Vol. 11, No. 2 Printed in U.SA.

INFECTION AND IMMUNrrY, Feb. 1975, p. 309-312 Copyright 0 1975 American Society for Microbiology

Antigenic Response to Topically Applied Proteins EDWARD F. HARRISON,* MILTON E. FUQUAY, AND WALTER A. ZYGMUNT Department of Biochemistry, Mead Johnson Research Center, Evansville, Indiana 47721 Received for publication 7 October 1974

Six different proteins varying widely in molecular weight, ribonuclease, lysostaphin, ovalbumin, penicillinase, collagenase, and Varidase were tested for their ability to induce circulating antibody formation in rabbits after repeated topical application of the proteins in a water-soluble gel vehicle. After a 12-week exposure period, significant hemagglutinin titers were noted in rabbits treated with ovalbumin, lysostaphin, or ribonuclease; markedly elevated, passive cutaneous anaphylaxis-reacting sera were obtained only from collagenase- or lysostaphin-treated animals. Precipitin antibodies as evidenced by gel diffusion were also found in sera from collagenase- and lysostaphin-treated animals. Topical application of penicillinase was only marginally effective and Varidase was totally ineffective in eliciting a positive circulating antibody response. In all cases, topical application of proteins for periods in excess of 3 weeks was required for induction of circulating antibody formation.

In view of the paucity of information (6, 9, 15) on the antigenic response in animals to topically applied proteins, it appeared desirable to determine whether production of antibodies could be induced through multiple topical application of various proteins to intact rabbit skin. Significant antibody titers, as evidenced by hemagglutination and passive cutaneous anaphylaxis (PCA), resulted from repeated topical application of five different proteins varying in molecular weight from 13,000 to 109,000. Varidase, a mixture of proteins of microbial origin, when applied topically failed to induce production of circulating antibodies. MATERIALS AND METHODS Animals. Groups of three to five young adult, male albino rabbits (Belted Dutch strain), caged individually and fed a commercial pellet diet, were used in these studies. The hair was carefully clipped and removed from the back of each animal. Any animal with abraded skin was discarded. Proteins tested and method of application. With the exception of lysostaphin (a mixture of proteins elaborated by Staphylococcus staphylolyticus [16]), which was produced within the Mead Johnson Research Center, the other five proteins were obtained from reliable commercial sources and represent the highest purity of material readily available. Ovalbumin (5x crystallized), penicillinase, and ribonuclease (5x crystallized) were obtained from Nutritional Biochemicals Corp., and collagenase was obtained from Agricultural Biologicals Corp. Varidase, a mixture of streptokinase and streptodornase with an assay value of 4,319 U/g, was kindly provided by Lederle Laboratories.

All proteins were formulated at a 1% concentration in an aqueous semifluid gel containing hydroxyethylcellulose, glycerin, phosphate buffer, and parabens as a preservative. The final pH was approximately 7.5. On a daily basis, Monday through Friday, for a total of 12 weeks each animal received a topical application to the intact skin (ca. 25 cm2 in area) of 0.5 g of protein gel per kg of body weight. All animals were placed in appropriate holding boxes for 4 h after each treatment to restrict their movement and to minimize inadvertent loss of gel. After 4 h, the rabbits were returned to individual cages where food and water were provided ad libitum. Serology. Appropriate blood samples were taken by cardiac puncture at the start of the experiment and at 3-week intervals thereafter. All blood samples were allowed to clot, and the sera were decanted and frozen at -20 C until assayed. Hemagglutination antibody titers were determined

using specific protein-sensitized sheep erythrocytes (14). Tannic acid was employed as a coupling agent for fixing the protein to the erythrocytes. This reaction was carried out at protein concentrations of 100 ug/ml in 0.15 M tris(hydroxymethyl)aminomethanesodium chloride buffer, pH 6.4. After centrifugation and washing, the sensitized cells were suspended at a concentration of 2% in 0.85% saline (containing ca. 1.0% fresh rabbit serum). Twofold serial dilutions in 0.85% saline (with volumes of 0.5 ml in series of 10 tubes each) were made with individual test sera and mixed with an equal volume of the sensitized erythrocytes (0.5 ml). The hemagglutination titers were determined after a 24-h incubation period at room temperature. PCA tests in guinea pigs, using a modification of the procedures of Ovary (10), were performed on those serum samples which gave a positive hemagglutination titer. Electric clippers were used to remove the 309

310

INFECT. IMMUN.

HARRISON, FUQUAY, AND ZYGMUNT

hair from the backs of male guinea pigs. A depilatory (Nair, Carter Products, New York) was employed to remove the remaining hair, and the skin was then washed thoroughly with warm water. The following day 0.1-ml portions of undiluted and diluted sera (1:4, 1:8, and 1:16) were injected intracutaneously into the backs of guinea pigs. Four hours later, each animal received an intravenous injection of a mixture of 0.5 ml of 0.5% Evans blue plus 0.5 ml of a 1.0% protein solution. The sites of the intracutaneous injections were observed for positive PCA reactions 1 h after administration of the protein-dye injection. When a positive PCA test (a minimum of 5 mm in diameter) was obtained with the 1:16 dilution of serum, the test was repeated with higher serum dilutions so that a definite PCA end point in each case was obtained. Ouchterlony gel diffusion assays (3, 5) for detection of precipitating antibodies were performed on all 12-week serum samples. The specific protein was dissolved in 0.15 M phosphate buffer, pH 7.2 (0.1 mg/ml), and was used as the antigen in the center well. Test sera were placed in wells located approximately 5 mm from the center well. All gel plates were prepared with 0.5% Oxoid Ionager no. 2 (Consolidated Laboratories, Inc.) in 0.15 M phosphate buffer, pH 7.2. A positive response was indicated by a visible band of precipitate formed at the antigen-antibody interface after a 48-h incubation period at room temperature.

RESULTS AND DISCUSSION Table 1 shows the reciprocal hemagglutinin antibody titers expressed as the mean for each test group. Significant antibody titers were obtained only from animals treated with ovalbumin, lysostaphin, and ribonuclease. Interestingly, topical administration of penicillinase elicited a very poor hemagglutinin antibody response.

Regarding stimulation of cell-bound antibodies, as evidenced by the PCA reaction, collagenase and lysostaphin were the only proteins which evoked a significant antibody response. That is, these serum samples gave a positive PCA reaction at a serum dilution of at least 1:4 (Table 2). Whereas the magnitude of the PCA antibody response obtained with lysostaphin was gradual over the time course from 6 to 12 weeks, stimulation of the PCA antibody response with collagenase appeared to be an all-or-none phenomenon at the 6- to 9-week interval. Secondly, antibody formation to lysostaphin was further stimulated from week 12 to week 15. By week 18, however, no PCA antibodies were detected in the sera from lysostaphin-treated animals. It should be pointed out that animals received the antigenic insult only through week 12. Unfortunately, sera from collagenase-treated animals were not available beyond week 12.

Relative to the detection of circulating antibodies by the Ouchterlony procedure, at the completion of the 12-week testing period sera from lysostaphin and collagenase rabbits were the only ones which elicited a positive precipitating antibody response (Table 3). Although differences were clearly discernible in the ability of a given protein to stimulate antibody formation within an experimental group, the magnitude of these differences at the 9- and 12-week intervals was approximately twofold with all six proteins, e.g., in the case of hemagglutinating antibodies. A similar order of magnitude in differences was noted in the sera from individual rabbits in each experimental group regarding stimulation of cell-bound antibody formation as evidenced by the PCA reaction. Finally, the fact that the purification process employed for lysostaphin resulted in a mixture of at least four different proteins, only one of which was indeed lytic to the cell walls of coagulase-positive staphylococci, led us to repeat the hemagglutination assays using, inTABLE 1. Hemagglutination titers obtained after repeated topical application of proteins to rabbits Mean reciprocal of hemagglutination titer (weeks of application)

Mol wt

Protein

0

Ribonuclease

13,400 32,000 44,000 50,000 47,600a 109,000

Lysostaphin Ovalbumin Penicillinase Varidase Collagenase

8 0 5 6 0

0

3

9

6

12

34 128 298 170 0 57 937 2,170 10 429 1,024 3,072 18 18 18 37 0 2 1 2 0 8 3 2

aApproximate molecular weight. TABLE 2. Mean reciprocal of rabbit serum dilutions eliciting a positive PCA response in guinea pigs Weeks on test

Protein

Treatment period 0

3

6

9-12

Post-drug observation period 14 15 16 18

Ribonuclease 0 0.3a 0.3 0.3 0.3 1.0 1.0 Lysostaphin 0 0 3.4 6.7 16.9 208 0 Ovalbumin 0 0 0 0 3.0 5.3 0 Penicillinase 0 0 0 0 0 Varidase 0 0 0 0 0 Collagenase 0 0 0 204 260 a Mean reciprocal of serum dilution producing positive PCA reaction.

VOL. 11, 1975

TOPICAL APPLICATION OF PROTEINS

TABLE 3. Detection ofprecipitating antibodies by gel diffusion in rabbit sera after topical application of proteins for 12 weeks Protein

No. of animals with sera showing positive precipitin bandsa

Ribonuclease ....... Lysostaphin ....... Ovalbumin ....... Penicillinase ....... Varidase ....... Collagenase .......

0/3 3/5 0/2

0/3 0/5 2/3

a Number of animals with positive sera over total number of animals exposed to specific protein(s).

stead, a purified staphylolytic fraction (2:-fraction) of lysostaphin. The resulting data clearly indicated that the mean values for the reciprocal of the hemagglutinating antibody titers were definitely lower when assayed with the 2-fraction as opposed to the crude lysostaphin preparation. In the former case, 6-, 9-, and 12-week values of 22, 67, and 294 were obtained. By week 15 and 18, however, the mean reciprocal values for the hemagglutinin titers had decreased to 28 and 8, respectively. Although repeated topical administration of lysostaphin to rabbit skin resulted in the induction of significant circulating antibody titers, such was not the case in human nasal staphylococcal carrier studies treated intranasally three or four times daily for 7 to 14 days with a 0.5% solution of lysostaphin in saline. Martin and White (8) found no increased sensitivity to intradermal challenge with lysostaphin 4 to 9 months after therapy. The presence of hemagglutinating antibody was detected in one subject prior to therapy, and low titers of precipitating antibody in pretreatment sera from several subjects were also observed. In part, the data suggested that the naturally occurring precipitins may have been directed against one of the contaminating antigens rather than the major component in lysostaphin, lytic peptidase. PCA studies in guinea pigs were also negative for all sera. Harris et al. (4), in their study on the use of lysotaphin in staphylococcal carriage in infants and children, found that in only one patient was there a suggestion of induced antibody formation. In this specific instance, the serum showed a slight rise in hemagglutination titer, and a positive reaction to cutaneous hypersensitivity was noted in the guinea pig PCA reaction. No clinical evidence of sensitization, however, was noted in either study in any of the 42 staphylococcal nasal carriers (4, 8).

311

Regarding possible mechanisms involved in percutaneous absorption (2, 13), Scheuplein (13) proposed a mathematical model for studying this phenomenon using human skin. He concluded that transient diffusion occurred primarily through hair follicles and ducts and that a steady state of diffusion took place primarily through the intact stratum corneum. Walzer (15) demonstrated significant percutaneous absorption of antigens, using finely triturated cottonseed and peanut meal suspended in a petrolatum base, in the rhesus monkey and in man. Absorption of the antigen was determined on the basis of a reaction developed at a passively sensitized site in a distant area. A greater frequency of cutaneous immediate allergy reactions was also noted among atopic individuals than normals upon intranasal exposure to crystalline bovine ribonuclease (12). Purchase and Steyn (11), however, found in the rat no percutaneous absorption of aflatoxin, a potent hepatocarcinogen, dissolved in acetone or dimethyl sulfoxide. Similarly, Kastin et al. (6) found that dimethyl sulfoxide failed to enhance the poor percutaneous absorption of certain polypeptides, vasopressin and adrenocorticotropin, in the rat and melanocyte-stimulating hormone in the frog. Baer and Godfrey (1), however, did report significant and rapid percutaneous absorption in guinea pigs of sensitizing and nonsensitizing substances related to the active principle of poison ivy. These substances though, pentadecylcatechol (a sensitizer) and pentadecylveratrole (a non-sensitizer), are not proteinaceous. Finally, Marzulli et al. (9) have reported that relative to man the back skin of the following animal species is more permeable: monkey, dog, cat, horse, rabbit, goat, guinea pig, and mouse (listed in increasing order of permeability). Thus, it is not surprising that topical administration of lysostaphin stimulated circulating antibody formation in rabbits but not in man. Leskowitz and Ovary (7) studied the possible relationship between molecular weight of antigen and the ability to elicit PCA reactions in the guinea pig. Using such antigens as ribonuclease, egg albumin, bovine serum albumin, human alpha-globulin, etc., and passive sensitization via an intracutaneous route, on intravenous challenge with antigen at a given antibody level the amount of antigen needed to elicit a reaction increased with the molecular weight of a given protein. In the present studies, there appeared to be a suggestion of a trend regarding the size of protein molecules and their ability to induce the formation of circulating antibodies

312

HARRISON, FUQUAY, AND ZYGMUNT

INFECr. IMMUN.

lated to the active principle of poison ivy. Adv. Biol. after prolonged topical application of proteins. Skin 11:161-165. In the case of hemagglutinins, topical applica- 2. Bartek, M. J., J. A. LaBudde, and H. I. Maibach. 1972. tion of three different proteins, ribonuclease, Skin permeability in vivo: comparison in rat, rabbit, lysostaphin, and ovalbumin, with molecular pig and man. J. Invest. Dermatol. 58:114-123. weights of 44,000 or less was found to signifi- 3. Feinberg, J. G. 1957. Identification, discrimination and quantification in Ouchterlony gel plates. Int. Arch. cantly enhance antibody titers. Of these three Allergy 11:129-152. proteins, however, the smallest in size, ribonu- 4. Harris, R. L., A. W. Nunnery, and H. D. Riley, Jr. 1968. Effect of lysostaphin on staphylococcal carriage in clease, was also the least antigenic. Topical infants and children, p. 110-112. Antimicrob. Agents application of penicillinase resulted in only Chemother. 1967. marginal stimulation of hemagglutinin titers, 5. Hayward, B. J., and R. Augustin. 1957. Quantitative gel and both Varidase (a mixture of streptokinase diffusion methods for assay of antigens and antibodies. and streptodornase) and collagenase showed no Int. Arch. Allergy 11:192-205. 6. Kastin, A. J., A. Arimura, and A. V. Schally. 1966. enhanced hemagglutinin titers. Topical absorption of polypeptides with dimethylsulfInterestingly, only collagenase, an enzyme oxide. Arch. Dermatol. 93:471-473. with a molecular weight in excess of 100,000, 7. Leskowitz, S., and Z. Ovary. 1962. The relation between and lysostaphin when topically applied were molecular weight of antigen and ability to elicit passive cutaneous anaphylaxis. Immunology 5:1-10. able to elicit significant PCA reactions in the R. R., and A. White. 1967. The selective activity guinea pig. Moreover, presence of precipitating 8. Martin, of lysostaphin in vivo. J. Lab. Clin. Med. 70:1-8. antibodies via Ouchterlony gel diffusion was 9. Marzulli, F. N., D. W. Brown, and H. I. Maibach. 1969. also seen only in sera from rabbits treated with Techniques for studying skin penetration. Toxicol. Appl. Pharmacol. 3(Suppl.):76-83. the same two proteins. In view of the detection Z. 1958. Immediate reactions in the skin of of precipitating antibodies and positive PCA 10. Ovary, experimental animals provoked by antibody-antigen reactions with sera from collagenase-treated interaction. Prog. Allergy 5:459-508. rabbits, failure to detect hemagglutinins in the 11. Purchase, I. F., and M. Steyn. 1973. Absence of percutaneous absorption of aflatoxin. Toxicol. Appl. Pharsame sera makes the collagenase tanned sheep macol. 24:162-164. erythrocyte reagent highly suspect. 12. Salvaggio, J. E., J. J. Cavanaugh, F. C. Lowell, and S. These experiments clearly demonstrate perLeskowitz. 1964. A comparison of the immunologic cutaneous absorption of certain proteins as responses of normal and atopic individuals to intranasally administered antigen. J. Allergy 35:62-69. evidenced by the presence of circulating antiR. J. 1967. Mechanism of percutaneous bodies in the sera from treated rabbits. Topical 13. Scheuplein, absorption. II. Transient diffusion and the relative application of proteins for periods in excess of 3 importance of various routes of skin penetration. J. weeks, however, was required before significant Invest. Dermatol. 48:79-88. antibody titers could be detected. Differences 14. Stavitsky, A. B. 1954. Micromethods for the study of proteins and antibodies. I. Procedure and general were evident among the proteins relative to applications of hemagglutination and hemagglutinatheir ability to induce hemagglutinin formation tion-inhibition reactions with tannic acid and proteinand cell-bound, PCA-reacting and -precipitattreated red blood cells. J. Immunol. 72:360-367. 15. Walzer, A. 1940. Cutaneous absorption. I. A direct ing antibodies. LITERATURE CITED 1. Baer, H., and H. Godfrey. 1971. Absorption through the skin of sensitizing and nonsensitizing substances re-

technic for demonstrating the percutaneous absorption of antigens. Arch. Dermatol. Syphilol. 41:692-698. 16. Zygmunt, W. A., and P. A. Tavormina. 1972. Lysostaphin: model for a specific enzymatic approach to infectious disease. Prog. Drug Res. 16:309-333.

Antigenic response to topically applied proteins.

Vol. 11, No. 2 Printed in U.SA. INFECTION AND IMMUNrrY, Feb. 1975, p. 309-312 Copyright 0 1975 American Society for Microbiology Antigenic Response...
NAN Sizes 0 Downloads 0 Views