Published Ahead of Print 26 December 2013. 2014, 88(6):3161. DOI: 10.1128/JVI.02927-13. J. Virol. 

Magnusson and Lennart SvenssonClaudia Istrate, Marie Hagbom, Elena Vikström, Karl-Eric Diarrhea

ofMotility but Not Permeability at the Onset Rotavirus Infection Increases Intestinal

http://jvi.asm.org/content/88/6/3161Updated information and services can be found at:

These include:

REFERENCEShttp://jvi.asm.org/content/88/6/3161#ref-list-1at:

This article cites 44 articles, 16 of which can be accessed free

CONTENT ALERTS more»articles cite this article),

Receive: RSS Feeds, eTOCs, free email alerts (when new

http://journals.asm.org/site/misc/reprints.xhtmlInformation about commercial reprint orders: http://journals.asm.org/site/subscriptions/To subscribe to to another ASM Journal go to:

on May 26, 2014 by U

niversity of Western O

ntariohttp://jvi.asm

.org/D

ownloaded from

on M

ay 26, 2014 by University of W

estern Ontario

http://jvi.asm.org/

Dow

nloaded from

Rotavirus Infection Increases Intestinal Motility but Not Permeabilityat the Onset of Diarrhea

Claudia Istrate,a,b Marie Hagbom,b Elena Vikström,c Karl-Eric Magnusson,c Lennart Svenssonb

Grupo de Virologia, Unidade de Microbiologia Médica, Centro de Malária e Outras Doenças Trópicais, Instituto de Higiene e Medicina Tropical, Universidade Nova deLisboa, Lisbon, Portugala; Division of Molecular Virology, Department of Clinical and Experimental Medicine, University of Linköping, Linköping, Swedenb; Division ofMedical Microbiology, Department of Clinical and Experimental Medicine, University of Linköping, Linköping, Swedenc

ABSTRACT

The disease mechanisms associated with onset and secondary effects of rotavirus (RV) diarrhea remain to be determined andmay not be identical. In this study, we investigated whether onset of RV diarrhea is associated with increased intestinal permea-bility and/or motility. To study the transit time, fluorescent fluorescein isothiocyanate (FITC)-dextran was given to RV-infectedadult and infant mice. Intestinal motility was also studied with an opioid receptor agonist (loperamide) and a muscarinic recep-tor antagonist (atropine). To investigate whether RV increases permeability at the onset of diarrhea, fluorescent 4- and 10-kDadextran doses were given to infected and noninfected mice, and fluorescence intensity was measured subsequently in serum. RVincreased transit time in infant mice. Increased motility was detected at 24 h postinfection (h p.i.) and persisted up to 72 h p.i inpups. Both loperamide and atropine decreased intestinal motility and attenuated diarrhea. Analysis of passage of fluorescentdextran from the intestine into serum indicated unaffected intestinal permeability at the onset of diarrhea (24 to 48 h p.i.). Weshow that RV-induced diarrhea is associated with increased intestinal motility via an activation of the myenteric nerve plexus,which in turn stimulates muscarinic receptors on intestinal smooth muscles.

IMPORTANCE

We show that RV-infected mice have increased intestinal motility at the onset of diarrhea, and that this is not associated withincreased intestinal permeability. These new observations will contribute to a better understanding of the mechanisms involvedin RV diarrhea.

While rotavirus (RV) is well established as a major cause ofsevere acute gastroenteritis in young children all over the

world, the mechanisms behind diarrhea and vomiting are stilllargely unknown. The fluid loss due to diarrhea may be caused byseveral mechanisms (1, 2). High concentrations of poorly absorb-able compounds may create an osmotic force causing a loss offluid across the intestinal epithelium (osmotic diarrhea). Secre-tory diarrhea is characterized by an overstimulation of the intes-tinal secretory capacity not coupled to an inhibition of fluid ab-sorptive mechanisms. Furthermore, if the barrier function of theepithelium is compromised by loss of epithelial cells or disruptionof the tight junction, hydrostatic pressure in blood and lymphaticvessels may cause water and electrolytes to accumulate in lumen(exudative diarrhea) (1). In most types of diarrhea, more than oneof these pathophysiological mechanisms are involved.

Several hypotheses have been forwarded to explain RV-in-duced diarrhea, including a functional role of the virus-encodedenterotoxin NSP4 (3–5), malabsorption secondary to failingtransport of electrolytes and/or glucose/amino acids (6), villusischemia (7–10), and stimulation of the enteric nervous system(ENS) (11) and of enterochromaffin (EC) cells (12). At the cellularand tissue levels, a new understanding of the disease mechanismsis beginning to emerge (2). Intestinal permeability and motilityare the two physiological mechanisms most rarely investigated,but currently they are being implicated in diarrheal syndromes (2,13). Intestinal permeability has only rarely been investigated inhumans infected with RV (14). In the early 1980s, Stintzing andcoworkers (15) assessed the intestinal permeability in young chil-dren with RV infection using polyethylene glycols (PEG) of dif-ferent molecular weights, and a significantly low urinary recovery

of PEG was noted. A similar observation was made in adults bySerrander and coworkers (16). Moreover, Johansen et al. (17)found that children with acute RV infection excreted less PEG intheir urine than healthy children and children with enteropatho-genic Escherichia coli (EPEC) infections. In vitro studies, on theother hand, have shown that RV can increase the permeability ofpolarized human epithelial Caco-2 cells (18), probably due to areorganization of the tight junction proteins claudin-1, occludin,and ZO-1 (19). Furthermore, the RV enterotoxin NSP4 inducesparacellular leakage in polarized epithelial cells and prevents lat-eral targeting of ZO-1 (20). In line with this, it has also been dem-onstrated in Ussing chamber experiments that the electrical tissueconductance is increased in RV-infected intestines (11, 21).

Intestinal motility has only occasionally been studied duringRV infection. It is well established that stimulating the vagal effer-ent nerves to the gut increases intestinal motility via a direct effecton the intestinal musculature and/or via an excitation of the in-testinal cells of Cayal (22). Atropine, the classical muscarinic re-ceptor antagonist, can block these effects. It is also well known that

Received 5 October 2013 Accepted 20 December 2013

Published ahead of print 26 December 2013

Editor: S. López

Address correspondence to Lennart Svensson, lennart.t.svensson@liu.se.

C. I. and M. H. contributed equally to this work.

Copyright © 2014, American Society for Microbiology. All Rights Reserved.

doi:10.1128/JVI.02927-13

March 2014 Volume 88 Number 6 Journal of Virology p. 3161–3169 jvi.asm.org 3161

on May 26, 2014 by U

niversity of Western O

ntariohttp://jvi.asm

.org/D

ownloaded from

naturally occurring opioids can be used as antidiarrheal agents,but their adverse central nervous effects make them undesirable asremedies. Loperamide, an opioid receptor agonist, differs fromother similar drugs by binding predominantly to intestinal tissue,and when given orally, it gains almost no access to the centralnervous system (CNS). Its antidiarrheal action is thought to resultfrom diminished intestinal motility by interfering with the myen-teric plexa (22, 23). Furthermore, loperamide can inhibit intesti-nal secretion caused by bacterial enterotoxins (24, 25). RegardingRV infections, Yamashiro and coworkers (26) found a significanteffect of the stool score on days 3 to 5 of treatment. Meta-analysesconfirmed that patients treated with loperamide are less likely tohave diarrhea 24 h after treatment and present with a shorterduration of diarrhea and fewer stools than patients in the placebogroup (27).

Previous human permeability and motility studies have beenperformed at a time when children already have documented RVdiarrhea. Thus, it has remained unresolved whether the effects onmotility and permeability are early or late host responses follow-ing diarrhea rather than the actual cause of diarrhea onset.

The objective of this study was to investigate whether and howintestinal permeability and motility are altered at the onset of RVdiarrhea. To address this issue, different-sized fluorescent dextrandoses were given to RV-infected and noninfected adult and infantmice to determine both permeability and transit kinetics. Intesti-nal motility was also studied after administration of the opioidreceptor agonist loperamide or the muscarinic receptor antago-nist atropine. Our study gives an important new understanding ofthe mechanisms triggering diarrhea.

MATERIALS AND METHODSAnimals. RV-naive BALB/c, 8-week-old adults and 5- to 7-day-old infantmice (B&K Laboratories, Sollentuna, Sweden) were used. They werehoused in standard cages with free access to food and water. Pregnantfemales were transferred to individual cages 1 week before the expectedday of birth, and offspring remained with their mother during the exper-imental period. The animal ethics committee in Stockholm approved theexperimental protocol (N291/010, N289/09).

Rotavirus infection and criteria for diarrhea. Mice (adults and in-fants) were orally infected with 10 �l/animal (100� the dose causingdiarrhea in 50% of animals [DD50]) of wild-type murine RV (EDIMstrain) as previously described (28). Only infant mice were examined fordiarrhea, since adult mice have previously been shown not to developdiarrhea (29). Adult mice were infected to investigate whether anychanges in motility and permeability occur in the absence of diarrhea.Diarrhea was defined as a liquid yellow stool revealed by gentle abdominalpalpation as described elsewhere (11, 30). The percentage of mice withdiarrhea was calculated as the number of pups with typical diarrheal stools(loose, yellow, liquid feces) divided by the total number of pups in thesame group. Mice were confirmed to have diarrhea before being includedin the treatment studies with loperamide and atropine.

Motility studies. In the motility experiments, adult (n � 6 to 13 ateach time point) and infant (n � 5 to 13 at each time point) BALB/c micereceived orally 10 �l of 4-kDa FITC-dextran probe (FD-4; Sigma) at 6, 12,24, 48, 72, and 96 h p.i. The probe was also given to noninfected mice. Theadult mouse dose was 0.5 mg/mouse, and the infant mouse dose was 0.25mg/mouse. FITC-dextran was dissolved in Milli-Q water, and the solu-tions were freshly prepared before each experiment. For time points of 24h p.i. and later, diarrhea was confirmed in all infected infant mice thatreceived the probe. After 30 min for adults and 15 min for infant mice, theanimals were sacrificed and the whole intestines, from stomach to rectum,were used for determination of transit time by UV light measurement bytaking photos using an FITC-specific filter (filter 520DF30; Bio-Rad) in a

ChemiDoc system (ChemiDoc XRS; Bio-Rad). The front part of the mainaccumulating FITC-dextran was defined from the photo, and the softwareprogram Illustrator CS6 was used to more exactly measure the intestinallength and migration of the FITC-dextran probe.

Intestinal transit was calculated at different time points on how far theFITC-dextran probe has passed as a percentage of the entire length of theintestine, from the pylorus to the rectum. Motility experiments were per-formed on two different occasions.

Antidiarrheal drug experiments. Loperamide hydrochloride and at-ropine sulfate salt monohydrate were purchased from Sigma (loperamide,no. 34014; atropine, no. A0257) and dissolved in Milli-Q water. Lopera-mide (10 mg/kg) or atropine (6 mg/kg) was administered orally to adultand infant mice at 40 and 44 h p.i. Four hours after the second dose, at 48h p.i., mice received 4-kDa FITC dextran and were then sacrificed after 15(infant) and 30 (adult) min for determination of transit time. In infantmice the antidiarrheal effect of loperamide and atropine was checked at 4h after the first dose and 4 h after the second dose (48 h p.i.) of treatment,a time point associated with severe diarrhea of almost all mice. In theloperamide experiment, mice also were checked for diarrhea on the fol-lowing day and received a final dose of loperamide at 64 h p.i., the effect ofwhich was investigated 4 h later.

Intestinal permeability studies. The transmural passage into theblood of fluorescent 10- and 4-kDa dextran was used to assess the barriercharacteristics of infant and adult mice infected with RV from 6 up to 96h p.i. compared to noninfected animals. The barrier properties were de-scribed in two ways: (i) by a permeability index equal to the ratio betweenthe relative uptake of fluorescent 10- and 4-kDa dextran and (ii) by thefold increase in either marker compared to the basal value for noninfectedanimals. The values are given as fluorescence intensity (FI). The relativepermeation of the 10- and 4-kDa probes was used to calculate a permea-bility index (R) through their relative fluorescence intensities, F10 and F4

in blood/serum, i.e., R � F10/F4.Briefly, RV-naive BALB/c mice, noninfected and RV infected, received

orally a mixture of 4-kDa FITC-dextran (adult mouse dose of 2.5 mg/mouse; infant mouse dose of 0.7 mg/mouse) and 10-kDa rhodamine-dextran (adult dose, 5.0 mg/mouse; infant dose, 1.4 mg/mouse). The fluo-rescently conjugated dextran (4-kDa FITC-dextran; no. FD-4; Sigma-Aldrich) and R881 (10-kDa rhodamine-dextran [4-kDa FITC-dextran;no. FD-4; and 10-kDa rhodamine-dextran; no R881; Sigma-Aldrich]) wasfreshly prepared in Milli-Q water before use. Blood was collected in serumtubes (BD Microtainer tubes; no. 365951) 3 h after tracer administrationand centrifuged for 5 min at 1,677 � g. The serum was diluted 1/50 inMilli-Q water, and the fluorescence intensities of FITC (494/518 nm) andrhodamine (565/580 nm) were measured with a fluorescence spectropho-tometer (LS-3B; Perkin-Elmer, Waltham, MA).

Histology. Specimens (0.5 cm) were taken from the duodenum, themiddle part of the jejunum, and the lowest part of ileum from adult andinfant mice, noninfected and infected, at 48 h p.i. The specimens werefixed in 4% phosphate-buffered formaldehyde (no. 02176; Histolab, Swe-den) and then embedded in paraffin. From each specimen, sections of5-�m thickness were cut and placed onto glass slides (SuperFrost PlusMenzel no. 06400; Histolab, Sweden). The paraffin was removed through2 steps of 5-min incubations in tissue clear solution (code 1466; Sakura,Tokyo, Japan), and the tissues were hydrated through a series of ethanol(EtOH) baths (99.5% EtOH twice for 5 min each, 95% EtOH for 5 min,70% EtOH for 5 min) followed by a final step in phosphate-buffered saline(PBS). Slides were then stained with hematoxylin and eosin for 5 min,rinsed in tap water, and mounted with Pertex mounting medium (no.00814; Histolab, Sweden) and a cover glass.

Statistical analysis. Data were analyzed with nonparametric two-tailed Mann-Whitney test and Kruskal-Wallis test using Prism Software5.0 (GraphPad, San Diego, CA) or Fisher’s exact test. Values were consid-ered significant at P � 0.05.

Istrate et al.

3162 jvi.asm.org Journal of Virology

on May 26, 2014 by U

niversity of Western O

ntariohttp://jvi.asm

.org/D

ownloaded from

RESULTSRotavirus increases intestinal motility in adult and infant mice.Alterations in intestinal transit/motility have been associated withseveral forms of gastrointestinal diseases, promoting the develop-ment of pharmaceutical drugs that attenuate intestinal motility.For RV infection it still remains unclear whether RV could in-crease intestinal transit and thereby contribute to diarrhea.

Therefore, to obtain a global view of the possible involvementof increased motility in RV diarrhea, we infected infant and adultmice with murine RV at 10 �l/animal (100� DD50 diarrhea doses)and characterized the kinetics of intestinal transit of a 4-kDaFITC-dextran probe. We observed that the probe passed along theentire intestine within 15 min in infected infant mice at 48 h p.i.,more rapidly than for noninfected mice (P � 0.001) (Fig. 1A andB). Interestingly, the peak of transit at 48 h p.i. coincided with thetime of the most severe diarrhea symptoms in pups, as 12 out of 13mice had diarrhea (Table 1). RV infection in mice was confirmedby eosin and hematoxylin staining of tissue sections of differentparts of the small intestine (Fig. 2). The characteristic vacuoliza-tion of the epithelial cells was seen on villi but not in the crypts.

We next investigated whether adult mice, despite not respond-ing with diarrhea following RV infection, have altered intestinalmotility. Theoretically, RV infection could stimulate the myen-teric plexus and motility in the absence of diarrhea (2). Indeed, wefound that the infection in adult mice did not resulted in increasedintestinal motility, as determined by the transit of FITC-dextran(Fig. 1C). However, a slight effect was found at 24 h p.i., a timepoint typically associated with onset of diarrhea in pups.

Atropine, a muscarinic receptor antagonist, attenuates rota-virus motility and diarrhea. The motility studies with atropineoral treatment, using a dose of 6 mg/kg, were performed as de-scribed in Materials and Methods. We observed a significant de-crease of intestinal motility in infected infant and adult micetreated with atropine compared to nontreated mice at 48 h p.i.(Fig. 3A and B). In noninfected infant mice, atropine had a modesteffect on motility (Fig. 3A). Since the effect of atropine was alsoobserved in noninfected mice (Fig. 3A and B), it did not seemlikely to be virus specific.

We also investigated whether atropine could attenuate RV-induced diarrhea in infant mice. Only mice with confirmed diar-rhea were included, and atropine was given at 40 and 44 p.i. The

FIG 1 Intestinal motility is increased at the time point of severe diarrhea, 48 h p.i. (n � 5 to 13). (A) Illustration of the transit of fluorescent FITC-dextran in theentire intestine of EDIM-infected and noninfected infant mice at 15 min after dextran treatment. FITC-dextran was visualized by a UV spectrophotometer(Chemi-Doc with an FITC-specific filter; Bio-Rad) and was identified as a fluorescence-dependent white color reaching different intestinal segments. (B and C)Transit of FITC-dextran in infant (B) and adult (C) mice 15 and 30 min after dextran treatment, respectively. Onset of diarrhea in infant mice occurred between24 and 48 h p.i. and was most severe at 48 h p.i., with a yellow, loose form. Values are based on the transit length of FITC-dextran as a percentage of the total lengthof the intestine. ***, P � 0.001 by Kruskal-Wallis test.

TABLE 1 Opioid receptor agonist loperamide attenuates rotavirusdiarrhea

Infant mouse type (n)

No. of mice suffering diarrhea at:

40 h p.i. 44 h p.i. 48 h p.i. 64 h p.i. 68 h p.i.

EDIM infected,untreated (13)

12/13 12/13 12/13 12/13 12/13

EDIM infected,loperamide treated (9)

8/9 0/8 0/8 6/8 2/8

Noninfected, loperamidetreated (6)

0/6 0/6 0/6 0/6 0/6

Noninfected, untreated(2)

0/2 0/2 0/2 0/2 0/2

Rotavirus Permeability and Motility

March 2014 Volume 88 Number 6 jvi.asm.org 3163

on May 26, 2014 by U

niversity of Western O

ntariohttp://jvi.asm

.org/D

ownloaded from

effect of the 2 doses of atropine was evaluated at 48 h p.i. In themock-treated group 6/7 still had diarrhea, in contrast to only 3/11in the atropine-treated group (P � 0.05 by Fisher’s exact test)(Table 2).

In summary, we conclude that the muscarinic receptor antag-onist can attenuate the increased intestinal motility and reducediarrhea.

Stimulation of �-opioid receptors in the myenteric plexa at-tenuates rotavirus diarrhea and motility. To further address howRV affects intestinal motility, we treated infected mice with lop-eramide, an opioid receptor agonist known for its effect of atten-uating intestinal motility by interfering with the myenteric plexaof the ENS. Infant and adult mice were treated with two doses ofloperamide (10 mg/kg, orally, 10 �l/mouse) at 40 and 44 h p.i. At48 h p.i., mice were given 4-kDa FITC-dextran, and after 15 and 30min in infant and adult mice, respectively, mice were sacrificed forintestinal transit analysis (n � 5 to 13 in each group) (Fig. 4A to

C). We observed a decrease in intestinal motility in RV-infectedinfant mice treated with loperamide compared to nontreated in-fected mice at 48 h p.i. (Fig. 4A and B). The FITC-dextran probewas flushed from the untreated infected mice (Fig. 4A, line 4) butwas delayed in loperamide-treated and infected mice (Fig. 4A, line5). Giving loperamide to noninfected infant mice had a modesteffect on motility (Fig. 4B). No significant effect on intestinal mo-tility was observed in infected adult mice treated with loperamidecompared to infected nontreated adults at 48 h p.i. (Fig. 4C). Thedrug again had only modest effect on noninfected adult mice (Fig.4C). We conclude that opioid receptor agonists can attenuate theincreased intestinal motility in RV-infected mice.

We next investigated whether loperamide could attenuate RV-induced diarrhea using the same experimental protocol as that inthe transit studies, i.e., infant mice were given doses of loperamideat 40 and 44 h p.i. In the mock-treated group, 12/13 had diarrheaat 40 h p.i., which was maintained during the course of the exper-iment until termination at 68 h p.i. (Table 1), and among infectedmice 8/9 had diarrhea at 40 h p.i., before administration of lop-eramide (Table 1). Only the 8 mice with confirmed diarrhea wereincluded in the loperamide-treated group. After the first dose (40h p.i.), none of them displayed diarrhea at 44 h p.i. (Table 1), andthey showed no sign of diarrhea following administration of asecond dose (48 h p.i.). The effect of loperamide was highly sig-nificant (P � 0.01 by Fisher’s exact test).

To study the pharmacological kinetics of loperamide, treatedmice were also inspected 20 h after administration of the seconddose (64 h p.i.), and surprisingly, 6/8 mice displayed diarrhea.When all eight mice once again were given a single dose of lopera-mide, only 2/8 mice exhibited diarrhea 4 h later (Table 1) (68 hp.i.), an effect that was statistically significant (P � 0.05 by Fisher’sexact test). In summary, we conclude that the opioid receptor

FIG 3 Atropine decreases the intestinal motility in EDIM-infected infant andadult mice at 48 h p.i. Transit of FITC-dextran in infant (A) and adult (B) mice15 and 30 min after dextran treatment, respectively. FITC-dextran was visual-ized by a UV spectrophotometer (Chemi-Doc with an FITC-specific filter;Bio-Rad) and was identified as a fluorescence-dependent white color reachingdifferent intestinal segments. Values are based on the transit length of FITC-dextran as a percentage of the total length of the intestine. **, P � 0.01 byMann-Whitney test.

TABLE 2 Muscarinic antagonist atropine attenuates rotavirus diarrhea

Infant mouse type (n)

No. of mice suffering diarrhea at:

40 h p.i. 44 h p.i. 48 h p.i.

EDIM infected, untreated (7) 7/7 7/7 6/7EDIM infected, atropine treated (11) 11/13 2/11 3/11

FIG 2 Histology of the small intestinal epithelium of EDIM-infected and noninfected infant mice. Tissue sections are stained with hematoxylin and eosin. (A andB) Jejunum at 48 h p.i. showing the characteristic vacuolization of cells in the middle and top of villi of a mouse with diarrhea and rotavirus in the stool. (C)Noninfected tissue from the duodenum.

Istrate et al.

3164 jvi.asm.org Journal of Virology

on May 26, 2014 by U

niversity of Western O

ntariohttp://jvi.asm

.org/D

ownloaded from

agonist loperamide can abolish diarrhea in a time-dependentmanner in this RV animal model.

The intestinal permeability to 4- and 10-kDa dextran probesis unaffected at the onset (24 to 48 h p.i.) of rotavirus-induceddiarrhea in mice. The fluid loss in diarrhea most likely is multi-factorial and may, in the case of RV infection, include distur-bances in intestinal permeability. Previous human studies inves-tigating intestinal barrier functions (15, 17) all were conducted ata time point at which subjects already had diarrhea. Thus, it re-mains unclear whether permeability changes are a late host re-sponse following diarrhea and/or the result of extensive lesions ofthe epithelia rather than the actual cause of diarrhea onset.

To address this issue, we orally administered 4-kDa FITC-dex-tran and 10-kDa rhodamine-dextran to infant and adult mice at 6,12, 24, 48, 72, and 96 h p.i. Adult and infant animals were bled at3 h after tracer administration to assess the transmural passage ofthe probes. The use of two different-sized probes allowed us toinvestigate in more detail whether the intestinal barrier was influ-enced by the RV infection. The plasma concentration of the twoprobes, as reflected in their optical densities, was estimated duringthe course of infection as a measure of their intestinal permeabil-ity. A ratio between the permeability of the two probes was alsodetermined (Table 3), as well as the change of permeability toeither compound (fold increase) compared to the correspondingnoninfected animal.

We found that infected infant, in contrast to adult, mice re-

sponded with a significant increase in permeability for both the 4-and 10-kDa probes at 6 and 12 h p.i. (P � 0.01; Mann-Whitneytest). The increase for the 4-kDa probe (Fig. 5A and B) was par-ticularly dramatic. In contrast, at 24 h p.i. and up to 96 h p.i., thepermeability to the 4-kDa probe was attenuated compared to thatof noninfected mice, and a transient peak for the 10-kDa probewas observed at 72 h p.i. The most important observation was thatno increased permeability was observed for the 4-kDa or the 10-kDa probes at 24 or 48 h p.i. (Fig. 5A and B and Table 3), timepoints when mice exhibited diarrhea, enterocyte vacuolization(28), and villus atrophy in the jejunum part of the small intestine(Fig. 2A and B). Intestinal lesions (vacuolization and atrophy)were also observed in the duodenum and ileum, as well as apicallyin the epithelium, but not in crypts. This suggests that these path-ological changes were not sufficient to induce increased permea-bility for 4- and 10-kDa dextran probes.

A reduced surface area should principally affect the permeabil-ity of the 10- and 4-kDa probes to the same extent and yield aparallel decline of either marker in the blood. However, if there isa perturbation of the junctional integrity, the paracellular uptakeof the larger 10-kDa probe might be enhanced relative to thesmaller 4-kDa probe (termed the 10-kDa/4-kDa ratio) (31). In theearly phase of infection, the 10-kDa/4-kDa ratio was balanced,based on a 20-fold increased for the 4-kDa probe and an 18-foldincreased for the 10-kDa probe. However, a pronounced 11-foldincrease in the 10-kDa/4-kDa ratio was observed at 24 h p.i and a17.9-fold increase at 72 h p.i., caused by the significant tighteningof the epithelium for the 4-kDa probe (Table 3). Interestingly, at48 h p.i., when the clinical signs of diarrhea are obvious, the 10-kDa/4-kDa ratio was 4.5 and was caused by the decreased perme-ability of the 10-kDa probe (Table 3).

To get further insight into the mechanisms associated with theonset of diarrhea, we investigated whether adult infected mice notresponding with diarrhea from the small intestine (28) had a per-turbed intestinal barrier. Accordingly, a set of experiments iden-tical to the infant mouse permeability studies was performed inadult mice. The permeability of the 4-kDa probe was decreased attime points 6, 12, and 24 h p.i. (Fig. 6A). A limited permeabilityincrease for the 4-kDa probe was observed at 48 and 72 h p.i. butdid not reach statistical significance. Rather, there was a signifi-cantly reduced permeability at 24 and 96 h p.i. (P � 0.05; Mann-Whitney test) (Fig. 6A). There were no clear changes observed inthe intestinal permeability of the 10-kDa probe (Fig. 6B) except at12 h p.i. (P � 0.05; Mann-Whitney test). The 10-kDa/4-kDa ratiowas highest from 6 to 24 h p.i. and then at 96 h p.i. (Table 4). It is

FIG 4 Loperamide decreases intestinal motility in EDIM-infected infant andadult mice at 48 h p.i. (A) Transit of fluorescent FITC-dextran in the entireintestine of EDIM-infected and noninfected and treated and nontreated infantmice at 15 min after dextran treatment. FITC-dextran was visualized by a UVspectrophotometer (Chemi-Doc with an FITC-specific filter; Bio-Rad) andwas identified as a fluorescence-dependent white color reaching different in-testinal segments. (B and C) Transit of FITC-dextran in infant (B) and adult(C) mice at 15 and 30 min after dextran treatment, respectively. Values arebased on the transit length of FITC-dextran as a percentage of the total lengthof the intestine. P � 0.05 (*) and P � 0.01 (**) by Mann-Whitney test.

TABLE 3 Permeability of dextran probes in rotavirus-infected infantmice

Time point(h p.i.)

10-kDa probe 4-kDa probe

10-kDa/4-kDaratio n

Foldincrease FIa

Foldincrease FI

6 1.8 18.1 (0.9) 20.1 28.2 (8.6) 0.6 612 1.4 14.5 (1.3) 8.6 12.0 (2.6) 1.2 724 1.0 9.9 (2.1) 0.6 0.9 (0.2) 11.0 648 0.6 5.9 (0.4) 0.9 1.3 (0.3) 4.5 972 1.4 14.3 (1.2) 0.6 0.8 (0.0) 17.9 696 0.5 4.7 (1.0) 1.9 2.6 (0.9) 1.8 6Noninfected 1.0 10.1 (0.3) 1.0 1.4 (0.4) 7.2 6a FI, mean fluorescence intensity. Data in parentheses are standard errors of the means.

Rotavirus Permeability and Motility

March 2014 Volume 88 Number 6 jvi.asm.org 3165

on May 26, 2014 by U

niversity of Western O

ntariohttp://jvi.asm

.org/D

ownloaded from

interesting that the increased ratio before 48 h p.i. seemed to bedue to reduction of permeability for the 4-kDa probe, while theuptake of the 10-kDa probe remained virtually unaffected.

DISCUSSION

Several disease mechanisms, such as malabsorption, altered per-meability, motility disturbances, stimulation of enterochromaffincells, activation of the ENS, and the RV NSP4 toxin, have beenassociated with fluid loss in RV diarrhea (2, 3, 5–8, 12, 13). Someof these mechanisms may indeed occur after the onset of diarrhearather than being the actual cause. In this study, we have investi-gated how intestinal transit (motility) and epithelial permeabilityare related to the onset of RV diarrhea. These two important dis-ease-associated mechanisms have not been studied together in theearly phase of RV diarrhea.

Motility of the small intestine, as in all parts of the digestive

tract, is controlled predominantly by excitatory and inhibitorysignals from the ENS; however, they are modulated by inputs fromthe CNS, whereas intestinal secretion is controlled mainly by in-trinsic nerves and in particular by the submucosal plexus (22). Theautonomic nervous system control of motility is exerted mainlyvia the release of acetylcholine and the effect on muscarinic recep-tors. Hence, it is possible to block this motor response by atropine,as also demonstrated in this study. Considering in vivo studies inseveral experimental models of acute diarrhea, atropine has failedto attenuate the rate of fluid secretion in the small intestine.

In the present study, gut motility was determined by studyingthe intestinal transit of the 4-kDa FITC-dextran probe in infantand adult mice. Interestingly, in infant mice, the onset of diarrheaoccurred concomitantly with a high intestinal transit of FITC-dextran at 48 h p.i. (Fig. 1A and B). To get further insight into thepathophysiological mechanisms responsible for the increased mo-

FIG 5 Intestinal permeability is not increased at the time for onset of diarrhea or at the time of the most severe diarrhea in infant mice. Permeability of 4-kDaFITC-dextran (A) and 10-kDa rhodamine-dextran (B) through the intestinal epithelium into serum at different time points p.i. was measured by spectropho-tometry at different fluorophore wavelengths (494/518 and 565/580 nm). Values are given as fluorescence intensity (FI). P � 0.001 (***), P � 0.01 (**), and P �0.05 (*) by Mann-Whitney test.

FIG 6 Intestinal permeability is not increased at the most critical time point of EDIM infection in adult mice. Permeability of 4-kDa FITC-dextran (A) and10-kDa rhodamine-dextran (B) through the intestinal epithelium into serum at different times p.i. was measured by fluorescence at wavelengths specific for eachfluorophore (494/518 and 565/580 nm). Values are given as fluorescence intensity (FI). *, P � 0.05 by Mann-Whitney test.

Istrate et al.

3166 jvi.asm.org Journal of Virology

on May 26, 2014 by U

niversity of Western O

ntariohttp://jvi.asm

.org/D

ownloaded from

tility, we treated noninfected and RV-infected mice with the�-opioid receptor agonist loperamide, attenuating the activity ofthe myenteric plexus (23), and with atropine, blocking muscarinicreceptors on smooth muscles. Loperamide differs from other opi-oid analogues, as it binds predominantly to intestinal tissue. Thus,when given orally it has virtually no access to the CNS. Lopera-mide is a widely used antidiarrheal agent, its effects being attrib-uted to an inhibitory action on smooth muscle tone and peristal-sis, attenuating both cholinergic and noncholinergic influences onintestinal smooth muscles (32, 33). The effect of loperamide wasnot restricted to infant mice, as adult noninfected and infectedmice responded with attenuated intestinal motility, strongly sug-gesting that the effect is not virus specific. The efficacy was verypronounced, since the diarrhea in infant mice was completelyabolished in 8/8 mice 4 h after a single dose. When giving a seconddose, it appeared that the effect was time dependent, since 6/8mice exhibited diarrhea 20 h after the second dose. In humans,maximal plasma loperamide concentration is achieved within 3 to5 h in humans (34), which corresponds to the significant effectobserved within 4 h in the mouse model. It has also been proposedthat loperamide, apart from its inhibitory action on smooth mus-cle tone, also can stimulate absorption of fluid, electrolytes, andglucose; moreover, it can reverse the prostaglandin E2 (PGE2) andcholera toxin-induced secretion to absorption (24). In previousstudies, loperamide has been administered to individuals with on-going diarrhea; therefore, the question of whether increased gutmotility is associated with the onset of diarrhea or a systemic hostresponse following diarrhea is still unresolved.

The effect of atropine on infected infant mice after 2 doses (6mg/kg) at 40 and 44 h p.i. determined a clear reduction in motilityas well as reduced diarrhea, where only 3/11 mice had diarrheaafter atropine treatment compared to 6/7 in the mock-treatedgroup.

In conclusion, our results obtained with atropine, the classicalmuscarinic receptor antagonist, and loperamide, a �-opioid re-ceptor agonist, clearly suggest that ENS is involved in the motilityresponse to rotavirus infection. We found earlier that the fluid lossin RV diarrhea is at least partly induced by activation of the ENS(11).

Adult mice were explored to determine whether motilitychanges were associated with water and electrolyte loss (i.e., diar-rhea) from the small intestine or whether motility changes oc-curred independently of diarrhea. Previous studies have shownthat the small intestine of infant mice (11, 35) but not adult mice(28) respond with electrolyte and water secretion (diarrhea) fol-

lowing infection with murine RV. In diarrhea-resistant adult micethere was a most modest increase of intestinal motility at 24 h p.i.(Fig. 1C) but no overall increased motility during the course ofinfection.

The participation of prostaglandins (PGs) in RV diarrhea maybe another factor explaining the different time courses for secre-tion and motility. It has been shown that PGs are produced underthe influence of microorganisms and have immune-modulatory,anti-inflammatory, as well as proinflammatory actions (36).Moreover, they can stimulate water secretion (37–40), an effectthat can be blocked by drugs attenuating nerve activity, such ashexamethonium, a nicotinic receptor blocker, or lidocaine, a localanesthetic (41), indicating that nerves are involved. Loperamide isalso a potent inhibitor of intestinal fluid secretion induced byPGE2 and cholera toxin. In children with RV gastroenteritis, ele-vated levels of PGE2 and PGF2 in both plasma and stools have beenfound, and treatment with the COX inhibitor aspirin reduced theduration of diarrhea (26). Moreover, intravenous infusion of PGcauses abdominal cramps (42), inhibits intestinal motility (43,44), and stimulates water secretion (24, 38, 45). In addition, RVinduces COX2 mRNA and secretion of PGE2 already within 8 hp.i. (46). These observations make it tempting to speculate that thedecreased motility with remaining diarrhea 72 to 96 h p.i. is in-duced by increasing concentrations of PGs. The action of PGs onsmooth muscles, whether they act directly on the muscle cells ormodulate of the release of neurotransmitters (47), is not yet fullyunderstood.

In several in vitro animal and human studies, both increasedand decreased permeability have been reported. Regarding RV,only a few human studies have been done (15–17); all of thesestudies were conducted on children already infected with RV, andthe time course for conceivable permeability changes was notstudied.

The use of a combination of two markers of different sizes tomeasure changes in the small intestinal permeability has proved tobe a sensitive and specific approach (16, 48). Interestingly, in-fected infant mice responded with a significant increase in perme-ability for both probes at 6 and 12 h p.i. (Fig. 5A and B), whereasadults showed an increase only for the 10-kDa probe at 12 h p.i.(Fig. 6A and B). The increased permeability at 6 and 12 h p.i. wasdue to the administration of the fecal debris but was not associatedwith diarrhea or increased motility. Furthermore, oral adminis-tration of a 10% fecal suspension from uninfected animals did notresult in diarrhea. An obvious conclusion from this unexpectedobservation is that increased permeability per se is not sufficient tocause diarrhea, since at these early time points mice do not havediarrhea. The onset of diarrhea (24 to 48 h p.i.) was not associatedwith increased permeability to the two dextran probes despitesignificant lytic lesions in the small intestine at 48 h p.i. (Fig. 2Aand B). This conclusion is further supported by our observationthat even adult mice have similar pathological lesions at the sametime points but no diarrhea (28). Apparently, RV-induced path-ological lesions per se in the small intestine (Fig. 2) are not suffi-cient to account for permeability changes or electrolyte changesduring RV infections. A similar observation was reported by Hey-man and coworkers (49), who studied transepithelial fluxes ofhorseradish peroxidase (HRP; molecular mass of about 44 kDa)from mucosa to serosa in EDIM-infected mice in an Ussing cham-ber. During RV diarrhea (days 2 to 8), no change in mucosal-to-serosal permeability to HRP was observed. On the other hand,

TABLE 4 Permeability of dextran probes in rotavirus-infected adultmice

Time point(h p.i.)

10-kDa probe 4-kDa probe

10-kDa/4-kDa ratio n

Foldincrease FIa

Foldincrease FI

6 1.1 8.7 (0.3) 0.4 1.1 (0.2) 7.9 512 1.3 9.9 (0.4) 0.4 1.2 (0.2) 8.3 524 0.9 6.7 (0.6) 0.3 1.0 (0.1) 6.7 1048 1.0 7.7 (0.2) 1.2 3.7 (2.5) 2.1 572 1.1 8.6 (0.2) 2.6 7.9 (3.6) 1.1 596 1.1 8.2 (0.4) 0.3 0.8 (0.0) 10.3 5Noninfected 1.0 7.7 (0.6) 1.0 3.0 (1.0) 2.6 7a FI, mean fluorescence intensity. Data in parentheses are standard errors of the means.

Rotavirus Permeability and Motility

March 2014 Volume 88 Number 6 jvi.asm.org 3167

on May 26, 2014 by U

niversity of Western O

ntariohttp://jvi.asm

.org/D

ownloaded from

during convalescence (days 9 to 14), when virus was no longerpresent in the mucosa, there was a 10-fold increase in HRP trans-port. They concluded that RV causes a transient rise in the gutpermeability during convalescence (49). In agreement with thisobservation, we found a significant transient rise in permeability72 h p.i. in both infant (P � 0.01; Mann-Whitney test) and adultmice.

In infant mice, the 10-kDa/4-kDa ratio was decreased early ininfection, at 6 and 12 h p.i., and at 48 h p.i. This reduced ratio at thetime of diarrhea was apparently due to a reduction of the penetra-tion of the 10-kDa probe. A most interesting observation was thatthe 4-kDa probe had reduced uptake from 24 to 72 h p.i in infantmice compared to uninfected infant mice. A similar observationwas obtained for the 10-kDa probe at 24 to 48 h p.i in infant mice.These characteristics likely reflect both the surface area absorptionand the barrier properties of the different probes sizes as describedby the 10-kDa/4-kDa ratio.

In conclusion, we show that RV-infected mice have increasedintestinal motility at the onset of diarrhea, and this is not associ-ated with any increased intestinal permeability. These new obser-vations may contribute to a better understanding of the mecha-nisms involved in RV diarrhea.

ACKNOWLEDGMENT

This work was supported by Swedish Research Council (LS) 320301.

REFERENCES1. Field M. 2003. Intestinal ion transport and the pathophysiology of diar-

rhea. J. Clin. Investig. 111:931–943. http://dx.doi.org/10.1172/JCI18326.2. Hagbom M, Sharma S, Lundgren O, Svensson L. 2012. Towards a

human rotavirus disease model. Curr. Opin. Virol. 2:408 – 418. http://dx.doi.org/10.1016/j.coviro.2012.05.006.

3. Estes MK 2003. The rotavirus NSP4 enterotoxin: current status and chal-lenges, p 207–224. In Desselberger U, Gray J (ed), Viral gastroenteritis.Elsevier, Amsterdam, the Netherlands.

4. Dong Y, Zeng CQ, Ball JM, Estes MK, Morris AP. 1997. The rotavirusenterotoxin NSP4 mobilizes intracellular calcium in human intestinalcells by stimulating phospholipase C-mediated inositol 1,4,5-trisphosphate production. Proc. Natl. Acad. Sci. U. S. A. 94:3960 –3965.http://dx.doi.org/10.1073/pnas.94.8.3960.

5. Ball JM, Tian P, Zeng CQ, Morris AP, Estes MK. 1996. Age-dependentdiarrhea induced by a rotaviral nonstructural glycoprotein. Science 272:101–104. http://dx.doi.org/10.1126/science.272.5258.101.

6. Lundgren O, Svensson L. 2001. Pathogenesis of rotavirus diarrhea. MicrobesInfect. 3:1145–1156. http://dx.doi.org/10.1016/S1286-4579(01)01475-7.

7. Lundgren O, Svensson L. 2003. The enteric nervous system and infec-tious diarrhea, p 51– 67. In Desselberger U, Gray J (ed), Viral gastroenteri-tis. Elsevier, Amsterdam, the Netherlands.

8. Ramig RF. 2004. Pathogenesis of intestinal and systemic rotavirus infec-tion. J. Virol. 78:10213–10220. http://dx.doi.org/10.1128/JVI.78.19.10213-10220.2004.

9. Osborne MP, Haddon SJ, Spencer AJ, Collins J, Starkey WG, Wallis TS,Clarke GJ, Worton KJ, Candy DC, Stephen J. 1988. An electron micro-scopic investigation of time-related changes in the intestine of neonatalmice infected with murine rotavirus. J. Pediatr. Gastroenterol. Nutr.7:236 –248. http://dx.doi.org/10.1097/00005176-198803000-00014.

10. Osborne MP, Haddon SJ, Worton KJ, Spencer AJ, Starkey WG, Thorn-ber D, Stephen J. 1991. Rotavirus-induced changes in the microcircula-tion of intestinal villi of neonatal mice in relation to the induction andpersistence of diarrhea. J. Pediatr. Gastroenterol. Nutr. 12:111–120. http://dx.doi.org/10.1097/00005176-199101000-00021.

11. Lundgren O, Peregrin AT, Persson K, Kordasti S, Uhnoo I, Svensson L.2000. Role of the enteric nervous system in the fluid and electrolyte secre-tion of rotavirus diarrhea. Science 287:491– 495. http://dx.doi.org/10.1126/science.287.5452.491.

12. Hagbom M, Istrate C, Engblom D, Karlsson T, Rodriguez-Diaz J, BuesaJ, Taylor JA, Loitto VM, Magnusson KE, Ahlman H, Lundgren O,

Svensson L. 2011. Rotavirus stimulates release of serotonin (5-HT) fromhuman enterochromaffin cells and activates brain structures involved innausea and vomiting. PLoS Pathog. 7:e1002115. http://dx.doi.org/10.1371/journal.ppat.1002115.

13. Michelangeli F, Ruiz MC 2003. Physiology and pathophysiology of thegut in relation to viral diarrhea, p 23–50. In Desselberger U, Gray J (ed),Viral gastroenteritis. Elsevier, Amsterdam, the Netherlands.

14. Kukuruzovic R, Robins-Browne RM, Anstey NM, Brewster DR. 2002.Enteric pathogens, intestinal permeability and nitric oxide production inacute gastroenteritis. Pediatr. Infect. Dis. J. 21:730 –739. http://dx.doi.org/10.1097/00006454-200208000-00007.

15. Stintzing G, Johansen K, Magnusson KE, Svensson L, Sundqvist T.1986. Intestinal permeability in small children during and after rotavirusdiarrhoea assessed with different-size polyethylene glycols (PEG 400 andPEG 1000). Acta Paediatr. Scand. 75:1005–1009. http://dx.doi.org/10.1111/j.1651-2227.1986.tb10331.x.

16. Serrander R, Magnusson KE, Sundqvist T. 1984. Acute infections withGiardia lamblia and rotavirus decrease intestinal permeability to low-molecular weight polyethylene glycols (PEG 400). Scand. J. Infect. Dis.16:339 –344. http://dx.doi.org/10.3109/00365548409073958.

17. Johansen K, Stintzing G, Magnusson KE, Sundqvist T, Jalil F, MurtazaA, Khan SR, Lindblad BS, Mollby R, Orusild E, Svensson L. 1989.Intestinal permeability assessed with polyethylene glycols in children withdiarrhea due to rotavirus and common bacterial pathogens in a develop-ing community. J. Pediatr. Gastroenterol. Nutr. 9:307–313. http://dx.doi.org/10.1097/00005176-198910000-00008.

18. Svensson L, Finlay BB, Bass D, von Bonsdorff CH, Greenberg HB. 1991.Symmetric infection of rotavirus on polarized human intestinal epithelial(Caco-2) cells. J. Virol. 65:4190 – 4197.

19. Dickman KG, Hempson SJ, Anderson J, Lippe S, Zhao L, Burakoff R,Shaw RD. 2000. Rotavirus alters paracellular permeability and energymetabolism in Caco-2 cells. Am. J. Physiol. Gastrointest. Liver Physiol.279:G757–G766.

20. Tafazoli F, Zeng CQ, Estes MK, Magnusson KE, Svensson L. 2001.NSP4 enterotoxin of rotavirus induces paracellular leakage in polarizedepithelial cells. J. Virol. 75:1540 –1546. http://dx.doi.org/10.1128/JVI.75.3.1540-1546.2001.

21. Isolauri E, Kaila M, Arvola T, Majamaa H, Rantala I, Virtanen E,Arvilommi H. 1993. Diet during rotavirus enteritis affects jejunal perme-ability to macromolecules in suckling rats. Pediatr. Res. 33:548 –553. http://dx.doi.org/10.1203/00006450-199306000-00002.

22. Furness J (ed). 2006. The enteric nervous system. Blackwell Publishing,Inc., Oxford, United Kingdom.

23. Ooms LA, Degryse AD, Janssen PA. 1984. Mechanisms of action ofloperamide. Scand. J. Gastroenterol. Suppl. 96:145–155.

24. Sandhu BK, Tripp JH, Candy DC, Harries JT. 1981. Loperamide: studieson its mechanism of action. Gut 22:658 – 662. http://dx.doi.org/10.1136/gut.22.8.658.

25. Watt J, Candy DC, Gregory B, Tripp JH, Harries JT. 1982. Loperamidemodifies Escherichia coli, heat-stable enterotoxin-induced intestinal se-cretion. J. Pediatr. Gastroenterol. Nutr. 1:583–586. http://dx.doi.org/10.1097/00005176-198212000-00023.

26. Yamashiro Y, Shimizu T, Oguchi S, Sato M. 1989. Prostaglandins in theplasma and stool of children with rotavirus gastroenteritis. J. Pediatr. Gastro-enterol. Nutr. 9:322–327. http://dx.doi.org/10.1097/00005176-198910000-00010.

27. Li ST, Grossman DC, Cummings P. 2007. Loperamide therapy for acutediarrhea in children: systematic review and meta-analysis. PLoS Med.4:e98. http://dx.doi.org/10.1371/journal.pmed.0040098.

28. Kordasti S, Istrate C, Banasaz M, Rottenberg M, Sjovall H, Lundgren O,Svensson L. 2006. Rotavirus infection is not associated with small intes-tinal fluid secretion in the adult mouse. J. Virol. 80:11355–11361. http://dx.doi.org/10.1128/JVI.00152-06.

29. Ward RL, McNeal MM, Sheridan JF. 1990. Development of an adultmouse model for studies on protection against rotavirus. J. Virol. 64:5070 –5075.

30. Kordasti S, Sjovall H, Lundgren O, Svensson L. 2004. Serotonin andvasoactive intestinal peptide antagonists attenuate rotavirus diarrhoea.Gut 53:952–957. http://dx.doi.org/10.1136/gut.2003.033563.

31. Menzies I. 1984. Transmucosal passage of inert molecules in health anddisease, p 527–543. In Skadhauge E, Heintze K (ed), Intestinal absorptionand secretion. MTP Press, Boston, MA.

32. Van Nueten JM, Janssen PA, Fontaine J. 1974. Loperamide (R 18 553),

Istrate et al.

3168 jvi.asm.org Journal of Virology

on May 26, 2014 by U

niversity of Western O

ntariohttp://jvi.asm

.org/D

ownloaded from

a novel type of antidiarrheal agent. Part 3. In vitro studies on the peristalticreflex and other experiments on isolated tissues. Arzneimittelforschung24:1641–1645.

33. Nakayama S, Yamasato T, Mizutani M. 1977. Effects of loperamide onthe gastric and intestinal motilities in the guinea pig and dog. NihonHeikatsukin Gakkai Zasshi 13:45–53. http://dx.doi.org/10.1540/jsmr1965.13.45.

34. Killinger JM, Weintraub HS, Fuller BL. 1979. Human pharmacokineticsand comparative bioavailability of loperamide hydrochloride. J. Clin.Pharmacol. 19:211–218. http://dx.doi.org/10.1002/j.1552-4604.1979.tb01654.x.

35. Wolf JL, Cukor G, Blacklow NR, Dambrauskas R, Trier JS. 1981.Susceptibility of mice to rotavirus infection: effects of age and administra-tion of corticosteroids. Infect. Immun. 33:565–574.

36. Scher JU, Pillinger MH. 2009. The anti-inflammatory effects of prosta-glandins. J. Investig. Med. 57:703–708. http://dx.doi.org/10.231/JIM.0b013e31819aaa76.

37. Sandhu JS, Fraser DR. 1983. Effect of dietary cereals on intestinal per-meability in experimental enteropathy in rats. Gut 24:825– 830. http://dx.doi.org/10.1136/gut.24.9.825.

38. Finck AD, Katz RL. 1972. Prevention of cholera-induced intestinal secre-tion in the cat by aspirin. Nature 238:273–274. http://dx.doi.org/10.1038/238273a0.

39. Jacoby HI, Marshall CH. 1972. Antagonism of cholera enterotoxin byanti-inflammatory agents in the rat. Nature 235:163–165. http://dx.doi.org/10.1038/235163a0.

40. Vane JR. 1971. Inhibition of prostaglandin synthesis as a mechanism ofaction for aspirin-like drugs. Nat. New Biol. 231:232–235. http://dx.doi.org/10.1038/newbio231232a0.

41. Brunsson I, Sjoqvist A, Jodal M, Lundgren O. 1987. Mechanisms underly-

ing the small intestinal fluid secretion caused by arachidonic acid, prostaglan-din E1 and prostaglandin E2 in the rat in vivo. Acta Physiol. Scand. 130:633–642. http://dx.doi.org/10.1111/j.1748-1716.1987.tb08186.x.

42. Bergstrom S, Duner H, von Pernow EUB, Sjovall J. 1959. Observationson the effects of infusion of prostaglandin E in man. Acta Physiol. Scand.45:145–151. http://dx.doi.org/10.1111/j.1748-1716.1959.tb01686.x.

43. Thor P, Konturek JW, Konturek SJ, Anderson JH. 1985. Role of pros-taglandins in control of intestinal motility. Am. J. Physiol. 248(Part 1):G353–G359.

44. Bennett A, Eley KG, Scholes GB. 1968. Effect of prostaglandins E1 and E2on intestinal motility in the guinea-pig and rat. Br. J. Pharmacol. 34:639 –647. http://dx.doi.org/10.1111/j.1476-5381.1968.tb08493.x.

45. Karim SM, Adaikan PG. 1977. The effect of loperamide on prostaglandininduced diarrhoea in rat and man. Prostaglandins 13:321–331. http://dx.doi.org/10.1016/0090-6980(77)90011-9.

46. Rossen JW, Bouma J, Raatgeep RH, Buller HA, Einerhand AW. 2004.Inhibition of cyclooxygenase activity reduces rotavirus infection at a post-binding step. J. Virol. 78:9721–9730. http://dx.doi.org/10.1128/JVI.78.18.9721-9730.2004.

47. Wennmalm A. 1978. Prostaglandin-mediated inhibition of noradrena-line release. II. Dual mechanism behind its frequency-dependence. ActaPhysiol. Scand. 102:199 –204.

48. Weaver LT, Chapman PD, Madeley CR, Laker MF, Nelson R. 1985.Intestinal permeability changes and excretion of micro-organisms instools of infants with diarrhoea and vomiting. Arch. Dis. Child. 60:326 –332. http://dx.doi.org/10.1136/adc.60.4.326.

49. Heyman M, Corthier G, Petit A, Meslin JC, Moreau C, Desjeux JF. 1987.Intestinal absorption of macromolecules during viral enteritis: an experimen-tal study on rotavirus-infected conventional and germ-free mice. Pediatr. Res.22:72–78. http://dx.doi.org/10.1203/00006450-198707000-00017.

Rotavirus Permeability and Motility

March 2014 Volume 88 Number 6 jvi.asm.org 3169

on May 26, 2014 by U

niversity of Western O

ntariohttp://jvi.asm

.org/D

ownloaded from