Modulation of intestinal epithelial permeability by chronic small intestinal helminth infections

Increased permeability of the intestinal epithelial layer is linked to the pathogenesis and perpetuation of a wide range of intestinal and extra‐intestinal diseases. Infecting humans with controlled doses of helminths, such as human hookworm (termed hookworm therapy), is proposed as a treatment for many of the same diseases. Helminths induce immunoregulatory changes in their host which could decrease epithelial permeability, which is highlighted as a potential mechanism through which helminths treat disease. Despite this, the influence of a chronic helminth infection on epithelial permeability remains unclear. This study uses the chronically infecting intestinal helminth Heligmosomoides polygyrus to reveal alterations in the expression of intestinal tight junction proteins and epithelial permeability during the infection course. In the acute infection phase (1 week postinfection), an increase in intestinal epithelial permeability is observed. Consistent with this finding, jejunal claudin‐2 is upregulated and tricellulin is downregulated. By contrast, in the chronic infection phase (6 weeks postinfection), colonic claudin‐1 is upregulated and epithelial permeability decreases. Importantly, this study also investigates changes in epithelial permeability in a small human cohort experimentally challenged with the human hookworm, Necator americanus. It demonstrates a trend toward small intestinal permeability increasing in the acute infection phase (8 weeks postinfection), and colonic and whole gut permeability decreasing in the chronic infection phase (24 weeks postinfection), suggesting a conserved epithelial response between humans and mice. In summary, our findings demonstrate dynamic changes in epithelial permeability during a chronic helminth infection and provide another plausible mechanism by which chronic helminth infections could be utilized to treat disease.


INTRODUCTION
The intestinal epithelial layer is a continuous cellular monolayer that forms a selectively permeable barrier separating the intestinal lumen from the lamina propria.It facilitates the absorption of nutrients while limiting the entry of harmful microbes and other antigens.Increased permeability of the epithelial layer is linked to the pathogenesis and perpetuation of a wide range of intestinal and extra-intestinal diseases, including inflammatory bowel disease, coeliac disease, rheumatoid arthritis, multiple sclerosis and diabetes. 1ight junctions, a group of cytosolic proteins that bind intestinal epithelial cells together, are rate limiting for epithelial permeability.Tight junction proteins include claudins, occludin, zonula occludens (encoded by TJP-1) and tricellulin (encoded by MARVEL2). 1 Claudins are either barrier sealing, which reduce epithelial permeability (e.g.claudin 1, 3, 5, 8 and 20), or pore forming, which increase permeability (e.g.claudin 2 and 15). 2 A greater expression of the other tight junction proteins reduces epithelial permeability.Another important determinant of epithelial permeability is myosin light-chain kinase (encoded by MYLK ) which phosphorylates the actin-myosin ring, triggering its contraction to increase epithelial permeability. 18][9] This forms part of the coordinated effector response designed to expel the helminth, collectively referred to as "weep and sweep." 10 Some helminth species establish chronic infections in their host. 11A chronic helminth infection is characterized by greater immunoregulation and a moderated T h 2 immune response. 12,134][15] These observations suggest that helminths could decrease permeability in their chronic infection phase.To date, changes in epithelial permeability in a helminth's chronic infection phase have not been examined.
Alterations in epithelial permeability during a helminth infection may have implications for human health.For example, elevated intestinal permeability could contribute to the negative health effects experienced by individuals with a high infection burden. 11Conversely, evidence from epidemiological and preclinical studies, and to a lesser degree from human clinical trials using controlled doses of helminths, suggests that helminths may protect their host against diseases linked to heightened epithelial permeability. 3This implies that helminths might decrease permeability under specific conditions.The effect of controlled helminth infections on epithelial permeability in humans has not been assessed.
This study examines changes in epithelial permeability in a chronic small intestinal helminth infection using the preclinical helminth model (Heligmosomoides polygyrus) and humans experimentally challenged with a controlled dose of the human hookworm Necator americanus.In the preclinical model, changes during the acute infection phase were assessed at 7 and 14 days postinfection (d.p.i.), while changes in the chronic infection phase were assessed at 42 d.p.i. 5 In the human participants, the acute and chronic infection phase was defined as 8 and 24 weeks postinfection, respectively. 12Findings from this study demonstrate dynamic changes in epithelial permeability that provide a mechanism through which helminths could either worsen human health or protect the host against disease.

An acute H. polygyrus infection increases epithelial permeability
Epithelial permeability and tight junction protein messenger RNA (mRNA) expression were assessed in C57BL/6 mice acutely infected with H. polygyrus.An increase in epithelial permeability to 4-kD fluorescein isothiocyanate (FITC)-dextran was observed 7 d.p.i.(Figure 1a).Consistent with this finding, in the jejunum (this helminth's intestinal niche) of mice infected with H. polygyrus for 7 days, claudin-2 mRNA expression increased and tricellulin (encoded by Marveld2) expression decreased (Figure 1b, Supplementary figure 1a).At 14 d.p.i., while only a trend toward an increase in epithelial permeability was observed (Figure 1a), significant changes in the expression of tight junction proteins associated with increased permeability persisted in the jejunum (Figure 1d, Supplementary figure 1c).There was no change in the expression of tight junction proteins in the colon in the acute infection phase (Figure 1c, e, Supplementary figure 1b and d).

A chronic H. polygyrus infection decreases epithelial permeability
Tight junction protein mRNA expression and epithelial permeability were assessed in the chronic infection phase (42 d.p.i.).An ongoing H. polygyrus infection was confirmed in all infected mice at 6 weeks postinfection (mean intestinal worm burden = 16 worms, range 7-23).At this time point, no significant differences in the mRNA expression of tight junction proteins were observed in the jejunum between infected and uninfected mice (Figure 2a, Supplementary figure 1e).In the colon, a significant increase in the mRNA expression of the barrier-forming claudin, claudin-1, was observed in infected mice (Figure 2b, Supplementary figure 1f).Staining for claudin-1 in the colon using immunofluorescence supported the increase in claudin-1 protein expression in infected mice at this time point (Figure 2c, d).Consistent with these findings, infected mice also demonstrated a significant decrease in epithelial permeability to 4-kD FITC-dextran in the chronic infection phase (Figure 2e).Importantly, given the dosing of FITC-dextran is weight based, there was no difference in weights between infected mice and their age-matched controls which could explain this finding (Supplementary figure 2).

Experimental hookworm infection alters epithelial permeability in humans
The effect of a chronic human hookworm infection on intestinal epithelial permeability was assessed in humans.Four healthy individuals (median age 33 years, 3 female) were infected with 30 N. americanus larvae.Intestinal epithelial permeability was assessed using the multisugar permeability test at baseline (before infection), 8 weeks postinfection (acute infection phase) and 24 weeks postinfection (chronic infection phase).This test involves ingesting four sugars (lactulose, rhamnose, sucralose and erythritol) and then measuring the concentrations of these sugars in urine collected over the 24 h following ingestion.Epithelial permeability is assessed in the small intestine (lactulose-to-rhamnose ratio in the 0-2-h urinary fraction), colon (sucralose-to-erythritol ratio in the 8-24-h urinary fraction) and whole gut (sucralose-to-erythritol ratio in the 0-24-h urinary fraction).
7][18] To examine this in humans, the T h 2 immune response was assessed in the four participants undergoing assessment of epithelial permeability by measuring peripheral blood eosinophils and fecal eosinophilic cationic protein (ECP), a protein released during degranulation of eosinophils.During the acute infection phase at week 8, all participants experienced an increase in peripheral blood eosinophils and fecal ECP compared with baseline (median change 3.6 9 10 9 L À1 , IQR 2.9-5.9,P = 0.125 and 1667.0 lg g À1 , IQR 823.9-7359.0,P = 0.125, respectively) highlighting a systemic and intestinal T h 2 immune response to the invading helminth (Figure 3b, Supplementary table 1).
Participant 2 experienced the greatest increase in peripheral blood eosinophils and fecal ECP (Figure 3b, Supplementary table 1) and demonstrated the largest increase in small intestinal permeability in the acute infection phase (Table 2 and Figure 3a).This finding may indicate that the T h 2 immune response also drives an (c) Self-reported symptoms for each of the four participants infected with human hookworm.Symptoms were assessed using a modified version of the Talley Gut Symptom Questionnaire. 34Symptoms were self-graded by the participant as absent, mild (nagging or annoying), moderate (strong negative influence on daily living) or severe (disabling).LRR, lactulose-to-rhamnose ratio; SER, sucralose-to-erythritol ratio.
increase in small intestinal permeability during a helminth infection in humans.Interestingly, participant 2 also experienced the worst gastrointestinal symptoms (diarrhea and abdominal pain) in the acute infection phase, suggesting a possible association between gastrointestinal symptoms and changes in epithelial permeability in individuals infected with helminths (Figure 3c).

DISCUSSION
][5][6] Furthermore, whether changes in epithelial permeability observed in preclinical models translate to humans has not been examined. 3This study demonstrates that mice acutely infected with the helminth H. polygyrus exhibit increased expression of jejunal claudin-2 mRNA and downregulation of jejunal tricellulin mRNA, and an increase in intestinal epithelial permeability.By contrast, mice chronically infected with H. polygyrus display increased colonic expression of the barrier-forming tight junction protein claudin-1 and decreased epithelial permeability.Consistent with these preclinical data, all four humans experimentally challenged with the hookworm N. americanus demonstrated an increase in small intestinal permeability in the acute infection phase and a decrease in colonic and whole gut permeability in the chronic infection phase.
Humans experimentally challenged with helminths commonly experience diarrhea in the early stages of the infection (typically weeks 4-8 postinfection), which resolves when the helminth establishes a chronic infection. 12,19The changes in epithelial permeability observed in this study could explain this pattern of symptoms.Increased epithelial permeability, as observed in the acute infection phase, causes diarrhea as a result of sodium and water moving into the intestinal lumen. 1 This process forms an important component of the T h 2-mediated "weep and sweep" response aimed at expelling the helminth. 10As helminths enter their chronic infection phase, the T h 2 immune response dissipates, epithelial permeability normalizes and gastrointestinal symptoms resolve.In contrast to humans experimentally challenged with helminths, individuals living in endemic areas are often reinfected with helminths, leading to a sustained T h 2 immune response. 20This could cause persistent elevation in epithelial permeability leading to sustained gastrointestinal symptoms and the negative health effects experienced by humans continuously reinfected with high burdens of helminths. 11xcessive intestinal epithelial permeability can allow the entry of microbes, microbial products and other toxins into the body.This process is proposed to cause or contribute to the pathogenesis of a range of human diseases. 3,21Data from epidemiological, preclinical and, to a lesser extent, human clinical studies, suggest that helminths may protect their host from developing many of the same diseases that are associated with excessive intestinal epithelial permeability, including inflammatory bowel disease, rheumatoid arthritis, multiple sclerosis and metabolic disease. 22This study's observation that epithelial permeability decreases in mice and humans chronically infected with helminths provides a potential mechanism through which helminths could confer this protection.
Alterations in the colonic microbiome could potentially account for the observed increase in colonic claudin-1 expression in mice, and the possible reduction in human colonic epithelial permeability during the chronic infection phase, while small intestinal tight junction protein expression remains stable.Previous studies have shown that small intestinal helminths can stimulate the production of short-chain fatty acids by the host's colonic microbiome. 14,23,24Short-chain fatty acids are demonstrated to facilitate the assembly of tight junction proteins, including upregulating the expression of claudin-1. 25,26hile this mechanism could explain the current study's findings, the interaction between helminth infection, colonic microbiome alterations and short-chain fatty acid production was not directly examined.Importantly, the possible reduction in colonic permeability in a helminths' chronic infection phase could be harnessed to treat colonic diseases caused by increased epithelial permeability, such as ulcerative colitis. 19,27Future, larger-scale human studies are required to confirm this finding and should explicitly investigate potential mechanisms.
This study used oral sugars to assess epithelial permeability in mice and humans; however, other factors altered during a helminth infection could also influence the movement of these compounds from the intestinal lumen into the systemic circulation.For example, infection with helminths changes the properties of the mucus layer, and changes in the mucus viscosity could alter the diffusion of FITC-dextran through colonic mucus. 3,28Furthermore, helminth infections increase intestinal motility which could accelerate the transit of these compounds through the intestine and reduce their absorption into the systemic circulation. 5,10Other techniques to assess epithelial permeability could help overcome these issues.One such technique is measuring permeability in ex vivo samples, although it should be noted that removing tissues from their natural context can also influence permeability.Alternatively, endoscopic techniques can be used for measuring permeability directly in human patients.However, this approach often involves invasive procedures, which may be less desirable to patients. 29,30n conclusion, the observed changes in epithelial permeability during a chronic helminth infection have important health implications for humans.Increased epithelial permeability in the acute infection phase may persist in some individuals living in endemic areas experiencing repeated helminth infections, leading to negative effects on the host's health.Conversely, the observed decrease in epithelial permeability in the helminths' chronic infection phase provides a potential mechanism through which certain species of helminths could protect their hosts against disease.While larger human studies are needed to confirm these findings, these initial findings suggest that there is a potential for controlled doses of chronically infecting helminths, such as human hookworm, to treat patients with excessive intestinal epithelial permeability.

Mice and H. polygyrus infection
All animal procedures were approved by the Victoria University Animal Ethics Committee, Wellington, New Zealand.Specific-pathogen-free C57BL/6 mice were bred and housed at the Biomedical Research Unit, Malaghan Institute of Medical Research, Wellington, New Zealand.All the mice used in this study were 8-to 10-week-old female mice.Mice were maintained on standard rodent chow and acidified water ad libitum, with a 12-h light-dark cycle.
The techniques used for H. polygyrus maintenance and infection are described previously. 31Mice were orally infected with approximately 200 larvae.Worm burdens were assessed by counting the number of worms present in the proximal small intestine using the Baermann technique. 31In brief, the proximal half of the small intestine was removed, opened longitudinally and incubated in phosphate-buffered saline for 12 h before collecting and counting adult worms.
Mice were killed and dissected at 7 and 14 d.p.i. to analyze changes in the acute infection phase of the helminth, while 42 d.p.i. was used to analyze changes in the chronic infection phase. 5Age-matched uninfected mice were used as controls.

Human experimental N. americanus infection
Four healthy individuals with no history of allergies, autoimmune or metabolic diseases and with no previous helminth infections (other than Enterobius vermicularis), were infected with 30 N. americanus larvae.Participants were followed up for 24 weeks postinfection with regular visits for symptom reporting, serum and fecal samples and multisugar permeability tests (discussed later).This study was conducted in Wellington, New Zealand, and was approved by the Health and Disability Ethics Committee (Ethics reference 19/CEN/81), and registered on the Australian New Zealand Clinical Trials Registry (ACTRN12619001129178).
Necator americanus larvae were developed from eggs isolated from stool samples provided by human donors infected for this purpose.The larvae were repeatedly washed in an iodine solution before testing for morphological integrity and viability/motility by an experienced technician using dissecting microscopy.Aliquots of 30 larvae were stored in 200 mL of deionized water contained in small glass microtubes and kept at approximately 25°C and protected from light for up to 1 week before inoculation.A dose of 30 larvae was used for each participant, administered through two dressings, each containing 15 larvae applied on each forearm.Detection of patency was performed through the presence of N. americanus eggs in the feces of infected individuals using the McMaster technique from week 6 postinfection.

Multisugar permeability test
Segmental permeability of the intestinal tract was assessed using the validated multisugar test performed at baseline (within 2 weeks before infection), 8 (acute infection phase) and 24 (chronic infection phase) weeks following the infection. 32,33n the 3 days before the test, as well as during the test, participants were asked to refrain from strenuous physical exercise; smoking and consumption of alcohol, nonsteroidal anti-inflammatories and artificial sweeteners.From 5 PM the evening before the test, the participant only ate food from a provided list of foods that did not contain the sugars of interest and then fasted overnight.Each test was performed at the same time of the day.Participants provided a baseline urine sample and then ingested a mix of water-soluble sugar probes.This included 1 g lactulose (Douglas Pharmaceuticals Ltd, Auckland, New Zealand), 0.5 g L-rhamnose (Sigma-Aldrich, St. Louis, MO, USA), 1 g sucralose (Brenntag, Auckland, New Zealand) and 1 g erythritol (NOW Foods, Bloomingdale, IL, USA), dissolved in 200 mL of warm tap water.After ingestion, participants collected all urine passed over the following 24 h in three separate fractions: 0-2 h, 2-8 h and 8-24 h.During the first 2 h of urine collection, participants were asked to refrain from any food or drink, except for water (250 mL h À1 ).At 2-and 6-h after sugar ingestion, participants ate either 2 pieces of plain white bread or 340 g of plain rolled oats with water (participants ate the same item in subsequent tests).At 8 h after sugar ingestion, participants were able to eat only from the same provided list of foods until completion of the test.Throughout the test, participants were encouraged to drink water (250 mL h À1 ).Urine collections could occur in their workplace or at home, and the urine samples were kept cool or refrigerated.
After the collection periods, volumes of urine fractions were determined, urine was centrifuged at 1800 g for 10 min at 4°C and aliquots of supernatant were frozen at À80°C until analysis.Sugar probes were analyzed by the hydrophilic interaction liquid chromatography with tandem mass spectrometry.Small intestinal permeability was determined by 0-2-h urine lactulose-to-L-rhamnose ratio, colonic permeability by sucralose-to-erythritol ratio in 8-24-h urine and whole gut permeability by sucralose-to-erythritol ratio in 0-24-h urine.

Symptom questionnaire
Possible symptoms attributable to a hookworm infection were assessed using a modified version of the Talley Gut Symptom Questionnaire. 34Symptoms were graded as mild (nagging or annoying), moderate (strong negative influence on daily living) and severe (disabling) and were self-reported by each participant.

Assessment of fecal eosinophil degranulation products
For extraction of ECP from feces the granule protein extraction protocol was used as described elsewhere. 35Feces samples were self-collected and immediately frozen at À80°C.On the day of processing, feces samples were thawed overnight in the refrigerator or at room temperature for 1 h.Feces samples were weighted and diluted five times in extraction buffer consisting of phosphate-buffered saline, pH 7.4, supplemented with 10-mmol/L ethylenediaminetetraacetic acid, 0.2% N-acetyl-N,N,N-trimethylammonium bromide (Sigma-Aldrich, St. Louis, MO, USA), 20% glycerol (Sigma-Aldrich, St. Louis, MO, USA), 0.05% Tween-20 and 1% bovine serum albumin.The mixture was homogenized using a homogenizer mixer until a homogenous solution was obtained.After incubation at 4°C for 30-45 min and mixing, the homogenate was centrifuged at 4000 g for 30 min at 5°C.The supernatant was aliquoted and frozen at À20°C for later enzyme-linked immunosorbent assay analysis.For each sample, a separate tube of the diluted homogenate was weighed and centrifuged.By weighing the pellet obtained after discarding the supernatant, a measure of semidry weight was obtained.Fecal ECP was measured by enzyme-linked immunosorbent assay (Diagnostics Development AB, Uppsala, Sweden) according to the manufacturer's protocol.The levels of markers in feces were adjusted for water content and expressed as micrograms per gram of semidry feces. 19

FITC-dextran permeability test
After a 4-h fast from food and water, FITC-dextran 4 kDa (Sigma-Aldrich, St. Louis, MO, USA) dissolved in phosphate-buffered saline (100 mg mL À1 ) was administered to each mouse (44 mg 100 g À1 body weight) by oral gavage.After 4 h, blood was collected in tubes (Microvette 500 Serum Gel, Sarstedt, N€ umbrecht, Germany) and the serum was separated as per the manufacturer's instructions.The concentration of FITC-dextran was determined using a fluorimeter with an excitation wavelength of 485 nm and an emission wavelength of 528 nm.Serially diluted FITC-dextran was used to establish a standard curve. 36,37munofluorescence staining and quantification A 1-cm segment of midcolon (3 cm proximal to the anorectal junction) was fixed in 10% neutral buffered formalin and embedded in paraffin using standard histological techniques.Paraffin sections (5 lm) were cut using a paraffin microtome, attached to glass slides and incubated overnight at 70°C.Formalin-fixed sections were deparaffinized in stepwise baths of xylene, hydrated in decreasing concentrations of ethanol and then stained for claudin-1 polyclonal antibody (PA5-16833; Thermo Fisher Scientific, MA, USA), followed by goat anti-rabbit Alexa 555-conjugated antibody (A32732; Thermo Fisher Scientific, MA, USA), and DAPI (4 0 ,6-diamidino-2-phenylindole), using standard indirect immunofluorescence techniques.
Images were acquired using an inverted microscope (IX83; Olympus, Japan) equipped with a laser scanning confocal head (FV3000; Olympus, Japan) and a 209 objective 0.75 NA using FluoView software (version 2.4.1).The images were acquired using a 405 laser (50 mW) to excite DAPI dye and a 561 laser (40 mW) to excite AF555 dyes; detection of configuration of 450/80 and 606/80 was used, respectively.Images were recorded using 0.3107 lm pixel size and ensuring matching detector linearity.
Image analysis was performed by a blinded researcher using QuPath software (verson 0.4.2). 38 Individual cells were identified using nuclear/cytoplasm segmentation.To ensure only epithelial cells were selected, a minimum threshold intensity of AF555 was determined to detect only positively stained cells.The same algorithm was applied to all images.The mean fluorescence intensity within selected cells from 5 to 10 images per mouse (409 magnification) was determined to quantify claudin-1 protein expression.

RNA isolation and real-time qPCR
RNA was isolated from 0.5 cm of the mid-jejunum (10 cm distal to the gastroduodenal junction) or mid-colon (4 cm proximal to the anorectal junction) using RNeasy Mini Kit (Qiagen, Hilden, Germany).The quality and concentration of the RNA were assessed via Nanodrop (NanoDrop One C; Thermo Fisher Scientific, MA, USA).Then, 2000 ng of total RNA was used for complementary DNA synthesis using a high-capacity RNA-to-complementary DNA kit (Applied Biosystems, Thermo Fisher Scientific, MA, USA).Quantitative real-time PCR was performed using SYBR Green PCR Master Mix (Applied Biosystems, Thermo Fisher Scientific, MA, USA) using the Real-Time PCR System QuantStudio 7 instrument (Applied Biosystems, Thermo Fisher Scientific, MA, USA) according to the manufacturer's instructions.Forward and reverse primer sequences are displayed in Supplementary table 2. Rplp0 was determined to be the most stable housekeeping gene in the jejunum and colon in the acute infection phase compared with uninfected age-matched mice, so was used to normalize gene expression (Supplementary figure 3). 39,40Changes in gene expression are expressed as fold difference to uninfected age-matched mice following the 2 DDCt method.

Statistical analysis
Preclinical data are represented as mean AE standard error of the mean.The two-tailed Student's t-test was used for two-way comparisons.Human data are displayed as median AE IQR.The Wilcoxon matched-pairs signed-rank test was used to compare paired data.P < 0.05 was considered statistically significant.Statistical analysis was performed using GraphPad Prism version 9.5.1 (GraphPad Software, La Jolla, CA, USA).

Intestinal permeability in helminth infections
TC Mules et al.

Figure 1 .
Figure 1.Mice infected with 200 Heligmosomoides polygyrus larvae were assessed for changes in intestinal tight junction protein messenger RNA (mRNA) expression and epithelial permeability in the acute infection phase (7 and 14 d.p.i.).(a) Epithelial permeability in infected mice compared with age-matched controls 7 and 14 d.p.i.Data are from 10 mice/group from two pooled experiments with each experiment represented by a different symbol.(b) and (c) Tight junction protein mRNA expression in the (a) jejunum and (b) colon of infected mice relative to age-matched controls (dashed line) 7 d.p.i.Data are from 7 to 10 mice/group from two pooled experiments with each experiment represented by a different symbol.Data from control mice are displayed in Supplementary figure 1a and b.(d) and (e) Tight junction protein mRNA expression in the (d) jejunum and (e) colon of infected mice relative to age-matched controls (dashed line) 14 d.p.i.Data are from 7 to 10 mice/ group from two pooled experiments with each experiment represented by a different symbol.Data from control mice are displayed in Supplementary figure 1c and d.Rplp0 was used as the housekeeping gene.Data were analyzed with the unpaired Student's t-test.Mean AE standard error of the mean.Significant changes are indicated by an *. *P < 0.05, **P < 0.01.d.p.i., day postinfection.FITC, fluorescein isothiocyanate.

Figure 2 .
Figure 2. Mice infected with 200 Heligmosomoides polygyrus larvae were assessed for changes in intestinal tight junction protein expression and epithelial permeability in the chronic infection phase (42 d.p.i.).(a) and (b) Tight junction protein messenger RNA (mRNA) expression in the (a) jejunum and (b) colon of infected mice relative to age-matched controls (dashed line) in the chronic infection phase.Data are from 7 to 10 mice/ group from two pooled experiments.Data from control mice are displayed in Supplementary figure 1e and f.Rplp0 was used as the housekeeping gene.(c) Mean claudin-1 fluorescence intensity/cell in infected and uninfected mice in the chronic infection phase.Data are from five mice/group from two pooled experiments with each experiment represented by a different symbol.(d) Representative immunofluorescence images stained for claudin-1 in the colon of infected and uninfected mice in the chronic infection phase.Claudin-1 is indicated in magenta, DAPI (4 0 ,6-diamidino-2-phenylindole) is indicated in green.(e) Epithelial permeability in infected mice compared with age-matched controls in the chronic infection phase.Data are from 15 mice/group from two pooled experiments with each experiment represented by a different symbol.Data were analyzed with the unpaired Student's t-test.Mean AE standard error of the mean.Significant changes are indicated by an *. **P < 0.01.FITC, fluorescein isothiocyanate.

Figure 3 .
Figure 3. (a) Changes in small intestinal, colonic and whole gut permeability in the four participants infected with human hookworm, as measured by the multisugar permeability test preinfection (baseline) and in the acute (8 weeks postinfection) and chronic (24 weeks postinfection) infection phase.The dashed line represents the median.(b) Peripheral blood eosinophils and fecal eosinophilic cationic protein (ECP) levels in the four participants infected with human hookworm.The dashed line represents the median.(c)Self-reported symptoms for each of the four participants infected with human hookworm.Symptoms were assessed using a modified version of the Talley Gut Symptom Questionnaire.34 Symptoms were self-graded by the participant as absent, mild (nagging or annoying), moderate (strong negative influence on daily living) or severe (disabling).LRR, lactulose-to-rhamnose ratio; SER, sucralose-to-erythritol ratio.

Table 1 .
Age, gender and fecal egg count 24 weeks following hookworm infection (eggs g À1 feces) in the four participants undergoing assessment of epithelial permeability.

Table 2 .
Changes in small intestinal, colonic and whole gut epithelial permeability from baseline in four participants infected with human hookworm, as measured by the multisugar permeability test preinfection and in the acute (8 weeks postinfection) and chronic (24 weeks postinfection) infection phase.The Wilcoxon matched-pairs signed-rank test was used for statistical analyses.IQR, interquartile range.