-

R.A. BUSCH

MFR PAPER 1296

Enteric Red Mouth Disease(Hagerman Strain)

ABSTRACT - Enteric red mouth (ERM) disease of salmonid fishes is reviewedin terms of description of the etiological agent, pathogenesis and diagnosis of theclinical condition, and epizootiological considerations pertinent to its effectivecontrol. Recent studies defining the asymptomatic carrier state of the disease arediscussed. Enteric red mouth disease is shown to establish an asymptomaticcarrier state infection in the lumen of the lower intestine in 50 to 75 percent ofepizootic survivors 60 to 65 days postinfection. This chronic carrier state is main­tained for more than 102 days in the experimental population. A cyclical intestinalshedding pattern develops with a periodicity of 36 to 40 days and serves to causeregular reinfection and mortality.

A serological presumptive screening test for ERM is described. The system isshown to be sensitive and specific and readily adapted to the rapid screening oflarge numbers of individual small volume sera for presumptive evidence ofspecificdisease association at a minimum cost and without sacrifice of fish over 15 cm inlength. Based upon serological screen field data, ERM is shown to consistentlydevelop a 2.0 to 2. 7percent carrier incidence among populations following epizoo­tic infection. The humoral immune response of rainbow trout to various antigenicpreparations of the ERM bacterium is examined. Various soluble protein fractionsare shown to be more immunogenic than lipopolysaccharide fractions. A particu­late cell wall antigen and several other killed whole cell antigens resulted in hightiters while a heat-killed whole-cell preparation is shown to be nonimmunogenic.Serum agglutinin titers declined at a monthly rate of 9.8 percent following inductionand were detectable for more than 9 months. Evidence is given to support thetheory that serum agglutinins in trout are IgM-like macroglobulins.

cause a similar clinical syndrome.McDaniel (1971) proposed the nameHagerman red mouth or HRM, describ­ing its original range. In 1975 the FishHealth Section of the AmericanFisheries Society chose to remove thisonus on the Hagerman Valley and cal­led it enteric red mouth disease, indicat­ing that the causative agent is one of thetwo major enteric pathogens of fish infresh water: the other being Ed­wardsiella tarda. Enteric red mouthdisease has also been referred to in theliterature as red mouth disease, red ventdisease, and bacterial hemorrhagic sep­ticemia, but in many instances, authorswere describing a clinical syndromethat was often of diverse or multipleetiology rather than the etiologicalagent itself.

In 1966, ERM was found to be en­demic to the Hagerman Valley of Idahoas well as the States of California,Nevada, Arizona, and Colorado. Whenone studies the epidemiology of thisgeographically isolated disease and itsdissemination, one often finds that itwas introduced with subclinically in­fected stocks of fish acquired from anarea where it is endemic. As an exam­ple, in the middle and late 1960's, thetrout industry in the Hagerman Valleyshipped large numbers of asymptomat­ically infected live fish to various west­ern states for grow-out and fish-out op­erations. During that time the diseasewas widely disseminated, and in recentyears economically important geo­graphical range extensions and epizoo­tics of ERM were documented to occurdue to transporation of asymptomati­cally infected stocks into previouslydisease-free areas (Wobeser, 1973). By1970, the originally recognized rangeof occurrence had expanded to includeAlaska, Oregon, Utah, Washington,and Wyoming. More recently, it hasbeen reported from British Columbia,Montana, Nebraska, Ohio, Saskatche­wan, and Tennessee. The disease hassince been reported from various otherplaces outside of North America, par­ticularly Italy, but these reports, to myknowledge, have not been serologicallyconfirmed. It is significant to note thatERM still remains comparatively lim­ited in its geographical distribution on

EPIDEMIOLOGY

R. A. Busch is with the Fish Pathol­ogy Laboratory, School of NaturalResources, Humboldt State Univer­sity, Arcata, CA 95521. Present ad­dress: Rangen Research Hatchery,Rt. 1, Hagerman, JD 83332.

The ERM bacterium was originallyreferred to as the "RM bacterium."This terminology tended to confuse itwith other bacterial pathogens that

Enteric red mouth (ERM) diseasewas first recognized as a cause of mor­talitieS in rainbow trout, Salmogairdn!eri, in the late 1950's in theHagerman Valley of Idaho. Sub­sequently, the disease was studied byRoss et al. (1966) and Rucker (1966).These workers were unable to give theetiological agent a more definitivetaxonomic position, but based uponmorphological, biochemical, andserological data, they placed it in thefamily Enterobacteriaceae.

42 Marine Fisheries Review

a worldwide watershed basis and,therefore, demonstrates an excellentpotential for preventative managementand an eradication type of program incomparison to other more ubiquitousfish pathogens.

ECONOMIC EFFECTS

Some causative agents of fish dis­eases are ubiquitous in their occurenceand even free living in certain instances,while other pathogens are fairly fastidi­ous and geographically isolated, as isthe case with ERM. This will have adefinite practical bearing on how we aregoing to prevent, manage, treat, andhandle stocks offish infected with ERMand reduce the adverse economic im­pacts of the disease on the trout indus­try .

In the Hagerman Valley, commer­cial fish farms located along a 35-kmstretch of the Snake River canyon pro­duce about 80 percent of the commer­cial rainbow trout in the United States.These are private and highly compete­tive operations. It is difficult to get ac­curate figures, but this productionamounts to over 9,000 metric tons (t) ofrainbow trout a year. One hatchery thathas just been put into operation has anannual production capacity of 2 ,250 t oftrout based upon a flow of 9.20 m3 persecond of spring water from the aquiferof the Snake River plain.

Araji (1972) did an economic studyon the impact of disease on the troutindustry in the Hagerman Valley andestimated an annual loss of approxi­mately $750,000 or roughly IO percentof the total production costs. I haveseen loss estimates as high as 35 percentof production costs and up to $2 V2 mill­ion annually depending upon the opera­tion and management involved. It isinteresting to note a recent study byKlontz and King (1975) on the exactnature of these losses. If you were tofollow losses due to disease in egg, sacfry, swim-up and fingerling stages, andon up to marketable 15- to 30-cm fish,the total number of mortalities aregreatest in the egg through swim-up frystages and decrease as the fish get older.However, while the mortalities in mar­ketable size fish represents only 14 per-

March /978

cent of the total mortalities, it amountsto 76 percent of the costs in dollars. Themortality in the small fish in thesehatcheries is mainly attributed to bacte­rial gill disease and othermanagement-related conditions as wellas infectious pancreatic necrosis, apresently untreatable viral disease. Incomparison, losses in the marketablesize fish are attributed to ERM andother treatable bacterial diseases. Thisconcept is of primary importance in theconsideration of the economics of dis­ease prevention and treatment in thesestocks.

CAUSATIVE ORGANISM

Morphologically, the ERM bac­terium is a 1.0 x 2.0 to 3.0 I-tm, gramnegative, monotrichously flagellatedrod which forms a smooth, circular,raised, entire, nonflourescent, nonpig­mented colony on nutrient agar with abuterous type of growth. It is non­aerogenic, fermentative, cytochromeoxidase negative, catalase positive, andtypical of enteric bacteria in general andis indole negative, methyl red positive,Voges-Proskauer negative, and citratepositive. This brings it rather closelydown to the Erwinia group of the familyEnterobacteriaceae. Serologically, an­tigenic similarities have been foundwith the somatic 0 antigens of atypicalErwinia groups 26 and 29 (Ross et aI.,1966; Cisar, 1972).

MORTALITY ANDTRANSMISSION

Enteric red mouth disease has beenknown to cause mortality in all troutand salmon. The most susceptible hostseems to be rainbow trout, with mortal­ity commonly running from 25 to 75percent during the course of an un­treated epizootic. Populations of brook,Salvelinus fontinalis, and brown trout,Salmo trutta, seem to be more resistant,with 5 to 10 percent mortality common.The bacterium has not been isolatedfrom warm-water fish, even under ex­perimental laboratory induction. Thedisease is commonly endemic and sur­vives for long periods of time in organi­cally rich waters and, consequently,lends itself quite well to enzootic infec-

tion under intensive culture situations.It is also commonly found as a clini­cally asymptomatic carner-state infec­tion.

Transmission of ERM occursthrough the water from feces of infectedfish. Pathogenesis of infection variesfrom peracute to chronic dependingupon the temperature, stress, species,age, and so forth. Peracute to acuteinfection usually occur in the spring andearly summer, usually in young-of­the-year fish, during periods of risingwater temperatures, and increasedhandling stress. Mortalities usuallycommence 4 to 8 days following expo­sure and run between 50 and 70 percentduring a 30- to 60-day course of clinicalinfection. Acute to subacute infectionsusually occur in yearling fish in the falland early winter with declining watertemperatures. Mortality usually runsabout 10 to 50 percent in a 2- to6-month period. Chronic infection re­sults in very low levels of mortality,approximately IO percent. However,these losses are often incurred in veryvaluable market-sized or mature broodstock fishes and, economically, a IOpercent mortality can be significant.

CLINICAL SIGNSAND PATHOLOGY

The clinical pathology of ERM isquite characteristic of bacterial hemor­rhagic septicemias in general. Bydefinition, ERM could best be referredto as a bacterial hemorrhagic sep­ticemia of salmonid fishes, caused bythe ERM bacterium, having a very welldefined geographical range of distribu­tion and endemic occurrence, and cap­able of inflicting severe mortality.

The common external clinical man­ifestations include subcutaneoushemorrhaging along the base of the finsand in or about the oral cavity and anus.Exophthalmos is brought about by tis­sue edema. This condition is noted tooften start unilaterally and develop to abilateral involvement (Fig. I). His­topathological examination dem­onstrates an edematous type of lesion ofthe choroid gland of the eye with anintraocular accumulation of fluid. Thisedema and increased intraocular fluid

43

pressure induces the exophthalmiccondition and eventually results in rup­ture of the eye with ensuing lens opacityand blindness (Fig. 2). Infected fishwill physiologically darken in color andare easily identified in populations sus­pected as infected, particularly as car­riers.

Internal gross pathology is distin­guished by petechial hemorrhage of thevisceral organs and tissues includingthe liver, pancreas, adipose tissues,swim bladder, various coelomic mesen­teries, and body musculature. Grosstissue edema is noted in the kidney,liver, and spleen along with hemor­rhagic reddening of gonadal tissues andthe distal ends of the pyloric caeca(Figs. 3-6). The gross pathology of thelower intestine is probably the mostsignificant clinical diagnostic sign ofERM. The intestine becomes inflamed,flaccid, translucent, hemorrhaged, anddistended with a serosanguineous yel­low mucoid material consisting of nec­rotized intestinal mucosa heavilyloaded with the pathogen.

The histopathology is typical of ahemorrhagic septicemic type of infec­tion and, in the acute form, the bac­terium is commonly found in theperipheral blood (Fig. 7). Thehematological picture is characterizedas an acute macrocytic, hypochromicanemia with leukopenia, resulting fromnecrotic destruction of hematopoieticand reticuloendothelial tissues, and anincreased clotting time as well as aplasma aproteinemia attributed toglomerular nephritis and necrosis of theintestinal mucosa. This condition re­sults in a decrease of the plasma colloidosmotic pressure producing gross sys­temic tissue edema as well as disruptionof the ionic balance.

The acute form of the disease is dis­tinguished by loss of capillary struc­tural integrity, tissue edema, and loss ofosmotic and ionic homeostasis. Thiscondition results in a rapid course ofinfection of 4-10 days with minimumgross clinical pathology. The chronicform is characterized by localized tis­sue hemorrhage and necrosis, a declinein nutritional condition, and secondaryinfection. Petechiation results from aloss of capillary integrity and erythro-

44

Figure I.-Bilateral exophthalmos in a rain­bow trout, Sa/rna gairdneri, due to gross tis­sue edema caused by acute septicemic infec­tion with enteric red mouth disease.

Figure 2.-Ruplure of the eye, lens opacity,and blindness in a rainbow trout, Sa/rnagairdneri, due to increased intraocular fluidpressure and edema caused by acute sep­ticemic infection with enteric red mouth dis­ease.

Figure 3.-GeneraJized petechiation of vis­ceral organs and tissues of a rainbow trout,Sa/rna gairdneri, due to acute septicemic in­fection with enteric red mouth disease.

cytic congestion of capillary beds andblood sinusoids (Fig. 8). The intestinaltract demonstrates a progressive nec­rosis and sloughing of the mucosa pro­ceeding down to the stratum compac-

Figure 4.-Petechial hemorrhage in the liverof a rainbow trout, Sa/rna gairdneri, due toacute septicemic infection with enteric redmouth disease.

Figure 5.-Hemorrhagic plaques in the swimbladder of a rainbow trout, Sa/rna gairdneri.due to an acute septicemic infection with en­teric red mouth disease.

Figure 6.-Petechial hemorrhage of the later­al musculature and visceral mesentery of arainbow trout, Sa/rna gairdneri, due to anacute septicemic infection with enteric redmouth disease.

tum (Fig. 9). This situation is highlyconducive to secondary infection andcomplication of the clinical picture bymultiple etiologies.

Marine Fisheries Review

r

remammg there for more than 102days. The relative intestinal recoveryrate and presence of the pathogenwithin the infected populations wasalso shown in Figures 12 and 13 to be ofa cyclical nature, with a periodicity of36 to 40 days. Only 25 to 50 percent ofthe asymptomatic carrier infections es­tablished would be clinically identifiedfor inspection and certification pur­poses by using classical attempts at iso­lation from the kidney and spleen dur­ing the low points of the cycle.

Cyclical Nature ofCarrier Infections

On the basis of pathogen recovery,gross pathological changes, and mortal­ity rates, it appears that a regular 36- to40-day cycle of intestinal shedding ofthe ERM bacterium occurs and pre­cedes the recurrence of systemic in­volvment and mortality by 3 to 5 days.This type of cyclical shedding couldprecipitate continuous recurrent mortal­ity in a naturally infected susceptiblepopulation' throughout the year. The ac­tual periodicity and mortality levels ofthese shedding cycles would be alteredby seasonal variations in water temper­atures, loading factors, handling, andother stresses as well as natural resis­tance and immunity of the population.This hypothesis is substantiated by theobservations of McDaniel (1971), whodemonstrated cyclical mortality pat­terns of similar periodicity to occurthroughout the year in a large, untreatedhatchery population chronically in­fected with ERM disease (Fig. 14).

30

25

OJ F M A M J J A SON 0 JMONTH

Figure 14.-Monthly mortality rates attrib­uted to endemic enteric red mouth diseaseinfection of an untreated hatchery populationof rainbow trout. SalmI! gairdneri. (fromMcDaniel. 1971).

March /978

Many demonstrated aspects of theclinical course of induced ERM infec­tion have the classical appearance of anacquired protective immune response.In particular, the decrease in mortalityand relative rates of gross pathologicalchanges ocurred 14 to 21 days postin­fection. This time period correspondsto the expected lag phase andlogarithmic induction period of a pro­tective immune response at 14.5°C.Electrophoregram analysis demon­strated a quantitative serum macro­globulin response as shown in Figures10 and II. However, this quantita­tive change in serum macroglobulincould not be detected as agglutinatingor precipitating humoral antibodies,indicating the possible involvement ofa protective monovalent or incompleteimmune globulin fraction, possiblyfunctioning in an opsonizing or cyto­philic capacity. An induced cellularimmunity is most probably responsiblefor the rapid clearance of the pathogenfrom the reticuloendothelial tissues andits localization in the lumen of the lowerintestine where it is comparatively re­moved from macrophage and immuno­globulin activity.

This insight into the establishment ofa clinically asymptomatic carrier infec­tion and its demonstrated cyclical na­ture of shedding could find direct appli­cation to management practices of thedisease in terms of the timing of hand­ling stress, loading factors, and other"controllable stress." The use of an­tibiotic or chemotherapeutic agentswith a high retention time in the lumenof the gut is indicated rather than othermore rapidly absorbed agents for treat­ment of the carrier condition. Detecting.the condition by classical isolation andidentification methods (particularly ifthe lower intestine is not sampled) isdifficult, and this may seriously affectsurveillance and certification opera­tions. The cyclical nature of the carrierinfection also indicates the need for apaired sampling at a IS-day interval toachieve maximum diagnostic ef­ficiency.

Identifying Carriers

Of major consequence to currentprograms of inspection, certification,

and control is the ability to efficientlydiagnose clinical carrier infections.Present methods of classical isolationand identification of the pathogen arewholly inadequate in terms of sensitiv­ity, logistics, time frame, and eco­nomics. Programs of this type often in­volve large numbers of fish which mustbe certified free of particular diseaseagents but are often highly valuable andcannot be sacrificed as in the case ofbrood stocks.

Development of the PassiveAgglutination Test

As a result of my experience with theU.S. Public Health Service Center forDisease Control's sylvatic plague sur·veillance program, I have developed ahighly sensitive and specific means ofserological disease surveillance requir­ing a minimum of lethal sampling. Theprocedure is based upon a passive orindirect agglutination test using micro­titer techniques.

The antigenicity of the ERM or­ganism was examined in detail. A pH6.4 boiled aqueous extraction of awashed 18-hour ERM culture wasfound to be the most antigenically com­plete soluble fraction of the organism.Many different substrate particle sys­tems, including latex, bentonite, char­coal, fresh erythrocytes, and fixederythrocytes were incorporated into thetest. Maximum sensitivity was ob­tained with fresh citrated sheep eryth­rocytes which were tanned and sensi­tized with 10 mg% antigen protein atpH 6.4. Test sera were diluted in nor­mal physiological saline with normalrabbit serum added at I: 100. Other sub­strate particle systems were found to besomewhat less sensitive but could belyophilized and stored for long periodsof time.

Figure 15 demonstrates the aggluti­nation patterns achieved. Note the verysharp end points and negative buttons.Titers are easy to pick out, and the testis well adapted to either screening ortitering test sera. In addition to the pas­sive agglutination test, a paired inhibi­tion dilution series utilizing 20 mg%antigen protein added to the serum di­luent was run. The inclusion of the

47

·..-lMlI9I • ' '.·c

'-

,e

G

2G'G'00000000@\?S(.)GO 08GG-00000008(i)

oOOOOOC@

0000000000 000

-0

Figure 15.-Typical results of the micromodified passiveagglutination test demonstrating diffuse patterns of positiveagglutination and negative buttons as well as end point titersof sera di Jution series.

Figure 16.-Basic materials necessary for the performance of the micromod­ified passive agglutination test applicable for screening of large numbers ofindividual small volume sera for the presumptive serological evidence ofspeci fie disease association.

Marine Fisheries Review

DCAARIERTITERS

OCARRIERTITERS

NQNCLINICAL

NQNCUNICAL

INFECTIONION

~

~ 5

OCARRIER ~ 4TITERS

Cl.INICAL

(JCARRIERTITERS

CLINICAL

INFECT

r

I ~ ,

Figure l8.-Natural distribution of serumpassive agglutination titers from hatcherypopulations of rainbow trout, Sa/rnogairdneri. surviving a recent clinical epizoo­tic of enteric red mouth (ERM) disease (left)and those not clinically infected but grown inan endemic area of ERM infection (right). assampled in October 1972.

Figure 17.-Natural distribution of serumpassive agglutinin titers from hatchery popu­lations of rainbow trout, Sa/rna gairdneri.surviving a recent clinical epizootic of entericred mouth (ERM) disease (left) and those notclinically infected but grown in an endemicarea of ERM infection (right), as sampled inSeptember 1972.

titers for the infected station stocks anda 7.7 percent incidence for all otherstations combined. Positive serum tit­ers during this period ranged from 1: 16to 1:32,768. The actual individual titerdistributions are summarized in Figure17 for the two populations. A definitenatural split in high and low level titersoccurred. High level natural titers of1:512 or greater (shaded areas of Fig.17) constituted 19.4 percent of all posi­tive titer (shaded and unshaded com­bined) from the clinically infected sta­tion and 30.8 percent of all positivetiters from all other nonclinically in­fected stations combined during thefirst sampling date.

As summarized in Figure 18, an 8.4percent incidence of positive titers forthe infected station and an 8.8 percentincidence of positive titers for the otherstations were found at the second sam­pling date. When the natural break be­tween the high and low level titers wasexamined for the second set of data,high titered sera were absent from theinfected station and constituted 21.3percent of the positive sera from theother stations combined.

It was postulated that the naturalbreak between high titered and low ti­tered sera is indicative of the differencebetween subacute to chronic clinical in­fections and the presence of an estab­lished carrier population, respecti vel y.

paired inhibition test gave completecontrol over any nonspecific agglutinat­ing activity for each sample tested,

This type of test lends itself to therapid field screening of large numbersof individual small volume sera for pre­sumptive evidence of specific diseaseassociation at a minimum cost andwithout sacrifice of fish over 15 em,Utilizing an unsophisticated hand­operated system costing $100, areasonably good technician can screenabout 500 fish in an 8-hour day againsta variety of diseases (Fig, 16), The testcan also be readily automated for evengreater efficiency, Results can be readimmediately following centrifugationor after 2-4 hours of incubation,

Field Application of PassiveAgglutination Test

In the fall of 1972, the passiveagglutination procedure was field testedin the Hagerman Valley. The fishstocks of eight stations having a historyof ERM, but no recent epizootics, and asingle station undergoing an activeepizootic were examined. Stationstocks were first sampled in September,just at the end of the usual seasonal peakin mortality attributed to ERM disease,and then reexamined 60 days later inNovember.

Initial serological screening resultedin a 38.2 percent incidence of positive

48

demonstrate a constant level of carrierincidence which is consistent withlevels of carrier incidence reported forother related diseases.

The overall results from the infectedstation, as shown in Figures 17 and 18,indicate an initially high rate of infec­tion with a proportionally largenumber, 6.3 percent, of high titeredpresumptive carrier fish being present.The second sampling indicated a 31percent decrease in positive screen ti­ters and the complete lack of any hightitered group of sera. This situationcould be indicative of a resistant recov­ering population in which true carrierstates had yet to be established. Thisdata could also be attributed to the factthat the second sampling was essen­tially a different lot of fish from thoseoriginally sampled, even though thestation was the same.

INDUCED HUMORALIMMUNE RESPONSEIN RAINBOW TROUT

262,144

'.096II::

2,048

W 1,024

t::"2.....2>6

ZQ '2.

..... ..ctz "..... ,.::l...JC)C)

ct

00 20 '0 120

TIME (days)

classical carrier states of infection.Such a hypothesis is supported by thefact that the relative rates of occurrenceof the high titered grouping remainedconstant in the combined stations sam­pled 60 days apart as shown in theshaded areas of Figures 17 and 18. The2.7 percent and 2.0 percent respectiveproportions of the populations seem to

262,14 4

TIME (days)

4,096

II:: 2,0 4 8

W 1,024.......... '"z ".0 "...... ..ctZ '2 -.>-BAE antigen..... -<>-ASS antigen::l...J -o-ASP antigenC) o :~~:~~NTRATIONC)

ct

131,012 131,072

'0 6~,~36 '06~,536 -.>-"U U

~ 32,768 ~ 32,168 -o-EW(

.~ 16,384a.

~CWEU ·u 16,384

~ 8,192 ~ 8,192 o :~~II~fs\RATION

This hypothesis is based upon the prem­ise that continued presence of thepathogen over a long period of time, asoccurs in an established carrier infec­tion, is necessary to induce high levelsof serum agglutinins. The lower titeredgrouping would then include those sur­vivors which still may harbor thepathogen but have not yet established

Figure 21.-The passive serum agglutinin response of rainbowtrout, Sa/rna gairdneri, to the administration of particulate antigenpreparations of the enteric red mouth disease bacterium. (AP, alumprecipitate; ew, cell wall; FK, Formalin killed; PK, phenol killed.)

Figure 19.-The serum passive agglutinin re­sponse of rainbow trout, Sa/rna gairdneri, tothe administration of water soluble antigenicpreparations of the enteric red mouth diseasebacterium. (BAE, boiled aqueous extract;ASS. ammonium sulfate supernate; ASP.ammonium sulfate precipitate.)

Figure 20.-The serum passive agglutinin re­sponse of rainbow trout, Sa/rna gairdneri, tothe administration of organic solvent solubleantigenic preparations of the enteric redmouth disease bacterium. (PE, phenol ex­tract; EWE, ether-water extract; eWE,chloroform-water extract.)

Induction of a specific immune re­sponse in rainbow trout to immun­ogenic preparations of the ERM bac­terium was examined in terms of itsnature and dynamics.

Figures 19 and 20 demonstrate thedynamics of a humoral agglutinin re­sponse in rainbow trout to extendedparenteral exposure to soluble antigenicpreparations of the ERM bacterium.Protein-based water soluble extracts areshown in Figure 19 and carbohydrate­based organic solvent extracts areshown in Figure 20. Various prepara­tions included a boiled aqueous extract(BAE), ammonium sulfate supernate(ASS), ammonium sulfate precipitate(ASP), phenol extract (PE), ether­water extract (EWE), and chloro­form-water extract (eWE). Humoralpassi ve agglutinati ng anti bodiesagainst protein-based soluble antigenswere first detected 13 days after intia­tion of the first of two series of weeklyinjections. Specific passive agglutina­tion titers to organic solvent extractswere initially detected in 28 days. Allserum agglutinin titers were shown torise throughout the initial period of an­tigenic stimulation, but when injections

200'80'60

-fr AP

-0- cw

-<>- FK

-0 PK

o :~~:~SNTRATION

'4080 100 120

TIME (days)604020

262,144

131,072

_ 6~,~36

832 ,768

~ 16,384

.~ 8,192

..; 4,096

0: 2,048

~ 1,024 !~ ~12}

Z 2'6

o 12S

!;:;: 64

Z 32

t=~ 16..JClCl<[

March /978 49

TIME (days)

Figure 22.-Passive agglutinin and quantita­tive macroglobulin response of rainbow trout,Sa/rna gairdneri, serum to aqueous and or­ganic solvent extracts of the enteric red mouthdiesease bacterium.

01~

Figure 24.-Gel filtration chromatographicfractionation and passive agglutinin activitydetermination of hyperimmune mammalian(rabbit) serum proteins.

TIME-day 120

E 0 C 8 A

TIME-day 42TIT<

toe B A

26t ~128 ~

I" .;

32 ffi..~ 4-0 '6 ~

~ 3-0 • ~

~ 2-0 .. ~~ z~ 1·0 l i=

"00o!--=''''''''--~'''''''''''''""--:''''=-'''''''''oo''--::''',,'''o--:,.~o---"o ~ELUTION VOLUME (V.J~ml ~

lIME- dayOTITER- nlqolivl

Figure 23 .-Sequential electrophoregrams ofpooled sera from hyperimmunized rainbowtrout, Sa/mo gairdneri. demonstrating serummacroglobulin response. Letters indicatemajor serum protein peaks based upon rela­tive electrophoretic mobility.

g MACROGLOBULIN ,RACTION

fraction. Electrophoretic separation ofpooled hyperimmune trout sera gener­ally indicated a quantitative increase ofthe anodic macroglobulin "F" fractionand the development of multiple peaksin this region during the course of con­tinued antigenic stimulation as shownin Figure 23. The gradual appearance ofmultiple macroglobulin peaks has alsobeen described by Evelyn (1971).

Analysis of gel filtration fractions ofanti-ERM hyperimmune rabbit serumwith the passive agglutination test indi­cated the presence of agglutinating an­tibodies in both the 188-198 and 78fractions of the rabbit serum as indi­cated in Figure 24. However, specific

Z2·0 W

fo­DII:

"8 Q..

otgo

E

'·4

: 2'2

."

........."

....

-<>- ~g~~~tl~ ~~~~~~~S[~ ORG. SOL. f)c TRACT

AGGlUTlNI"t RESPQHSE

o ~~~I'~~;TPATION

.• - ~~~~~ACT 1-2

.•. ORG. 501... EXTRACTMACROGlOElUl.IN

32,715

0 8 ,192

"e".09tla.~2,()48

-',024

6~.~36i__-

16,38

ffi :U2

~ ".zzi= 32::J

~ lIS\!)<t •

hyperimmunized with soluble antigensof ERM demonstrated an increase intotal macroglobulin (fraction F) serumprotein during continued antigenicstimulation but no direct correlationcould be made to the presence of serumpassive agglutinins (Fig. 22). However,the macroglobulin response is seen tobe directly proportional and responsiveto antigenic stimulation and resemblesa classical humoral immune responseincluding what appears to be a trueanamnestic response at 106 days. Thismacroglobulin response could befurther evidence of a functionally pro­tective monovalent cytophylic or op­sinizing globulin not detected byagglutination or precipitation tech­niques.

Electrophoretic analysis of the timesequence sampling of pooled sera fromhyperimmunized rainbow trout pro­duced a normal seven protein compo­nent pattern, as shown in Figure 23,consisting of: A prealbumin and albu­min complex grouped as peak "A"; asecond component, labeled "B,"analogous to an alpha-l peak of lipo­protein; three pseudoglobulin peaks,analogous to alpha-2, beta-I, andbeta-2 fractions and called "C," "D,"and "E"; and a euglobulin or gammacomplex of peaks labeled as the "F"

'Reference to trade names or commercial firmsdoes not imply endorsement by the NationalMarine Fisheries Service, NOAA.

were discontinued, titers to lipo­polysaccharide antigens declined whiletiters to protein antigens continued torise. Freund's complete adjuvant wasused in the first two injections of allpreparations.

A second series of weekly injections,commencing on experimental day 98,was shown to have an inductive effecton specific titers of all antigens but notto the extent of a true anamnestic re­sponse. Maximum serum passiveagglutinin titers of 1:65,531 were ob­tained from a protein-based antigen(BAE) at 106 days. Comparativelypoor responses were obtained from thelipopolysaccharide preparations. TheEWE and CWE antigens were ex­tremely toxic to rabbits due to theirlipopolysaccharide endotoxin compo­nents but were found to have no toxiceffect upon trout at the levels adminis­tered.

Rainbow trout were also injectedwith various particulate antigen prep­arations of ERM including an alum pre­cipitate (AP), washed cell wall suspen­sion (CW), and tFormalin-killed (FK)and phenol-killed (PK) whole cells.Humoral passive agglutinins to theseparticulate antigens were detected at 2 Ito 28 days into the first series of weeklyinjections as shown in Figure 2 I. TheCW preparation induced the highestpassive agglutinin response with amaximum titer of 1:262,144 in 84 days.The FK and AP antigens gave much thesame response. The PK preparation in­duced a poor serum passive agglutininresponse while a heat-killed prepara­tion failed to elicit any type of a passiveagglutinin response at all. The highpassive agglutination titers induced byvarious particulate antigens seemed toreach a common maximum titer ofI:4,096 following discontinuation ofinjections. From this point, the varioustiters declined similarly at an averagerate of 9.8 percent per month and weredetectable for more than 9 months.

Quantitative analysis of seriai elec­trophoregrams from rainbow trout

50 Marine Fisheries Review

agglutinin actIVIty was found in onlythe macroglobulin fraction of thehyperimmune trout serum as shown inFigure 25. No shifts in specific immuneactivity among effective molecularweight globulin fractions were found tooccur in the salmonid sera during ex­tended periods of immunization. Someconsistent but seemingly nonspecificactivity was also found in the prealbu­min fraction of the trout serum.

SUMMARY

The ERM bacterium has been shownto be a highly infectious and economi­cally significant pathogen of salmonidfishes, particularly under the stresses ofintensive culture. Due to its present lim­ited but expanding geographical dis­tribution, it is of primary importancethat preventative measures be taken tolimit the further dissemination of thedisease. Such preventative measuresand treatment should take into accountthe presented findings concerning theepidemiology and pathogenesis of the

Figure 25.-Gel filtration chromatographicfractionation and passive agglutinin activitydetermination of hyperimmune rainbowtrout, Sa/ma gairdneri, serum proteins.

disease and, in particular, the asymp­tomatic carrier state of infection. Pro­grams of inspection, certification, andsurveillance, based upon presumptiveserological screening proceduresadapted to large numbers of individualsmall volume sera with minimum lethalsampling, give promising results fordisease diagnosis, surveillance, and de­tection of clinically inapparent carrierinfections.

LITERATURE CITEDAraji, A. A. 1972. An economic analysis of the

Idaho rainbow trout industry. AE Ser. 118,Dep. Agric. Econ., Coil. Agric., Univ. Idaho,Moscow. 9 p.

Cisar, J. O. 1972. Properties of anti-Aeromanassa/manicida antibodies from juvenile cohosalmon (Oncharhynchus kisulch). Ph.D.thesis, Oreg. State Univ., Corvallis. 136 p.

Evelyn, T. P. T. 1971. The agglutinin response insockeye salmon vaccinated intraperitoneallywith a heat-killed preparation of the bacteriumresponsible for salmonid kidney diesase. J.Wildl. Dis., 7:328-335.

American Fisheries Society, Fish Health Section.1975. Suggested procedures for the detectionand identification of certain infectious diseasesof fishes. U.S. Fish Wildl. Servo 60 p.

Klontz, G. W., and J. G. King. 1975. Aquacul­ture in Idaho and nationwide. Dep. Water Re­sour., State Idaho, Boise. 86 p.

McDaniel, D. W. 1971. Hagerman redmouth: anew look at an old fish problem. Am. Fish.Farmer and U.S. Trout News, 1:14.

Ross, A. J., R. R. Rucker, and W. H. Ewing.1966. Description of a bacterium associatedwith redmouth disease of rainbow trout (Sa/magairdneri). Can. J. Microbiol. 12:763-770.

Rucker, R. R. 1966. Redmouth disease of rain­bow trout (Sa/ma gairdneri). Bull. Off. Int.Epizoot. 65:825-830.

Wobeser, G. 1973. An outbreak of red mouthdisease in rainbow trout (Sa/ma gairdneri) inSaskatchewan. J. Fish. Res. Board Can.30:571-575.

March 1978

MFR Paper 1296. From Marine Fisheries Review, Vol. 40, No.3, March 1978.Copies of this paper, in limited numbers, are available from 0822, User Ser­vices Branch, Environmental Science Information Center, NOAA, Rockville,MD 20852. Copies of Marine Fisheries Review are available from the Superin­tendent of Documents, U.S. Government Printing Office, Washington, DC20402 for $1.10 each.

51