Coinfection with malaria alters the fecundity and within-host persistence of an intestinal nematode
Luc Bourbon, Aloïs Dusuel, Emma Groetz, Mickaël Rialland, Benjamin Roche, Bruno Faivre, Gabriele Sorci, Chao Yan, Chao Yan, Chao Yan, Chao Yan
TL;DR
Malaria coinfection increases the number of nematode eggs excreted and extends the infection period in mice, likely due to a weakened immune response.
Contribution
This study demonstrates that malaria coinfection alters nematode fecundity and persistence through Th2 immune suppression.
Findings
Coinfected mice excreted more nematode eggs than singly infected mice.
Coinfection extended the patent period of the nematode, increasing cumulative egg excretion.
Th2 immune response suppression in coinfected hosts mediated the increased egg excretion.
Abstract
Infections with soil transmitted helminths (STHs) are highly prevalent in humans living in the intertropical region. While, in most cases, STHs can establish chronic infections, the dynamics of the infection can be altered when other parasites exploit the same host. These changes can have consequences in terms of the health of the host, the epidemiology of the disease (e.g., the duration of the infection and the inter-host transmission success) and the fitness of the parasite. Here, we investigated if the coinfection with Plasmodium yoelii alters the dynamics (fecundity and with-host persistence) of the murine nematode Heligmosomoides polygyrus. We found that, compared to single infected mice, coinfected hosts excreted more worm eggs, while the worm biomass in the intestine did not differ between single infected and coinfected mice. Moreover, the increase in egg excretion was also…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Fig 1
Fig 2
Fig 3
Fig 4
Fig 5
Fig 6
Fig 7
Fig 8- —http://dx.doi.org/10.13039/501100001665Agence Nationale de la Recherche
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsParasites and Host Interactions · Helminth infection and control · Malaria Research and Control
Introduction
Soil-transmitted helminths (STHs) are agents of highly prevalent parasitic diseases in most countries of the intertropical region, associated with poverty and unfavorable socioeconomic conditions [1,2]. According to the World Health Organization, 24% of the world’s population suffers from infection with STHs (c.a. 1.5 billion humans) [3]. Infection with STHs includes several species of nematodes that differ in terms of transmission mode, duration and intensity of the infection, or morbidity. STHs are transmitted either through the ingestion of food and/or water that has been contaminated by the eggs of the helminths (roundworms and whipworms) or through the penetration of the host skin by the infective larvae (hookworm). STHs usually establish chronic infections that can last for years [2]. Parasite burden is a key driver of the morbidity of the infection [4] and severe symptoms involve hookworm-induced anemia in children and pregnant women, roundworm-induced intestinal obstruction, tissue damage during larval migration and impaired physical development [5,2], although there is still a paucity of data [6,7]. Strongyloides stercoralis can also cause damage due to its unique capacity (among STHs) to shortcut the environmental stage of the life cycle and directly reinfect the host, especially in immune compromised individuals [8]. However, for the vast majority of STHs, infection involves the contact with transmissible stages in the external environment; therefore, high egg excretion in the feces of infected people is a key factor contributing to maintain a reservoir of infective stages from which humans can get infected. Mass treatment with anthelmintic drugs can reduce the prevalence of the infection and alleviate the symptoms of the disease [9,10], but the consistency of the benefits has been debated [11,12]. Moreover, worm resistance to drugs raises concerns over the sustainability of anthelmintic mass treatment on the long-term [13–16]. This stresses the importance of understanding the environmental factors (including host immunity) that modulate the production of transmissible stages and the associated transmission success and infection risk [17].
Geographic areas with high prevalence of helminthiases are also endemic for major infectious diseases, such as malaria or AIDS and therefore people can harbor multiple infections involving macro and microparasites [18–21]. Coinfection between STHs and malaria has the potential to exacerbate the severity of the disease symptoms and substantial effort has been devoted to understand under which circumstances STHs might aggravate Plasmodium virulence [22–24]. On the contrary, the effect of coinfection on the dynamics of STHs has been barely addressed, and clearly merits further investigation.
The reasons underlying a possible effect of coinfection on STH dynamics are manifold. In particular, during the acute phase of infection, Plasmodium polarizes the host immune response towards a type 1 immunity [25], while control of helminth infection requires a type 2 response orchestrated by Th2 cytokines such as IL-4, IL-5 or IL-13 [26–30]. Therefore, when hosts are infected with microparasites that elicit a Th1 response, they should be less able to resist a helminth infection [31–33], implying that when exposed to helminth infective stages, the probability of successful infection should be higher in coinfected hosts. Finally, previous infection might debilitate the host and limit its access to resources, offering a more favorable ground for subsequent infections, independently of the host capacity to mount any parasite-specific immune response [34].
Although the dichotomy between Th1 and Th2 responses provides a useful framework for analyzing and predicting the effects of coinfection between microparasites and helminths, it should be acknowledged that the real world is more complex than this simple model suggests. This is because microparasites and helminths can induce the production of Th1 and Th2 responses at different stages of the infection [35]. Therefore, the interplay between the Th1/Th2 balance and coinfection should be considered as a dynamic process that depends on the timing and order of microparasite and helminth infections. Previous work has shown that hosts previously infected with helminths suffer from more severe disease symptoms when subsequently infected with Plasmodium [36–38], or other intracellular parasites, such as Toxoplasma gondii [39]. These results are also in agreement with the idea that initial polarization of the immune response towards a Th2 response makes the host more susceptible to Plasmodium [40]. Therefore, it is important to understand whether the predicted changes in helminth infection dynamics also depend on the order of infection [41,42].
Here, we used an experimental model involving two murine parasites [the intestinal nematode Heligmosomoides polygyrus (hereafter Hp) and Plasmodium yoelii (hereafter Py)] to investigate if coinfection had any effect on the fecundity of the nematode, and if this effect was driven by the host immune response. In addition, we investigated if coinfection had any long-term effect on Hp, by assessing the within-host persistence and the cumulative number of eggs excreted.
We predicted that Hp infecting hosts harboring Py should have higher infection success and therefore excrete more eggs in the host feces. However, if coinfected hosts have an impaired immune response towards Hp, accounting for an enhanced egg shedding, the higher investment into reproductive investment should incur a cost for worms in coinfected hosts. Therefore, we also predicted that the positive effect of coinfection should be transitory, resulting in similar cumulative number of excreted eggs in single infected and coinfected hosts.
Materials and methods
Ethics statement
Experiments were approved by the Comité d’éthique Grand Campus Dijon of the University of Burgundy and were authorized by the French Ministry of Higher Education and Research under the numbers 33492 and 47487.
Experimental animals and infections
C57BL/6JRj female mice (7–10 weeks old) were purchased from Janvier Labs (Le Genest-Saint-Isle, France), housed in cages containing 5 individuals under pathogen-free conditions, and maintained under a constant temperature of 24°C and a photoperiod of 12h:12h light:dark with ad libitum access to water and standard chow diet (A03-10, Safe, Augny, France). All mice were acclimatized to the housing conditions during, at least, 7 days prior to the start of the experiments, were monitored twice a day to check health status, and euthanized by cervical dislocation under anesthesia with isoflurane either if they reached previously defined end points or at specific days (14, 21 and 35) post-infection (p.i.) for terminal collection of blood and organs.
Mice were infected with Heligmosomoides polygyrus bakeri by oral gavage with L3 larvae (350 larvae suspended in 0.2 ml of drinking water) and with Plasmodium yoelii 17XNL by intra peritoneal injection (i.p.) with 5 x 10^5^ infected red blood cells (iRBC) suspended in 0.1 ml PBS. To obtain Hp L3 larvae, feces from donor infected mice were mixed with distilled water and charcoal, spread on a Whatmann paper laid on a wet paper towel and placed into a Petri dish. Petri dishes were stored at room temperature in the dark. L3 larva were collected after 11 days by washing out the back side of the Whatmann paper, the paper towel and the bottom of the Petri dish with distilled water. L3 larvae were then isolated by centrifugation (100 rpm, 10 min, 4°C), washed twice, and kept at 4°C in distilled water until use (see [43] for a detailed description of the methods to culture Hp).
Experimental groups
Mice were randomly assigned to 5 experimental groups (10–15 mice per group). A control (non-infected) group of mice was sham infected (i.p. injection of 0.1 ml of PBS and oral gavage with a 0.2 ml of drinking water); one group (single Hp) received a single Hp infection; one group (Py-14 + Hp) was infected with Py and 14 days after with Hp; one group (Py-28 + Hp) was infected with Py and 28 days after with Hp; one group (Hp-28 + Py) was infected with Hp and 28 days after with Py. The timings of infection in the coinfection groups correspond to specific stages in the life cycles of the two parasites. Day 14 and day 28 post Py infection approximately correspond to the peak of parasitemia and the end of the acute phase of the infection, respectively. Day 28 post Hp infection corresponds to the stage when all larvae have molted into adults, adults have emerged into the intestinal lumen and the infection has reached a chronic stage. The whole experiment was repeated twice, although not all traits were measured in all mice of the two replicates. Additionally, ten mice per group (excluding the non-infected group) were monitored up to 99 days post Hp infection to assess the effect of coinfection on Hp within-host persistence and the cumulative egg shedding.
Hp egg shedding and infection burden
To assess the fecal egg count (FEC, parasite eggs per gram of feces), mice were individually placed for 1 hour in clean cages; 200 mg of feces were then collected and disposed in tubes containing 2.5 ml of water. Feces were mechanically broken up and 5 ml of salted water (75% saturation, 0.27g NaCl per ml of water) were added to allow egg floatation. Eggs were counted in a McMaster chamber using an optical microscope (Eclipse E400, Nikon, Tokyo, Japan) under magnification x150. To assess infection burden, the intestines were removed, transferred in tubes containing a deep freeze solution (4.2g of sorbitol, 100ml of 0.9% NaCl solution and 39ml of glycerol) and stored at -80° C. Intestines were then dissected and all worms found were collected, washed with water and disposed in 1.5 ml tubes (that had been previously weighed using a precision balance (MCA6.6S-2S00-M, Sartorius, Göttingen, Germany) (± 1 µg) with 500 µl of absolute ethanol. The opened tubes were subsequently placed in an incubator at room temperature during 2 – 3 days. When dry, tubes were weighed again and the worm biomass assessed as the difference between the second and first measurement of tube weights.
Plasma and organ processing
Plasma was separated from total blood by centrifugation (7 min 3000 rpm 4°C), aliquoted and stored at -80°C. About one third of the spleen was cut and immediately frozen in liquid nitrogen for RT-qPCR, and the remaining was kept in cold PBS before the flow cytometry staining procedure.
IL-13 administration and neutralization
Mice (8 weeks) were infected with Hp and randomly split into three groups at day 28 p.i.: one group (n = 8) received three i.p. injections of 2 µg recombinant mouse IL-13 (rIL-13) (PeproTech 210-13-50UG, Gibco, Grand Island, New York, USA) in 200 ml of PBS; one group (n = 7) received three i.p. injections of 100 µg anti-mouse IL-13 monoclonal Ab (clone eBio1316H, Invitrogen, Carlsbad, California, USA) in 200 ml of PBS; one group (n = 7) received three i.p. injections of 100 µg control IgG1 (clone eBRG1, Invitrogen) in 200 µl of PBS. Mice were injected at days 28, 30 and 32 post-infection.
RNA extraction and RT-qPCR for IL-4 and IL-13 gene expression
Spleens were homogenized in TRIzol Reagent (Invitrogen, Carlsbad, California, USA) under strong agitation using 0.5 mm glass beads and Precellys 24 Touch homogenizer (Bertin Technologies, Montigny-le-Bretonneux, France). RNA extraction was performed following the manufacturer’s instructions. RNA concentration was measured with NanoPhotometer N50 (Implen, Munich, Germany). Reverse transcription was performed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, California, USA) from 1.5 µg total RNA. Quantitative PCR were performed with PowerUp SYBR Green Master Mix (Applied Biosystems) on QuantStudio 3 Real-Time PCR System (Applied Biosystems). We used two housekeeping genes (β-actin and GAPDH); β-actin provided more consistent values (less intra-group variability) and therefore we used it as housekeeping gene. Primer sequences are reported in the supplementary material (S1 Table).
Plasma cytokine quantitation
Concentrations of plasma IL-5 and IL-13 were measured by multiplex using a Mouse Luminex Discovery Assay (LXSAMSM, Bio-Techne, Minneapolis, Minnesota, USA) according to the manufacturer’s instructions. Samples were analyzed with a Bio-Plex 200 system (Bio-Rad, Hercules, California, USA) in the ImaFlow Facility (US58 BioSanD, Dijon, France).
Spectral flow cytometry
Spleens were homogenized with a 70 µm cell strainer and washed two times with PBS to obtain a single-cell suspension. Red blood cells were removed from suspensions with eBioscience RBC Lysis Buffer (Invitrogen) (3 min incubation at room temperature). Live cells were counted with viability trypan blue dye and dispatched in 96-well V bottom plates to obtain 1.10^7^ cells/well. Cells were then stained with LIVE/DEAD Fixable Blue Dead Cell Stain Kit (Invitrogen) in PBS for 30 min at 4°C, incubated in stain buffer with BD Fc Block (BD Pharmingen, Franklin Lakes, New Jersey, USA) for 15 min at room temperature (0.5 µg/well), and stained with surface antibodies in brilliant stain buffer (see S2 Table for details on the markers used) for 30 min at 4°C. Cells were then fixed and permeabilized with eBioscience GATA-3 CD4 Transcription Factor Staining Buffer Set (Invitrogen) prior intracellular staining in brilliant stain buffer (S2 Table) for 30 min at room temperature. Cells were resuspended in stain buffer, filtered with 100 µm nylon mesh and analyzed on a 4-laser Cytek Aurora spectral flow cytometer (Cytek Biosciences, Fremont, California, USA). Events were gated and analyzed with SpectroFlo v3.0.0 software (Cytek Biosciences) (see S1 Fig for the gating strategy).
Statistical analyses
The effect of coinfection on the number of eggs excreted (FEC) from day 11 to day 35 p.i. was analyzed using a linear mixed model (LMM) with a normal distribution of errors and identity link. FEC were ln-transformed to reduce skewness. Given that FEC were repeatedly measured for the same individuals over the course of the infection, mouse id was included as a random intercept. The model included the treatment group (single Hp, Py-14 + Hp, Py-28 + Hp), time p.i. (continuous variable), and the two-way interaction as fixed effects. We also conducted post-hoc comparisons of LS-means with Bonferroni correction of p values (p_adj_ = p/k, where k is the number of comparisons).
The effect of timing of Py infection on FEC (ln-transformed) was also analyzed using a LMM with mouse id declared as a random effect, and timing (pre Py infection, post Py infection) as a fixed effect. Since Py was inoculated at day 28 post Hp infected, we ran a model where the response variable was the FEC (ln-transformed) of single Hp infected mice, and the fixed effect was the timing (pre day 28 p.i., post day 28 p.i.). This model included the mouse id as a random effect.
The effect of coinfection on worm biomass was analyzed using a two-way ANOVA that included treatment (single Hp, Py-14 + Hp, Py-28 + Hp), time p.i. (21, 35 day p.i.) and the two-way interaction as explanatory variables.
The effect of coinfection on IL-4 and IL-13 gene expression was analyzed using two-way ANOVAs. The response variable was the Δ Ct (with β-actin as reference gene) and the treatment (non-infected, single Hp, Py-14 + Hp, Py-28 + Hp), time p.i. (14, 21, 35 day p.i.), and the two-way interaction were the explanatory variables.
The effect of coinfection on the level of IL-5 and IL-13 in plasma (at day 14 and 21 p.i. analyzed together) was assessed using a non-parametric test due to the high frequency of zeros, especially in the non-infected group. We ran the omnibus test (Kruskal-Wallis test) with the treatment (non-infected, single Hp, Py-14 + Hp, Py-28 + Hp) as the explanatory variable. For the pairwise comparisons, p values were adjusted using the DSCF method [44].
The effect of coinfection on the population of Th2 cells was analyzed using a generalized linear model with a beta-distribution of errors and a logit link. We used the beta-distribution because the response variable was a proportion (GATA-3^+^ cells among CD4^+^ T lymphocytes). The model included the treatment (non-infected, single Hp, Py-14 + Hp, Py-28 + Hp), time p.i. (14, 21, 35 day p.i.), and the two-way interaction as explanatory variables. Overdispersion was checked by computing the χ²/df ratio.
The persistence of the infection was analyzed using the Kaplan Meier estimator. We ran two models. In the first model, we compared the single Hp infected group and the two coinfection groups where Hp followed Py infection (Py-14 + Hp, Py-28 + Hp); in the second model, we compared the single Hp infected group to the coinfection group where Py followed Hp infection (Hp-28 + Py). Significance was assessed using a Log-Rank test and p values for pairwise comparisons were Sidak corrected [p_adj_ = 1 – (1 – p)^k^, where k is the number of comparisons]. We considered that mice that shed no eggs during at least two consecutive sampling dates had cleared the infection.
For the analysis of cumulative egg shedding, we added the number of eggs excreted during each of the 14 sampling dates (from day 12 to day 99 p.i.) for each mouse in each experimental group. These sums were then divided by the mean of the sums for the single Hp infection group, indicating thus the cumulative egg excretion with respect to single infection. These values were analyzed using one-way ANOVAs. We ran two models. In the first one, we compared the single Hp infected group and the two coinfection groups where Hp followed Py infection (Py-14 + Hp, Py-28 + Hp); in the second model, we compared the single Hp infected group to the coinfection group where Py followed Hp infection (Hp-28 + Py).
Finally, the IL-4 and IL-13 gene expression at day 99 p.i. was analyzed using one-way ANOVAs. First, we ran an ANOVA where we compared the Ct values (mean values over two technical replicates) over time (14, 21, 35, 99 day p.i.); then, we ran a one-way ANOVA where we compared the Δ Ct values (with β-actin as reference gene) at day 99 p.i. between single Hp infected and coinfected (Py-14 + Hp, Py-28 + Hp) mice.
For all linear models requiring a normal distribution of errors, we checked the distribution of model residuals and we drew the Q-Q plot, and found no departure from normality. For the gene expression of the Th2 cytokines, we also computed the log_2_ fold change as to better visualize the magnitude of the difference between groups.
Not all response variables were measured in all mice resulting in differences in sample size. Moreover, missing samples (for instance when mice did not defecate during the time they were placed in the individual cages), or technical issues during the processing of the samples (for instance during the RNA extraction and qPCRs) also contributed to differences in sample size across groups.
All the statistical analyses and figures were done using SAS Studio. Data and codes are available in DRYAD [45].
Results
Effect of coinfection on Hp egg shedding and infection burden
There was no time-dependent effect on egg shedding over the day 11 to day 35 p.i. (Table 1 and Fig 1a). However, there was a difference in the average FEC between groups, with mice in the Py-14 + Hp group shedding significantly more eggs compared to single Hp infected hosts (Table 1 and Fig 1a).
Table 1: LMM with a normal distribution of errors and identity link function investigating the changes in FEC [ln(eggs/g feces)] in single Hp infected (reference group) and coinfected (Py-14 + Hp, Py-28 + Hp) mice over time.
Coinfection increases egg excretion independently of the infection burden.a) FEC [ln(eggs/g feces)] over the course of the infection for single Hp infected and coinfected (Py-14 + Hp, Py-28 + Hp) mice. LS-means were significantly different between the single Hp and Py-14 + Hp groups (Bonferroni corrected p < 0.0001), but not between single Hp and Py-28 + Hp groups (Bonferroni corrected p = 0.3043) (n = 28 individuals in the single Hp group, 25 individuals in the Py-14 + Hp group and 25 individuals in the Py-28 + Hp group). Dots represent the raw data, the lines represent the fit of the LMM, and the shaded area around the line the 95% CI. b) Dry biomass of adult Hp collected at day 21 and 35 p.i. in the intestinal lumen of single Hp infected and coinfected (Py-14 + Hp, Py-28 + Hp) mice. A two-way ANOVA showed that there were no statistically significant differences between the treatments nor between times p.i. (treatment, F2,40 = 0.34, p = 0.7136, time p.i., F1,40 = 0.67, p = 0.4175). The two-way interaction was not statistically significant neither (F2,38 = 0.16, p = 0.8489). Dots represent individual observations; boxes represent the interquartile range (IQR); the horizontal line within each box indicates the median; the larger dot indicates the mean; whiskers extend to the most extreme values within 1.5 × IQR from the quartiles and points beyond the whiskers indicate outliers. Numbers above the x-axis indicate the sample size. c) Changes in FEC [ln(eggs/g feces)] following the infection with Py at day 28 post Hp infection (dotted vertical line). A LMM showed that FEC significantly increased between the pre and post Py infection (F1,91 = 32.92, p < 0.0001, n = 17 individuals and 109 observations). Dots represent the raw data, the lines represent the fit of the LMM, and the shaded area around the line the 95% CI. d) Changes in FEC [ln(eggs/g feces)] up to day 28 and after day 28 p.i. (dotted vertical line) in single Hp infected mice. A LMM showed that FEC did not differ between pre and post day 28 p.i. (F1,82 = 0.27, p = 0.6049, n = 28 individuals and 111 observations). Dots represent the raw data, the lines represent the fit of the LMM, and the shaded area around the line the 95% CI.
As all mice were infected with the same number of infective larvae and not allowed to be reinfected, this increase in FEC might result from two different non-exclusive processes. First, the environmental conditions provided by a Py infected host might improve the survival of L3 larvae during the prepatent period, resulting in a larger number of adult worms emerging in the intestinal lumen and releasing eggs. Second, more favorable environmental conditions encountered by adult worms in Py infected mice might allow females to produce and release more eggs. Therefore, according to the first hypothesis, the increase in FEC in coinfected hosts should stem from a larger number of adult worms. We compared the biomass of adult worms retrieved at day 21 and 35 p.i. between single Hp infected and coinfected (Py-14 + Hp and Py-28 + Hp) mice and found no difference between groups (Fig 1b). Therefore, this result suggests that the increase in egg excretion in coinfected mice is not due to a larger infra-host parasite population size, but rather to an adjustment of worm fecundity.
To go further in the investigation of the process underlying the rise in egg shedding, we assessed whether the infection with Py might induce an increase in FEC when adult worms are already releasing eggs in the intestinal lumen. To this purpose, we infected mice with Hp and after 28 days they were infected with Py and we looked at any change in the number of eggs excreted pre and post Py infection. We found that FEC increased following the Py infection (Fig 1c). As a control, we also compared whether there was a similar increase in FEC post day 28 in single Hp infected mice and found no difference (Fig 1d).
Therefore, overall, these results show that coinfection, and particularly when Hp infected mice that were at the peak of the acute infection with Py (Py-14 + Hp), produced an increase in worm fecundity, and that this is a plastic response that is induced when the within-host environment changes following the Py infection.
Immune mechanisms modulating the plastic adjustment of worm fecundity
We then investigated whether the inhibition of the Th2 response due to the infection with Py might account for the observed increase in egg shedding of coinfected mice. We assessed the expression of Th2 cytokine genes in the spleen and found that Hp infection induced an overexpression of both IL-4 and IL-13 genes (Table 2 and Fig 2a-2d). However, IL-4 gene expression was significantly downregulated in Py-14 + Hp and in Py-28 + Hp mice compared to single Hp infected hosts (Fig 2a and 2b). IL-13 gene expression was also downregulated in Py-14 + Hp mice compared to single Hp infected hosts, whereas the difference was not significant between Py-28 + Hp and single Hp infected mice (Fig 2c and 2d).
Table 2: Two-way ANOVAs investigating the changes in IL-4 and IL-13 gene expression in non-infected, single Hp infected and coinfected (Py-14 + Hp, Py-28 + Hp) mice at day 14, 21 and 35 post-infection.
Coinfection downregulates the gene expression of Th2 cytokines.a) Δ Ct values of IL-4 relative to β-actin at day 14, 21 and 35 p.i. in single Hp and coinfected (Py-14 + Hp, Py-28 + Hp) mice. IL-4 gene expression was significantly downregulated in coinfected groups compared to the single Hp group (Bonferroni corrected p < 0.0001 and p = 0.0499, respectively). b) log2 fold change in IL-4 expression relative to non-infected hosts at day 14, 21 and 35 p.i. in single Hp and coinfected (Py-14 + Hp, Py-28 + Hp) mice; c) Δ Ct values of IL-13 relative to β-actin in spleen at day 14, 21 and 35 p.i. in single Hp and coinfected (Py-14 + Hp, Py-28 + Hp) mice. IL-13 gene expression was significantly downregulated in the Py-14 + Hp group compared to the single Hp infection group (Bonferroni corrected p = 0.0006), whereas the difference between Py-28 + Hp and single Hp was not statistically significant (Bonferroni corrected p = 0.0864). d) log2 fold change in IL-13 expression relative to non-infected hosts at day 14, 21 and 35 p.i. in single Hp and coinfected (Py-14 + Hp, Py-28 + Hp) mice. Dots represent individual observations; boxes represent the interquartile range (IQR); the horizontal line within each box indicates the median; the larger dot indicates the mean; whiskers extend to the most extreme values within 1.5 × IQR from the quartiles and points beyond the whiskers indicate outliers.
We also assessed the amount of circulating IL-5 and IL-13 in plasma. Due to zero-inflated distributions, we used a non-parametric test. For the concentration of IL-5, the omnibus test was highly significant (Kruskal-Wallis, χ²_3_ = 30.59, p < 0.0001), and the pairwise comparisons showed that non-infected mice had lower levels of IL-5 compared to all infected groups (Fig 3a). The level of IL-5 in plasma was also lower in Py-14 + Hp mice compared to single Hp infected mice (Fig 3a); however, there was no significant difference between Py-28 + Hp and single infected mice (Fig 3a). For IL-13, the omnibus test was also highly significant (Kruskal-Wallis, χ²_3_ = 21.67, p < 0.0001), and the pairwise comparisons showed that both coinfection groups had significantly lower IL-13 levels than single Hp infected mice (Fig 3b). Interestingly, the level of IL-13 in plasma did not differ between the non-infected group and the two coinfection groups (Fig 3b).
Coinfection reduces the level of Th2 cytokines in plasma.a) Level of IL-5 (pg/ml) in plasma of non-infected, single infected and coinfected (Py-14 + Hp, Py-28 + Hp) mice. Pairwise comparisons showed that non-infected mice had lower levels of IL-5 compared to all infected groups (all DSCF corrected p’s < 0.006). The level of IL-5 in plasma was also lower in Py-14 + Hp mice compared to single Hp infected mice (DSCF corrected p = 0.0354), but not between Py-28 + Hp and single Hp infected mice (DSCF corrected p = 0.6405). Numbers above the boxes indicate the sample size. b) Level of IL-13 (pg/ml) in plasma of non-infected, single infected and coinfected (Py-14 + Hp, Py-28 + Hp) mice. The pairwise comparisons showed that both coinfection groups had significantly lower IL-13 levels than single Hp infected mice (DSCF corrected p = 0.0301 and p = 0.0162, respectively). The level of IL-13 in plasma did not differ between the non-infected group and the two coinfection groups (DSCF corrected p = 0.2890 and p = 0.1513, respectively). Numbers above the boxes indicate the sample size. Dots represent individual observations; boxes represent the interquartile range (IQR); the horizontal line within each box indicates the median; the larger dot indicates the mean; whiskers extend to the most extreme values within 1.5 × IQR from the quartiles and points beyond the whiskers indicate outliers.
We finally checked whether coinfection affected the expansion of Th2 cells compared to single Hp infection. To this purpose, we ran a generalized linear model with a beta-distribution of errors because the response variable was the proportion of GATA-3^+^ cells among CD4^+^ T cells. The model showed a highly significant interaction between treatment and time p.i. (Table 3). While the proportion of Th2 cells was low and stable over time for the non-infected group, single Hp infected mice had higher proportions of Th2 cells compared to coinfected hosts at day 14 and 21 p.i., with the difference vanishing at day 35 p.i. (Fig 4).
Table 3: Generalized linear model with a beta-distribution of errors and a logit link function investigating the changes in the proportion of GATA-3+ cells among CD4+ T lymphocytes.
Proportion of GATA-3+ cells among CD4+ T lymphocytes at day 14, 21 and 35 p.i. for non-infected, single Hp infected and coinfected (Py-14 + Hp, Py-28 + Hp) mice.Post-hoc comparisons of LS-means showed that the proportion of Th2 cells differed between single Hp infected and Py-14 + Hp coinfected hosts at both day 14 and 21 p.i. (Bonferroni corrected p = 0.003 and 0.0183, respectively). The proportion of Th2 cells also differed between single Hp infected and Py-28 + Hp coinfected hosts but only at day 21 p.i. (Bonferroni corrected p < 0.0001). At day 35 p.i., there was no difference between the single Hp infected and any of the coinfection group (Bonferroni corrected p’s > 0.05). Dots represent individual observations; boxes represent the interquartile range (IQR); the horizontal line within each box indicates the median; the larger dot indicates the mean; whiskers extend to the most extreme values within 1.5 × IQR from the quartiles and points beyond the whiskers indicate outliers. Numbers above the boxes indicate the sample size.
Manipulation of the Th2 response
If the rise in the number of excreted eggs results from the down regulation of the Th2 response in Py infected mice, we expect that experimentally manipulating the Th2 response in the absence of Py infection should reproduce the results reported above.
To this purpose, we infected mice with Hp and at day 28 p.i., they were treated with neutralizing IL-13 monoclonal antibodies, with rIL-13, or with IgG1 as a control, and we assessed whether the treatments produced a shift in egg excretion. For control mice, the injection of IgG1 did not produce any change in egg excretion (Fig 5a). However, as expected, the anti-IL-13 antibody treatment produced an increase in egg excretion (Fig 5b); while the injection of rIL-13 induced a reduction in egg shedding (Fig 5c). Therefore, neutralizing IL-13, which plays a role in the orchestration of the Th2 response, at day 28 post Hp infection, reproduced the results obtained in coinfected mice.
Treating single Hp infected mice with recombinant IL-13 or anti-IL-13 antibodies alters egg excretion.a) FEC [ln(eggs/g feces) of mice treated with IgG1 (three injections at day 28, 30 and 32 p.i.). A LMM showed no change between the pre and post injection egg excretion (F1,64 = 0.17, p = 0.6857, n = 7 individuals and 72 observations). b) FEC [ln(eggs/g feces) of mice treated with anti-IL-13 antibodies (three injections at day 28, 30 and 32 p.i.). A LMM showed that egg excretion increased after the treatment (F1,68 = 15.21, p = 0.0002, n = 7 individuals and 76 observations). c) FEC [ln(eggs/g feces) of mice treated with rIL-13 (three injections at day 28, 30 and 32 p.i.). A LMM showed that egg excretion decreased after the treatment (F1,79 = 24.99, p < 0.0001, n = 8 individuals and 88 observations). Dots represent the raw data, the lines represent the fit of the LMM, and the shaded area around the line the 95% CI.
Hp persistence time and cumulative egg excretion in single and coinfected hosts
The rise in the number of excreted eggs might be a transitory adjustment to the immune environment provided by hosts coinfected with Py. Moreover, if there is an energetic trade-off between early and late investment in egg production, over the long run, the cumulative egg excretion might be similar in single Hp infected and in coinfected hosts. To investigate this question, we monitored egg excretion over a 99-day period in single Hp infected and coinfected mice (Py-14 + Hp, Py-28 + Hp). Contrary to the prediction, we did not find that the rise in worm fecundity was transitory, since coinfected mice overall excreted a higher number of eggs compared to single infected hosts (Fig 6a). This result was corroborated by the analysis of the probability of within-host persistence over time, that showed that single infected mice cleared the infection faster compared to the coinfection groups (Fig 6b). Therefore, when Hp infects hosts that had been previously infected with Py, the polarization of the immune response allows the worm to persist for longer and to shed more eggs in the environment.
Coinfection increases cumulative egg excretion and improves within-host persistence.a) Cumulative egg excretion in single Hp infected and coinfected (Py-14 + Hp, Py-28 + Hp) hosts. A one-way ANOVA showed a highly significant difference between groups (Treatment, F2,26 = 6.14, p = 0.0066), and post-hoc comparisons showed that the two coinfection groups had higher cumulative egg excretion compared to single Hp infected mice (Bonferroni corrected p = 0.0181 and p = 0.0156, respectively). Dots represent individual observations; boxes represent the interquartile range (IQR); the horizontal line within each box indicates the median; the larger dot indicates the mean; whiskers extend to the most extreme values within 1.5 × IQR from the quartiles and points beyond the whiskers indicate outliers. Numbers above the boxes indicate the sample size. b) Probability of within-host persistence in single Hp infected and coinfected (Py-14 + Hp, Py-28 + Hp) hosts. The Log-Rank test indicated a highly significant difference between groups (Log-Rank test, χ²2 = 11.01, p = 0.0041, n = 10 single Hp, n = 10 Py-14 + Hp, n = 9 Py-28 + Hp), with single Hp infected hosts clearing the infection faster than both groups of coinfected mice (Sidak corrected p = 0.0063 and p = 0.0364, respectively). Crosses indicate censored values.
We then investigated whether the downregulation of the Th2 response in coinfected hosts persisted long enough to possibly account for a higher cumulative egg excretion and longer within-host persistence of Hp in coinfected hosts. Overall, as expected, both IL-4 and IL-13 genes were less expressed at day 99 than at day 14, 21 or 35 post Hp infection, as shown by higher Ct values (S2 Fig). However, the difference in gene expression between groups persisted at day 99 p.i. (Fig 7), with coinfected hosts having a downregulated expression of the two Th2 cytokine genes.
Coinfection downregulates the gene expression of Th2 cytokines at day 99 post-infection.a) Δ Ct values of IL-4 relative to β-actin at day 99 p.i. in single Hp and coinfected (Py-14 + Hp, Py-28 + Hp) mice. A one-way ANOVA showed a highly significant difference between groups (F2,26 = 9.64, p = 0.0007), and post-hoc comparisons of LS-means showed that Δ Ct were significantly higher in Py-28 + Hp coinfected hosts compared to single Hp infected mice (Bonferroni corrected p = 0.0005), but did not differ between single Hp infected and Py-14 + Hp coinfected hosts (Bonferroni corrected p = 0.0900). b) log2 fold change in IL-4 expression relative to single Hp infected hosts at day 99 p.i. in coinfected (Py-14 + Hp, Py-28 + Hp) hosts. c) Δ Ct values of IL-13 relative to β-actin at day 99 p.i. in single Hp and coinfected (Py-14 + Hp, Py-28 + Hp) mice. A one-way ANOVA showed a highly significant difference between groups (F2,26 = 10.79, p = 0.0004), and post-hoc comparisons of LS-means showed that both coinfection groups had higher Δ Ct values compared to single Hp infected hosts (Bonferroni corrected p = 0.0341 and p = 0.0003, respectively). d) log2 fold change in IL-13 expression relative to single Hp infected mice at day 99 p.i. in coinfected (Py-14 + Hp, Py-28 + Hp) hosts. Dots represent individual observations; boxes represent the interquartile range (IQR); the horizontal line within each box indicates the median; the larger dot indicates the mean; whiskers extend to the most extreme values within 1.5 × IQR from the quartiles and points beyond the whiskers indicate outliers. Numbers above the x-axis indicate the sample size.
In the absence of reinfection, the patent period (the period during which hosts keep shedding parasitic eggs) depends on two parameters: parasite mortality (background and immune-dependent mortality), and host mortality. In a previous work, we showed that coinfection between Hp and Py increases the severity of the disease in an asymmetrical manner, the disease becoming more severe only when Py infects hosts already harboring Hp [38]. We therefore investigated whether host mortality that occurs in coinfected hosts when Py follows Hp, might offset any potential benefit in terms of more favorable immune environment. To explore this question, we first compared the Hp within-host persistence in single infected mice and in mice that were first infected with Hp and at day 28 with Py, and indeed found that the two groups had similar within-host persistence, although for different reasons (parasite mortality in the single Hp infection group and host mortality in the coinfection group) (Fig 8). However, despite similar within-host persistence, coinfected hosts cumulatively excreted substantially more eggs than single infected hosts (Fig 8b). Therefore, coinfection with Py incurred a net positive effect on cumulative egg excretion, even when coinfection may lead to host mortality.
Coinfection improves cumulative egg excretion even when host mortality sets a limit to the duration of the patent period.a) Probability of within-host persistence in single Hp infected and Hp-28 + Py coinfected hosts. Persistence did not differ between groups (Log-Rank test, χ²1 = 0.042, p = 0.8369, n = 10 single Hp, n = 9 Hp-28 + Py). Crosses indicate censored values. b) Cumulative egg excretion in single infected and coinfected (Hp-28 + Py) hosts. A one-way ANOVA showed that egg excretion was significantly higher in coinfected hosts compared to single infected mice (F1,17 = 31.83, p < 0.0001). Dots represent individual observations; boxes represent the interquartile range (IQR); the horizontal line within each box indicates the median; the larger dot indicates the mean; whiskers extend to the most extreme values within 1.5 × IQR from the quartiles and points beyond the whiskers indicate outliers.
Discussion
Micro and macroparasites elicit different facets of the host immune system that have reciprocal inhibitory effects [32,46–50]. Therefore, hosts harboring a microparasite infection should offer a more favorable ground for the infection with helminths. For instance, mice infected with the gastrointestinal nematode Nippostrongylus brasiliensis and coinfected with the protozoan Toxoplasma gondii had lower Th2 responses as assessed by Th2 cytokine production in the mesenteric lymph nodes and the spleen, and higher and prolonged nematode egg excretion [51]. In a slightly different system of concomitant infection with Nippostrongylus brasiliensis and Plasmodium chabaudi, Hoeve et al. [31] also reported an impaired Th2 response. Similar results were also reported in mice coinfected with Plasmodium chabaudi and Hp [32]. In agreement with these previous results, we found that Py infection weakened the Th2 response and this correlated with higher parasite fecundity and longer patent period (within-host persistence). On the contrary, we did not find strong evidence suggesting that previous infection with Py improves the infectivity of larvae, defined as the probability of molting into reproductive adults, since the dry mass of adult worms retrieved in the intestinal lumen did not differ between single infected and coinfected mice. Asymmetrical effects of altered Th2 response due to coinfection with microparasites have already been reported in similar systems. For instance, mice infected with Hp and subsequently infected with Leishamia infantum resulted in higher Hp egg shedding, while the larval infectivity was not affected by the coinfection [52]. Enhanced Hp fecundity, independently of larval survival, has also been found in mice that had been previously infected with Toxoplasma gondii [33]. We also provided additional evidence indicating that coinfection had a direct effect on worm fecundity. Indeed, when mice harboring a chronic infection with Hp were subsequently infected with Py they excreted a higher number of nematode eggs in the feces. This result unambiguously shows that worms were able to rapidly adjust their investment into egg production, because at day 28 post Hp infection all larvae have molted into adult worms and in the absence of reinfection, the population size of reproductive worms cannot increase. Whatever the demographic parameters affected by the coinfection, these results also show that the increase in egg shedding does not depend on the order of infection, since we found that mice that were either infected first with Py or with Hp excreted more nematode eggs compared to single infected hosts.
One of the central tenets of life history theory is that when resources are limited, increased investment into reproductive effort at a given stage (age) should result in decreased investment into reproductive effort at later stages (ages) or increased mortality rate [53]. Based on this reasoning, we predicted that the enhanced egg shedding that we observed from day 11 to day 35 p.i. should be paid later on, possibly resulting in similar cumulative egg excretion in single infected and coinfected mice. Contrary to this prediction, we found that coinfected mice consistently excreted more eggs over the long term compared to single infected hosts. We also showed that worms in coinfected hosts had longer within-host persistence and therefore a longer patent period allowing them to cumulatively shedding more eggs in the environment. The absence of a trade-off suggests that the host immune response strongly constraints the expression of the life history traits of Hp (fecundity and within-host persistence) and that relaxing these constraints (e.g., in coinfected hosts) allows the nematodes to release more propagules over time.
As mentioned above, coinfection can exacerbate the severity of the disease symptoms compared to single infection, and possibly incurring host death, which obviously sets a limit to the patent period. In a previous work, we showed that the probability of mortality of coinfected hosts strongly depends on the order of infection [38]. When hosts are first infected with Py and subsequently with Hp, there is no host mortality and in this case the duration of the patent period is only constrained by parasite mortality (background + immune mediated mortality). However, when hosts are first infected with Hp and subsequently with Py, exacerbation of disease symptoms can incur host mortality. In this case, the duration of the patent period does not depend only on parasite mortality but also on host mortality. Under this scenario, it is therefore possible that the benefit of increased egg shedding in coinfected hosts is offset by the reduction of the patent period due to host mortality. However, contrary to this hypothesis, we found that despite similar within-host persistence in single Hp infected and coinfected mice (when Py infects last), due to host mortality in the coinfection group, the cumulative egg excretion was still substantially higher in coinfected hosts. This result therefore shows that the net benefit of coinfection for Hp propagule production is consistent across different coinfection scenarios, even when host mortality sets a limit to the patent period.
We showed that the differences in fecundity and within-host persistence are likely to be driven by the immune environment provided by the host. In agreement with previous results, we found that Hp infection induced a Th2 response characterized by the upregulation of Th2 cytokine genes, higher levels of circulating Th2 cytokines and an expansion of Th2 cells (GATA-3^+^ CD4^+^ T cells) in the spleen. However, coinfection consistently weakened the anti-Hp immune response, since coinfected hosts had downregulated Th2 cytokine gene expression, reduced levels of circulating Th2 cytokines and reduced expansion of Th2 cells. We interpreted these inhibitory effects as the consequence of the polarization of the immune response towards anti-Py effectors (Th1), known to have inhibitory effects on the Th2 response (e.g., [33]). Alternatively, condition-dependent effects might also contribute to explain the rise in Hp fecundity, if coinfected hosts suffer from worst body condition. To disentangle these two hypotheses, we conducted an experiment where Hp infected mice were treated with anti-IL-13 antibodies, in the absence of Py infection. Therefore, if the rise in Hp fecundity is mediated by immune effects, we should expect that neutralizing IL-13 should reproduce the results observed in coinfected hosts. Indeed, we found that treating Hp infected mice with anti-IL-13 monoclonal antibodies increased egg excretion, while treatment with rIL-13 reduced egg shedding. Therefore, these results further show that the adjustment of fecundity does not reflect a parasite response to poor host condition, caused by the coinfection, and corroborate previous work showing that IL-13 and its receptor are important component of the anti-Hp response [54]. Manipulation of the Th2 or the Th1 response using a variety of approaches has provided consistent results, similar to those reported here. For instance, blockade of IFN-γ with anti-IFN-γ antibodies restored the Th2 response in a coinfection model between Hp and Plasmodium chabaudi [32], while treating mice with IFN-γ increased Hp fecundity [55]. Blockade or supplementation with IL-4 also had the expected results on Hp fecundity and infection persistence [56,57]. Manipulation of IL-4 had no effect on the larval stage [57] which is also in agreement with the reports of no effect of coinfection on larval survival. IL-4 and IL-13 have also been shown to play a role in the expulsion of adult worms by promoting peristaltic movements and the contractility of intestinal smooth muscles [58].
Interestingly, we found that the difference between single infected and coinfected hosts in the expression of Th2 cytokine genes persisted well after the Py infection had been cleared. Of course, the amount of IL-4 and IL-13 mRNA was lower at day 99 p.i. compared to day 14 or day 21 p.i., but the expression of both genes were still downregulated in coinfected hosts. Although a long lasting anti-Hp Th2 response has already been reported up to day 70 p.i. [59], we cannot affirm that the longer persistence and higher cumulative egg excretion is due to a permanent downregulation of the Th2 response in coinfected hosts. Further work should elucidate this hypothesis and the possible mechanisms underlying a prolonged inhibition (once Py has been cleared) of the Th2 response in coinfected hosts.
It is tempting to use these results to predict the epidemiological consequences of malaria coinfection on STHs in humans. While work on model systems can provide valuable insights to inform future studies on more complex systems, extrapolating these results to human helminthiases in malaria-endemic areas is not straightforward. Instead, we suggest viewing this work as proof of concept that malaria coinfection can alter the epidemiology of STHs via immune-mediated effects. In particular, we propose that malaria coinfection may positively impact the prevalence of STH infection, as coinfected hosts contribute to maintaining a larger reservoir of transmissible stages from which people can become infected. Similarly, we can speculate about the potential public health consequences of malaria treatments. Preventing and treating Plasmodium infection is paramount, given the significant mortality burden of malaria (610,000 deaths in 2024) [60]. However, protecting the population against malaria may also reduce the prevalence of helminth infection, if coinfection increases the reservoir of transmissible stages in the environment. Further research is required to confirm this in human helminth/malaria coinfections. A final public health issue refers to the possible impairment of the efficacy of antimalarial vaccines in coinfected hosts. However, although some studies have reported a reduced vaccine efficacy in helminth-infected hosts (e.g., [61,62]), current evidence suggests a more complex pattern, depending on the species of the coinfecting helminth and the Plasmodium antigen [63–66].
Limitations
We would like to acknowledge a few limitations of this work. First, our assessment of the Th2 response was limited to the systemic level (IL-4 and IL-13 gene expression in the spleen, IL-5 and IL-13 levels in plasma, GATA-3^+^ CD4^+^ T cells in the spleen). As Hp and Py do not share the same resources or infection site, the most likely mechanism underlying the observed effects of coinfection is systemic immune responses. Nevertheless, assessing local immune effectors in the intestinal compartment should provide valuable confirmatory evidence for the results reported here. Second, the manipulation of the Th2 response only involved the neutralization or the administration of IL-13. Other experimental approaches involving other cytokines (e.g., IL-4) or immune signaling pathways (e.g., STAT6) could also be informative. Third, our assessment of Hp life history traits does not refer to individual worms but to the whole infra-host nematode population. Indeed, for obvious reasons, it is not possible to monitor individual worm fecundity and survival. Moreover, we only used a proxy of cumulative egg excretion since we measured FEC on a weekly basis. Similarly, our conclusion of no trade-off between early and late fecundity, resulting in a higher cumulative egg excretion in coinfected hosts, does not preclude the existence of trade-offs involving other traits. For instance, higher investment into egg production might compromise larval survival in the external environment or larval infectivity, ultimately decreasing the transmission success of worms from coinfected hosts. Further work should elucidate these possible trade-offs. Finally, we only used female mice and further work should confirm whether these results are sex-specific or also apply to male hosts.
Conclusion
In conclusion, these results highlight the potential impact of malaria coinfection (and, more broadly, microparasite coinfection) on the excretion of eggs and the persistence of gastrointestinal nematodes within hosts. They show that coinfection affects key epidemiological and demographic traits, generating positive feedback whereby coinfected hosts maintain a large reservoir of transmissible stages in the environment. This, in turn, sustains a high risk of coinfection. If these results can be extrapolated to other systems involving human microparasites and STHs, they suggest that, in addition to the direct benefit for the host, targeting the coinfecting microparasites might also indirectly reduce helminth transmission success by reducing the number of transmissible stages found in the external environment and, consequently, reducing the risk of infection. Nevertheless, while these results provide proof of concept that malaria coinfection can alter STH epidemiology, further work is needed to establish the extent to which they can be extrapolated to human helminthiases in malaria-endemic areas.
Supporting information
S1 TablePrimer sequences used for the RT-qPCR of IL-4 and IL-13 and the two reference genes (β-actin and GAPDH).(DOCX)
S2 Table7-color spectral flow cytometry panel used for the staining of mouse splenocytes.(DOCX)
S1 FigGating strategy used in flow cytometry analysis to identify Th2 cell subset within mouse total splenocytes.(1) Time parameter to exclude microfluidic fluctuations. Selection of (2) splenocytes without debris and (3) single cells. Selection of (4) live LIVE/DEAD-Blue^−^ and immune CD45^+^ cells. Selection of (5) CD19^-^CD3^+^ T cells. Selection of (6) CD8^-^CD4^+^ helper T cells. Selection of (7) GATA3^+^ Th2 cell subset within CD4^+^ T lymphocytes.(PDF)
S2 FigMean Ct values of IL-4 and IL-13 increased at day 99 post-infection.a) Ct values (mean of two technical replicates) for IL-4 at day 14, 21, 35 and 99 p.i. in single Hp infected and coinfected (Py-14 + Hp, Py-28 + Hp) mice. A one-way ANOVA showed a highly significant effect of time p.i. (F_3,101_ = 103.32; p < 0.0001). Pairwise comparisons of LS-means showed that values at day 99 p.i. were significantly higher than values at day 14, 21 and 35 p.i. (all Bonferroni corrected p’s < 0.0001). b) Ct values (mean of two technical replicates) for IL-13 at day 14, 21, 35 and 99 p.i. in single Hp infected and coinfected (Py-14 + Hp, Py-28 + Hp) mice. A one-way ANOVA showed a highly significant effect of time p.i. (F_3,100_ = 46.64; p < 0.0001). Pairwise comparisons of LS-means showed that values at day 99 p.i. were significantly higher than values at day 14, 21 and 35 p.i. (all Bonferroni corrected p’s < 0.0001). Dots represent individual observations; boxes represent the interquartile range (IQR); the horizontal line within each box indicates the median; the larger dot indicates the mean; whiskers extend to the most extreme values within 1.5 × IQR from the quartiles and points beyond the whiskers indicate outliers. Numbers just above the x-axis indicate the sample size.(PDF)
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Brooker S. Estimating the global distribution and disease burden of intestinal nematode infections: adding up the numbers--a review. Int J Parasitol. 2010;40(10):1137–44. doi: 10.1016/j.ijpara.2010.04.004 20430032 PMC 3034165 · doi ↗ · pubmed ↗
- 2Loukas A, Maizels RM, Hotez PJ. The yin and yang of human soil-transmitted helminth infections. Int J Parasitol. 2021;51(13–14):1243–53. doi: 10.1016/j.ijpara.2021.11.001 34774540 PMC 9145206 · doi ↗ · pubmed ↗
- 3World Health Organisation. Soil-transmitted helminth infections. 2023 [cited 2025 Feb 18]. https://www.who.int/news-room/fact-sheets/detail/soil-transmitted-helminth-infections
- 4Else KJ, Keiser J, Holland CV, Grencis RK, Sattelle DB, Fujiwara RT, et al. Whipworms and roundworms infections. Nature Review Disease Primers. 2020;6:44.10.1038/s 41572-020-0171-332467581 · doi ↗ · pubmed ↗
- 5Lustigman S, Prichard RK, Gazzinelli A, Grant WN, Boatin BA, Mc Carthy JS, et al. A research agenda for helminth diseases of humans: the problem of helminthiases. P Lo S Negl Trop Dis. 2012;6(4):e 1582. doi: 10.1371/journal.pntd.0001582 22545164 PMC 3335854 · doi ↗ · pubmed ↗
- 6Campbell SJ, Nery SV, Doi SA, Gray DJ, Soares Magalhães RJ, Mc Carthy JS, et al. Complexities and Perplexities: A Critical Appraisal of the Evidence for Soil-Transmitted Helminth Infection-Related Morbidity. P Lo S Negl Trop Dis. 2016;10(5):e 0004566. doi: 10.1371/journal.pntd.0004566 27196100 PMC 4873196 · doi ↗ · pubmed ↗
- 7Raj E, Calvo-Urbano B, Heffernan C, Halder J, Webster JP. Systematic review to evaluate a potential association between helminth infection and physical stunting in children. Parasit Vectors. 2022;15(1):135. doi: 10.1186/s 13071-022-05235-5 35443698 PMC 9022337 · doi ↗ · pubmed ↗
- 8Tamarozzi F, Martello E, Giorli G, Fittipaldo A, Staffolani S, Montresor A, et al. Morbidity Associated with Chronic Strongyloides stercoralis Infection: A Systematic Review and Meta-Analysis. Am J Trop Med Hyg. 2019;100(6):1305–11. doi: 10.4269/ajtmh.18-0895 30963990 PMC 6553888 · doi ↗ · pubmed ↗