HAL Id: hal-00901251https://hal.archives-ouvertes.fr/hal-00901251

Submitted on 1 Jan 1980

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

SWINE DYSENTERY COMPARISON OFEXPERIMENTAL DISEASES PRODUCED BY

INFECTION WITH COLONIC MUCOSA OR WITHTREPONEMA HYODYSENTERIAE, FRENCH

STRAINS, AND OF ” NATURAL” DISEASEJ.P. Raynaud, G. Brunault, J. Philippe, P. Brizard

To cite this version:J.P. Raynaud, G. Brunault, J. Philippe, P. Brizard. SWINE DYSENTERY COMPARISON OF EX-PERIMENTAL DISEASES PRODUCED BY INFECTION WITH COLONIC MUCOSA OR WITHTREPONEMA HYODYSENTERIAE, FRENCH STRAINS, AND OF ” NATURAL” DISEASE. An-nales de Recherches Vétérinaires, INRA Editions, 1980, 11 (1), pp.69-87. �hal-00901251�

SWINE DYSENTERYCOMPARISON OF EXPERIMENTAL DISEASES

PRODUCED BY INFECTION WITH COLONIC MUCOSAOR WITH TREPONEMA HYODYSENTERIAE, FRENCH STRAINS,

AND OF « NATURAL» DISEASE

J.P. RAYNAUD G. BRUNAULT J. PHILIPPE

P. BRIZARD

Agricultural Research and Deve%pment Station, Pfizer lnternational, 37400 Amboise, France

Résumé

DYSENTERIE PORCINE. COMPARAISON DES MALADIES OBSERVÉES APRÈS INOCULA-TION AVEC DES MUQUEUSES INFECTEES OU AVEC TREPONEMA HYODYSENTERlAE,SOUCFiES FRANÇAISES, ET LA MALADIE « NATURELLE » - L’inoculation orale de broyats demuqueuse et de matières d’animaux en dysenterie, ou de souches pathogènes de Tieponemahyodysenteriae aussi bien que la contamination spontanée en loges infectées, provoquent enmoyenne l’apparition d’une dysenterie chez 359 sur 406 porcelets SPF. Quel que soit le mode decontamination utilisé, la morbidité est élevée : après contamination en loges, 75 porcelets sur 83sont dysentériques ; après une seule ingestion de matériel contaminé, 263 sur 280 animaux sontatteints ; et après inoculation de T. hyodysenteriae, 25 sur 33. Le taux de mortalité observé chezles animaux dysentériques est toujours supérieur à 80% : respectivement, 48 sur 60, 132 sur 154et 13 sur 14. Peu d’animaux guérissent spontanément. La durée moyenne d’incubation varieselon le mode de contamination, entre 9 et 13 jours. Les animaux atteints de la forme aiguë meu-rent 16 à 23 jours après contamination, et ceux atteints de la forme chronique, 29 jours après.L’augmentation du nombre de Tiepomena, de Campylobacter et de Balantidium observée aprèsle début de la maladie, est approximativement la même pour les trois modes de contamination.Cette maladie est caractérisée par une alternance de « dysenterie au sens strict » et de diarrhéemucoïde, cette dernière étant plus importante et plus fréquente pour les cas de guérison sponta-née ou les formes chroniques.

Swine dysentery causes considerable lossesto swine production because of deaths, redu-ced feed efficiency, and expenses for chemo-therapy (Harris and Glock, 1975). This disease« costs » 30 to 50 millions of US dollars peryear to the US. It has been identified in all themain swine producing countries in recent

years (Roncalli and Leaning, 1976). In many ofthese countries the incidence of swine dysen-tery is growing with the increased size of unitsand with the system of intensive production.Swine dysentery appears first of all in its

acute form. Afterwards, this classical type is

replaced by a sub-acute or a chronic form.The etiology of the disease has not been

completey clarified. It was first described inthe US (Whiting et al., 1921) ; its infectiousand transmissible nature was demonstratedafter the ingestion by healthy pigs of the sto-mach and colon contents of sick animals

(Whiting, 1924). A gram negative germ thatwas curved and microaerophilic was isolatedbut the disease transmitted was a benign diar-rhea (Doyle, 1944). An acute dysentery was

transmitted to 50 out of 60 pigs by oral inocu-lation with Vibno coli (now CampylobacterJincorporated in gastric mucins (James andDoyle, 1947). But the reproduction of thedisease with isolated vibrios was unsuccessfulin Great Britain (Deas, 1960). Then, interestwas aroused in the spirochaetes found in thecolon of dysenteric animals (Terpstra et al.,1968) ; the ingestion of large anaerobic spiro-chaetes induced successfully the disease inswine (Taylor and Alexander, 19711. This largespirochaete, Treponema hyodysenteriae, wasisolated ; alone, its ingestion causes dysenteryin two out of four pigs, and in association withVibrio !CampylobacierJ coli, dysentery in 12

out of 15 pigs (Harris et al., 1972a). Virulentstrains of T. hyodysenteriae cause severe

dysentery in normal, specific pathogen free(SPF) or minimal disease (MD) pigs, but donot affect germ-free pigs. To be really patho-genic, T. hyodysentenae must be associatedwith other anaerobic germs, commensal in thecolon (Lysons et al., 1978). In addition, T.

hyodysenteriae is always present in diseasedanimals and it is also possible to observe at anabnormal level, regularly, a gram negativebacillus : Campylobacter ssp, irregularly, twoprotozoa : a large size Balantidium and a smallsize Trichomonas. « T, hyodysentenae has aprimary role as an etiological factor of swinedysentery, (Harris and Glock, 1975). Othercommensal germs, i.e., Campylobacter spp,strict anaerobic Bacteroides sp., Fusobacte-rium sp. and large ciliate Balantidium, whichare found in normal animals shoud be conside-

red as opportunists and they can increase theseverity of the disease ».To check the efficacy of drugs used for the

prevention of swine dysentery we have wor-ked since 1972 (Raynaud etal., 1972) using so-called « classical » protocols. Then we perfor-med a new model of experimental swine

dysentery with continuous pen contaminationof experimental animals by « seeder pigs »,i.e., a dysenteric animal is permanently main-tained in a « contaminated » pen. « Sentinelanimals » or « detectors » are normal, healthypiglets put in the contaminated pen to showthat the pen is really contaminated. In this arti-cle we report our observations concerning theswine dysentery obtained in SPF pigs afterpen contamination, after oral infection usinginfected material from acutely dysentericdonors, or after oral infection using strains ofT hyodysenteriae.

Material and Methods

1. Animals

In these experiments, 405 young weanedswine from farms stocked with specific pathogen free (SPF) strains without any specificallydefined disease antecedents were used.

Piglets having an average weight of 19 kg(range 13 to 28 kg) were used. Table 1 showsthe distribution of animals between the diffe-rent experiments.

All animals were fed ad libitum with a com-mercial mash. During the two weeks prece-ding the start of experimentation, an

« entrance in piggery » feed was given, andduring the experimental period animals recei-ved a « fattening » feed (table 2). No starvingwas followed.

2. HousingThe experiments with pen contamination or

oral infection were conducted in a 16-pen pig-gery having forced ventilation and air-

conditioning. Each pen (2 m x 2.25 m) has aslatted wooden floor (2 m x 1 m), a concretefloor of 1.25 m x 1 m, and an automatic fee-der and drinker in an area of 1 m x 0.5 m.

The experiment with T. hyodysenteriaeinfection was conducted with 3 to 5 pigletsput in wire pens of a flat deck in a buildingthermostatically controlled for temperature,and programmed automatic ventilation.

3. Clinical observations and samplingsIndividual weights and feed consumed are

noted each week and at onset of dysentery.Each day the condition of feces is examined

and noted : consistency (normal, semi-liquid,liquid), color, composition (normal or no ano-maly, presence of mucous or glairs, presenceof bloodl. In our experiments we used the fol-lowing definitions :- dysentery stricto sensu : the feces are abnor-mal in consistency (liquid or semi-liquid) withblood associated or not with mucous.- mucoid diarrhea : the feces are of abnormalconsistency with mucous but no blood.- diarrhea : the feces are semi-liquid or liquidbut no other anomaly is found.The clinical forms which are seen in experi-

mental swine dysentery follow those descri-bed by Harris and Glock (1975) :- Acute dysentery : with large amounts ofmucus and flecks of blood in the feces. As thediarrhea progresses, watery stools containingblood, mucus, and shreds of white muco-fibrinous exudate are seen with concurrent

staining of the rear quarters.- The feces in chronic forms often contain wellmixed dark blood leading to .so-called blackscours.

The ultimate cause of death is associatedwith dehydration, acidosis and hyperkalemia.On each animal fecal samples are taken

before experimental infection, at first signs ofdisease and at death. Samples taken on liveanimals are examined immediately : Balanti-dium motile trophozoites lose their motilityafter 2 h between 25 and 37° C (non motilecysts are difficult to diagnose).

4. Post-mortem examinations

In all cases of swine dysentery observedduring these experiments, we found the grosslesions which are typical of so-called acute orchronic forms, and always limited to thecaecum-colon-rectum areas (the small intes-tine is always normal in appearance). In theacute form, hyperemia, congestion andoedema of the mucosa could be found uni-formly on the entire tract or in localised areas.When localised, some necrotic areas surroun-ded with fibrin exudation are found (theFrench description is o typhlocolite n6cro-hemorragique (-Espinasse, 1971). It is inthese foci that clumps of Balantidium are

counted in very large numbers. The mucouscould be present as a uniform layer on the sur-face of the mucosa. In areas it is also someti-

mes found to be viscous and/or mucofibri-nous. In the chronic form, blood or hemorrha-ges are not found. The mucosa is thickenedand the lesions are exclusively of the mucous,necrotic and mucofibrinous forms.

5. Laboratory examination of the feces

Routinely the following examinations are

completed in the laboratory :- counts of Balantidium trophozoites, Tricho-monas and Treponema- semi-quantitative appreciation (using a scalefrom 0 to + + + +) of Treponema and

Campylobacter.To be noticed is the fact that for the exami-

nation and counting of Balantidium motile tro-phozoites, it is essential to take a sample on

live animals and to examine it immediately(maximum delay : 2 h at a temperature of bet-ween 25 and 37° C), otherwise the trophozoi-tes become non motile cysts which are diffi-cult to diagnose.

For the detection of microorganisms, theseroutine technics are not well adapted, not

being sensitive enough.

Description of the routine technic for labora-tory examination of the feces

A sample of 1 g of faecal material is suspen-sed in 9 ml of saline.- Of the homogeneized suspension

0.1 ml is used to count Balantidium motile

trophozoites (and Trichomonas)

0.05 ml is spread on a slide, the smear isstained with Vago (after fixation with a 2%solution in water of Merbromin during 5 minand washing with water, the slides are stainedwith a 2% solution in water of methyl violetl.10 to 20 fields are examined.

Treponema hyodysenteriae (and large spiro-chaetes) and Campylobacter sp. are noted ona scale from 0 to + + + + .- After 10 minutes of sedimentation.

0.01 ml taken at 1 cm below the surface ofthe supernatant are put on a slide and coveredwith a coverslip, examined with phase micros-cope at x 400 ; 10 fields are examined.

Only large typical Treponema hyodysente-riae with typical motility are counted (an ave-rage of 1 organism/field : 4.5 x 106/g).A similar sample is used from the superna-

tant - 1 cml - is poured on Columbia agar with400 pg/ml of Spectinomycin and 7 % V/Vcitrated bovine blood. Petri dishes are kept inGas Pak jars (20% C02, 80% H2) for 3 to 4days at 42° C. This selective isolation technicas described by Kinyon et al. (1976) is verysimple and effective. The strain of Treponemahyodysenteriae give a typical completeIbetalhemolysis. With dilutions, the same cul-ture is used for numbering so-called « viableTreponema hyodysenteriae ».

Sensitivity levels and significance of theresults

a) Treponema counts per gram ; one Trepo-nema per field gives a count of 4.5 x 106/g.As 10 fields are counted, the sensitivity level ofour routine technic is about 4.5 x 105/g.The selective isolation technic described by

Kinyon et al. (1976) is able to detect manyorganisms as the sensitivity level is about 50

Treponema per gram. The Treponema hyody-sentenae type spirochaetes are referred to asType I by Taylor and Alexander 119711, and aso large type » by Harris and Glock (1975).They are 0.29 to 0.38,um in diameter and 7.1 Ito 7.3 iim long.

b) We also counted the Spirochaetes whichwere not this a large typq », but small spiro-chaetes which are the non pathogenic Trepo-nema. They are 0.20 to 0.22,um in diameterand 3.3 to 3.6 t!m long. The morphology of allthese spirochaetes has been extensively des-cribed by Elazhary et al., (1978). These« small » spirochaetes are not usually presen-ted on dysenteric animals where the Trepo-nema hyodysentenae type are recorded.

c) Balantidium motile trophozoites and Tri-

chomonas counts per gram ; the sensitivitylevel of our routine technic is about 10/g.

d) Stained smear (Vago) : We detect about103 organisms/g (rated 1:). The values arehalf for Treponema than for Campylobacterand counts are about 1 to 2 x 10!/g for + , 2to 4 x 107/g for + +, 4 to 8 x 107/g for+ + + , and 1 to 2 x 108/g for + + + + . Allratings given to animals are averaged and thisforms the Vago index average.

6. lsolation of Treponema hyodysenteriaestrains

After sedimentation of a one tenth suspen-sion of fecal material or colonic mucosal scra-

pings in saline, 1 ml of the supernatant is pou-red on two Petri dishes. Some drops are puton the first, and routine streaks for isolatedcolonies are done on the second. After 3 to 4

days incubation at 42° C in a Gas Pak jar, theagar surface on the edges of completeo beta ii hemolysis are examined by phasecontrast microscopy. Only large T. hyodysen-teriae with typical morphology and motility arecounted. Sub-cultures are completed at 4-dayintervals until a pure culture is obtained whichis then used for infection (no other sub-cultureis rlnne before infection).

7. Experimental infection or contamination

Disease transmitted in contaminated pens.

We used a new model with oral infection +

continuous pen contamination (Raynaud et

a/., 1980) which can be summarized with thefollowing procedure :Phase 1: .-4 weeks infection + contamination of all ani-mals and supplementation of some animals.

Day - 5 to - 3 : the « supplemented » ani-mals are fed with supplemented feed.

Day 0 : oral individual infection with hemorra-gic parts of colonic mucosa scrapings andintestinal contents of animals affected withacute swine dysentery.Day 0 to day 25 : each day the supplementedanimals are fed with the drugs being experi-mented, mixed in their feed. Each night theanimals of each supplemented pen are trans-ferred to a contaminated pen (always thesame) (fig. 1 l.Day 25 : the supplement is withdrawn fromthe feed and the pen contamination is stop-ped.Day 30 : the remaining live animals are submit-ted to a transportation stress (1 day’s travel ina truck).

Phase 2 :4 weeks of control for possible swine dysen-tery relapses.

During phase 1, one normal healthy animalwhich has not been orally infected is added toeach contaminated pen every night ; this ani-

mal is used as a sentinel to reveal the infection

existing in the pen.The characteristics of the disease shown by

the a sentinel » animals are presented here asa reference to « natural » disease obtained infarms in contaminated pens or contact withcontaminated animals.

Four series of experiments were carried outaccording to this type of contamination bet-ween May 1976 and February 1978.

Oral infection with colonic contents andmucosal scrapings

Each pig received orally by means of a

syringe, 100 ml of suspension with infectivematerial. The piglets were not fasted beforeinfection.

The suspension of infective material is usedeither immediately, or within a short time afterpreparation : « fresh material », or after deepfreezing and thawing : « deep frozen mate-rial ».

The infective material consists of mucosal

scrapings and intestinal contents of one ani-mal affected with acute swine dysentery. Theyare collected during slaughter or at death.

This material is mixed with two parts of asaline solution and put in a blender which min-ces and churns into a homogenous mixture inone minute. The suspension is sieved on a 800micron screen. The fresh suspension is putinto plastic sachets each containing 100 g, theaverage oral dose for one piglet. The sachetsare put in liquid nitrogen for instant deep free-zing, then kept at - 80° C. We have kept infec-tive material in these conditions without anyapparent loss of pathogenicity for about oneyear. When being used, the sachets are tha-wed in water at 50° C.

Four series of experiments using freshmaterial were carried out from May 1976 toDecember 1976 and three with deep frozenmaterial from July 1976 to August 1976.

lnoculation of T. hyodysenteriae strains

The strains, isolated according to the tech-nique described by Kinyon et al. (1976), camefrom field cases originating in the center ofFrance (Touraine) for strains 1 and 3, and thenorth of France for strains 2 and 2’. Strains 1and 3 came from different animals whereasstrains 2 and 2’ were isolated from the same

animal, strain 2 during acute symptoms andstrain 2’ after spontaneous cure (Windsor(1979), reported also field cases cured sponta-neously after arrival in the laboratory).

Each of the T. hyodysenteriae strains selec-ted was submitted to complete sub-culture at4-day intervals until a pure culture was obtai-ned. The contents of five to ten dishes of asame strain were administered orally bysyringe to each piglet. Large T. hyodysente-.riae were counted by phase contrast micros-copy. The dose administered to each pig con-tained 109 to 1010 germs (microscopy) or 108to 109 colony forming units (culture). Starva-tion of animals was only applied for the nightpreceding oral infection.A series of experiments with each strain

was carried out from November 1976 to March1977.

Results

7. Laboratory examination of the feces in nor-mal SPF swine.

Fecal material from 56 normal healthy SPFswine was sampled before use of the animalsin experiments. Laboratory examinationresults are presented in table 3.

For these animals we present results of

« large » Treponema and « small » spirochae-tes. They are very numerous and found in thesame range as those counted by Elazhary et al.(19781. Treponema hyodysenteriae were notdetected microscopically, but Campylobacter

were found in 73% of the animals and Balanti-dium coli in 27% with ranges of 0 to 7800/g.Trichomonas spp. are infrequent.The « normal » flora has micro-organisms

which are known to be opportunists, andwhich could participate in the dysentery, byincreasing the severity of the disease.

2. Swine dysentery obtained by pen contami-nation

Clinical results : .’

In the four series of pen contamination

experiments, the proportions of animals thatdeveloped swine dysentery were successively13/14 (93%), 6/6 (100%), 38/45 (84%) and18/18(100%). The average day of onset of thedisease also varied according to the series andwas respectively 13.5 + 1.6 days (average ±standard error), 19.0 + 2.7, 14.4 ± 1.2 and10.9 + 1.2. Detailed values for each series donot permit a correlation between the day ofonset of swine dysentery and the percentageof animals that die from it or are cured. Of the83 animals submitted to pen contamination,75 (90.4% developed acute dysentery. Duringthe first days of disease, all clinical cases werein the form of dysentery stricto sensu. No ani-mal developed a chronic form of swine dysen-tery at start. The average day of onset of thedisease was 13.6 + 0.7 days after start of con-tamination.

Complete evolution of the disease observedon 60 animals showed that in 47 animals thatdied from the disease, the dysentery strictosensu phase developed during 7.6 ± 1.2 days,

and in 12 animals that were spontaneouslycured, during 9.7 ± 1.1 days (table 4). Theaverage day of death or cure is thus directlyrelated to the average day of onset of thedisease, but the duration of the disease is

practically the same whatever the contamina-tion time. In animals that died with swinedysentery, the average time of death is 22.7 ±2.3 days after start of contamination, after 7.6± 1.2 days of dysentery stricto sensu and 2.5 5:t 1.1 days of mucoid diarrhea. In animals thatwere spontaneously cured, dysentery strictosensu lasted a little longer, 9.7 + 1.1 days, butbefore the cure, 38.8 + 2.9 days after conta-mination, we noted a high number of days ofmucoid diarrhea, 6.8 ± 1.8.

Parasitological and bacteriological resultsIn regard to results of the examination of

fecal samples (table 5) at the onset of thedisease, all the dysenteric animals showed alarge number of Treponema and a high level ofCampylobacter ; Balantidium protozoa wereonly found in one third of the animals, and Tri-chomonas were not found at all. Results atdeath are different : all animals excreted Tre-ponema and Campylobacter at a higher level ;Balantidium protozoa and even Trichomonaswere also more numerous and found in moreanimals.

3. Or2/ infection with colonic ’contents andmucosal scrapingsClinical results, « fresh )) material

In the four series of experiments using oralinfection with fresh materials, the proportionsof animals that developed swine dysenterywere successively 55 out of 60 infected ani-mals (91.7% 32/33 (97%), 33/34 (97.1%) 1and 33/37 (89.2% 1 among animals wheredisease was allowed to evolve completely, theproportions that died from swine dysenterywere successively 25/30 (83.3%), 12/14(85.7%), 22/24 (91.7%) and 19/21 (90.5%).As the morbidity and mortality rates did notdiffer significantly from one series to another,only the overall average results of the fourseries are given (table 61. However, it appea-red that the material became more and morevirulent during the series. The incubationperiod decreased successively from 11.3 ±0.8 to 10.3 + 1.1, 6.4 ± 0.5 and 5.6 + 0.5days ; the percentage of spontaneous curesalso decreased : 16.7 (5/30), 14.3 (2/ 141, 8.3(2/24) and 9.5 (2/21) ; death occurred sooner16.6 ± 1.8 ; 17.4 ± 2.6 ; 11.8 + 1.0 and9.5 + 0.9 days.

Clinical results, « deep frozen )) material .’

However, in the three series using deep fro-zen material, no difference was noticed in

regard to morbidity, mortality, incubationperiod, percentage of spontaneous cures ortime of death (average results, table 6).

Average clinical results on 280 animals : .’The average clinical results show that mor-

bidity and mortality observed after infectionby fresh or deep frozen material are compara-ble ; respectively 151 out of 164 animals

(92.1 % versus 109 out of 116 (94%) ; ; 78/79(87.6%) versus 56/67 (80.6%) (table 6). Theclinical results obtained after oral infection byfresh or deep frozen material can therefore bepooled. As was observed in the differentseries carried out with fresh material, a slightdifference in virulence seems to appear bet-ween the fresh material and the deep frozenmaterial in regard to the infection parameters,incubation period, day of spontaneous cure ordeath (table 7). Average values could outlinethe experimental disease obtained after oralinfection with colonic contents and mucosalscrapings. The infection was successful on263 animals out of 280 (94%1. The dysenterystricto sensu occurred 9.8 days after infectionand 260 animals (99%) presented an acuteform at onset. Of the 156 animals where thedisease was allowed to evolve completely, 125died in acute form. Only 3 animals presented achronic form at onset and 7 out of 156 died in

chronic form. The average rate of sponta-neous cure was 15%, i.e., 24 out of 156 ani-mals. Notwithstanding the issue or the form ofthe disease, the incubation period was in thesame range (9.0, 10.0 and 13.1 days), as wasthe duration of the dysentery stricto sensu

(5.5, 4.7 and 6.0 days) ; the only major diffe-rences were in the death time (16.0 days forthe acute form versus 28.6 days for chronicform) and the duration of mucoid diarrhea (2.5versus 13.9 days) (table 81.

Parasito%gical and bacteriological resultsWhen the disease started all the animals

excreted both T. hyodysenteriae and Campy-lobacter (table 9). In average, each animalshowed 95 x 106 live T. hyodysenteriae per

gram of fecal material (range : 9 to 225 x 106),a Vago index of 2.7 (range 1 to 4) for Trepo-nema sp. and 2.2 (range 1 to 4) for Campylo-bacter. The average number of Balantidiummotile trophozoites per animal was 4200 orga-nisms per gram of fecal material (up to

62,000). Trichomonas were only found in 27animals out of 44 with counts up to 900,000organisms per gram of fecal material. On theday of death, the values were higher for Tre-ponema and Balantidium and about the samelevel for Trichomonas and Campylobacter.

4. Oral infection with T. hyodysenteriaestrains

Clinical results :

Three of the four strains used, strains 1, 2and 3, induced an acute form of dysentery inrespectively 4 animals out of 6 inoculated, 13 3out of 18 and 7 out of 9. The day of apparitionof the disease was comparable for the threestrains, 10.5 + 1.3 days, 11.6 + 1.5 and14.0 + 1.7. The fourth strain, 2’, inoculatedin 9 animals, induced only one case of swinedysentery 27 days after inoculation, and thisanimal cured spontaneously 19 days later aftertwo days of dysentery stricto sensu andwithout mucoid diarrhea. This strain whichwas isolated from an animal having curedspontaneously had a much attenuated viru-lence. Results from this strain have not beenincluded in the overall average results. In fact,the inoculation of strains 1, 2 or 3 of T. hyody-senteriae induced an acute form of swine

dysentery in 24 out of 33 animals (72.7%) anda chronic form in one animal only (3%). Theaverage incubation period was 12.1 + 1.0

days. Of the animals which were allowed tohave a complete evolution of the disease, onlyone animal out of 14 (7%) showed a sponta-neous cure, the other 13 (93%) were founddead with acute forms (table 10).

Parasitological and bacterio%gical results : .’Before inoculation, T. hyodysenteriae were

not detected, but after inoculation all the ani-mals showing clinical symptoms of swine

dysentery excreted this organism in large

numbers (table 11). Campylobacter were seenat a low level in 50% of the animals, andBalantidium too were present on 24% of theanimals.

These organisms could play a role in enhan-cing the pathogenicity of Treponema aloneand combined they could provoke a verysevere dysentery.On animals dead in acute dysentery, the

counts in T. hyodysenteriae, Balantidium, andCampylobacter are very high on almost all thepigs. As previously seen, Trichomonas is onlyirregularly present.

Discussion

In three different experiments we compared- dysentery obtained in healthy animals infec-ted by exposure to contaminated pens. Thedisease obtained is then similar to the « natu-ral » disease obtained in farms.- dysentery obtained in animals orally infectedwith colonic mucosa from acutely affecteddonors (the oral infection being done withcolonic mucosa scrapings and contents).- dysentery obtained in animals infected withoral Treponema hyodysenteriae pathogenicstrains.

1. Clinical results

The average clinical results (morbidity, mor-tality and clinical data) are summarized in

figure 2, and they cover respectively 83, 280

and 33 animals, i.e., a total of 396 animals. Onthe morbidity alone, T. hyodysenteriae infec-tion was slightly less pathogenic than theother two modes of infection : 73% of animalswith swine dysentery (versus 93 or 90%), and3% of the cases were chronic at onset insteadof 1 or 0%.

If we consider the day of onset of dysenteryin all the affected animals, the disease occur-red in about the same number of days, 10, 12 2or 14 days after infection. However, this figureis less precise for pen contamination (14 days)as we only know the day the animals were putin the contaminated pen, but not really whenthe infection started.

For those animals who will die in acuteform, the disease is exactly comparable irres-pective of the type of infection (same numbersof days with dysentery stricto sensu, withmucoid diarrhea). In average the day of deathis close (16 to 23 days post infection).

For the animals who will die in chronic form,the first two parameters are the same (daynumber of onset ; number of days with dysen-tery stricto sensu), but a significant differencewas found in the number of days with mucoiddiarrhea. This symptom is, in fact, the onlytrue difference between the acute and chronicforms of dysentery.

For the spontaneously cured disease, the

dysentery occurred slightly later (13 to 14 daysinstead of 9, 10, 10, 13 days), the dysenterystricto sensu was equivalent (6 to 10 days ins-tead of 5, 6, 8, 8 days), and the mucoid diar-rhea was comparable. But again, the self-cured disease is the same whether given bypen contamination or oral infection.

In regard to clinical signs, it is easy to diffe-rentiate acute, chronic and spontaneouslycured dysentery, but for two latter it is difficultto ascertain a real difference.

Acute dysentery is the same when given bypen contamination, oral infection or T. hyody-senteriae infection.

2. Reproducibility of the experimental dysen-tery model with colonic scrapings and con-tents

Comparison of results obtained previously(1970-1971) and today. We infected similaranimals in equivalent conditions with freshmaterial in 1970 and 1971 and the results werepublished later (Raynaud et al., 1972). Theycould be compared with those obtained in thefirst series of 1976 experiments. These compa-rative clinical results are presented in table 12.If we compare the clinical results obtained inthese two series of experiments we realize that

the disease we gave in 1970-1971 is exactlycomparable to the disease we give today. Ofcourse the source of infective material is diffe-rent as it originates from field cases 6 yearsapart ; but the onset of dysentery was in 10 to11 days, the percent of spontaneously curedanimals 18 or 17 ; the day of dysentery onseton animals dead with dysentery was thesame : 10 days, and the day number of death16 or 17.

It is surprising to show that the infectivematerial we have in hands today is the same asthe one we had six years earlier and that the

piglets respond to this infective material in thesame way.

3. Results from rr freezing versus freshmaterial

These results are different from those pre-sented by Kinyon et al. 119761. They state« freezing of dysenteric colonic specimens hasbeen shown to be detrimental to the survivalof T. hyodysenteriae N. But our conditions ofwork are also different as deep freezing is

applied here. In our experience, the deep free-zing of infective material is recommended tostore the disease in a laboratory. It is veryeffective if the original comes from a severedisease, rich in Treponema sp.

4. Non pathogenic strain of « large beta

hemolytic Treponema u

According to Harris et al., (1978) cc pathoge-nic and non pathogenic T. hyodysenteriaemay be separated on the basis of hemolyticpattern ». But in this experiment a typical betahemolytic large Treponema sp. was not

pathogenic. This could be the same sort ofdeviation recorded by Taylor (1979) under thestatement « the ready identification of patho-genic and non pathogenic organisms has yetto be achieved although many isolates resem-ble classical T. hyodysenteriae in their antige-nicity and cultural characters and can be safelyclassed with that organism ». This in factcould mean that recognizing and isolating atypical T. hyodysenteriae is not fully suffi-

cient ; its pathogenicity should also be asses-sed. Authors have completed a rapid, econo-mical and reliable model to check this patho-genicity with an in vivo multiple, ligated loopsof swine colon model (Whipp et al., 1978).

Again, in our experiment the only non patho-genic T. hyodysenteriae strain was isolatedfrom a spontaneously cured animal. We donot know the significance of this observation.

Disease given by T. hyodysenteriaeOn conventional or SPF swine, pathogenic

strains of T. hyodysenteriae produce a severe

disease ; in our experimental conditions 76%of the infected animals exhibited an acute or

chro,nic dysentery. Of the animals which wereallowed to have a complete evolution of thedisease only 7% showed a spontaneous curei.e. 93% were found dead with acute dysen-tery. These findings are in accordance withthose of the majority of authors who haveworked with similar protocols. But the diseasewe obtained was in fact a very severe one.Only Fernie et al., (1975) reported an entirelydifferent experience « typical swine dysenterywas reproduced in healthy pigs by feedingpure cultures of Campylobacte! together withspirochaetes but not feeding pure cultures ofthe spirochaetes... alone ». These results arepresented as still valid today by Windsor(1979) but they are questionable as it is todayknown that the dysentery can be experimen-tally produced by the infection of germfreepiglets when spirochaetes are associated witha number of strict obligate anaerobes which

were absent from the Fernie et al. (1975) com-bination (Meyer and Simon, 1975).And it is important for a valid comparison

between these different experimental results,to realize that the normal flora, before infec-tion of the animals we used, contained syste-matically Campylobacter and frequentlyBalantidium. It is known that Campylobactercomplicates or aggravates the disease givenby T. hyodysenteriae alone (Hamdy andGlenn, 19741. It is also suspected that Balanti-dium could « enhance the severity of lesionsdue to spirochaetes » (Morales et a/. 1976).This is also our experience (Raynaud et al.,1972). Consequently, the presence of Campy-lobacter on some or many animals even if nor-

mally harmless and non pathogenic, couldmake a very severe typical dysentery when the eanimals are infected with T. hyodysenteriae.Different results could be expected if the« normal flora » (containing the secondaryorganisms) is different at time of infection.

5. Laboratory results

The help of the laboratory is useful in exami-ning swabs on Vago colored slides ; this per-mits to show up the presence of Treponemasp. and Campylobacter sp. at an abnormallyhigh level. It should be noted that before con-tamination, on normal animals, there is some-times a low level (1 = +) of Campylobacter.This germ is known to be present - at a verylow level - in healthy animals. Treponema sp.can also be found in healthy carriers, but todemonstrate this, special techniques must beused, and in our conditions of work it is excep-tionally found in healthy animals.

The examination of fresh fecal samples canconfirm the presence of Treponema sp. as along spiral-shaped and motile filament of cha-racteristic appearance. Before confirming thatit is T hyodysenteriae, this germ must be iso-lated and cultured to show its pathogenicity.In our experience a T. hyodysenteriae, like

organisms with typical complete hemolysis,could be non pathogenic.

Fresh fecal samples also make it possible tocount Balantidium coli motile trophozoiteswhich become more numerous and more fre-

quent when the diseased animal is near deathwhere it is found in 83% of animals. The roleof this protozoon is then shown by the necro-tic appearance of the lesions, or their evolu-tion towards a chronic form.

Trichomonas are only occasionally found :counting them does not seem to hold muchinterest either for the diagnosis or for the pro-gnosis of the dysentery.We consider that to appreciate the severity

and evolution of dysentery it is important tocount Treponema and Balantidium trophozoi-tes in the feces, and to estimate the impor-tance of Campylobacter on a smear. These arethe major micro-organisms which are easy tocount.

To illustrate these points, some results withthe three different contaminations are presen-ted in figure 3. The Campylobacter index onstained smears shows an increase during thedisease, on start or on death. The Treponemaindex on stained smears shows a significantincrease during the disease ; the results areeasier to interpret as no organism was foundon normal animals. The Treponema counts aresignificantly modified by the experimentaldisease and Balantidium counts are uniformlymodified too.

No difference was seen between the threemodes of contamination, i.e., the experimen-tal disease is very similar when dysentery is

given by pen contamination, oral or T. hyody-senteriae infections, but we noticed a relativedecrease of Treponema counts or indexes

during the disease with pen contamination(« natural disease »1. We have no explanationfor this relative loss of Treponema voided inthe feces when the disease becomes moresevere, i.e., closer to death.

Also, on figure 3 we see a parallel trend :similar results when Treponema counts onfresh smears are compared with Treponemaindex on stained smears.

Conclusions

With the « normal » SPF or conventional

young swine which were used in these experi-ments since 1970, the dysentery obtained is

stable when clinical and pathological results

are considered. The swine have Campylobac-ter spp. and Balantidium coli as permanentorganisms in the intestinal flora. This could

explain the severity of the disease obtainedeither after oral inoculation of T. hyodysente-riae strains or colonic mucosa and contentsfrom donors with acute dysentery, or aftercontamination in pens.The disease is mainly acute, and only a

small percentage of the animals are self-cured.In regard to the clinical features, experimentalor « natural » swine dysentery is characterizedby an alternation of days with dysenterystricto sensu and days with mucoid diarrhea,the latter becoming more important and morefrequent in the self-cure or chronic diseases.

Changes of the intestinal flora are very

important, and it is easy to demonstrate

occurrence of Tieponema strains and increaseof « small spirochaetes », Campylobacter andBalantidium.

The model of experimental disease is easyto standardize with oral inoculation of colonicmucosa and scrapings and easy to keep in thelaboratory as deep frozen material. It is also

relatively easy to check the possible contami-nation of pens by using « sentinel » animals.

However, large beta hemolytic Treponemastrains are not always pathogenic if selected

by morphology and cultural characteristics. Itseems difficult to ascertain I hyodysenteriaeotherwise than by the experimental infectionof SPF piglets, but we had no opportunity toexperiment the newly described technic of« ligated colonic loops » which could avoidthe troublesome handling of animals.

Accepted for publication September 17 th 1979.

Acknowledgements

The authors express their gratitude to Dr.E.B. Patterson, Scientific Director Agricultu-ral, Pfizer International, for having encoura-ged this work and its publication. They alsothank for their competence and interest M. J.Taillant, Farm Manager, and M. P. Lejeau,Foreman.

Summary

Oral inoculation of colonic mucosa scrapings and intestinal contents of animals affected withswine dysentery, or of pathogenic strains of Treponema hyodysenteriae, as well as spontaneouscontamination in infected pens caused in average swine dysentery to appear in 359 out of 409SPF piglets. The morbidity is high irrespective of the method of contamination : after pen conta-mination, 75 out of 83 piglets were dysenteric ; after only one ingestion of contaminated matter,265 out of 280 animals were ill ; and after inoculation of T. hyodysenteriae, 25 out of 35. Mortalityin dysentery was always higher than 80 %, respectively 48 out of 60 animals, 132 out of 154, and13 out of 14. Very few animals were self-cured. The average incubation period varied accordingto the mode of contamination between 9 and 13 days. Animals having contracted and acute formof the disease died 16 to 23 days after contamination, and those with a chronic from 29 days, alsoafter contamination. The increase in the number of Treponema, of Campylobacter and of Balan-tidium observed after the onset of the disease was approximately equivalent for all three modesof contamination. This disease was characterized by an alternance of dysentery stricto sensu andof mucoid diarrhea, the latter occurring more frequently in cases of self-cure and in chronicforms.

References

DEAS D.W., 1960. Observations on swine dysentery and associated vibrios. Vet. Rec., 72, 65-69.DOYLE, 1944. Cited by Harris and Glock, 1975.ELAZHARY M.A.S.Y.. BERTHIAUME L&dquo; ROY R.S., ALAIN R., LAGACE A., 1978. Morphological and

quantitative study of spirochaetes in feces of normal and infected SPF pigs during the incubation periodof swine dysentery. Can. J. Comp. Med., 42, 302-309.

ESPINASSE J., 1971. Dim6tridazole et typhlocolite nécrohémorragique du porc. Essai d’interprétation. Cah.Med. V6t., 40, 95-99.

FERNIE D.S., GRIFFIN R.M., PARK R.W.A., 1975. The possibility that Campylobacter (Vibrio) coli and Tre-ponema hyodysenteriae are both involved in swine dysentery. Br. Vet. J., 131, 335-338.

HAMDY A.H., GLENN M.W., 1974. Transmission of swine dysentery with Treponema hyodysenteriae andVibrio coli. Am. J. Vet. Res., 35, 791-797.

HARRIS D.L., GLOCK J.G., 1975. Swine dysentery in Diseases of swine, 4th ed., 541-553, Iowa State Uni-versity Press, Ames.

HARRIS D.L., GLOCK R.D., CHRISTENSEN C.R., KINYON J.M., 1972a. Swine dysentery. I. Inoculation ofpigs with Treponema hyodysenteriae (new species) and reproduction of the disease. Vet. Med., 67,61-64.

HARRIS D.L., KINYON J.M., MULLIN M.T., GLOCK R.D., 1972b. Isolation and propagation of spirochetesfrom the colon of swine dysentery affected pigs. Can. J. Comp. Med., 36, 74-76.

HARRIS D.L., KINYON J.M., GLOCK R.D., SONGER J.G., EGAN I.T., 1978. Cultural detection of swine in-fected with pathogenic Treponema hyodysenteriae. Proceed. 5th Congr. lnternat. Pig Veterinary Socie-ty Zagreb, M3.

JAMES, DOYLE, 1947. Cited by Harris and Glock, 1975.

KINYON J.M., SONGER J.G., JANC M., HARRIS D.L., 1976. Isolation and identification of Treponemahyodyset?teriae aid to the diagnosis and treatment of swine dysentery. Proceed. l9th annu. Meet. Am.Assoc. Vet. Lab. Diagn., Miami Nov. 1976, 189-196.

LYSONS R.J., HALL G.A., ALEXANDER T.J.L., BEW J., BLAND A.P., 1978. Aetiological agents and pa-thogenesis of swine dysentery. Proceed. 5th Congr. lt7ternat.Pig Vet. Soc., Zagreb, M1.

MEYER R.C., SIMON J., 1976. Swine dysentery : an etiologic study. Proceed. 4th Congr. lnternat. Pig Vet.Soc. Ames, L4.

MORALES G.A., AYCARDI E., BELTRAN L.E., 1976. Swine dysentery caused by Treponema hyodysente-riae in Colombia. Proceed. 4th Congr. lnternat. Pig Vet. Soc. Ames, L5.

RAYNAUD J.P., BRUNAULT G., PATTERSON E.B., 1980. A new swine dysentery model for evaluation ofdrug prophylaxis : model development. Oral infection plus pen contamination. ln press.

RAYNAUD J.P., RENAULT L., MAIRE C., VAISSAIRE J., 1972. Reproduction experimentale de I’enteritehbmorragique (dysenterie) du porc en France. Etude de la maladie exp6rimentale en phase aigue. Rev.Med. V6t., 123, 729-754.

RONCALLI R.A., LEANING W.H.D., 1976. Geographical distribution of swine dysentery. Proceed.

/. P. V. S., Ames, L17.TAYLOR D.J., 1979. Swine dysentery. Vet. Pract., 1, 4-9.TAYLOR D.J., ALEXANDER T.J.L., 1971. The production of dysentery in swine by feeding cultures contai-

ning a spirochaetes. Br. Vet. J., 127, Iviii - Ixi.

TERPSTRA J.I., AKKERMANS J.P.W.M., OUWER KERK H., 1968. Investigations into the etiology of vi-brionic dysentery (Doyle) in pigs. Neth. J. Vet. Sci., 1, 5-13.

WHIPP S.C., HARRIS D.L., KINYON J.M., SONGER J.G., GLOCK R.D., 1978. Enteropathogenicity testingof Treponema hyodysenteriae in ligated colonic loops of swine. Arn. J. Vet. Res., 39, 1293-1296.

WHITLING, 1924. Cited by Harris and Glock, 1975.WHITLING et al., 1921. Cited by Harris and Glock, 1975.WINDSOR R.S., 1979. Swine dysentery. Vet. Annu., 19, 89-96.