SEROLOGICAL DIFFERENTIATION OFPOPULATIONS OF SOCKEYE SALMON,Oncorhynchus nerka

Marine Biological Laboratory

LIBRARYOCT 2 -1958

WOODS HOLE, MASS,

SPECIAL SCIENTIFIC REPORT- FISHERIES No. 257

UNITED STATES DEPARTMENT OF THE INTERIOR

FISH AND WILDLIFE SERVICE

EXPLANATORY NOTE

The series embodies results of investigations, usually of restricted

scope, intended to aid or direct management or utilization practices and as

guides for administrative or legislative action. It is issued in limited quantities

for Official use of Federal, State or cooperating agencies and in processed formfor economy and to avoid delay in publication

.

United States Department of the Interior, Fred A. Seaton, SecretaryFish and Wildlife Service, Arnie J. Suomela, Commissioner

SEROLOGICAL DIFFERENTIATION OF POPJLATION OFSOCKEYE SALMON, ONCORHYNCHUS NERKA

By

G. J- RidgwayU. S. Fish and Wildlife Service

andJ. E. Cushing and G. L. Durall

University of California, Santa Barbara

Contribution No. 1 to research conducted withthe approval of the United States Section of theInternational North Pacific Fisheries Commission.The Commission, established in 1953 by the Inter-national Convention for the High Seas Fisheries ofthe North Pacific Ocean, coordinates the researchof the member nations, Japan, Canada, and theUnited States. The resulting investigations providedata to the Commission for use in carrying out itsduties in connection with fishery conservationproblems in the North Pacific Ocean. Publication ofthis scientific report has been approved by theUnited States Section of the Commission.

Special Scientific Report --Fisheries No. 257

Washington, D. C.

August 1958

ABSTRACT

As part of a study applying serological methods to

the distinction of salmon races , the normal serums ofhuman beings, cattle, horses, sheep, pigs, goats, brownbullheads, and other salmon were tested for agglutinativeactivity with the erythrocytes of sockeye salmon. Onlyhorse serum and pig serum reacted strongly with the salmoncells. With some of the pig serums tested, consistentdifferences were found in the strength of reaction of the

cells of individual sockeye salmon. The frequencies ofindividuals displaying the various strengths of reactionwere found to differ significantly between areas of originof the salmon tested. When techniques of population ge-netics were used, evidence for the possible genetic con-trol of the observed differences was obtained.

CONTENTS

Page

Material and Methods 1

Experimental Results and Discussion 2

Acknowledgements h

Summary h

Literature Cited 5

Addendum 5

SEROLOGICAL DIFFERENTIATION OF POPULATIONS OF SOCKEYE SALMON,

ONCORHYNCHUS NERKA

This is a report on the results of

a collaborative investigation^/ of the

differentiation of sockeye salmon ( Oncor-hynchus nerka Walbaum) populations byserological methods. The aim of the work

was to estimate the potential value of

agglutinins found in the nonimmune serums

of animals (so-called "natural antibodies")

for detecting racial variations amongsalmon erythrocyte antigens. Such aggluti-nins have been used successfully in researchon the erythrocyte entigens of man, cattle,

sheep, chickens, and other warm-bloodedanimals. (Cf. Cushing and Campbell 1957for a general account of these reactions).They have also been used to demonstrateindividual variations in the erythrocyteantigens of skipjack Katsuwonis pelamisLinnaeus (Cushing, 1956) and brown bull-heads, Ictalurus n. nebulosus Le Sueur(Cushing and Durall 1957 ) , and otherfishes, Suyehiro (19^9)-

Variations in erythrocyte antigensprovide favorable material for racialanalyses, since whenever studied theirgenetic determination has been found to

be simple and direct. Also their speci-ficity has not been found to be subjectto environmental influences. Evidencefor the genetic determination of suchvariations in fish has recently been pro-vided by Hildeman (195b) , and the po-tential utilization of these antigens in

fishery research has been discussed byCushing and Sprague (1953 ), and byRidgway (1957)-

1/ This collaboration was made possiblethrough agreements reached by the U. S.

Fish and Wildlife Service, the Regentsof the University of California, and

the Office of Naval Research. The re-

search was supported in part by a con-tract with the Office of Naval Research.

MATERIALS AND METHODS

Samples of whole salmon blood were

obtained by field crews from the majorsalmon-producing rivers of North America

and were either shipped immediately by

commercial air transport or brought

directly to the laboratory. These samples

were collected by bleeding salmon from

the tail into open-mouth screw-cap bottleswithout preservative or anticoagulant.

The largest collections were made in the

rivers of Bristol Bay and at various

localities in the Columbia River System.

The clotted bloods (approximately 5 to 50

ml.) were transported in bottles in insu-

lated picnic jugs containing ice. They

were usually received from 1 to 3 days

after collection and were kept refriger-

ated at 5 to 10°C. until used. Under these

conditions most of the samples remained

useful for a week or two.

Suspensions (1.5 percent) of fresh

cells were prepared on the day of use by

washing 1-5 ml. aliquots of whole blood

in three 20-ml. changes of Alsever'ssolution. All washings were carried out

in a refrigerated centrifuge , and the cell

suspensions were kept refrigerated except

when in immediate use. Individual tests

were performed by placing two drops of

suspension with two drops of suitable

serum dilution in 10 x 70 ml. test tubes.

After 15 minutes ' incubation at room

temperature, the tubes were centrifuged

for 1 minute. The sedimented cells were

resuspended by lightly tapping the tube

and the degree of agglutination was scored

in the conventional manner (k indicating

very strong, 3 strong, 2 moderate, 1 light

and + barely detectable agglutination with

indicating no agglutination).

The influence of variables such as

the speed of centrifuging, age of cells,

temperature, etc., were considered and

controlled. Efforts were made to control

the objectivity and accuracy of the method

by testing the same sample on different

days and making the readings without

knowledge of the sample sources. Readings

were also made on the same material by

independent observers. Such controls sub-

stantiated the comparisons of different

tests, provided a basis for evaluating the

limits of experimental error in these tests,

and suggested refinements in technique for

control of such error.

EXPERIMENTAL RESULTS AND DISCUSSION

Several serums from humans^/, cattle

,

horses, sheep, pigs, goats and fish, all

preserved by freezing, were titrated

against sockeye cells. Included were a

number of cross matches (approximately 700)

of cells and serums (1 in k) between sock-

eye samples from various localities. All

of the latter gave negative results, an

observation that is in contrast to those

that have been made on the brown bullhead,

the white croaker ( Genyonemus lineatus,

Ayres) and skipjack (Cushing and Durall

1957)- Some brown bullhead serums (o out

of 16) were found to react strongly with

sockeye cells at 1 in k dilution. However,

these serums did not differentiate among

sockeye from different areas. There also

appeared to be no relation between the

positive reactions and the blood group of

the brown bullhead involved.

It is of interest that bovine serum

which proved so useful in detecting indi-

vidual differences in chickens and in skip-

jack reacted very poorly with sockeye cells.

Pig and horse serums were found to

agglutinate sockeye cells more strongly

than the other serums examined. As ti-

trations of several single pig serums with

the cells of different fish showed differ-ences in the strength of reactions withthese cells, attention was concentrated

on this observation. While the optimum

temperature of agglutination for this

serum was approximately four degrees, it

was practical to work only at room temper-

atures which consistently remained at 22°

to 2U°C.

2/ Human typing serums kindly supplied by

Hyland Laboratories, Los Angeles,California.

A standardized procedure was adopted

for the comparison of readings of differentsets of samples. This procedure consistedof testing the cells of each fish againstfour dilutions (1:16, 1:32, l-.bk and 1:128)of heat- inactivated (56*C for 3° minutes)serum from a single pig designated W.P. 4.

The scores of the readings at the latterthree dilutions were added to give a singletotal score. Table 1 presents an exampleof the protocol just described. Figure 1presents a summary of scores from many suchprotocols, which combines the samples intothree groups. The first of these consistsof samples taken from the rivers of BristolBay, the second of samples taken from riversof the eastern Pacific coast from KodiakIsland to the Fraser River, the third ofsamples taken in the Columbia River System.An analysis of variance was performed(Table 2), which shews that the threegroups differ significantly. This analysiswas performed by the method given inSnedecor, 1957, page 283, which takes intoaccount the differences in variances. Thisis the first time that it has been possibleto demonstrate a statistically significantserological difference between populationsof individuals within a single species offish.

It should be borne in mind that the

salmon in each of these areas are homingtoward a variety of separate and distinctspawning grounds. In spite of this it

seems reasonable that genetic exchangesmay be more frequent within the Columbiaand Bristol Bay groups than between them,and that this would plausibly account forthe differences observed.

Consideration of all the individualscores of the above groups suggests thatthese form a trimodal distribution (Figure1). Some statistical justification forthe validity of this suggestion was ob-tained by applying a test for bitan-gentiality (Haldane 1952).

Additional evidence of antigenicheterogeneity was obtained from absorptionexperiments. Samples of the serums of

several pigs were absorbed with strong andweakly reactive cells. In general, twocapacities for antibody removal were ob-served. Several absorptions indicatedthat weaker reacting calls (Bristol Bay)were less effective in removing antibodiesthan stronger reacting cells (Columbia River).

One exceptional serum was discovered.

This serum appeared to have antibodies that

were relatively more amenable to fraction-ation. Absorption of this serum with the

cells of a Bristol Bay fish left consider-able antibody for the cells of a ColumbiaRiver fish while absorption with the cells

of the Columbia River fish removed all the

antibody for both types. The small amountof this serum that was available preventeda systematic expansion of this observation.However, the occurrence of this serum showsthat a study of the serums of other pigsmay prove to be profitable.

The discovery that the erythrocytesof salmon from different geographic areasvary significantly with respect to theirreactions with antibodies in pig serumraises the question of the nature of themechanism controlling this variation.While the answer to this question awaitsfurther study, several sources of evidenceindicate that a genetic mechanism probablyexists. First, the erythrocyte antigensof mammals and birds have been found to begenetically determined wherever investi-gated. Second, Hildeman (1956) has shownthat certain erythrocyte antigens of gold-fish have hereditary determinants. Third,different species of salmon, (Ridgway andKLontz, 1957) as well as other species offish, (Suyehiro, 19^-9 > Cushing and Sprague,

1953); show characteristic species speci-ficities with respect to their erythrocyteantigens. This suggests that the antigenicspecificities involved are geneticallydetermined.

An analysis of the frequency distri-butions of the antigenic variations insockeye salmon populations was made withthe Hardy-Weinberg formula (Srb and Owen,

1953) in an attempt to obtain furtherinformation concerning the possible ge-netic background for these variations.The Hardy-Weinberg formula relates thefrequencies of allelic genotypes to thefrequencies of allelic genes in a randomlymating population, in terms of the binomialexpansion. It was through the use of thisformula that Bernstein determined theexact method of inheritance of the humanABO groups (see Race and Sanger, 1954) andit is an indispensable tool in the study ofthe population genetics of blood groups inman and animals.

Our interpretation of the data pre-

sented is based upon the, assumption thatthe observed variations in strength ofagglutination, expressed in terms of high,intermediate and low scores, are due to

two primary causes.

The first is that the active aggluti-nin in serum WP4 has a lower affinity fora qualitatively different but related anti-gen on the surface of the lower scoringcells than for the antigen on the higherscoring cells. Since our absorption re-sults indicate that these antigens must beserologically related, because they reactwith the same agglutinin, we assume thatthey are determined by allelic genes.Differences in the strength of agglutinationbrought about by the reaction of one typeof antibody with different antigens de-termined by allelic genes are well document-ed in the literature on blood groups (Race

and Sanger, 1954).

The second primary cause of variationmust be postulated to account for the inter-mediate reactions. This is assumed to bea dosage effect causing a stronger reactionwith cells of individuals homozygous for a

particular gene than those of individualsheterozygous for this gene. Dosage effectsexhibited by blood grouping reagents arealso well documented in the literature; in

fact, the scoring system used by Race,Sanger and Lehane (1953) in their study of

the dosage effect in the human Duffy bloodgroups serves as a model for our scoringsystem.

The apparent trimodality of thedistribution of agglutination scores(Figure 1) served as a basis for assigningindividual fish to one of three categories,rated as strong, intermediate and weakreactors. These categories, as explainedabove, may be postulated to correspond to

three genotypes determined by a singlegene and its allelic alternates. Strongreactors are considered as homozygous forthis gene (symbolized AA), intermediatereactors as heterozygous (Aa) and weakreactors as lacking this gene (aa). Thenumbers of fish occurring in each categoryare compared with those expected on thebasis of the Hardy-Weinberg formula in

Table 3.

Two groups of fish drawn from spawn-ing populations were also studied. Thesewere taken from the Okanogan River and the

Little Wenatchee River which are the

major spawning areas for Columbia Riversockeye salmon. Since these two popu-lations were examined at a later date andunder somewhat different conditions, it

is necessary to assign the individual fishto the three categories on the basis of the

modes present in the population, ratherthan according to standard score values.For this reason, the present analysis mustonly be considered as a guide toward furtherstudies under more refined conditions.

It is of interest that the Mendelianpopulations taken from spawning groundsgave data that appear to fit the Hardy-Weinberg equation quite well, while twoout of three populations that were knownto be mixtures of breeding stocks did not,nor did a composite of all populations(Table 3)- In the two cases where thenumber of fish was sufficient in all cate-gories for the application of the ChiSquare test these conclusions appear to bestatistically sound. (Okanogan River X =

0.106 d.f. = 2, "composite population"X = 61. 9 d.f. = 2).

Although the close agreement of theactual Bristol Bay data with the Hardy-.Weinberg Law may possibly be due to chancealone, the results may tentatively beascribed to the several populations in thissystem being genetically homogeneous withrespect to the relative frequencies of "A"and "a". While the reasons for and impli-cations of this homogeneity remain to bedetermined, three not mutually exclusivepossibilities suggest themselves. Thefirst is that the various "parent-stream"stocks, while genetically isolated, havenot diverged with respect to the frequen-cies of these genes as they obtained in acommon ancestral stock. The second is

that the various "parent-stream" stocksare not sufficiently isolated to allow fordivergence of the frequencies of the twoalleles among them. The third is thatidentical selection pressure in theseparate populations may impose identicalpoints of genetic equilibrium.

The failure of the Columbia Riverand Pacific Coast samples to conform tothe Hardy -Weinberg formula can be ex-plained by the fact that each representsa mixture of populations differing withrespect to the frequencies of A and a.

This explanation receives support from

consideration of the good fit of samplesobtained on the spawning grounds of theupper Columbia after the Wenatchee andOkanogan runs had separated from eachother. In these cases, where fairly largesamples were taken from known Mendelianpopulations, the distribution of scoresdoes conform to the Hardy-Weinberg law.

In view of these various considerationsthere seems to be good reason to postulatethat the variations in erythrocyte antigenswhich we are comparing are geneticallycontrolled by a simple allelic system ofgenes, the frequencies of which may varyin racial stocks of different geographicorigins

.

ACKNOWLEDGEMENTS

The authors wish to express theirsincere appreciation to the faculty of the

Department of Microbiology, University ofWashington for making laboratory spaceavailable for this investigation; to the

many fishery biologists who assisted withthe sample collections; to the BiometricsUnit, Pacific Salmon Investigations forperforming the analysis of variance; to

Dr. Lucian Sprague for several of the

animal serums; and to Mr. G. W. Klontz fortechnical assistance with some of the

experiments.

SUMMARY

1. In over 700 cross matches between the

cells and serums of samples fromseveral localities, no evidence wasobtained for the existence of naturalisoagglutinins in sockeye salmon.

2. The normal serums of several humans,cattle, horses, sheep, pigs, goatsand brown bullheads were tested foragglutinative activity with the

erythrocytes of sockeye salmon. Onlyhorse serum and pig serum reactedstrongly with the salmon cells.

3. Consistent differences were found in

the strength of reaction of some pigserums with the cells of individualsockeye salmon.

The frequencies of the various strengths

of reaction, which were expressed as

scores, were found to differ significantly

between areas of origin of the salmon

tested.

k. An hypothesis to explain the possiblegenetic determination of these differ-ences was presented and tested withthe Hardy-Weinberg law.

LITERATURE CITED

RIDGWAY, G. J. , AND G. W. KLONTZ1957- Manuscript in preparation.

SNEDECOR, G. W.

1956. Statistical methods. 5th Ed.The Iowa State College Press,.

Ames, Iowa.

SRB, A. M., AND R. D. OWEN1953- General genetics. W. H. Free-

man & Co., San Francisco.

CUSRTNG, J. E.

195b. Observations on serology of

tuna. U.S. Fish and WildlifeService, Special ScientificReport—Fisheries No. I83.

14 pp.

CUSHTNG, J. E., AND D. H. CAMPBELL1957- Principles of immunology.

McGraw-Hill, New York.

CUSHTNG, J. E. , AND G. L. DURALL1957- Isoagglutination in fish. Amer.

Nat.. 91 : pp. 121-126.

CUSHTNG, J. E. , AND L. SPRAGUE1953- Agglutinations of the erythro-

cytes of various fishes byhuman and other sera. Amer.Nat. 87: pp. 307-315-

HALDANE, J. B. S.

1952. Simple tests of bimodality andbitangentiality. Ann.Eugenics 16, pp. 359-364.

HILDEMAN, W. H.

1956. Goldfish erythrocyte antigensand serology. Science 124,pp. 315-316'.

RACE, R. R. , AND R. SANGER1954. Blood groups in man. 2nd Ed.

CC. Thomas, Springfield, 111.

Publ.

SUYEHIRO, YASUOI949. On the agglutination of the

bloods of fishes. Bull.

Physiograph. Sci. Res. Inst.

Tokyo Univ. 2, pp. 42-50. (in

Japanese).

ADDENDUM

Since the preparation of this manu-script we have tested somewhat largersamples from Bristol Bay and the ColumbiaRiver, taken in the 1957 season, and haverepeated the distinction with considerableprecision. In 1956, 86 percent of BristolBay sockeye salmon tested had scores lessthan 3-5 while 63 percent of those fromthe Columbia River had scores greater than

3.5. In 1957, 92 percent of the BristolBay fish tested had scores less than 3-5while 79 percent of those from the ColumbiaRiver had scores greater than 3- 5- Thisyear to year stability provides furtherevidence for the genetic control of thecharacters determined by this method.

RACE, R. R. , R. SANGER, AND D. DELANE1953- Quantitative aspects of the

blood-group antigen fya. Ann.

Eugenics 17, pp. 255-2o6.

RIDGWAY, G. J.

1957> The use of immunological tech-niques in racial studies. U.S.Fish & Wildlife Service, SpecialScientific Report—FisheriesNo. 208, pp. 39-^3-

h £

vH

•H«aS

oo

a4)

opCO

m

&

ON

COH

H

VOH

ITS

COH

CM

CM

co no H

ON

CO

t—

VO

ITS

ro

(M

4"

-*

-*

oo cm

o

o

CO CM +1

m cm +1

ro cm +|

oo cm +1

ro h O

J" ro ro H

j- ro CM H

j- m h o

CM

CM

-4" CM

PO rH

J" CO

H O

H O

CO CM +1 O

H O

+ 1 O

H O

ro h +1 o

CO CM +1 O

CO CM +1 O

PO CM +1 O

ro cm +1 o

CM CM +1 O

c

Table 2. --Frequencies of Agglutination Scores ofSockeye Salmon

io

r?1)

g•H

"g

J!

O

g•H

ft

2

0)

H

00

a>Hoa<b

I

a)

a -ft

s

1

3

T3

P -"

o

6

0) fta cj

Pi

tt'

T3

H OJod £> <m XJp S o oo SOS ri

do•HPa)

ftoft

ONCM

r—

d

t—

CVI

cj

CO

CO

ON

ON

r—ITS

J-CJ

CO

OJ

t-

ON

no

ONCJ

CJ

COUN

o

10

MBL WH01 Library - Serials