Intraperitoneal Injection of Lymphomatous Nerve Tissue into Resistant or Susceptible Chickens

Intraperitoneal Injection of Lymphomatous Nerve Tissue into Resistant or Susceptible Chickens

Intraperitoneal Injection of Lymphomatous Nerve Tissue into Resistant or Susceptible Chickens K. B. DEOME Division of Veterinary Science, University o...

938KB Sizes 0 Downloads 0 Views

Intraperitoneal Injection of Lymphomatous Nerve Tissue into Resistant or Susceptible Chickens K. B. DEOME Division of Veterinary Science, University of California Berkeley, California

INTRODUCTION HE successful selection of two strains of birds differing in their resistance to spontaneous lymphomatosis under the conditions of the parent flock (Hutt, Cole, and Bruckner, 1941, and Taylor et al., 1943), raised the question of whether or not they would retain their relative resistance when exposed to conditions that would produce a large increase in the incidence of the disease. To answer this question it would be necessary to increase the incidence within the parent flock without introducing birds or inocula from other flocks against which the strains had not been selected. This could be accomplished if the incidence within the parent flock could be increased by injecting them with lymphomatous material from affected members of the same flock. The experiments reported in this paper were designed to answer two questions: (1) could the incidence of lymphomatosis among the susceptible and resistant birds be increased by injecting them with lymphomatous material from other members of the same flock; (2) would the two strains retain their relative resistance with respect to one another when the incidence of the disease had been increased in both of them? Data are presented in such a manner as to permit simultaneous evaluation of several levels of exposure and susceptibility.

T

EXPERIMENTAL PROCEDURE

The birds used in these experiments were Single Comb White Leghorns from the flock maintained by the Division of Poultry Husbandry, University of California, Berkeley, where, since 1933, a program, of selection for and against spontaneous lymphomatosis has been carried out. Two strains of birds have been produced: one relatively susceptible to the spontaneous disease; the other relatively resistant to it. They will be referred to as susceptible and resistant, respectively. A description of the breeding program, the management, and the yearly incidence of lymphomatosis are reported in a previous paper (Taylor et al., 1943). The data presented in this paper were obtained from 18 groups of birds, consisting of three susceptible and three resistant groups started each year during the years 1938 to 1940, inclusive. These groups were arranged as follows in each year of the experiment: a susceptible and a resistant control flock maintained at Poultry Husbandry; a susceptible and a resistant control group kept at Veterinary Science; and a susceptible and a resistant inoculated group kept at Veterinary Science. The distribution of the 18 groups, according to year, flock, and experimental procedure, is presented in Table 1. The Veterinary Science birds were full sisters and brothers

[381]

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

(Received for publication April 19, 1943)

382

K.

B.

DEOME

TABLE 1.—The corrected data arranged to show the effect of three levels of exposure and two levels of resistance on the incidence of lymphomatosis

resistance

Year Flock

Resistant Resistant Resistant

1938 1938 1938

Susceptible 1938 Susceptible 1938 Susceptible 1938 1939 1939 1939

Susceptible 1939 Susceptible 1939 Susceptible 1939 Resistant Resistant Resistant

1940 1940 1940

Susceptible 1940 Susceptible 1940 Susceptible 1940

i I: If

Lymphomatous deaths Total

Visceralf

Neuralf

Mean ChiNo. Per- age a t squared cent death*

Per- M e a n N o . cent age at death*

Ocularf

Mean PerNo- cent age at No. death*

Percent

Mean age a t death*

Cont. Cont. Inoc.

616 65 65

120 22 45

19.5 33.8 69.2

15.5 13.2 15.3

1

7.25 16.29

21 11 14

3.4 16.9 21.5

10.3 13.2 12.7

72 10 29

11.7 15.4 44.6

16.9 10.7 17.1

84 2 9

13.6 3.1 13.8

17.7 23.0 16.0

Cont. Cont. Inoc.

277 57 58

96 34 48

34.7 59.6 82.8

11.8 11.2 9.4

1 12.38 ) 7.40

43 29 40

15.5 50.9 69.0

9.5 11.6 9.0

49 14 16

17.7 24.6 27.6

12.6 15.7 11.2

24 3 6

8.7 5.3 10.3

13.6 8.0 8.0

Cont. Cont. Inoc.

488 56 69

54 5 43

11.1 9.8 62.3

16.2 10.4 12.0

1 0.29 j 37.23

8 3 15

1.6 5.3 21.7

14.8 14.0 9.6

26 2 27

5.3 3.6 39.1

17.1 5.0 8.2

21 0 10

4.3 0.0 14.5

23.6

Cont. Cont. Inoc.

231 121 99

84 80 87

36.4 66.1 89.9

13.2 11.8 5.7

j 28.18 J 13.98

34 52 61

14.7 43.0 61.6

10.6 10.7 5.5

25 33 45

10.8 27.3 45.5

12.7 11.3 6.5

38 16 0

16.5. 23.2 0.0

15.0 11.9

Cont. Cont. Inoc.

440 70 63

73 20 44

16.6 26.6 69.8

15.2 14.8 9.2

) 5.74 j 22.67

29 7 14

6.6 10.0 22.2

15.8 20.0 5.9

38 13 34

8.6 18.6 54.0

17.1 11.9 10.0

22 3 2

5.0 4.9 3.2

13.9 17.0 14.0

Cont. Cont. Inoc.

207 59 51

80 31 38

38.6 52.5 74.5

11.8 6.7 8.1

1 3.67 j 5.62

31 17 20

15.0 28.8 39.2

7.6 7.3 8.0

33 18 24

15.9 30.5 47.1

13.9 5.2 8.1

21 0 1

10.1 0.0 2.0

13.1

15.9

5.0

* Months after hatching. t Birds showing more than one type of lesion were included in more than one group. % Poultry Husbandry control. § Veterinary Science control. II Veterinary Science inoculated.

of the birds in the respective control groups at Poultry Husbandry. The birds kept at Veterinary Science were hatched on the premises from eggs obtained from the Poultry Husbandry flock. A separate incubator was provided for each strain. At hatching time the chicks of each strain were separated into inoculated and control groups in such a way that the progeny of any dam were about equally distributed among them. After segregation the groups were brooded and reared separately in units of 25 or less. The equipment used and the sanitary program followed were the same as those employed for the isolation of birds having diseases known to be infectious. None of the birds kept at Veterinary Science were killed until moribund or 24 months of age. The susceptible and resistant control groups at Poultry Husbandry were hatched, brooded, • and reared together. Material for inoculation was always taken from Poultry Husbandry birds to

avoid the possibility of introducing material against which the two strains had not been selected. The choice of donor birds was further limited to those showing only lesions of neural lymphomatosis. Lymphomatous nerves from the donor birds were cut to very small pieces with shears, ground with a pestle, and diluted to five times their original volume with physiological saline. The resulting inoculum contained great numbers of tissue fragments. In each case the inoculum was used immediately. Birds from the susceptible and resistant strains, previously chosen for injection, were given 0.S to 1.0 c.c. of the inoculum intraperitoneally during the first 7 days after hatching. A 16-gauge needle was used so that large tissue fragments could be introduced. The number of chicks from the susceptible and resistant groups injected with each inoculum was approximately equal so that the greater potency of a particular inoculum would not influence the results.

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

Resistant Resistant Resistant

1 t I

No. living 1 month Exp. or Proc. more

INTRAPERITONEAL INJECTION OF LYMPHOMATOUS NERVE TISSUE

RESULTS

Before discussing the results in detail a general view of the data should be obtained. When comparing the incidences of lymphomatosis in two populations, special care must be taken to ensure the validity of such a comparison. Among the possible sources of error, two are of special significance. The first of these is produced by the absence of a negative control which makes it necessary to compare the incidences in two experimental populations. The difference in percentage incidence may not appear large and yet the actual difference may be highly significant. Such differences must be evaluated by appropriate statistical methods. The second and greatest source of error is inherent in the age distribution of the lymphomatous and non-lymphomatous

deaths in the populations being compared. For example, the mean age at death from lymphomatosis in the Veterinary Science susceptible group during 1939 was 5.7 months, whereas that of the Poultry Husbandry control group was 13.2 months. If a considerable portion of the birds in both groups had died of other causes between the ages of 5.7 and 13.2 months it is obvious that the percentage of birds dead of lymphomatosis would have been reduced much more in the latter group than in the former. Differences of this type are present in all populations and their effects are rarely appreciated. The error due to death from other causes is easily avoided if the data are rearranged to show the number of birds in each population that would have died of lymphomatosis if none had died of other causes. This may be done by determining from the raw data the rate (percentage) of lymphomatous deaths in each three-month age period and applying this rate to the original population, including those which died of other causes. The use of a three-month age period is not mandatory, but was used here because it was found convenient. The corrected data will show a higher percentage of lymphomatous deaths than the raw data but the error due to deaths from other causes will be eliminated. Two additional differences between Poultry Husbandry and Veterinary Science birds should be recorded. All of the latter which had not died earlier were killed at 24 months of age, whereas many of the former lived beyond that age. To minimize this error, all Poultry Husbandry control birds living more than 24 months were eliminated from the data. In addition, no males were included in the Poultry Husbandry groups, whereas both males and females were kept at Veterinary Science. Since the sex ratio among birds dying of lymphomatosis at Veterinary Science

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

The 1938 and 1939 inoculated chicks received only one injection; whereas the 1940 birds were repeatedly inoculated, and the number of injections per bird depended upon the date of hatch. The first group hatched in 1940 received 13 inoculations over a period of 10 months, whereas the last group hatched received only S over a period of 3 months. Autopsies and diagnoses were done by the writer without knowledge of the particular group to which the specimen belonged so that bias in favor of any particular group would be eliminated. All obvious cases of lymphomatosis were diagnosed at autopsy and the questionable ones after microscopic preparations of the tissues had been studied. No cases were included about which there was any reasonable doubt. Total and differential counts of the blood cells were made twice each month on all birds at Veterinary Science during 1938 but since no leukemic birds were found, blood examinations were not made in 1939 and 1940.

383

384

K. B. DEOME

was approximately 1:1 this factor produced no known error. The number of birds living one month or more was selected as the base number upon which the total percentage of lymphomatous deaths was calculated, since no birds died of the disease during the first month of life. The corrected complete data are presented in Table 1, and summarized for the three years in Table 2.

the disease, and therefore the level of the conditions producing it, increased in the following order: the levels which prevailed among the Poultry Husbandry control groups, among the Veterinary Science control groups, and among the Veterinary Science inoculated groups. Since these levels represented the sum of the forces sufficient to produce increasing incidences of lymphomatosis in birds of the same susceptibility,

Distribution of all cases of lymphomatosis according to t h e types of lesions foundf

Total Levels of resistance

Levels of exposure Year Flock

Resistant Resistant Resistant

1938-1940 P . H . J 1938-1940 V.S.5 1938-1940 V.S.J

Susceptible 1938-1940 P . H . susceptible 1938-1940 v.s. Susceptible 1938-1940 v.s.

Exp. Proc.

No. living 1 month or more N o .

Percent

Mean age No. at death*

Percent

Ocular

Visceral

Neural Mean age No. at death*

Percent

Mean age No. at death*

Percent

Mean age at death*

Cont. 1,544 Cont. 191 Inoc. 197

247 16.0 47 24.6 132 67.0

15.6 12.8 12.2

58 23.5 21 44.7 43 3 2 . 6

13.6 15.7 9.4

136 5 5 . l ' 25 53.2 90 68.2

17.0 9.2 11.8

127 51.4 5 10.6 21 15.9

15.8 20.0 17.9

Cont. Cont. Inoc.

260 36.4 145 61.2 173 83.2

12.3 9.9 7.7

108 41.5 98 67.6 121 6 9 . 9

9.2 9.9 7.6

107 41.2 65 44.8 85 4 9 . 1

13.1 10.7 8.7

83 31.9 19 13.1 7 4.1

13.9* 9.9 6.5

715 237 208

* Months after hatching. t Birds showing more than one type of lesion were included in more than one group. J Poultry Husbandry control. § Veterinary Science control. |] Veterinary Science inoculated.

The data show that the incidence of lymphomatosis in the Veterinary Science control groups was higher than among the Poultry Husbandry control birds, and that the incidence in the Veterinary Science inoculated groups was higher than in either of the former. The order of incidence in each strain remained the same during each of the three years with the exception of the 1939 Veterinary Science resistant controls. Since in each year three groups of equally resistant birds from each strain showed different incidences of lymphomatosis when exposed to the three sets of conditions, it follows that the incidences within the groups would be measures of the relative lymphomatosis-producing power of the conditions to which the groups were exposed. Within each strain the incidence of

they will be referred to as levels of exposure and designated PHC, VSC, and VSI, respectively. Therefore, at a given level of susceptibility, the incidence of lymphomatosis increased with the level of exposure. The effect of inoculating lymphomatous tissue was very marked in both the susceptible and resistant strains. Significant increases in the incidence of lymphomatosis were produced each year. If the 1940 inoculated birds which were injected repeatedly are compared with the singly injected 1938 and 1939 groups no significant differences appear, indicating that repeated injections do not increase the incidence beyond that produced by a single injection. The influence of varying levels of susceptibility upon the total lymphomatous response can be estimated from the data in Tables 1 and 2. A comparison of the in-

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

TABI.E 2.—Corrected data combined for the years 1938, 1939, and 1940, showing the distribution of all cases of lymphomatosis according to the type of lesions found

INTRAPERITONEAL INJECTION OF LYMPHOMATOUS NERVE TISSUE

the combined data in Table 2. These data are expressed as percentages and are not usable for this purpose in that form. However by the use of the angular transformation (% = sin2 6) values can be obtained which are suitable for an analysis of variance. The effect of the levels of susceptibility within the levels of exposure and conversely the effect of the levels of exposure within the levels of susceptibility may be estimated by calculating the percentage of reduction in the mean square resulting from the removal of each of these sources of variance. The data, presented in Figure 1, show

BETWEEN LEVELS OF SUSCEPTIBILITY (df s i , sq = 354) rBETWEEN LEVELS OF EXPOSURE (df = 2, sq = 466)

TOTALdf = 5 Iq = 241 L

WlTHIN LEVELS OF SUSCEPTIBILITY (df = 4 , Sq » 241)

%ITHHT LEVELS OF EXPOSURE (df = 2, Stf s 16)

•BETWEEN LEVELS OF EXPOSURE (df = 2 , S ? = 466)

TOTAL — df - 5

rBETWEEN LEVELS OF SUSCEPTIBILITY (df = 1 , S3 - 35*)

s q =264 •WITHIN LEVELS OF EXPOSURE(df = 3 i * 3 - 129)

•WITHIN LEVELS OF SUSCEPTIBILITY (df = 2 , sq » 16) FIG. 1. Analysis of variance showing the reduction of the mean square due to each source of variance.

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

cidence of the disease among the susceptible and the resistant birds at any given level of exposure in any year will reveal consistently higher incidences among the susceptible birds. These differences are statistically significant except at the VSI level where the increased incidence in the 1938 group is of borderline significance and that in the 1940 group is insignificant. Reasons for the smaller differences at the VSI level will be presented later in this paper. The relative potency of the levels of susceptibility and the levels of exposure can be estimated by an analysis of variance of

385

386

K. B. DEOME

Resistant

Susceptible

FIG. 2. The percentage of lymphomatous deaths among the inoculated birds of each strain which could have been attributed to each level of exposure.

are subdivided to show the percentage of the lymphomatous deaths among the inoculated birds of each strain that could have been attributed to each level of exposure. The percentage of deaths attributable to the PHC and VSC levels were ap-

proximately twice as great in the susceptible as in the resistant groups, whereas the VSI level accounted for more than twice the proportion of lymphomatous deaths in the resistant as in the susceptible groups. It seems paradoxical that the number of lymphomatous deaths attributable to VSI should have been greater in the resistant than in the susceptible strain whereas the opposite was true of the VSC and PHC levels. The same type of situation, with minor yearly variations, can be demonstrated if the data for each year are plotted separately. An explanation of this situation can be found in the probable distribution of resistant and susceptible individuals in the two strains of birds used. It is unlikely that all of the individuals in either strain were equally susceptible to lymphomatosis. Available genetic information indicates that the situation is complex, and it may be safely assumed that many genetic factors influence the resistance of an individual to a given level of exposure. It follows that there are more individuals susceptible enough to become clinical cases at a given level of exposure in a susceptible than in a resistant population. In both populations there should be some individuals which approach 100 percent and others which approach 0 percent susceptibility while the vast majority is scattered in some fashion between these two limits. If we could test every individual, we could construct a frequency curve to show the exact distribution of birds in the two populations according to their degrees of resistance. Although no such tests are available, we can use the present data to construct a graphic representation of the probable situation. It must be understood that the exact shapes of frequency curves constructed from this sort of data are only as accurate as the assumptions upon which they are based. They will show only the relative distribution of in-

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

that the variance accounted for by differ241-16 ences in levels of exposure is '•— = 85 264 percent, whereas the variance attributable to differences in levels of susceptibility is 129-16 = 43 percent. Therefore, in the 264 present material differences in exposure between PHC, VSC, and VSI are of greater magnitude as a source of variance than differences in susceptibility between the susceptible and resistant strains. The amount of lymphomatosis attributable to any one level of exposure is the amount occurring at that level less its controls. These values for the VSI level were calculated from the combined data in Table 2 and presented as subdivided percentage bar diagrams in Figure 2. The bars

INTRAPERITONEAL INJECTION OF LYMPHOMATOUS NERVE TISSUE

dividuals in the two strains according to their degrees of resistance for the generations under consideration. It is obviously impossible to construct two different curves from the available data without assuming a particular shape for one of them. As a point of departure, a normal curve may be assumed to represent one of the strains. In Figure 3 the

enough to show increasing percentages of lymphomatous deaths, the areas under the curves must be divided to represent these portions. The dividing lines will be the levels of exposure. The combined data were obtained from Table 2 and the yearly data from Table 1. From the table of ordinates of the normal curve it is a simple matter to construct curves of the same area for the

I^Z&)}333ymo

OftDegree of resistance Susceptible

Resistaht V5I V5C ?HC

P e r c e n t a g e of t h e popu/dtioris s u s c e p t i b l e enough t o die of -lymphomatosis at the t h r e e levels of exposure Level of exposure

FIG. 3. Frequency curves showing the relative distribution of birds in the susceptible (solid line) and resistant (broken line) populations according to their degrees of resistance to lymphomatosis, and the percentage of each population which was susceptible enough to become clinical cases at the PHC, VSC, and VSI levels of exposure.

normal curves were assigned to the susceptible rather than the resistant populations, since, as is shown in Table 1, the percentage of lymphomatous deaths among the former were more uniform from year to year than among the latter. It must be clear that the area under the normal curve represents the whole susceptible population, whereas its shape represents the arbitrary distribution of the individuals in the population according to their degrees of resistance. As the actual populations of birds were divided by three levels of exposure into portions susceptible

resistant populations, using the same levels of exposure and the same base lines. The area under a curve from the extreme right to the line representing a given level of exposure will represent the portion of the whole population which was susceptible enough to die of lymphomatosis at that level of exposure. The frequency curves in Figure 3, constructed from the combined data, show a very different distribution of individual susceptibility in the resistant population than in the susceptible one. The curve for the resistant strain is skewed markedly to the left.

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

Combined D+U

387

388

K. B. DEOME

1338

Susceptible

FIG.

The seemingly paradoxical situation represented in the bar diagram, Figure 2, in which the percentage of lymphomatous deaths attributable to the VSI level of exposure was greater in the resistant than in the susceptible population, can be explained by the distribution of susceptible and resistant individuals within the populations. In two populations, such as those described by the frequency distributions in Figure 3, the portion of the populations susceptible enough to become clinical cases at the VSI level of exposure but not at the VSC level would be expected to be greater in the resistant than in the susceptible population while the opposite would be expected between the VSC and PHC levels, and below the PHC level. There were many individuals in both populations which were too resistant to become clinical cases at the VSI level of exposure, in fact 33.0 percent of the resistant and 16.8 percent of the susceptible group showed this order of resistance. It follows that a further significant increase in the incidence of lymphomatosis among the susceptible birds could be produced only by a

3. 1938.

very great increase in the level of exposure. There would be a very small chance of obtaining significant differences between the susceptible and resistant groups at levels above VSI. It is not surprising that the differences obtained, at the VSI level were not always statistically significant. Figure 3 presents similar curves constructed from the data for the individual years 1938, 1939, and 1940. An examination of the curves will show that the yearly variations in the lymphomatous response can be explained on the basis of varying levels of exposure and different relative distributions of individuals within the populations according to their degrees of resistance. In the 1939 generation, for example, the resistant group showed no greater response to the VSC than to the PHC level of exposure. In two populations such as those described by the frequency curves in Figure 3 for 1939, one would expect to find very few birds whose level of resistance would have permitted them to become clinical cases at the VSC but not at the PHC level. Similarly in populations of the type described for 1940 there would be a

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

Resistant

.0% Degree of resistance 3V.7 *0(%i Percentage of the populations _ susceptible enough S < * t o Ji« of J9.5 • lymphomatosis 33.8 • 4t the t h r e e _l._69.i • levels of j$Q exposure '•"'Level of exposure

INTRAPERITONEAL INJECTION OF LYMPHOMATOSIS NERVE TISSUE

389

l«9

.0% D e g r e e of resistance -J
-66.1

5usccptible MOX*RcsUtdnt 007.*-

FIG. 3. 1939.

small chance of obtaining a significant difference in the incidence of lymphomatosis between the susceptible and resistant populations at the VSI level of exposure. Actually no significant difference was found. If the distribution of individuals accord-

ing to their degrees of resistance in the susceptible and resistant populations was due to genetic factors, as seems probable, then the curves for each year should describe the distribution of genotypes within the two populations. These correspond to

mo / * • >

0% Degree of resistance Susceptible-

iOOtResistant • I00JL. 100%*-

-38.6+07* Percentage of -52.5 the pODtt/dttons •7^.5susceptible enough of •+&U-•to die . /6.6 + lymphomatosis 7 dt'the three 26.6—levels of —69.8 » exposure »0ZJ ~*"i*Uev*\ of exposure VSI V5CPHC FIG. 3. 1940

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

the populations Susceptible enough ^ t - t o rft< of « ll.l * lywiphomdtosis a t the t h r * e 9.8 levels of *—-I—r—6JU3 Jtl**?0'"™ >o%L evel of exposure V5J VSC PHC 8M

390

K. B. DEOME

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

the expected genotypic distributions if exposure above which a small increase will many factors determined the degree of re- produce a large response. This point is the sistance. It must be recalled that the de- threshold of resistance for a given populagree of skewness of the curves cannot be tion and indicates the degree of resistance predicted from the present data, but the of the individuals susceptible enough to beskewness of the curve for the resistant come clinical cases at that level. The population is correct in relation to that threshold for the susceptible and resistant of the susceptible group. groups can be estimated from the bar diaThe yearly variations in the distribution grams in Figure 2. It is considerably below of individuals according to their degrees of the PHC level of exposure in the former resistance were probably due to genetic group and above the PHC level in the latdifferences between the generations of birds. ter. The threshold concept has added sigThe greatest yearly variation would be ex- nificance when applied to the frequency pected in the resistant strain. An inspec- curves in Figure 3. In each curve the tion of the frequency distributions in Figure threshold is on the right side of the mean 3 will make this clear. In the resistant and below the point of inflection of the group, the level of exposure was low and curve. It is the possession of a high the distribution skewed to the left with the threshold that makes the resistant populamost resistant genotypes concentrated at tion valuable for practical purposes. the left end. The range of genotypes repreThere is a general relationship between senting sufficient resistance to prevent birds the mean age at death from lymphomatosis from becoming clinical cases at a low level and the level of exposure. The trend is best of exposure was so great that there was seen in the combined data presented in only a small chance of selecting a popula- Table 2 where the groups of birds from the tion possessing the most resistant geno- susceptible and resistant populations are types. In the susceptible group, on the arranged according to increasing levels of other hand, the susceptible genotypes were exposure. At a given level of susceptibility limited to a relatively small range at the the mean age at death decreases regularly right end of the distribution and were with the increasing exposure, and at a given clearly marked by becoming clinical cases level of exposure the mean age at death inat a low level of exposure. Under these con- creases as the degree of susceptibility deditions there was a much better chance of creases. Both increased susceptibility and selecting a population possessing the most increased exposure tend to decrease the susceptible genotypes. The hypothesis is mean age at death from lymphomatosis supported by the observation that selection and the action of these two factors is cumudid not appreciably decrease the incidence lative. This relationship is illustrated of lymphomatosis in the resistant strain graphically in Figure 4 where the percent(Taylor et al., 1943) and that the suscep- age of lymphomatous deaths in each threetible strain responded more uniformly than month age period is plotted against time. the resistant one to the three levels of exUp to this point the discussion has dealt posure during the three-year period for with all types of lymphomatosis. The relawhich corrected data are available. The data tionship between the three types of the show that the greatest variations in redisease and the levels of exposure and sussponse were in the resistant populations at ceptibility were not entirely consistent from the VSC and VSI levels. year to year. Generally, the percentage of In populations like those under con- birds with neural lesions at each level of exsideration there is usually a certain level of posure was greater in the susceptible than in

INTRAPERITONEAL INJECTION OF LYMPHOMATOUS NERVE TISSUE

391

ing to integrate these factors a thorough understanding of each must be obtained. 30 The three levels of exposure were three sets of conditions which produced increas20 ing percentages of lymphomatosis when 10 birds from a population of given susceptibility were exposed to them. The first of o 3 6 ? /2 15 18 2/ 2# these was the PHC level or the level in the •M Age ih months n parent flock. This level should have been s fairly constant since the flock was well o Susceptible St fa in isolated, few birds were introduced, and the E -3 L«v«ls of exposure management was uniform from year to year. The VSC level appeared among the control birds raised at Veterinary Science in spite of added sanitary precautions and the more rigid isolation of birds in small groups. Apparently these precautions were outweighed by other factors. Birds from 3 6 9 12 JS 18 21 2¥ or other flocks were kept on the premises for A$t ih months short periods of time and the inoculated FIG. 4. The rate of death from lymphomatosis birds with their higher incidence of the in each three-month age period among birds of disease were also kept here. Whether either the susceptible and resistant strains at the PHC, of these factors was responsible for the VSC VSC, and VSI levels of exposure. level cannot be determined at present. the resistant populations. The same was The VSI level was attained when the true of the birds with visceral lesions at the birds under VSC conditions were inocuPHC and VSC levels but not at the VSI lated. The mechanism by which inoculation level. Lymphomatous birds in the resistant increases the incidence of lymphomatosis population showed a higher proportion of is not clear. No new material was introvisceral than neural lesions. In the susduced into the group by inoculation because ceptible group the situation was reversed. the inocula were always prepared from .The relative proportion of neural and vismembers of the two strains concerned. ceral cases remained about the same in the Since the control groups of both strains had •resistant group at the three levels of exlymphomatosis, it follows that inoculation posure, whereas in the susceptible population the proportion of the lymphomatous must either have increased the number of birds showing neural lesions increased birds exposed, or exceeded the threshold slightly with the level of exposure. In of resistance in the exposed birds so that general the proportion of lymphomatous more of them became clinical cases. In birds showing ocular lesions tended to de- either case it appears that the difference between VSI and VSC or PHC must have crease with increasing levels of exposure. been quantitative and as shown above accounted for 85 percent of the variance in the DISCUSSION The incidence of lymphomatosis found incidence of the disease among birds of among the birds under discussion was the similar susceptibility. A satisfactory exresult of the interaction of two variable planation of the differences between the factors: exposure at three levels, and sus- three levels of exposure must await further ceptibility at two levels. Before attempt- experimentation. Resistant Strain

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

392

K. B. DEOME

resistant to become clinical cases at the increased level of exposure, should offer a better basis for selection than the survivors at a low level of exposure. It is probable that any attempt to increase resistance at the present PHC level of exposure would continue to fail since only a very small portion of the individuals in the population would reveal their genotypes by becoming clinical cases. This is particularly true since the population is skewed toward the resistant side. The skewness of the population, the low level of exposure, and the fact that the primary criteria of selection used were oriented on production rather than viability characters could easily account for the previous failures of selection to increase the average level of resistance. As the level of exposure increased, increasing portions of the population with susceptible genotypes would reveal themselves by becoming clinical cases, and the survivors would be increasingly valuable as resistant genotypes. In contrast to this, the PHC level of exposure has favored the selection of a susceptible strain because clinical cases at this low level clearly mark the families apt to contain the most susceptible genotypes. The influence of various levels of exposure and susceptibility on the Poultry Husbandry birds can best be visualized by referring to the frequency curves in Figure 3. At a given level of susceptibility, an increase in the level of exposure would produce an increased incidence of lymphomatosis. This increase would be small if the level of exposure was below the threshold of the population and very large if the threshold was exceeded. Further increase in the level of exposure would not significantly increase the percentage of clinical cases. The level would have to be very high to produce an incidence of 95 percent. At a given level of exposure an increase in susceptibility would have a similar effect. At a low level of exposure there would be

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

The two levels of susceptibility were produced by selection. In the susceptible strain selection was directed toward increasing the percentage of clinical cases. At a given level of exposure this is equivalent to decreasing the threshold of resistance of the population. The selection was successful as is shown by the increased incidence of lymphomatosis since selection began in 1933 (Taylor et al., 1943). Selection in the resistant strain was directed toward decreasing the percentage of clinical cases or increasing the threshold of resistance in the population. Selection in this direction was not uniformly successful (Taylor et al., 1943). The selection was not as rigorous as had been practiced in the susceptible group since several other criteria of selection were employed in addition to the incidence of lymphomatosis. Yearly variation in the disease would be expected in both groups but should be most noticeable in the resistant strain. This expectation is borne out by the data. It is likely that many genetic factors influence the level of resistance. The frequency curves constructed for the distribution of birds within the populations according to their susceptibility (Fig. 3) probably reflect the relative distribution of genotypes within those populations. Such a distribution would be expected from the type of selection practiced. Therefore, it should be possible to increase the average resistance of the resistant population still further by more rigorous selection. There are two ways in which more rigorous selection could be brought into practice. The first of these would be by adopting higher standards of selection without changing the exposure. The second could be carried out by increasing the level of exposure within the portion of the flock to be selected. This may be done by inoculating birds with lymphomatous material as was done in the experiments described in this paper. The survivors, that is those too

INTRAPERITONEAL INJECTION OF LYMPHOMATOUS NERVE TISSUE

393

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

no demonstrable change if a very resistant be encountered which will exceed the population became slightly less resistant, threshold of resistance of a very resistant but at the same level of exposure an equal strain, and when this happens a great increase in susceptibility in a susceptible many clinical cases should be expected. A population might produce a great many study of the frequency curves in Figure 3 will show that the value of a resistant cases. The influence of these factors is cumu- strain lies in its ability to tolerate relalative so that an increase in exposure and tively high levels of exposure without showdecrease in resistance would produce a very ing a great increase in the number of large increase in the incidence of lympho- clinical cases. These considerations are not matosis. Generally, as the incidence in- intended to discredit the practice of selectcreases the mean age at death decreases. ing for resistance to lymphomatosis but An "outbreak" of serious proportions at an rather to indicate the limits beyond which early age could result from the cumulative selection would not be effective in reducing effect of these two factors. It appears that the incidence of the disease. the lower the level of susceptibility and the The possibility of improving selection for higher the level of exposure, the more apt resistance to lymphomatosis by artificially a population is to have a high incidence of producing a high level of exposure is to be lymphomatosis at an early age. The pro- tested in our flocks. For the present it portion of the lymphomatous cases showing should be regarded as a promising working neural lesions should be high in this case, hypothesis subject to test. There is a posthe proportion showing ocular lesions low, sibility that selection under these conditions and the proportion showing visceral lesions may not produce a sufficiently high level of remain about characteristic of the popula- resistance to justify the danger of estabtion. lishing a high level of exposure within a The poultry breeder should find in these breeding flock. The situation may be simideductions information permitting him to lar to that existing in the development of evaluate his breeding program for resistance mammary tumors in mice. In the latter to lymphomatosis. If he has a high inci- case it was possible by selection to produce dence of the disease in his flock he should strains which rarely developed breast tuexpect selection to increase the resistance mors and others which had a great many of of his birds and decrease the incidence of them. These strains might be regarded the disease to a relatively low level beyond as resistant and susceptible respectively which further improvement would be very since it has been demonstrated that the difficult. On the other hand, a low inci- principal cause of tumor production is a dence of the disease in his flock would substance passed from mother to offspring indicate that further selection would main- through her milk (DeOme, 1940). When tain existing resistance but produce very young mice from the resistant strain are little improvement. He should not expect "exposed" to substances from high tumor birds from his resistant strain to be uni- strain foster-mothers, more than one-fourth formly resistant to the great variety of con- of them develop mammary tumors in spite ditions existing in other flocks,, for it is of a very high order of genetic resistance probable that the levels of exposure in to the condition, produced by many genother flocks differ both quantitatively and erations of selection and inbreeding by qualitatively from those existing in his flock brother to sister matings. The same amount in which the resistant birds were selected. of exposure would have produced a much It is probable that levels of exposure will higher incidence in the susceptible strain

394

K. B. DEOME

SUMMARY

A series of experiments is reported in which birds from two populations known to be relatively susceptible and resistant respectively to spontaneous lymphomatosis were subjected to three levels of exposure. The results were as follows: 1. At a given level of susceptibility the incidence of lymphomatosis increased as the level of exposure increased. 2. At a given level of exposure the incidence of lymphomatosis increased as the level of susceptibility of the populations of birds increased. 3. The effect of increasing levels of exposure and increasing levels of susceptibility on the incidence of lymphomatosis was cumulative.

4. The mean age at death from lymphomatosis decreased as the level of susceptibility or the level of exposure increased and the effect of these two factors was cumulative. 5. The incidence of lymphomatosis in groups of birds from both strains was increased by injecting them with lymphomatous material from members of the parent flock. Frequency curves showing the relative distribution of individuals within the two populations according to their degrees of resistance to lymphomatosis were constructed. Within the limits of accuracy of these distributions the following inferences can be drawn: 1. The relative resistance or susceptibility of the two strains was due to the action of numerous genetic factors. 2. It is likely that additional increases in the level of resistance could be achieved by increasing the level of exposure in the flock to be selected, but further selection at a low level of exposure probably would produce no increase in resistance unless extended over a very long period of time. 3. The value of a resistant population lies in the possession of a high threshold of resistance below which relatively large increases in the level of exposure produce small increases in the incidence of lymphomatosis. REFERENCES

Hutt, F. B., R. K. Cole, and J. H. Bruckner, 1941. Four generations of fowls bred for resistance to neoplasms. Poultry Sci. 20:514-536. Taylor, L. W., I. M. Leraer, K. B. DeOme, and J. R. Beach, 1943. Eight years of progeny-test selection for resistance and susceptibility to lymphomatosis. Poultry Sci. 22 :339-347. DeOme, K. B., 1940. The incidence of mammary tumors among low-tumor strain CS7Blk mice when foster-nursed by high-tumor strain A females. Am. Jr. Cancer 40:231-234.

Downloaded from http://ps.oxfordjournals.org/ at University of North Dakota on June 9, 2015

but the important fact remains that in spite of a very high level of resistance a level of exposure existed which caused more than one-fourth of them to become clinical cases. In addition, once members of the resistant strain had been exposed, they passed the agent on to their descendants for a number of generations (unpublished data). This case clearly illustrates the idea that even high levels of resistance can be overcome by high levels of exposure and shows the danger of artificially producing a high level of exposure against which the existing level of resistance was ineffective. It remains to be proved that any practical method of selection over a reasonable length of time will produce strains of birds sufficiently resistant that existing levels of exposure will not exceed their threshold of resistance. On the other hand, if desirable genotypes representing high levels of resistance are to be cumulated and maintained it is imperative that exposure be kept at a sufficiently high level to permit families possessing these genotypes to be identified.