to demonstrate which model induce a higher or lower inflammatory profile. This information is important for selecting the most appropriate model to be used in a research. It was not clear why the authors decided to test TNF-/- 313 and TNFRI-/- deficient mice considering that the statistical significance obtained for tnf was less evident in relation to data obtained for infy. Along with, differences were also found for IL-10, Cox-1 and Cox-2 for animals infected on day E8.5, and a higher amount of infy was maintained in animals infected on day E.10.5. Another obscure aspect of this experiment was the fact that only infection occurring at day E8.5 was chosen to be tested. Graphs demonstrating correlation between transcripts and parasitemia are very confusing; it is strongly recommended to remake them, for example, adding the infection day number into the obtained result to facilitate comprehension. Furthermore, some valuable analyses are absent in the work. It would be interesting to insert data related to pups’ viability at weaning (%fetuses), as well as hematocrit change in P. chabaudi chabaudi AS-infected B6 mice. An additional aspect that should be included is a more detailed description of histological analysis to compare if the new models exhibit pathological features associated with human gestational malaria.

Discussion

It would be interesting to discuss more the advantages of this new models compared to the ones already available for testing. Authors should go into greater depth when explaining findings related to Tnf and Il10 transcripts, increased only in the E8.5 infection group. These cytokines are associated with human malaria-induced preterm delivery that was observed in all experimental groups evaluated (E6.5, E8.5 and E10.5). The choice for testing TNF-/-and TNFR 403 -/- mice at E8.5 needs to be better explained. Furthermore, it would be relevant to describe which scientific answers could be answered through the use of each proposed model, highlighting differences between them. Finally, discussion of results that were not presented in the article should be avoided.

Minor comments:

Title:

Line 1: Please correct the word “midgestional” for “midgestational”

Abstract:

Line 30: Please correct the word “midgestional” for “midgestational”

Introduction:

Line 87 to 88: It would be interesting to add information about therapies available to protect Swiss Webster mice against pregnancy loss at midgestation;

Materials and methods:

Line 164: Insert a comma before “as previously described”

Line 198: Remove duplicate parentheses.

Results:

Line 202 to 203: More direct and objective title is needed;

Line 223 to 224: The same title was used;

Line 245 to 248: It would be interesting to add data showing a significant reduction in pup weight (by 0.7546 g, P = 0.0102);

Line 253 to 254: If placenta weight stagnates after E16.5, then we would expect the same value between days E16.5 and E17.5, and a lower value for E15.5 day.

Line 272: Legend from figure 4 better sums up what was written for ‘Placental inflammatory responses do not universally precede preterm delivery’ part. It is strongly recommended to modify this title.

Line 329: Please, remove ‘protective antioxidant defense molecules’ as it seems that they also cause cellular damage.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

22 Feb 2022

Reviewer #1: In this manuscript, the authors describe three murine models of placental malaria, by infecting B6 mice with P. chabaudi chabaudi at embryonic day 6.5, 8.5, or 10.5. The authors observed no intrauterine growth restriction, in contrast to other models of placental malaria, although they do report pre-term delivery. The authors also examined antioxidant and inflammatory gene transcripts, finding variations which are linked to the day of infection. Lastly, the authors have used mice deficient in TNF and its receptor to demonstrate that these do not drive pre-term delivery as a result of infection at embryonic day 8.5.

There are already a number of murine models of placental malaria, and the one described here provides an additional one which can be used to explore responses which result in pre-term delivery.

The statistical analysis methodology is appropriate considering the nature of the study and other recent work in the field.

There are, however, a few areas, which could be improved in the manuscript, listed below.

1. In lines 98 and 99, the authors state that infection during the second trimester is linked to poor outcomes of pregnancy and provide appropriate citations. They also claim that the mouse midgestation period (E6-12) is analogous to the human second trimester, but neither of the citations provided in that sentence support this. The human placenta is fully formed and exposed to maternal blood by the end of the first trimester, whereas the mouse placenta undergoes significant changes between days E6 and E12. These changes include chorioallantoic fusion around E8, and the fetuses are exposed to the full contents of maternal blood by E11.5-12. It is important that the authors clarify if they mean “the middle third of pregnancy” when calling the mouse midgestation and the human second trimester analogous, or if there are specific functional aspects.

Thank you for pointing this out. Significant changes have been made to the manuscript to be clearer about what other studies have described as the similarities and differences between human and mouse placental development, see lines 112-117. The 2nd half of mouse pregnancy is more closely aligned with the end of the first trimester of human pregnancy, so the manuscript has been modified to reflect that and we have included the appropriate citations.

2. Accounting for this series of changes in the mouse placenta during midgestation, it is pertinent for the authors to explain how the 3 models presented differ with respect to the phase of human disease they aim to recapitulate.

The reviewer’s comment is appreciated. The manuscript has been significantly modified to more clearly identify how the model described serves to recapitulate human malaria infection during pregnancy.

3. The authors use precipitous weight loss as an indicator of pre-term delivery (caption for figure 1 and lines 208-210), could this not also be an indicator for resorptions?

This point is well taken. However, we point out that number of embryo and embryo viability do not differ between infected and uninfected dams when they were euthanized one day prior to anticipated weight loss (S1 Table). This suggests, but doesn’t prove, that the weight loss observed in our model is related to preterm delivery. At these later times in gestation (E16.5-18.5), fetal loss typically presents as, and is used to model, stillbirth, rather than resorptions. Resorptions have been more commonly described as fetal loss during earlier times in gestation (~E7-E10). We’ve updated our manuscript to discuss this and included the appropriate citation.

4. The histological data used to determine placental parasitemia in Figure 2 would benefit from the inclusion of a representative image showing the difference in maternal and fetal sinusoids in infected mice.

We appreciate this suggestion and have now included some representative H&E images for each infection group in manuscript, see figure 4.

5. The caption for Figure 3 is difficult to follow e.g “…E6.5 infection group 4IP Dams, 32 placentae…”, Whereas the graph has labels corresponding to sacrifice date, and the number of data points do not appear to correspond (E15.5, 9 points).

We regret this oversight. The graphs have been updated for clarity.

6. The qPCR method states that pooled samples of minimum 4 placentas were used per dam (lines 167-169), whereas the caption of Figure 4 appears to indicate that each circle represents a placenta (rather than a pool) – this needs to be clarified.

The figure caption (now Figure 5) has been updated to clarify that each circle represents pooled placentae from individual dams in each group.

7. For Figure 5, it is not clear why placental parasitemia (measured by histology and averaged, or gene expression in the same samples) was not used instead of, or in addition to maternal parasitemia.

We have now analyzed the data looking at placental parasitemia as well and updated the figure (now figure 6) and the text to include the significant results. The data reporting placental parasitemia is only for a subset of animals, detected by counting on Giemsa-stained placental sections.

8. In lines 369 to 372, the authors state that these are the first models they know of which do not induce maternal mortality. Models using recrudescence (Sharma et 2016), specifically in multigravid mice (Marinho et al 2009) have also shown this.

We regret our failure to make our point clearly and appreciate the reviewer pointing this out. I’ve restated this sentence to clarify that our model is one that recapitulates certain aspects of human PM in primigravid hosts that lack immunity or without the need for antimalarial treatment to prevent mortality. This is meant to support the scenario in which malaria causes significant morbidity and impacts pregnancy but is not lethal to the infected mother, if left untreated.

9. The authors do not address the significant physiological differences in the human and mouse placenta within the manuscript, nor do they attempt to explore the potential limitations of their models. Adding this information to the discussion would allow the reader to determine whether the models are appropriate or not.

This is a very important point, and we appreciate the reviewer making it. We believe the manuscript is much improved with this information included.

Reviewer #2: Major comments:

Introduction:

Overall, it was not clear in the introduction the importance of performing comparative studies in plasmodium-infected mice across distinct time points during mid-pregnancy. The authors should emphasize all biological aspects that endorsed the study. Additionally, it was not clear why days E6.5, E8.5 and E10.5 were chosen to create the models, as well as the choice for B6 mice and Plasmodium chabaudi chabaudi AS plasmodium strain. These points should be taken into consideration to provide a more robust background.

We thank the reviewer for seeking clarification on our rationale. Because a pregnant woman can be infected during any point during pregnancy, it can be argued that initiating infection at various times points during gestation in a mouse model can provide important information. As we discuss elsewhere in the manuscript (Alicer, in the intro?), use of the B6/PccAS combination offers many benefits: induction of infection earlier in gestation has resulted in confirmation of the importance of several pathologies reported to occur in human placental malaria (cite relevant Poovassery papers, Sarr ox stress and Avery), and this model also allows, as we now show here, induction of infection at various stages of pregnancy that impact embryo and fetal health but do not induce maternal death, as is the case in untreated P. berghei infection.

Materials and methods

The statistical analysis applied in the study should be summarized, section is very long.

Done, as requested.

Results

Comparisons across E6.5, E8.5 and E10.5 groups in relation to mRNA expression of pro- and anti-inflammatory cytokine genes should be carried out to demonstrate which model induce a higher or lower inflammatory profile. This information is important for selecting the most appropriate model to be used in a research.

This is a great suggestion by the reviewer and we have performed univariate and multivariate regression analyses to address this point. See supporting tables 3-10.

It was not clear why the authors decided to test TNF-/- 313 and TNFRI-/- deficient mice considering that the statistical significance obtained for tnf was less evident in relation to data obtained for infy. Along with, differences were also found for IL-10, Cox-1 and Cox-2 for animals infected on day E8.5, and a higher amount of infy was maintained in animals infected on day E.10.5.

Since TNF has long been associated with adverse pregnancy outcomes in malaria and our lab has previously published work showing that TNF ablation rescues pregnancy compromise, we reasoned that TNF may be important in our new model of malaria-induced preterm delivery. In our previous publication, both TNF and IFNg deficiency rescued pregnancy outcomes in another model of malaria, future studies in our novel model will include infections in IFNg knockout mice. See Poovassery et al, 2009, “Malaria-Induced Murine Pregnancy Failure: Distinct roles of IFNg and TNF” (PubMed 19783682).

Another obscure aspect of this experiment was the fact that only infection occurring at day E8.5 was chosen to be tested.

This timepoint was chosen based on the Tnf transcript results shown in Fig. 4.

Graphs demonstrating correlation between transcripts and parasitemia are very confusing; it is strongly recommended to remake them, for example, adding the infection day number into the obtained result to facilitate comprehension.

This is an excellent point, we have reorganized this figure for clarity.

Furthermore, some valuable analyses are absent in the work. It would be interesting to insert data related to pups’ viability at weaning (%fetuses), as well as hematocrit change in P. chabaudi chabaudi AS-infected B6 mice.

Pup viability data is provided in S1 Table and hematocrit change is displayed in S1, S2, and S3 figures for both the observational study and the serial sacrifice study. Live pups were not produced in this study.

An additional aspect that should be included is a more detailed description of histological analysis to compare if the new models exhibit pathological features associated with human gestational malaria.

As the revised manuscript details, the time points in mouse gestation that were studied here correspond developmentally and functionally to human placental development in the first trimester and beginning of the second trimester. To our knowledge, only a few studies, which we cite, have examined malaria infection this early in gestation and none have reported placental histology. Thus, little is known regarding histological and pathological features of human gestational malaria at this early time in pregnancy. Nonetheless, this time is very important, as it is when maternal blood begins to enter and circulate in the definitive placenta.

We did conduct a histopathological analysis of placental necrosis as an indicator of placental dysfunction, but no differences were found between placentae from infected and uninfected dams.

Discussion

It would be interesting to discuss more the advantages of this new models compared to the ones already available for testing.

We appreciate this excellent suggestion to improve our manuscript and have done so.

Authors should go into greater depth when explaining findings related to Tnf and Il10 transcripts, increased only in the E8.5 infection group. These cytokines are associated with human malaria-induced preterm delivery that was observed in all experimental groups evaluated (E6.5, E8.5 and E10.5). The choice for testing TNF-/-and TNFR 403 -/- mice at E8.5 needs to be better explained. Furthermore, it would be relevant to describe which scientific answers could be answered through the use of each proposed model, highlighting differences between them. Finally, discussion of results that were not presented in the article should be avoided.

All excellent points made by the reviewer. We have clarified our use of TNF mice in lines 353-355. We have also speculated on some scientific answers that this proposed model may help to address in the discussion, lines 510-522.

Minor comments:

Title:

Line 1: Please correct the word “midgestional” for “midgestational”

Done as requested.

Abstract:

Line 30: Please correct the word “midgestional” for “midgestational”

Done as requested.

Introduction:

Line 87 to 88: It would be interesting to add information about therapies available to protect Swiss Webster mice against pregnancy loss at midgestation;

We are currently conducting studies to characterize this and our findings will be reported in a separate paper.

Materials and methods:

Line 164: Insert a comma before “as previously described”

Done as requested.

Line 198: Remove duplicate parentheses.

Done as requested.

Results:

Line 202 to 203: More direct and objective title is needed;

Done as requested.

Line 223 to 224: The same title was used;

Clarified, as requested.

Line 245 to 248: It would be interesting to add data showing a significant reduction in pup weight (by 0.7546 g, P = 0.0102);

This reduction was only observed in the context of litter size in the E10.5 group, so as number of viable pups increased, pup weights decreased. This phenomenon has been described before and the reference is provided.

Line 253 to 254: If placenta weight stagnates after E16.5, then we would expect the same value between days E16.5 and E17.5, and a lower value for E15.5 day.

We understand the confusion here and opted to remove this interpretation from the manuscript as it is not part of the core paper.

Line 272: Legend from figure 4 better sums up what was written for ‘Placental inflammatory responses do not universally precede preterm delivery’ part. It is strongly recommended to modify this title.

Title has been removed to adhere with PLOS One formatting standards.

Line 329: Please, remove ‘protective antioxidant defense molecules’ as it seems that they also cause cellular damage.

Done, as requested.

________________________________________

Submitted filename: LP paper_Response to Reviewers.docx

Click here for additional data file.

8 Mar 2022

A novel murine model of post-implantation malaria-induced preterm birth

PONE-D-21-23277R1

Dear Dr. Andrew,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Claudio Romero Farias Marinho, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

11 Mar 2022

PONE-D-21-23277R1

A novel murine model of post-implantation malaria-induced preterm birth

Dear Dr. Andrew:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Claudio Romero Farias Marinho

Academic Editor

PLOS ONE