AN AT TEMPT TO PRODUCE ACTIVE IMMUNITY IN THE HERMIT CRAB PAGURUS GRANOSIMANUS George H. F attison from rd Universit Hopkins Marine Station of Stanfo rove. INTRODUCTION There has been very little immunological work done on invertebrates relative to the large amount done on vertebrates. and especially on mammals. The discovery by Cantacuzéne (1912. 1925, 1934) that the European hermit crab, Eupagurus prideauxii, apparently develops immunity to the toxin of the sea anemone, Adamsia palliata, through close biological association with it suggests that the phenomenon of immunity might account for the ability of the local hermit crab, Pagurus granosimanus (Stimpson, 1859), to walk unharmed over large beds of anemones. To ascer- tain if immunity was actually protecting them, an attempt was made to immunize P. granosimanus to the toxic effects of an homogenate of the sea anemone, Anthopleura elegantissima. The manner of assay for immunity involved the determination of the median lethal dose (LD50) of a standard homogenate for a sample of the P. granosimanus population by the method of Finney (1962). This LDgo was then compared to the LDgo of the same homogenate determined for two groups of P. granosimanus previously injected with different sublethal doses of the homogenate. In the two latter cases median lethal doses significantly larger than that for the normal sample would be considered evidence for the development of immunity to the toxic effect of the homogenate, Complementary to the above experiment an attempt was made to increase resistance of P. granosimanus to anemone toxin by placing them for a period of time in an aquarium containing a high concentration of anemones. Again the median lethal dose was (2) computed and used as an indication of the tolerance of the group. MATERIALS & METHODS All the P. granosimanus used were collected at one time at Stillwater Cove, just north of Carmel beach, Carmel, California. The specific area of the cove in which the animals were collected had a low concentration of anemones of all kinds relative to other areas of the coast. About twenty-five small Anthopleura elegantissima were mas- cerated in 150 ml. of Ringer's solution and then homogenized. This standard homogenate was frozen for later use in separate bottles. A. The LDgo for normal crabs was determined as follows: seventy P. granosimanus were cracked out of the shells they were inhabiting and were divided into seven groups of ten individuals each irrespective of size. Each group was injected with a dif- ferent dilution of the standard homogenate, one group serving as a control. The control group was injected with only Ringer's solution which was used as the vector for the homogenate. The injections were made into the right side of the abdomen through a portion of the columellar muscle with a Yale one quarter ml. syringe with needle size +26. Each individual was weighed and a dose proportional to weight was given, namely, .Ol ml. of homogenate at the appropriate dilution per .2 grams body weight. Broken but usable shells were returned to the hermit crabs and the number dead in each group was recorded between twelve and twenty-four hours 0 (3) later. An individual which was moribund but moving either its first or second antennae was classified as alive; if no movement of antennae was observable the individual was recorded as dead. B. Two other groups of P. granosimanus were injected with sublethal doses of the standard homogenate which was diluted to 1:160 and 1:320 with Ringer's solution, i.e. they contained 144 Ag N/.200 ml and 72 Ag N/200 ml, respectively. After an incubation period of six days each group was challenged in the manner des- cribed above and the median lethal doses were computed. C. Seventy P. granosimanus were placed in an aquarium with approximately forty-five small Anthopleura elegantissima for seven days. At the end of this time they were challenged as previously described and the LD50 computed. RESULTS The doses, numbers of animals used, and number killed are presented in Tables 1 - 4. The values of the median lethal doses and their approximate 95% confidence intervals for the various groups are presented in Table 5. The values are expressed in micrograms of nitrogen per 200 mg. of body weight. If we regard the values of the median lethal doses as the mean tolerances, tests of significance can be performed between the mean of the normal sample and the means of each of the other samples. In each case the test was performed at as.05 and con- firmed the obvious fact that there is no apparent difference (4) between the means, i.e., the null hypothesis was accepted each time. Therefore, no indication of inducible immunity was demon¬ strated. DISCUSSION Using the method described there is no evidence for an in¬ crease in the mean tolerance of P. granosimanus to the toxic effects of sea anemone homogenate after exposure to these animals or their extracts. Since the null hypothesis has been rejected the power of the tests becomes important. The power was approximately computed for the test between the mean of the normal sample and the sample immunized at 114 g N2/.2 grams body weight. If it is assumed. for instance, that the actual difference between the means was 145 gN the power is about .18. The cause of this relatively low power for the relatively large sample size lies in the fact that a natural population heterogeneous for size, age, and physiological condition was sampled. While an attempt was made to use animals having a minimal exposure in the past, the actual history of these animals was not known and one can assume that this accounts for some of the variability of natural resistance. It seems most probablg that P. granosimanus is protected from anemone nematocysts in natural circumstances by the chitinous exoskeleton which covers the exposed portions of its body. Results which suggested that P. samuelis does not produce active immunity to anemone homogenate were obtained but they lacked (5) completeness and so could not be treated statistically. SUMMARY 1) An attempt was made to establish active immunity in Pagurus granosimanus to an homogenate of the sea anemone, A. elegantissima, with no success. 2) Keeping P. granosimanus in close contact with A. elegan- tissima did not increase its resistance to anemone homogenate. DOSE IN MICROGRAMS ON 460 307 230 184 164 144 0 DOSE IN MICROGRAMS OEN +60 307 230 134 164 4 4 0 TABLE F NUMBER OF ANIMALS USED 10 10 10 10 10 10 10 +2 TADLE NUMBER OE ANIMALS USED 5 7 10 NUMBER OF ANIMALS KILLED 10 3 3 0 NUMBER OF ANIMALS KILLED 4 4 — DOSE IN MICROGRAMS GEN +60 907 230 194 153 1 3 D DOSE IN MICROGRAMS OET 160 207 230 184 153 O TABLE HUMBER OF ANIMALS USED 8 6 8 ABLE I NUMBER OF ANIMALS USED 10 10 /0 10 10 /1. NUMBER OF ANIMALS KILLED 6 2 5 2 0 O NUMBER OF ANIMALS KILLEO 6 5 2 2 O NORNAL CRABS CRABS TAAUNIZED A AA ERTRACT NITROCEN PER 28 WEIORI CRABS TAADDI2BO AT 72% EYTRReT NITROCEN PER 2SWEERT CRABS EYPDSED TO ANEMONES NATUR ALLY TABLE 5 LDeo 341 233 314 238 959. CONFIDEUCE ENTERVALS 186 - 623 173 - 316 205 -48 192- 132 * LITERATURE CITED Cantacuzéne, J. (1912) Sur certains anticoup naturels observés pagurus prideauxii. C. R. Soc. Biol. Paris 73:663-664. chez . (1925) Action toxique des poisons d'Adamsia palliata sur les Crustacés Decapoda. C. R. Biol. Paris 92:1131-1133, 1133-1136, 1464-1466. (1934) L'action precipitante specifique du serum d'Eupagurus prideauzii. C. R. Biol. Paris 117:269-271. Finney, D. J. (1962) Probit Analysis - A Statistical Treatment of the Sigmoid Curve. Cambridge University Press, London. 4 LEGENDS Table 1. Table showing doses, number of animals used, and number of animals killed in the determination of the ID50 for the normal crabs. Table showing doses, number of animals used, and number Table 2. of animals killed in the determination of the LD5o for crabs immunized at 144 micrograms of extract nitrogen per .200 ml. body weight. Table showing doses, number of animals used, and number Table 3. of animals killed in the determination of the LD50 for crabs immunized at 72 micrograms of extract nitrogen per .200 ml. body weight. Table showing doses, number of animals used, and number Table 4. of animals killed in the determination of the LD5o for crabs naturally exposed to anemones. Table 5. Table showing values of LDgo and approximate 95% confi¬ dence intervals.