Specificity of mycobacterial sensitins.

The influence of various inoculation routes on the development and specificity of mycobacterial hypersensitivity reactions was investigated. The results indicated that the rate of development of skin test reactions was not determined by the inoculation route. Skin test reactions were seen within two hours after antigen injection and were often highly developed within 8-10 hours. The reactions were characterized by maximum appearance of inflammation at 24-30 hours followed by a decline in size beginning at 48 and 72 hours. Latex agglutination tests carried out with guinea pig sera 3 weeks after sensitization indicated the presence of circulating antibody at the time skin tests were performed. The sensitivity patterns induced by M. kansasii and M. tuberculosis, strain H37Ra induced responses in the order of PPD-S greater than PPD-Y greater than PPD-A. M. kansasii induced responses in the order of PPD-Y greater than PPD A greater than PPD-S. Cell extracts M. kansasii obtained by freezing-thawing and ultrasonic disruption was subjected to polyacrylamide (disc) electrophoresis. At least 10 bands could be distinguished and 4 major bancs were separated. Chemical analysis of the material eluted from the 4 major bands indicated that skin test activity was correlated with polysaccharide in association with protein. Skin test bio-assay data were analyzed by computer analysis. Values for parallelism of the log-dose response curve for the unknown and standard indicated that one band (K-3) contained active material of similar chemical constitution of that of PPD-Y. Another band (K-4), although active, exhibited non-parallelism suggesting the presence of other or different active material. Material isolated from the cell extract of M. kansasii by disc electrophoresis was not more specific than presently available PPD-Y. However, the studies indicated that the band fractions met the definition of "sensitin" in that the agents were capable of detecting delayed hypersensitivity without inducting the production of antibody.

SPECIFICITY OF MYCOBACTERIAL SENSITINS by Thomas Morgan Dietz A thesis submitted to the faculty of the University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Microbiology University of Utah August, 1966 This Thesis for the Doctor of Philosophy Degree by Thomas Morgan Dietz has been approved July, 1966 , ·;i�.ji ) i I I ij .. Reader, _j ;' J ) ./ 1 ." . ... j j ¥ , .---- ' . )j. ,/) . f: /- · ·v Supervisory Committee '- .' 'd" " \f"./ I '. . - Head, 'MaJor Department Dean, Graduate School The author wishes to express his appreciation to Drs. Stanley Marcus and Ernest Runyon for their assistance and encouragement during these studies and the preparation of this dissertation. Grateful acknowledgement is also made to Dr. Louis P. Gebhardt, Dr. Paul S. Nicholes, Dr. Douglas W. Hill and Dr. Bill B. Wiley for their helpful suggestions and criticisms throughout this work. Appreciation is extended to Mrs. Diane Ward and Mrs. Bonnie Perkins for their help in proofing and typing the thesis. The author wishes to thank his colleagues at the Veterans Administration Hospital James E. Brisbay, Kameron W. Maxwell and Mary Janis Smith for their assistance throughout these studies. Grateful acknowledgement is made to the Veterans Administration Hospital, Salt Lake City, Utah for the use of hospital facilities necessary to conduct this investigation. Finally, the author wishes to thank his wife Carol and his children, whose understanding and patience made this work possiole. The research reported herein has been supported by research grants from the National Institutes of Health (AI-K6-14-924, CA 07302). iii TABLE OF CONTENTS TABLES .•••••••••••••••••••.•••••••.••.••.•••.•.••••.••••••. vii FIGURES •.••..••••••..••••..•••...•.•.•.••••.•.••....•..••. viii INTRODUCTION. . . . . . • • • . . • . . . . • . . . . . • . . . . . . . . . . . . • . . . • . . . . . • . 1 REVIEW OF LITERATURE....................................... 2 I. II. III. IV. His torica1. • . • . . . . . . . • . . . . • . . . . • • . . . • . . • . . . . • . . . . . 2 Comparison tuberculosis .... 4 A. Medical Relationship ..•.••...•........•....... 4 B. Bacteriological Comparison .•.••••.••....•..... 4 C. Epidemiological and Antigenic Relationship .••. 5 D• S U111Illa ry. • . • • • • • . . . • • . . • . • . • • • • . . . • • • . • • • • • . • . • 8 Factors Influencing the Development of Delayed Hypersensitivity ....••..•....••...••...... 8 of~. kansasii with~. Mycobacterial Constituents .•...••......•.......... 16 A. Tubercu1oproteins .........•..•.•...•......••.• 16 B. Polysaccharides ...•......••.......•....•••.... 24 C. Lipids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 26 V. New Methods for Preparation of Mycobacterial Antigens .....••...•.•...............•..•.....•.... 29 VI. SU111Illary. • • • . • • • • • • . • . • • • . • . • • • • • • • • • . . • . • • . . . • • • .• 34 MATERIALS AND METHODS. • . • • • • . . . • . . . . • • . . . . • • • • . . . . . . • . . . • •. 35 I. Culture Growth .•.•.....•.....•...•.••....•.••....• 35 II. Harvest Technique .•.•..........••................. 35 III. Nitrogen Determination ..•.....•....••..•••••...••• 36 IV. Protein Determination ...•.•...•...••..••...•••.... 36 iv V. Carbohydrate Determination •••••••••••••••••••••• 38 VI. Nucleic Acid Determination ••••..•..•.••..•....•.. 38 VII. Acrylamide Electrophoresis ...•••..••....•......•. 38 A. Equipment .................•.................. 38 B. Reagents.. . . . . . . . . • . • . . . • . . • • • . . . • . . . . . . . . . .. 40 C. Preparation of Gels ..••.....•..•..•.•...••.•. 40 D. Loading of Column. . • . • . • . . • • •• •• . • • . . . . • . . . . • 41 E. Electrophoresis •.••....•••.•.•.••.•.•••...•.. 41 F• Des taining. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . 44 G• PAS S t a in. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 H. Feulgen Stain .....•.....•...••..•.•.•...•.•.. 45 I. Calculation of Rf Values .•••••••...••...••..• 45 J. Isolation of Protein Fractions ....•.....••••• 45 VIII. Guinea Pig Injection............................. 46 IX. Routes of Inoculation ..•..•.•.•.•.•••••••.•••...• 46 A. Subcutaneous. • . • . . . . . . • • • • . • • • • • • • • . • . • • . • . .. 46 B. Subcutaneous in Adjuvant ..•••..•..•..•.•.•... 46 C. Intradermal.................................. 47 D. Intratesticular ...•..•..••.••...••....•....•• 47 E. Intracardiac .••........•..•....•••..•..••.... 47 F. Aerosol...................................... 47 G. B"y Mou th. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 47 X. Skin Tes ts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 XI. Skin Tests with Fractions Obtained by Disc Electrophoresis.................................. 48 v t XII. Bio-Assay of Ant igens • • • . • . • • • . . • • . • • • • • • . . . . • • . . 48 XIII. Detection of Antibody....... • . . . . . • • . • • •• . . • • . . .. 48 EXPERIMENTAL RESULTS...................................... 50 I. Experiments Designed to Test the Influence of Inoculation Route on Specificity of Mycobacterial Antigens •..•••••.•.•••••••••••.••.••••• A. B. C. Rate of Development of Delayed Hypersensitivity .••.•.•..••••••••..••••.••..• 50 Degree of Sensitization for Various Inoculat ion Routes .•. ,........................ 63 Sensitivity Patterns Induced by kansasii and M. tuberculosis •••••••••••••• 63 Studies on Cellular Extracts, Cell Walls and Culture Filtrates............................ 67 ~. II. 50 A. M. Yields of Six Week Old Cultures of kansasii and M. tuberculosis .••.••••••.••. 67 Biochemical Analysis of Cellular Extracts, Cell Walls and Culture Filtrates ••.•••••••••• 67 Polyacrylamide Electrophoresis of Cellular Extracts, Cell Walls and Culture Filtrates... 72 Studies of Bands Isolated from Cell Extracts by Disc Electrophoresis Techniques .•.••.•••.••••. 74 A. Isolation of Bands .••..•.••••••.•••••.••••..• 74 B. Chemical Characterization of Bands ••••..•.••• 75 C. Bio-Assay of Band Activity ••••••.••..•••.•••. 79 D. Spec i fici ty of Bands......................... 86 E. Antigenicity of Bands K-3 and K-4 ....•.•.••.. 87 DISCUSSION................................................ 91 S~RY. 99 B. C. III. ••••••••••••••••••••••••••••••••••••••••• ••••••• •• BIBLIOGRAPHY. • • . . • • • • • • • . . . . • • . . • • . • • . . . . • . . . • • . . . . . . • . . .• 101 vi TABLES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. Twenty-four Hour Skin Reactions of Guinea Pigs Injected with ~n tuberculosis H37Raon~nnnnno,o~.onnono 64 Twenty-four Hour Skin Reactions of Guinea Pigs Injected with~. kansasii .••.••.••..•••.••.••.•.•.. 65 0 •• Fraction Yield from 6 Week Culture of M. kansasii in Modified Sauton Medium. . . •• .• . .• • . • . . . • . • • . • . . •• • . • 68 Fraction Yield from 6 Week Cultures of M. tuberculosis (H37Ra) in Modified Sauton-Medium •..•..•• 69 Results of Biochemical Analyses of Fractions Obtained from M. kansasii............................. 70 Results of Biochemical Analyses of Fractions Obtained from H37Ra ••.••••.•.•.•••.•••••••.••.•••••••• 71 Rf Values Calculated for Four Major Bands Isolated from Cell Extract of M. kansasii ••••.••.•.••• 77 Results of Biochemical Analysis of Four Major Bands Obtained from Cell Extract of M. kansasii •.••.•. 78 Pilot Study of Skin Reactions to Determine Approximate Potency of Test Preparations in Normal and~. kansasii Injected Guinea Pigs ••.••.•••...•••••. 80 Bio-Assay Data Obtained Using Band K-4 as Skin Test Antigen in~. kansasii Infected Guinea Pigs ••••.. 82 Bio-Assay Data Obtained Using Band K-3 as Skin Test Antigen in~. kansasii Infected Guinea Pigs •.•.•. 83 Bio-Assay Data Obtained Using Band Y-4 as Skin Test Antigen in~. kansasii Infected Guinea Pigs •••.•. 84 Potency of Bands K-4, K-3 and Y-4 Relative to Standard PPD-Y as Determined by Skin Test Reactions in M. kansasii Infected Guinea Pigs •.••.••..•.••.••... 85 Comparison of Reactions Elicited by Band K-4, PPD-Y and PPD-S in H37Ra Injected Guinea Pigs .•••..••..••••• 88 vii 15. 16. Comparison of Reactions Elicited by Band K-3, PPD-Y and PPD-S in H37Ra Injected Guinea Pigs ••.•••••••••••• 89 Results of Latex Agglutination Studies with Sera from Normal~ Skin Tested and ~n kansasii Infected Guinea Pigs........................................... 90 FIGURES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Flow Sheet for Isolation of Culture Filtrate, Cell Wall Extracts and Cell Extracts from~. kansasii •••••• 37 Photograph of Disc Electrophoresis Assembly Including Upper and Lower Buffer Tanks and Power Supply ••••••••• 39 Photograph Illustrating Photopolymerization of of Acrylamide Gels with the Use of a Fluorescent Light Source.......................................... 42 Photograph of Acrylamide Columns after Polymerization Showing Relative Positions of Lower Gel, Spacer Gel and Sample Gel........................................ 43 Rate of Development of Hypersensitivity Reactions. Mean (mm) for 5 Guinea Pigs Injected Subcutaneously with Strain H37Ra Suspended in Bayol Oil 3 Weeks Prior to Skin Testing •.••••.•••••••••••••••••••••.•.•• 51 Rate of Development of Hypersensitivity Reactions. Mean (rom) for 5 Guinea Pigs Injected Subcutaneously with Strain H37Ra 3 Weeks Prior to Skin Testing •••••.• 52 Rate of Development of Hypersensitivity Reactions. Mean (rom) for 5 Guinea Pigs Injected Intratesticularly with Strain H37Ra 3 Weeks Prior to Skin Testing •••••.• 53 Rate of Development of Hypersensitivity Reactions. Mean (rom) for 5 Guinea Pigs Injected by Cardiac Puncture with Strain H37Ra 3 Weeks Prior to Skin Testing.......................................... 54 Rate of Development of Hypersensitivity Reactions. Mean (rom) for 5 Guinea Pigs Injected Intradermally with Strain H37Ra 3 Weeks Prior to Skin Testing ••.•••• 55 Rate of Development of Hypersensitivity Reactions. Mean (rom) for 5 Guinea Pigs Subjected to an Aerosol of Strain H37Ra 3 Weeks Prior to Skin Testing •••.••••• 56 viii 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. Rate of Development of Hypersensitivity Reactions. Mean (mm) for 5 Guinea Pigs Injected Subcutaneously with ~. kansasii Suspended in Bayol Oil 3 Weeks Prior to Skin Tes ting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Rate of Development of Hypersensitivity Reactions. Mean (mm) for 5 Guinea Pigs Injected Subcutaneously with~. kansasii 3 Weeks Prior to Skin Testing ...•.•.. 58 Rate of Development of Hypersensitivity Reactions. Mean (mm) for 5 Guinea Pigs Injected Intratesticularly with M. kansasii 3 Weeks Prior to Skin Testing ...•.... 59 Rate of Development of Hypersensitivity Reactions. Mean (mm) for 5 Guinea Pigs Injected by Cardiac Puncture with M. kansasii 3 Weeks Prior to Skin Testing............................................... 60 Rate of Development of Hypersensitivity Reactions. Mean (mm) for 5 Guinea Pigs Injected Intradermally with~. kansasii 3 Weeks Prior to Skin Testing .....•.. 61 Development of Hypersensitivity Reactions. Mean (mm) for 5 Guinea Pigs Subjected to an Aerosol of M. kansasii 3 Weeks Prior to Skin Testing •••.•••.••.•. 62 Twenty-four Hour Skin Reactions of Guinea Pig Injected with H37Ra •••.••.••••.•.•.•••.••....•••.•...• 66 Twenty-four Hour Skin Reactions of Guinea Pig Injected with~. kansasii .••.•..•.•.•.•.•..•......•..• 66 Disc Electrophoresis of Cell Wall, Culture Filtrate and Cell Extract Material Isolated from M. kansasii. Amidoschwartz Protein Stain ..••..••.••.. 73 Disc Electrophoresis of Cell Extract of H37Ra. Amidoschwartz Protein Stain........................... 73 Disc Electrophoresis of Cell Extract of M. kansasii Showing 4 Major Bands or Areas Utilized as Skin Test Agents. Amidoschwartz Protein Stain .•...••.•........• 76 ix SPECIFICITY OF MYCOBACTERIAL SENSITINS INTRODUCTION A positive tuberculin reaction has been considered to be a sensitive indication of past or present infection with tubercle bacilli. It is now known that cross reactions to tuberculin induced by other mycobacterial infections are common. Production of skin test antigens from culture filtrate by precipitation with ammonium sulfate (PPD type, Affronti, 1961), trichloracetic acid (Magnusson, 1961), or with a combination 0f these precipitants (1\, Takeya et al., 1961) has been reported to yield products superior in specificity to old tuberculin (O.T.). However present evidence suggests that mycobacterial antigens are still not sufficiently specific to differentiate among the various mycobacterial infections of man. Two pulmonary pathogens, Mycobacterium tuberculosis and Mycobacterium kansasii, appear to be closely related in that infection of man with either of these organisms will often induce cross reactions to PPD antigens prepared from each species. While the homologous reaction will usually be larger, cross reactions obscure the interpretation of tuberculin reactions. It has been the purpose of this investigation to attempt to isolate more specific skin test antigens ~. from~. kansasii and tuberculosis by means of polyacrylamide gel electrophoresis. In addition, studies have been made of various factors (e.g. inoculation route) influencing the production and specificity of delayed hypersensitivity. REVIEW OF LITERATURE I. HISTORICAL The first mycobacterial skin test antigen as prepared by Robert Koch (1890) was not intended to be used as an aid in the diagnosis of tuberculosis but rather as a therapeutic agent. Subcutaneous injection of Koch's old tuberculin (O.T.) in the tuberculous patient induced fever, vomiting and local swelling and pain. The non-tuberculous patient failed to react. It was soon recognized that O.T. was of no therapeutic value; however its application as a diagnostic skin test antigen has been widely aCcepted for many years. Newer methods of administration eliminated the generalized reaction and caused, instead, only a localized reaction at the test site. These methods include a cutaneous test (Moro and Von Pirquet), a conjunctival test (Calmette), a modification of KochIs ormginal subcutaneous test (Harrlburger) and an intracutaneous test (Mantoux). The Mantoux test has survived as the most accurate and sensitive test. The use of the tuberculin test as a diagnostic tool for identifying arians. tubercu~ous cattle was recognized very early by veterin- The medical profession found the test of limited usefulness because practically everybody in the early part of the century reacted to tuberculin (e.g. reviews by Francis, 1959; Edwards and Edwards, 1960). As the incidence of tuberculosis declined, the tuberculin test became more useful. Doubts about the specificity of the tuberculin test arose in the 3 United States during the 1930's. Pulmonary calcification was often accompanied by a negative tuberculin test. This led to the investigations which showed Histoplasma capsula tum to be a common cause of pulmonary calcification. This organism was found in areas where the tuberculin test had seemingly failed (Edwards and Edwards, 1960). The tuberculin test has long been used as a means of detecting diseased cattle for disposal. With periodic testing and the prompt removal of reactors, the number of tuberculous herds decreased fairly rapidly. Some of the cattle which were slaughtered and subjected to post-mortem examination failed to show any evidence of tuberculous infection even though they had reacted to tuberculin. As the frequency of tuberculous animals decreased the proportion of "no-lesion" reactors gradually increased. Widespread occurrence of so-called saprophytic mycobacteria suggested to Some workers that these organisms might induce non-specific skin sensitivity. The works of many investigators have shown that acid-fast organisms other than~. tuberculosis might also be the cause of non- specific tuberculin skin sensitivity in man (Edwards, et a1., 1955; Edwards et a1., 1958; Singer and Rodda, 1961). These reports have been previously reviewed (Dietz, 1963). The discovery that mycobacteria other than M. tuberculosis do cause pulmonary disease clinically indistinguishable from tuberculosis stimulated increased study of these organisms. these pulmonary pathogens, ~. The most prevalent of kansasii, was characterized through its property of photochromogenicity, i.e. increased pigmentation 4 upon exposure to light (Buhler and Pollack, 1953; Hauduroy, 1955; Runyon, 1959). The similarities and differences of M. kansasii and M. tuberculosis will be reviewed in the following section. II. A. COMPARISON OF ~. KANSASII WITH ~. TUBERCULOSIS Medical Relationship Disease induced by ~. kansasii bears a marked similarity to tuberculosis in physical findings, roentgenographic appearance, pathology, presence of acid-fast bacilli in the sputum, and course of disease (Phillips and Larkin, 1964; Engbaek, et al., 1964; Lemaistre, 1963; and Pfuetze, et al., 1965). M. kansasii has been isolated from such sources as spinal fluid, spleen, liver, pancreas, testes and lymph nodes (Runyon, 1965). In addition cases of bone and joint involvement have been reported to be caused by a photochromogen (Klinenberg et al., 1965). These infections usually respond poorly to drug therapy (e.g. Engbaek et al., 1964; Lemaistre, 1963). However some physicians (Campagna and Greenberg, 1964) have reported good response with antituberculosis drugs. B. Bacteriological Comparison The most distinguishing feature of M. kansasii is the property of photochromogenicity. The colonies are intermediate in roughness, the cells are broad, banded and loosely corded. ~. tuberculosis, on the other hand, is rough, dry, eugonic and buff colored. M. tuberculosis is further characterized by tight serpentine cording of cells (Topley and Wilson, 1955; Runyon, 1965). These morphological 5 characteristics sometimes overlap, but two simple biochemical tests further distinguish between these species: kansasii shows high ~. catalase activity even when heated at 68 0 C while M. tuberculosis is negative (Kubica and Gleason, 1960). Tubercle bacilli produce niacin in amounts much greater than M. kansasii (Konno, 1958; Runyon et al., 1959). In addition to morphological and biochemical differences, animal pathogenicity studies indicate different virulence for each mycobacterial species. ~. kansasii injected subcutaneously usually does not produce progressive disease in guinea pigs (Runyon, 1959). However, intravenous injection of large doses have been shown to produce generalized granulomatosis in guinea pigs (Kubin, 1962). Mice inoculated intravenously or intraperitoneally with~. kansasii usually show lesions in lung, liver, spleen or kidneys (Runyon, 1959; Tsukamura and Tsukamura, 1964a). Typical~. tuberculosis will produce severe progressive disease with small inocula in both mice and guinea pigs. Efforts to immunize guinea pigs against tuberculosis with photochromogens has not been completely successful, although some protective effect has been demonstrated (Klugh and Pratt, 1962). The use of Adansonian classification, i.e. classification by comparison of physiological properties, grouped ~. kansasii with M. tuberculosis in the same branch among mycobacteriaceae. The criteria of significance were slow growth and feeble fermentation capacity (Bojalil and Trujillo, 1962). C. Epidemiological and Antigenic Relationship 6 Studies of skin sensitivities to mycobacterial antigens have been carried out by the group known as Tuberculosis Program Group of the United States Public Health Service. Antigens used were of the PPD type prepared by the method of Seibert (Affronti, 1961; Seibert and Glenn, 1941). Skin reactions of Navy recruits (Edwards, 1963) showed a higher reaction rate with skin test antigens prepared fromM. kansasii (13.1%) than with antigens prepared (8.6%). from~. tuberculosis A survey of healthy children in Baltimore (Walker and Patron, 1964) indicated a lower reaction rate when tested with M. kansasii antigens. prepared Studies of reactions in healthy populations with antigens from~. kansasii and~. tuberculosis indicated that cross reactions were frequent (Hsu, et al., 1963; Griffith et al., 1963), Unlike tuberculosis, the source of infection with M. kansasii is not known. Correlation studies of skin tests with gastric cultures indicated that although infections with photochromogens were widespread, resultant disease was relatively rare (Kendig, 1961). studies have revealed that contacts of patients infection have larger reactions to~, with~. Epidemiological kansasii kansasii antigen than do non- contacts, indicating possible transfer of infection or common exposure (Chapman et al., 1962). A smaller study of patient contacts (Eckman et al., 1961) indicated that family contacts had a significantly higher incidence of sensitivity to both tuberculin and photochromogen antigen; however, in no case was M. kansasii found to be transmitted by person-to-person contact. Evidence is still not conclusive with regard to the significance of person-to-person transmission. 7 Searches for possible reservoirs of infection have included raw milk (Chapman et al., 1965) and dogs (Leon et al., 1964). These sources have been implicated as possible reservoirs of infection, but further evidence is still required in order to determine a source in nature. Two varieties of M. kansasii with different clinical significance have been described (Wayne, 1962). It was suggested that one variant was capable of infecting man without causing disease, while the other usually was isolated from patients with disease. Studies of the skin reaction of patients with~. kanaasii disease and patients with tuberculosis at Suburban Cook County Hospital showed that ~. kansasii is antigenically similar to~. tuberculosis. Reactions to PPD antigens prepared from each species were very nearly the same size. However, reactions to the homologous antigen were larger than to the heterologous antigen (Edwards and Palmer, 1958). Numerous other reports substantiated the findings that infections with either M. tuberculosis or M. kansasii will induce cross reactions to skin test antigens of each species (Chapman, 1962; Martosh et al., 1962; Magnusson et al., 1961; Hsu et al., 1964). Further studies on the specificity of the tuberculin reaction were carried out by experimental infection in guinea pigs (Edwards et al., 1960; Edwards and Hopwood, 1962; Bjerkedal and Nativig, 1962; Tarshis and Frisch, 1952; Gernez, Rieux et al., 1962). These studies have also shown that infection with M. tuberculosis, virulent or avirulent strains, and BCG will induce skin sensitivity to various 8 III. FACTORS INFLUENCING ~ DEVELOPMENT OF DELAYED HYPERSENSITIVITY When a foreign protein enters the tissues of a host, certain 9 alterations take place such that upon subsequent injection of the same protein various forms of hypersensitivity will result. If the challenging injection is made intracutaneously, various types of reactions may occur. These include immediate inflammatory response or local skin anaphylaxis, Arthus reaction, and tuberculin or delayed hypersensitivity. This review is primarily concerned with delayed hypersensitivity but this phenomenon will be discussed by contrasting it with the immediate and Arthus hypersensitivities. Arthus and anaphylactic or immediate types of skin sensitivity differ with respect to amount and route of antigen entering the body on first and subsequent contacts, the degree of sensitization, and the character of the sensitizing antigen. However it is generally believed that the Arthus reaction represents an exaggerated localized form of anaphylactic sensitivity (Rich, 1951; Raffel, 1959; Carpenter, 1965). Immediate skin sensitivity is usually established by a single injection of a small amount of antigen. Testing with an intracutaneous injection of protein will induce evanescent edema and erythema. Repeated injections over a period of time will lead to more severe reactions s'uch as necrosis and sloughing of the skin. This intensified reaction is termed the A!thus phenomenon and is apparently mediated by circulating bivalent antibody (Raffel, 1959; Boyd, 1947; Carpenter, 1965). Classical immediate skin sensitivities are characterized by reactions of the "wheal and flare" type. Within one to fifteen minutes after intracutaneous or percutaneous inoculation of the test 10 antigen, the test area swells and develops erythema and induration. The reaction increases to a maximum within thirty minutes to several hours) after which the reaction fades and disappears within a few hours. The characteristic wheal and flare response has been attributed to histamine or histamine-like substances released from damaged cells (Lewis, 1927; Austin and Hunphrey, 1963). Some workers feel that the Arthus reaction is not related to histamine release but rather to mechanical damage to tissue rather than to individual cells (Crowle, 1962). As noted above, animals sensitized to give an immediate skin response have humoral or circulating antibody present in their serum. This antibody may be passively transferred to other animals by means of sera (Carpenter, 1965). The Arthus reaction is the most difficult of the immediate types to distinguish from the dermal reactions of the delayed or cellular type. The peak Arthus reaction usually develops rapidly b~t may recede slowly; depending on antibody concentration, the Arthus reaction may be acoompanied by induration and infiltrating cells such as macrophages and histiocytes typical of delayed reactions (Crowle, 1962; Gell, 1958; Leskowitz, 1960; Waksman, 1960). Delayed hypersensitivity is usually distinguished from the immediate type by the criteria that its occurrence is not correlated with the presence of circulating antibody and it cannot be passively transferred by means of serum. It appears to be firmly established that delayed skin sensitivity can be transferred regularly in animals and man by injection of lymphocytic cells obtained from sensitive donors (Chase, 1945; Kirchheimer and Weiser, 1947; 11 Waksman and Matoltsy, 1958; Lawrence, 1949). Transfer of delayed hypersensitivity by means of lymphocytic cells has been carried out with essentially no loss of specificity e.g. patients sensitized by transfer of cells from M. avium infected patients reacted maximally to tuberculin prepared Najarain and Feldman, 1963). from~. avium (Jensen et al., 1962; McCluskey et al.(1963) have reported that this type of passive transfer of lymphocytic cells induces nonspecific hypersensitivity. One group of workers have suggested that passive transfer of delayed hypersensitivity by means of sera does occur (Bhadarakom et al., 1963). This work has not been confirmed by others (Crowle, 1962; Rich, 1951). Gell (1959) reported that although delayed hypersensitivity differs from Arthus-type sensitivity in the time of reaction development, characteristic histological picture, and relation to circulating antibody, each criterion should be looked upon critically. As mentioned above, a mild Arthus reaction might be maximal at 2 to 4 hours and more intense reactions maximal at 24 hours (Waksman, 1960; Bauer and Stone, 1961). A typical delayed reaction is usually maximal within 24 to 36 hours in the guinea pig. During the first 6-8 hours, few or no macroscopic changes occur, although histological changes are evident at 4 hours (Gell, 1959; Follis, 1949). from the start. The Arthus reaction, on the other hand increases The tuberculin skin reaction in mice develops more rapidly, appearing in 5 or 6 hours and approaches a maximum in 8 to 12 20 hours (Crowle, 1959; 1960; 1964). Crowle found that tuberculin reactions in the mouse showed typical induration, but unlike the guinea pig, these reactions were not associated with erythema. reactions in rats appeared to be similar (Crowle, 1962). Skin Another worker (Belsey, 1966) has studied the tuberculin reaction in mice utilizing PPD antigen suspended in adjuvant. He found that reactions were maximal at 48-72 hours with induration often accompanied by erythema at 48 hours. Rich (1951) reported that the establishment of delayed hypersensitivity is dependent upon a number of factors including the number of injected bacilli, the native resistance of the host, and the route by which the bacilli enter the body. In the early 1930's it was thought that the only reliable way to induce tuberculin hypersensitivity was to infect animals with living bacilli. (1929) reported that delayed hypersensitivity to ~gg pienes white or horse serum could be induced in guinea pigs if these antigens were injected into tuberculous foci in the animals. These studies led to the use of water-in-oil emulsions as adjuvants in the production of antibody (Freund and McDermott, 1942; Freund, 1956). Other workers found that killed bacilli induced delayed hypersensitivity (Petroff and Stewart, 1925). Delayed skin reactivity was reported to develop faster if the sensitizing dose was large, whi~e with sufficiently smaller doses development was much delayed (Rich, 1951; Krause, 1916). The appearance of delayed hypersensitivity has been observed in infants 13 as early as three weeks after inoculation with killed tubercle bacilli (Goodwin and Schwentker, 1934). Studies on the immunizing capacity of phenol-killed mycobacteria in guinea pigs have indi.cated that while the appropriate dose would produce a rapid development of immunity, a too large dose inhibited this rapidly developing immunity (Tsukamura and Tsukamura, 1964b). Bjerkedal (1959) injected guinea pigs with small and large doses of virulent tubercle bacilli and concluded that the degree of tuberculin hypersensitivity was independent of the number of infecting bacilli. He further found that the individual guinea pig retained the same level of sensitiveness from a short time after infection until the beginning of the terminal stage of tuberculosis (a period of approximately 60 weeks for animals with a high survival time). Guinea pigs injected with small numbers of organisms have been reported to develop as much hypersensitivity as with heavy doses (Jespersen and Mackep-,:ang'" 1959). Studies of tuberculin sensitivity in man and guinea pigs have revealed that attenuated or avirulent organisms may induce delayed skin hypersensitivity (Meyer et al., 1956; Edwards et al., 1960; Dietz et al., 1963). Steenken et al. (1934) found that a virulent strain (H37Rv) of tubercle bacillus had a greater sensitizing capacity than an avirulent strain (H37Ra). They found that even if the strains were heat killed the same results were obtained. These results suggest that virulent bacilli contain a better sensitizing antigen. Other workers have reported that virulent bacilli contain 14 more organisms per unit weight than does the same weight of avirulent organisms (Kahn and Schwarzkopf, 1933). Thus if equal weights of each killed strain were injected, animals receiving virulent organisms actually received more bacilli hence more sensitizing antigen. Skin test antigens prepared from virulent and avirulent strains of tubercule bacilli show no differences in eliciting tuberculin reactions (Seibert and Morley, 1933). Route of inoculation has been reported to influence the rate of development and degree of sensitivity of delayed skin reactions. Several authors using killed and living bacilli found intravenous injection a less effective pathway for the establishment of delayed hypersensitivity than intraperitoneal, intratesticular or subcutaneous injections (Petroff et al., 1929; Branch and Cuff, 1930; Clawson, 1935). Branch and Cuff (1930) also reported that intramuscular injection induced poor sensitivity. Pappenheimer and Freund (1959) reported that a single injection of killed organisms (if mixed with adjuvant) induced maximal sensitivity whether administered by the intradermal, subcutaneous, intramuscular or intraperitoneal route. They reported further that in the absence of adjuvant, little delayed hypersensitivity was developed. Recent studies (Edwards et al., 1959; Takeya et al., 1961; Takeya et al., 1962; Magnusson, 1961; Magnusson, 1962).have shown that delayed reactions were engendered by subcutaneous, intramuscular and intradermal inoculation. Rich (1951) reported that induction of delayed hypersensitivity by ingestion of bacilli was difficult unless 15 given in high doses and then development was slower than by other routes. Lurie (1964) noted that rabbits developed allergic sensitivity by inhaling virulent t 11bercle bacilli. Other workers have reported a uniform and high level of delayed hypersensitivity in guinea pigs when large doses of organisms were inoculated intratesticularly (Johnson and Smith, 1964; Smith and Johnston, 1964). This brief survey of the literature reveals conflicting reports concerning the influence of inoculation route upon development of delayed hypersensitivity. The natural resistance to infection and inherent capacity to develop delayed hypersensitivity is characteristic for various animal species. The human being is capable of developing a much higher degree of hypersensitivity than is the guinea pig or rabbit. Positive reactions in the human being to a billionth of a milligram of PPD have been reported (Rich, 1951; Furcolow et al., 1941). Rats and mice have been reported to be nearly completely lacking in ability to produce a delayed response although eliciting an Arthus reaction (freund and Stone, 1956). Several other workers have reported delayed intracutaneous reactions with mice (Gray and Jennings, 1955; O'Grady 1957; Crowle, 1962; CroWle and Hu, 1964; Belsey, 1966). In the tuberculous chicken, the intracutaneous injection of tuberculin fails to evoke a local tuberculin reaction in most areas although marked reaction occurs to tuberculin injected in the skin of the wattle or the comb (Feldman, 1938). Temporary alterations in the physiological state of the host 16 may influence the rate of development of delayed hypersensitivity. Certain viral infections such as measles and influenza have been reported to markedly decrease cutaneous hypersensitivity (Bloomfield and Mater, 1919; Mitchell et al., 1935; Brody and McAlister, 1964). Dehydration has been reported to decrease cutaneous hypersensitivity (Pilcher, 1930). Delayed hypersensitivity has been reported to be diminished by administration of vitamins A and C (Birkhaug, 1939; Uhr et al., 1963). Pyridoxine-deficient guinea pigs showed severely depressed systemic reactivity with PPD (Trakatellis et al., 1963). Factors which may influence the development and interpretation of the delayed skin reaction include: repeated tuberculin tests at the same site (Arnason and Waksman, 1963), strength of the testing agent and other factors (Duboczy, 1964; 1965; Pepys, 1955; Maxwell et al., 1964). IV. A. MYCOBACTERIAL CONSTITUENTS Tuberculoproteins The search for the active principle of tuberculin has been carried on since the time of Koch. Koch's experiments (1890) led him to believe that the substance responsible for eliciting the specific toxic effect in the diseased animal was of protein nature. Hammerschlag (1891) was among the first to identify protein as one of the constituents of the tubercle bacillus. Weyl (1891) also isolated a protein and considered it to be a mucin or mucoprotein, differing however from other mucoproteins in that on heating with dilute acid, a reducing substance was not split off. Kuhne (1893) isolated a 17 "propeptone" (probably a polypeptide) from tubercle bacilli; he found however that the same substance could be isolated from the growth medium" Hofman (1894) obtained 6 protein fractions from extracts of tubercle bacilli. globulin. He called two of these albumin and Levene (1898) did not believe that the proteins were albumin or globulins but rather nucleo-proteins. proteins which he called nucleo-proteids. He isolated three Ruppel (1898) isolated a protein from the tubercle bacillus which he called "Tuberkulosaroine." He thought it was bound to a nucleic acid in the tubercle bacillus. Lowenstein and Pick (1911) claimed that the tuberculin product was destroyed by digestion with pepsin as well as with trypsin and therefore had the characteristics of a polypeptide. Pfeiffer, et al. (1911) and Mueller (1925) and Seibert (1926a) also found that pepsin as well as trypsin destroyed the active substance. The contention that the active tuberculin was of a nucleoprotein nature was refuted by the work of Toenniessen (1924). His preparation contained 12.3% nitrogen, no phosphorous, and no purine bases. Masucci and McAlpine (1930) made a tuberculin active preparation from the culture filtrate of~. tuberculosis by precipitating eight times by half saturation with ammonium sulfate, once with 5 volumes of 95% alcohol at pH 4.7, four times at pH 4.7, and finally treatment with barium hydroxide. Gough (1933) obtained an active preparation from tuberculin filtrate by first removing a precipitate at pH 5.0 and then adsorbing the active substance on benzoic acid from which it was freed with acetone. Phosphotungstic and sulfuric acid were 18 used to precipitate the active fractions. Numerous other preparations, considered to be protein have been isolated from extracts of tubercle bacillio Coghill (1926) obtained an active water soluble protein from bacilli previously treated by repeated extraction with cold ether. Dienes et ale (1925) using aqueous and weak alkaline extracts of tubercle bacilli precipitated the proteins with sodium sulfate, acid, or heat. Gough (1934) used a more delicate procedure for isolating active proteins. Living ice cold tubercle bacilli, grown on synthetic medium were extracted with a large excess of absolute alcohol. (1934) also attempted to iso~te in relatively undenatured form. Heidleberger and Menzel tle:proteins of the bacillary bodies They used living bacilli which had been frozen and dried in vacuum, and then killed by immersion in cold acetate buffer at pH 4.0 for 30 days. After this the cells were extracted 4"times with cold redistilled acetone, 3 times with purified anhydrous ether, dried in vacuo, ground for 10 days, reextracted 2 more times with cold ether, and dried. The cell residues were then successively extracted with buffers ranging from pH 4.0 to 11.0, and then finally 0.1, 0.2 and 0.5 N sodium hydroxide at room temperature. Each fraction was then precipitated with acetic acid to maximum flocculation. Among the many fractions obtained, two were identified as antigenic components. The work of Seibert and others (1926a, b) further indicated that the active fraction of tuberculin behaved somewhat like a normal protein. Half-saturation with ammonium sulfate, as well as complete 19 saturation, precipitated active protein fractions from tuberculin filtrates. Seibert (1926c) further found that a small amount of a protein with high tuberculin activity could be crystallized at an optimum pH of 4.9. Very fine needles formed and usually grouped themselves into burrs. The protein was redissolved and recrystallized 14 times with increasing potency. Seibert (1928) studied a potent tuberculin isolated from culture filtrate obtained from organisms grown upon a non-protein synthetic medium. The tuberculin was prepared by precipitation with trichlor- acetic acid (TCA). Activity was correlated with nitrogen content. Later work (Seibert, 1930) revealed that this protein, precipitated by TCA induced precipitin production. Changes in techniques e.g., ether extraction of the TCA precipitate yielded a much less antigenic product called purified protein derivative, PPD (Seibert, 1930; Seibert and Munday, 1932). Seibert and Morley (1933) found that the molecular size of the protein was between 2,000 and 34,000. A more accurate study conducted in Svedberg's laboratory (Seibert et al., 1938) using the ultracentrifuge, led to the conclusion that practically homogenous fractions with molecular weights of 16,000 and 32,000 could be isolated from tuberculin. skin test antigens. Both fractions were highly potent as It was shown that the molecule of 16,000 molecular weight did not stimulate the production of precipitin antibodies, nor elicit a typical Arthus reaction when injected into normal rabbits, whereas the 32,000 molecular weight molecule was antigenic. high~y The small molecule became antigenic after it was adsorbed 20 to aluminum hydroxide and charcoal, indicating that it acted like a haptene. Later work carried on by Seibert and co-workers revealed that the protein (PPD)contained impurities of nucleic acid and polysaccharide (Seibert and Dufour, 1940; Seibert, 1940). Further purification procedures were employed: precipitation with ammonium sulfate, using less heat, and carrying out the entire procedure in the cold room resulting in a more stable, potent less antigenic skin test agent (Seibert, 1940; Seibert, 1941a; Seibert and Glenn, 1941) than previously described (Seibert, 1930). Seibert and Glenn (1941) prepared the standard PPD (PPD-S) which is still employed and accepted as an international tuberculin standard. Seibert (194lb) summarized the findings concerning the relationship of protein and tuberculin activity as follows: 1) coagulable protein appeared in the culture medium at about the same time as tuberculin activity, 2) protein precipitants carried down the activity along with the protein, 3) when the protein fractions were purified the content of nitrogen paralleled the potency, 4) pepsin and trypsin destroyed the potency simultaneously with the breakdown of the protein molecule, 5) the highest potency accompanied the crystallizable protein and was shown to increase during 14 recrystallizations, 6) when protein precipitable by trichloracetic was found in tuberculin ultrafiltrates, there was also tuberculin activity in the filtrates. Otherwise there was never more than a trace of material of lower potency, 7) the highest potency accompanied protein molecules 21 of different sizes which were shown to be practically homogeneous in sedimentation and diffusion, 8) the active principle migrated with protein in electrophoresis. Although tuberculin activity was definitely correlated with the protein of the tubercle bacillus later studies revealed the presence of at least two polysaccharides and three different proteins (Seibert, 1949; Seibert et a1., 1955; Seibert and Soto-Figueroa, 1957). The polysaccharides were designated by roman numerals I and II, while the proteins were labeled A, B, and C. 4 gave the C fraction first. Simple precipitation at pH This was the least potent as a tuberculin. The B fraction had a somewhat similar mobility to the C fraction, however it was soluble at pH 4 and was precipitated on addition of 30% alcohol. The A "fraction had the lowest mobility and required 70% alcohol at pH 4-6 for precipitation. It was heavily contaminated with carbohydrate, in some cases the carbohydrate content was as high as 20%. Seibert and co-workers (Seibert, 1953; Seibert and Soto-Figueroa, 1957) further investigated properties of these fractions. fraction. The most potent material was always present in the A When used as a skin test antigen, it gave larger skin reactions in infected pigs per mg of nitrogen than::did PPD-S the C antigens gave less. while By the complement fixation test the C fraction appeared to be the most reactive antigen. Electrophoretic studies indicated that each of the groups of components designated as A, B, and C proteins probably consisted of more than a single protein with similar electrophoretic mobilities. 22 The question of the possible antigenicity of tuberculo-protein led to many experiments. Stacey (1955) reported that 4 components could be isolated from tubercle bacilli treated with a saturated solution of urea. One of the fractions was extensively purified and contained approximately 5% carbohydrate. This compound gave a typical tuberculin skin reaction in Mantoux positive humans and in tuberculous guinea pigs. complement fixation test. It also gave positive results in the This fraction yielded positive serologic reaction with sera from rabbits injected with the extract or with killed M. tuberculosis cells. Furthermore guinea pigs injected with the urea extract showed typical skin reactions when tested with old tuberculin. The product consisted mainly of one electrophoretic component which closely resembled Seibert's C protein. It was however more potent and similar to the A protein in eliciting skin reactions in tuberculous animals. Following Stacey's presentation, he was questioned by Raffel, Rich and Hanks concerning the statement that the urea-extracted protein induced delayed skin sensitivity. These men were critical of the finding that tuberculo-protein could induce delayed hypersensitivity. Raffel (1946, 1948), in a series of experiments, demonstrated that although injections of tuberculoprotein induced the production of humoral antibody, cellular antibody or delayed sensitivity was not produced unless certain lipid fractions were also injected. This work was corroborated by Crowle (1958). Smith and co-workers (Erickson and Smith, 1962; Smith et al., 1964) found that defatted mycobacteria induced a high degree of resistance 23 as well as delayed skin sensitivity if injected with Bayol-Arlacel adjuvant. Purified protein derivatives {PPD's} have been prepared from other mycobacterial species by the Seibert method (Affronti, 1961). Trichloracetic acid precipitation of heated culture filtrates yielded non-antigenic substances (i.e. substances that do not induce humoral antibody) capable of eliciting delayed hypersensitivity (Magnusson, 1961; Magnusson et al., 1961b; Magnusson, 1962). Magnusson has proposed that substances of this type be called "sensitins" (Magnusson, 1961). Takeya and associates employed a combination of ammonium sulfate and trichloracetic acid to precipitate a protein which they designated 11 which was also non-antigenic and capable of eliciting delayed hypersensitivity (Takeya et al., 1960; Takeya et al., 1961; Takeya et al., 1962; Sinnaka et al., 1962). A tuberculin active peptide (TAP) has been isolated from acetone-washed bacterial cells by precipitation with picric acid (Someya et al., 1963). All of the above preparations (PPD, TPT,1T, TAP) were reported to be relatively non-antigenic and capable of detecting delayed hypersensitivity. Their activity was specific in that homolQgous reactions were usually larger; however in each case cross reactions were observed. It may be concluded that the literature supports the concept that the skin test "active" fraction isolated from the tubercle bacillus is probably a mixture of proteins. Tuberculo-proteins are unusual proteins in that their resistance to harsh physical and chemical treatments is high. In addition they are considered by most 24 investigators to be either weakly or non-antigenic. B. Polysaccharides The polysaccharide components of the tubercle bacillus are complex in structure. Stacey (1955) broadly classified the polysaccharides into 3 groups as follows: a) those found in the culture filtrates; b) those found in combination with lipids and extracted with the fats and waxes with organic solvents; c) those occurring in the somatic region of the defatted cell. As previously mentioned the polysaccharides associated with tuberculins were extensively studied by Seibert and her associates (Seibert et al., 1949; Seibert, 1949; Seibert, 1950). It appears from their work that the polysaccharides were not concerned with the tuberculin skin reaction. The poly- saccharide termed polysaccharide I contained units of mannose, galactose and arabinose and was serologically active in a precipitin against rabbit and horse anti-tuberculous sera. By means of ultra- centrifuge studies it appeared that the molecule was an elongated ellipsoid in shape and had a molecular weight of about 9,000. saccharide II has also been iso~ted. Poly- It gave high precipitin titers with horse, rabbit and human antituberculous sera. It was also able to induce antibodies on injection into rabbits, although no protection was demonstrated. Polysaccharide II appeared similar to glycogen in that it had a high molecular weight, gave an opalescent aqueous solution and had high optical activity. It contained a low proportion of an amino sugar but was mainly a polyglucose. Polysaccharides have been isolated from the somatic components of defatted cells. It is 25 probable that some of these sugars are the ribose and deoxyribose constituents of the nucleic acids. have also been detected. Cellulose, glycogen, and pentosans The glycogens were serologically inactive in precipitin tests against tuberculosis antisera (Seibert, 1950). Heidelberger and Menzel (1939) isolated 2 serologically active polysaccharides together with serologically inactive polysaccharides. G1ucosamine and D mannose appeared to be absent from the specific compounds. There appeared to be some evidence that specificity was related to the pentose content of the polysaccharides (Burger, 1950). A heat stable polysaccharide when adsorbed onto sheep erythrocytes, rendered them specifically agg1utinab1e with antisera from patients with tuberculosis (Middlebrook and Dubos, 1948). Crowle (Crowle 1963; Crowle and Hu, 1965) isolated a waters'olub1e trypsin extract from tubercle bacilli which protected mice against tuberculosis as effectively as entire bacilli. The substance termed "immunogen" appeared to be a high molecular weight polysaccharide. Evidence indicated that the immunogen was a very poor inducer of humoral and cellular antibody. The immunogen was not active in detecting delayed hypersensitivity (Crowle, 1966). Chaparas and Baer (1963) isolated a carbohydrate component from tubercle bacilli which was capable of eliciting a tuberculin reaction indistinguishable from that elicited by protein fractions. Later work (Chaparas and Baer, 1964) revealed that the carbohydrate was contaminated with approximately 10 per cent protein; however digestion with the proteolytic enzyme Pronase indicated that the 26 carbohydrate component did exhibit tuberculin activity (Chaparas and Baer, 1965). It may be concluded that a fundamental disagreement exists currently concerning the capacity of specific polysaccharides to elicit the delayed hypersensitive response. On the other hand, there is general agreement concerning serologic reactivity of these polysaccharides. C. Lipids Three types of fatty acids have been found in mycobacteria. The straight chain fatty acids, principally palmitic and hexacosanoic acid were reported by Asselineau and Lederer (1935). No fatty acids higher than hexacosanoic acid seem to have been isolated from the mycobacteria. The second group consisted of methyl-branched fatty acids such as tuberculostearic acid, phthoic or phthienoic acid which contained from 23-31 carbon atoms. The third group were long-chain branched fatty acids, the mycolic acids (Stodola et al., 1938). These were ether-soluble hydroxyacids of molecular weight about 13,000 containing about 88 carbon atoms, and seemed characteristic for mycobacteria. Some strains of mycobacteria seem to contain only two mycolic acids, some 3 or 4. Most of the mycolic acids were present in the organism as esters of nitrogen and phosphorous containing polysaccharides. It was shown (Anderson and Creighton, 1939) that mycolic acids could be fractionated into two principle fractions, both of which were acid-fast and very similar in their solubilities. Anderson (1932) 27 described the isolation and properties of 4 wax fractions from the tubercle bacillus. Waxes A, B, and C were similar in constituents while wax D was distinctive in being composed of a lipopolysaccharide, an ester of mycolic acid with a nitrogen and phosphorus containing polysaccharide of unknown molecular weight. Wax A and wax B apparently played little or no role in the production of delayed hypersensitivity. However there was evidence that wax C and wax D did take part in the induction of delayed hypersensitivity. Analysis of wax D has shown it to contain 50 per cent mycolic acids, and 50 per cent nitrogenous polysaccharides (Willett, 1963). Choucroun (1947) found that treatment with paraffin oil removed from tubercle bacilli, two fractions which had the capacity to produce hypersensitivity or lesions. The "toxic" or lesion producing fraction contained the following elementary composition: carbon 53%, hydrogen 10%, nitrogen 1%, phosphorus 0.4%. No protein was detected. It was concluded that the "toxic" fraction was a polysaccharide ester of the mycolic acid isolated by Anderson. Chaucroun's secnnd fraction or "sensitizing lr fraction, proved to contain a protein component, very insoluble in the usual organic solvents. The "sensitizing tr fraction was capable of establishing true delayed type of hypersensitivity when injected in normal animals. Evidence was presented of increasing longivity and reduced lesions in tuberculous guinea pigs which had been treated with a mixture of the protein and 1ipo-carbohydrate complex (PMKo). It was found that the lipopolysaccharide complex contained a tripeptide with a single free amino group consisting 28 of aspartic acid, alanine and an unknown substance. This tripeptide was also found in the "toxic" fraction. Raffel (1946) showed that a purified wax fraction plus tubercu10protein was able to induce allergic sensitization to old tuberculin whereas neither fraction alone was able to do so if highly purified. In an experiment to determine the effects of the chemical constituents Raffel (1946) employed living BCG, bacilli killed by moderate heat, bacilli deprived of phosphatide and wax by ether-a1coho1-ch10roform extraction, and defatted bacilli with lipid re-added. phatide and a wax preparation were also tested. molecular weight protein was used. Crude phos- In addition a high These constituents were injected into guinea pigs which were subsequently tested for immediate and delayed hypersensitivity and resistance to infection with highly virulent strains of tubercle bacilli. The experiments were repeated several times with the same results. The heat-killed bacilli and the defatted bacilli plus wax alone were able to produce the delayed type of hypersensitivity with the complete absence of resistance. These results indicated that sensitization of the tuberculin type was due to the adjuvant effect of the bacillary wax. It was found that injection of large doses (up to 10 mg) of defatted bacilli failed to induce delayed hypersensitivity. Smith and Robertson (1962) found wax D and PMKo to be ineffective in immunizing guinea pigs against tuberculosis. Sabin et a1. (1931) found that both the highly purified phosphatide and wax fractions c~used the production of epithe10id cells, 29 giant cells, and tubercles. The phosphatide was more active in inducing this type of reaction. Takahashi et ale (1961) found that antiphosphatide was produced chiefly under conditions where tubercle bacilli might have undergone destrUction. The level of antiphospha- tide was found to reflect most faithfully the progression of experimental tuberculosis infection. These results were confirmed in a study of patients with tuberculosis (Takahashi, 1962). The significance of the phosphatide antibody is not known. There was no evidence that the presence of this antibody increased the resistance to tuberculous infection; nor has the phosphatide antigen been found to establish the hypersensitive state. v. NEW METHODS FOR PREPARATION OF MYCOBACTERIAL ANTIGENS The discovery that mycobacteria other than M. tuberculosis might induce tuberculin sensitivity has led many investigators to study the specificity of mycobacterial antigens. While several authors have utilized the same principles of preparation as previously described i.e., protein precipitation, particular emphasis has been placed upon optimum production of antigens with as little biochemical change as possible. In addition studies have been made concerning the preparation of cell walls and protoplasm as possible skin test and immunogenic antigens. Magnusson and co-workers investigated various aspects of tuberculin production. It was found that optimum production of tuber- culoprotein and tuberculin activity was obtained from 9 week old cultures, if the pH of the culture medium was maintained at 7.6 - 8.6 30 (Kim et a1., 1964; Magnusson et a1., 1964). Others (Castelnuovo, et al., 1964; Tepper, 1965) have found that the antigenic components of a culture filtrate varied with time of growth. The number of antigens increased until a maximum number was reached, after which certain antigens disappeared. It was suggested that more stable antigenic preparations could be isolated by extraction of whole tubercle bacilli. Kim et a1., (1963) reported that large amounts of protein could be isolated from cells, although the skin test activity was low. However others (Warfvinge, 1962) have isolated potent skin test antigens from cells. Various techniques have been employed for isolating cell fractions including: use of lytic enzymes (Wilson and Slepecky, 1962), pressure cells (Youmans and Youmans, 1964; Ribi et a1., 1959, 1965), freezing and thawing (Warfvinge, 1962) and mechanical disintegration (Kwapinski, 1963). Endotubercu1in, isolated from virulent living tubercle bacilli (Warfvinge, 1962) by freezing-thawing, sensitized guinea pigs to tuberculin, however, after lyophilization the sensitizing capacity was lost. At least 6 different fractions, obtained by precipitation exhibited tuberculin activity. The authors reported good specificity i.e. maximum response in persons infected with tubercle bacilli and no response in the non-infected individual. Endotubercu1ins prepared from other mycobacterial species were not studied. The use of mycobacterial cell walls, cell wall extracts. and protoplasmic extracts as immunizing agents has been well studied 31 (Ribi et a1., 1958; Ribi et a1., 1959; Larson et a1., 1963; Ribi et a1., 1965). These reports did not include studies of fractions as possible skin test antigens. Protoplasmic fractions were not as biologically active as cell wall preparations. A "cytoplasmic" particulate fraction (Youmans and Youmans, 1964) obtained by pressure disruption, was found to be immunogenic. This fraction was found to be quite labile to heat, freezing and thawing, sonic oscillation and the effect of nyp~tonic solutions. Numerous techniques have been described for the separation and purification of biological materials, e.g., see recent reviews (Weyer, 1966). A high degree of fractionation has been obtained with the'use of Sephadex (Pharmacia). Sephadex is a polymer of dextran which may be obtained in various degrees of cross-linkage. The degree of cross-linkage controls the pore size, and hence the fractionation ranges. Sephadex has been employed in the separation of mycobacterial antigens by several workers (Chaparas and Baer, 1964; Baer and Chaparas, 1963; G1enchur et a1., 1965). G1enchur et a1. (1965) isolated at least 9 fractions, presumed to be antigenic, from bacillary extract of avirulent tubercle bacilli. One of these fractions when utilized as a skin test antigen evoked an immediate or Arthus type reaction. Later reports (G1enchur et a1., 1966) indicated that 2 of 7 fractions tested were capable of eliciting delayed hypersensitivity. This work further confirmed the hetero- 32 geniety of PPD. At least 7 protein fractions were obtained by Sephadex and ion-exchange chromatography. Not all fractions were antigenic. Baer and Chaparas (1963) utilizing Sephadex chromatography in conjunction with dialysis procedures isolated tuberculin active substances from the culture filtrate of BCG. More recent studies (Chaparas and Baer, 1964) indicated at least 7 fractions were obtained by Sephadex fractionation. Gel diffusion studies with these fractions revealed up to 10 antigenic components found in each fraction. Knicker and LaBorde (1964) isolated at least 20 antigens from culture filtrates of mycobacterial species by means of ion-exchange chromotography. activity. These antigens were not tested for skin test Gel diffusion studies revealed that almost one-fourth of the antigens were species specific i.e. specific for the strain from which they were isolated. Yoneda (1964) reported successful purification of two antigenic proteins from the culture filtrate of M tuberculosis by Sephadex chromatography. A new technique of electrophoresis (disc electrophoresis) based upon discontinuities in the electrophoretic matrix has recently been described (Ornstein, 1964; Davis, 1964). Like Sephadex chromatography, separation of protein molecules is obtained by a molecular sieving process. Pore size is determined by cross-linkages formed during the polymerization of acrylamide and methylenebisacrylamide (Burns, 1966). Another aspect of disc electrophoresis, "thin starting 33 zones," regulates the formation of concentrated protein at a specific boundary (Ornstein, 1964) i.e. all the proteins in a sample are concentrated into very small discs, one stacked upon the other in order of decreasing mobility. The thin starting zones and sieving effect of the gel provide high resolution of protein samples. Ultracentrifugation and free-boundary electrophoresis studies have shown that disc electrophoresis s'eparated proteins both according to charge and size of the molecule (Williams and Reisfeld, 1964). Because of the relatively recent discovery of disc electrophoresis, many reports are concerned with modifications of equipment and influences of reagents (Raymond and Nakamichi, 1964; Weinstein and Douglas, 1964; Hjerten et al., 1965b; Matson, 1965). Separation of biological materials by disc electrophoresis has been attempted with much success. Soluble proteins of certain plants have been isolated (Steward and Barber, 1964), phosphorylase enzymes in algae have been studied (Fredrick, 1964), the proteins of brain and pineal tissue have been investigated (Pun and Lombrozo, 1964), and ribosome fractions have been isolated from!. coli by these means (Hjertein et al., 1965a). Affronti et al. (1965) studied various fractions of mycobacteria by means of disc electrophoresis. Culture filtrate and fractions obtained from filtrate by ammonium sulfate and alcoholic precipitation were analyzed. From ten to fourteen bands were obtained from ammonium sulfate fractions. nineteen bands were seen. At an optimum protein concentration, Polysaccharide fractions, when separated 34 by disc electrophoresis yielded two to three wide but distinct bands. It was concluded by these workers (Affronti et al., 1965) that polyacrylamide electrophoresis offered several advantages for separation of mycobacterial proteins: the system was reproducible, only small samples were required, and specific stains could be used for identification of the separated components. Another possible advantage is that the system could be adapted to large-scale or preparative production (Racusen and Calvanico, 1964; Jovin et al., 1964; Lewis and Clark, 1964). VI. SUMMARY It may be concluded that the literature supports the concepts that currently available mycobacterial skin test antigens are sensitive but not adequately specific and that continued search for more specific reagents is indicated. It may be further concluded that among presently available techniques, polyacrylamide gel electrophoreses might be a sensitive tool for the isolation of mycobacterial antigens. MATERIALS AND METHODS I. CULTURE GROWTH A strain of M. kansasii (P-16) and of~. tuberculosis (H37Ra) were obtained from the Veterans Administration Hospital, Salt Lake City, Utah. The organisms were grown in 1 liter volumes of modified Sauton liquid medium (Tsukamura and Abo, 1958). Cultures were shaken continuously during the first week of growth, after which they were allowed to stand for 6 weeks while pellicle formation occurred. Cultures were treated with phenol to a final concentration of 0.5 per cent and allowed to stand 72 hours. The formula for Sauton medium is as follows: Modified Sauton Medium Glycero 1. . . . . . . . . . . . . . . . . . . • . . . . .. Citric Acid ...•..•.......•........ Magnesium Sulfate ...•............. Dipotassium Phosphate ............. Ferric Ammonium Citrate ....•...... Sodium Glutamate .........•........ Water. . . . . . . . . . • . . . . . . . . . . . . . . . . .. I I. 50 ml 2.0 g 0.5 g 0.5 g 0.05 g 4.0 g to 1 Ii ter HARVEST TECHNIQUE The mycobacterial cells were separated from the culture medium by filtration. The filtrate was sterilized by Seitz filtration and concentrated to approximately one-tenth of the original volume by flash evaporation pervaporation. The cells were washed 3 times in distilled water and subjected to freezing and thawing (-60 0 C to 37 0 C) conditions at least 21 times. The cells were then subjected to ultrasonic disruption for 36 20 minutes at 20,000 cycles per second. The cell extract was obtained by centrifugation for 30 minutes at 10,000 rpm in an International Model HR-l centrifuge" The 858 centrifuge head was employed. The extract was concentrated by pervaporation and further clarified by filtration through membrane filters of 0.3 micron pore size. The remaining cells and cell walls were subjected to ball-mill grinding for 48 hours. The extract was concentrated by pervaporation and filtered through membrane filters of 0.3 micron pore size. fractions were stored in the refrigerator at 4 0 C. All The flow sheet for the outlined procedure is illustrated in Figure 1. III. NITROGEN DETERMINATION Total nitrogen was determined utilizing a standard microKjeldahl technique (Kabat and Mayer, 1961). One ml samples were digested by boiling in a solution of potassium sulfate, copper sulfate and sulfuric acid for 20 minutes. The digested samples were steam-distilled and allowed to collect in boric acid solution. The ammonium borate was titrated with standardized hydrochloric acid. IV. PROTEIN DETERMINATION Total protein was determined by a micro modification of the Lowry technique (Kabat and Mayer, 1961). A solution of alkaline copper tartrate was added to 0.5 ml of sample and allowed to stand for 10 minutes. Commercially available Folin-Ciocalteau reagent was added with vigorous mixing. 750 mu. Readings were made in a Beckman DU at A purified protein derivative solution containing 0.5 ug/ml Mycobacteria Cultivated on Modified Sauton Medium I ~ Filtration ~ , Culture Filtrate Freezing-Thawing (-60 0 C to 37 0 C) t ! Seitz Filtration i Ultrasound Disruption (20,000 CPS for 20 min.) Concentration t J, Refrigeration 1- Centrifqgation Extract t t Sterilization Concentration t Refrigeration I Cells-Cell Walls J,. Ball-Mill Grinding .j" Concentration ~ Sterilization t Refrigeration Fig. 1. Flow Sheet for Isolation of Culture Filtrate, Cell Wall Extracts and Cell Extracts from M. kansasii. V-'> -....J 38 was employed as a standard. V. CARBOHYDRATE DETERMINATION Carbohydrate was calculated as glucose using anthrone reagent (Koehler, 1952; Kabat and Mayer, 1961). Anthrone reagent was added to 2.5 ml of sample with vigorous mixing. for 10 minutes and allowed to cool. The solution was boiled Readings were made at 600 mu utilizing a Coleman Jr. spectrophotometer. VI. NUCLEIC ACID DETERMINATION Deoxyribonucleic acid (DNA) was determined by a method outlined by Seibert (1940) with the exception that the tubes were boiled for 15 minutes instead of incubating for 24 hours. Tests were performed by addition of diphenylamine reagent to 0.6 ml of sample; they were boiled for 15 minutes, and read at 600 mu. by comparison with a DNA standard. Values were determined This method has been reported to detect as little as 0.05 mg of DNA. VII. A. ACRYLAMIDE ELECTROPHORESIS Equipment A disc electrophoresis assembly (Canalco Model 12) consisting of upper and lower buffer tanks with electrodes was employed. A constant current, high voltage power supply (Canalco, Model 1400) was utilized as a power source (Figure 2). A fluorescent illuminator was employed to aid in polymerization of gels. Fig. 2. Photograph of disc electrophoresis assembly including upper and lower buffer tanks and power supply. Note descending dye front in columns. 40 B. Reagents Five stock solutions were prepared and stored in the refrigerator. Solution A C. 1 N Hydrochloric acid Tris (hydroxymethy1) Aminomethane N, N, N', N' tetramethy1ethy1ene diamine Water 48 m1 36.3 g 0.23 m1 to make 100 m1 Solution B 1 M Phosphoric acid Tris (hydroxymethy1) Aminomethane Water 48 m1 5.7 g to make 100 m1 Solution C Acrylamide monomer N, N', methy1enebisacrylamide monomer Water 30.0 g 0.8 g to make 100 ml Solution D Acrylamide monomer N, Nt, methylenebisacrylamide monomer Water 10.0 g 2.5 g to make 100 ml Solution E Riboflavin Water 4.0 mg 100 ml Preparation of Gels Lower or running gels were prepared by addition of 1 ·part solution A to 2 parts of solution C to 1 part water (1:2:1). An equal volume of 0.56 per cent ammonium persulfate was added to the above mixture. The best results were obtained when the ammonium persulfate was prepared fresh daily. Spacer or stacking gels were prepared by addition of 1 part B to 2 parts D to 1 part E (1:2:1). Sample gels were prepared by diluting 1 part spacer gel with 2 parts of the sample. 41 D. Loading of Column A rubber cap was placed over one end of a straight glass tube (65 mm by 5 mm i.d.). The tubes were then placed in a vertical position under the spring clamps in the slots of the fluorescent illuminator (Figure 3). The tubes were filled to approximately 1/2 inch from the top with 0.9 ml of lower gel. Water was carefully layered on top of the lower gel in order to form a flat meniscus. The lower gel was allowed to polymerize for 30 minutes after which the water was briskly shaken off. Two-tenths ml of undiluted spacer gel was added and layered with water as previously described. After 15 minutes the water layer was shaken off and sample gel added and allowed to polymerize for 15 minutes (Figure 4). E. Electrophoresis The columns were placed in the styrene grommets of the upper bath assembly and electrophoresis was carried out with a current of 5 milliamperes per column. Glycine buffer pH 8.3 was utilized. The formula for glycine buffer utilized throughout the thesis work is as follows: Glycine Buffer pH 8.3 lOX Tris (hydroxymethyl) Aminomethane ..• 6.0 g Glycine .•.•...•.......•....•.•••.... 28.8 g Water •••...••••...•........•.••.••.. to make 1 liter After the completion of electrophoresis the power supply was shut off and gel tubes removed from the upper bath. Gels were removed from the tube by gently rimming the tops of the gel with a 22 gauge hypodermic needle through which a moderate stream of water Fig. 3. Photograph illustrating photopolymerization of acrylamide gels with the use of a flourescent 1 ight source. ---- Samp leGe I ---- Spacer or Stacking Gel - - - Runn i ng or Lower Ge I Fig. 4. Photograph of acyrlamide column after polymerizatio~ showing relative positions of lower gel, spacer gel and sampl e gel. 44 flowed. Each gel was then placed in Amido-Black staining solution for at least one hour. Amido-Black Stain Amido-B lack. . . . . . • . • .. 3.3 g Acetic Acid ........•.. 300 ml Solution diluted 1:2 with distilled water before use. F. Destaining The gels were removed from the stain and washed twice in 7.5 per cent acetic acid. They were then placed in destaining tubes (eye dropper tubes) and the tubes placed in the grommets of the upper tank. Both upper and lower tanks were then filled with 7.5 per cent acetic acid and destaining was carried out at 12.5 milliamperes per tube. hour. The unbound dye was usually cleared in approximately one At the end of the destaining procedure the gels were removed and placed in small test tubes containing 7.5 per cent acetic acid. Gels were also stained with periodic acid Schiff stain (PAS) for glycoproteins and polysaccharides and Feulgen reagent for DNA. G. PAS S'fain On completion of the electrophoretic run, the gels were placed in 7.5 per cent acetic acid at room temperature for 1 hour. They were then placed in 0.2 per cent periodic acid in the refrigerator for 1 hour. The periodic acid was removed electrophoretically with 7.5 per cent acetic acid for 1 hour. in The gels were then placed Schiff's reagent until red bands appeared (usually 24 hours). The formula for preparation of the Schiff reagent follows: 45 Schiff's Reagent (Feulgen Reagent) Basic fuchsin ..•.••.•••..•....•.•..•... 0.5 g Potassium metabisulfate .•.•••••..••.•.. 1.0 g Water. " ~ ~ 100 ml 0 0 • " n , 0 " 0 " 0 0 " " 0" 0" 0 0 " , " " 0 n " Shake until straw colored; mix with activated charcoal (0.25-0.5 g); shake, filter and store in the refrigerator. H. Feulgen Staiq After completion of the 'electrophoretic run, the gels were placed in 20 ml of ice cold 1 N hydrochloric acid in the refrigerator for one-half hour. Gels were then placed in IN hydrochloric acid pre-heated and maintained at 60 0 C for 12 minutes. The gels were then placed in Schiff's reagent in a closed container at room temperature for 1 hour. I. Calculation of Rf Values The position of each band was expressed in terms of Rf values as used in conventional chromatography. Here the Rf value was considered as the ratio of the distance traveled by a particular fraction to that traveled by the tracking dye front. J. Isolation of Protein Fractions Unstained gels were divided into 4 major protein containing areas as predetermined by calculation of Rf values. Each area was placed in 10 ml of 0.066 M phosphate buffer pH 7.0 and allowed to stand in the refrigerator for 24 hours. The eluate was then filtered through a membrane filter (0.3 micron pore size) and placed in cellulose dialysis bags and dialyzed against 30 per cent polyvinylpyrrolidone (PVP). The protein was then reconstituted to one-half the original 46 volume, sterilized by membrane filtration and stored in the deep freeze. VIII. GUINEA PIG INJECTION Male Hartley strain guinea pigs, 400 to 500 grams, were obtained from a local supplier. Animals were injected with inocula prepared by suspending actively growing, 2 week old cultures in Dubos Broth. The suspensions were adjusted to contain approximately 1 mg/ml wet weight as determined by the Hopkins tube method. Dubos Broth Base Formula in grams per liter Tripticase •••••••••••.••••••••• Asparagine .•.••••.•••••••••.••• Monopotassium phosphate ••.••.•• Disodium phosphate •.••••••••.•• Ferric ammonium citrate ...••••• Magnesium sulfate •••.••..•••••• Calcium chloride ••••••.•••••••• Zinc sulfate .•.•.•.••••.•.••••• Copper sulfate ••.•.•••••••••.•• Polysorbate 80 •••••••••••••.••• IX. 0.5 2.0 1.0 2.5 0.05 0.01 0.0005 0.0001 0.0001 0.2 ROUTES OF INOCULATION Attempts at sensitization with the following routes of inoculation were carried out. A. Subcutaneous 1 mg bacilli injected in the back of the neck in a single dose. B. Subcutaneous in Adjuvant 1 mg bacilli suspended in Bayol oil injected in a single dose in the back of the neck. 47 c. Intradermal 1 mg bacilli inoculated in nuchal area in 4 injections of 0.25 ml each. D. Intratesticular 0.5 mg bacilli injected in each testicle. E. Intracardiac 1 mg bacilli injected by cardiac puncture under ether anesthetic. F. Aerosol A model A3 TRI-R airborne injection apparatus was utilized for aerosolization of animals. The guinea pigs were exposed to an aerosol containing droplet nuclei of less than 6 microns for 1 hour at an air flow of 33 liters per minute. G. .!!y. Mouth Animals were individually fed 1 mg of bacilli by means of a syringe and bulbed needle. x. SKIN TESTS Skin tests were performed on the shaved flanks of 5 guinea pigs 3 weeks after sensitization. Tests were made by the intradermal injection of 0.1 ml of 25 TU strength (0.0005 mg protein) Purified Protein Derivatives (PPD's). erythema. Reactions were read as mm of The PPD's were obtained through the courtesy of Dr. Lydia Edwards of the United States Public Health Service. 48 XI. ~ PPD Organism PPD-S PPD-Y PPD-A M. tuberculosis M. kansasii M. avium TESTS WITH FRACTIONS OBTAINED BY DISC ELECTROPHORESIS Skin tests were performed as outlined above. Six animals were sensitized by the injection of organisms suspended in Bayol oil. Reactions were as mm erythema 6, 12, 24 and 48 hours after test. XII. BIO-ASSAY OF ANTIGENS Potency of the isolated fractions were determined by comparing dilutions of the test samples with dilutions of the standard PPD. Assay design was a conventional 3 x 3,.i.e., 3 dilutions of antigen (standard and test) in 6 test sites. Some assays were carried out utilizing a Latin square design to eliminate differences between pigs and possible differences between sites. Other assays used a randomized block design to eliminate differences between pigs. Calculations were performed as outlined by the Pharmacopeia of the United States of America XVII. Computations were performed by computer analysis by and through the courtesy of Dr. E. Fingl, Department of Pharmacology, College of Medicine, University of Utah. XIII. DETECTION OF ANTIBODY The possible antigenicity of the mycobacterial fraction was tested by use of sensitized latex particles. Polystyrene latex (0.81 micron diameter) particles were obtained commercially. particles 49 were sensitized by addition of 2 parts Old-Tuberculin antigen (1:50) to 1 part latex (1:20). The suspensions were allowed to stand for 24 hours at 37 0 C and then washed twice in 001 per cent bovine serum albumin, pH S.4. Serologic tests were carried out using twofold dilutions of sera in 0.5 m1 of tris (hydroxymethy1) amino-ma1ete buffer. tube. One-tenth of a m1 of sensitized latex was added to each The suspensions were then incubated at 37 0 C for one hour, after which they were centrifuged at SOD G for 5 minutes. Agglutina- tion of the latex particles was easily visible when the tubes were viewed against a dark background with the aid of a fluorescent light (Dietz et a1., 1966). EXPERIMENTAL RESULTS I. EXPERIMENTS DES IGNED TO TEST !!!§ INFLUENCE OF INOCULATION ROUTE ON SPECIFICITY OF MYCOBACTERIAL ANTIGENS A. Rate of Development of Delayed Hypersensitivity Skin test reactions of guinea pigs sensitized by various inoculation routes were read 2, 4, 6, 8, 10, 12, 16, 18, 24, 48, and 72 hours after intradermal injection of the skin test agent. The average skin reactions of 5 guinea pigs are plotted for each time interval in Figures 5 through 16. Guinea pigs injected with~. tuberculosis (H37Ra) gave maximum reactions to PPD-S no matter the route. Maximum reactions were found to develop between 24 and 30 hours after injection. Generally, reactions developed rapidly, with reactions to the homologous antigen (PPD-S) greater than 10 rom within 8 hours after injection. After the peak reaction at 24-30 hours, reactions decreased slowly but steadily and at 72 hours were faded and difficult to read. Aerosol challenge with H37Ra failed to induce skin sensitivity (Figure 10). The rate of development of skin reactions when M. kansasii was injected was similar to that of M. tuberculosis. Maximum reactions were measured at 24-30 hours after the intradermal test injection. Reactions persisted, although faded, up to 72 hours after skin test injection. Aerosol challenge (Figure 16) induced positive reactions to PPD-Y which were maximum 8 hours after test, and faded at 48 hours. M. TUBERCULOSIS STRAIN H37Ra CHALLENGE ROUTE SUBCUTANEOUS IN BAYOL OIL ~ 20 ~ 8 ~ til ~ 15 H ,--- 8 u "'- F:t: PPD-S til ~ fj 10 ~ 5 5 10 15 20 25 30 35 40 TIME (HOURS) 45 50 55 60 Fig. 5. Rate of development of hypersensitivity reaction. (rom) for 5 guinea pigs injected 3 weeks before skin tests. 65 Mean 70 VI ..... M. TUBERCULOSIS STRAIN H37Ra CHALLENGE ROUTE SUBCUTANEOUS 1 20 ~ r:iI ......... ~ 15 H 8 CJ ,.:( ~ f) 10 ~ 5 PPD-Y 5 10 15 20 25 30 35 40 TIME (HOURS) 45 50 55 60 Fig. 6. Rate of development of hypersensitivity reaction. (rom) for 5 guinea pigs injected 3 weeks before skin tests. 65 Mean 70 VI N M. TUBERCULOSIS STRAIN H37Ra CHALLENGE ROUTE INTRATESTICULAR i ~ -S 20 r:t:I H 15 8 () ~ ~ ~ o 10 ~ 5 J// 5 10 PPD-Y 15 20 25 30 40 45 35 TIME (HOURS) 50 55 60 Rate of development of hyperse~sitivity reaction. Fig. 7. (rom) for 5 guinea pigs injected 3 weeks before skin tests. 65 Mean 70 VI w M.. TUBERCULOSIS - ~ 20 CHALLENGE I STRAIN H37Ra ROUTE INTRACARDIAC ~ ~r:r:1 - ~ 15 H E-t I () '\ PPD-S ~ PPD-Y ~ f2 r:a:.. 10 0 Ir ~ ------- \ 5 "-- 5 10 15 20 25 PPD-A 30 35 40 45 50 65 70 55 60 TIME (HOURS) Fig. 8. Rate of development of hypersensitivity reaction. Mean (rnm) for 5 guinea pigs injected 3 weeks before skin tests. VI +' M. TUBERCULOSIS STRAIN H37Ra CHALLENGE ROUTE INTRADERMAL ~ ~ 20 ~ ril ~ 15 H 8 U f:( ~ ~ 10 ~ 5 10 70 50 55 60 65 35 40 45 TIME (HOURS) Fig. 9. Rate of development of hypersensitivity reaction. Mean (rom) for 5 guinea pigs injected 3 weeks before skin tests. 5 15 20 25 30 U'1 U'1 M. TUBERCULOSIS STRAIN H37Ra CHALLENGE ROUTE AEROSOL ~ 20 ~ ~til - Z 15 o H E-t () < ~ ~ 10 ~ 5 5 10 15 20 25 30 35 40 45 TIME (HOURS) 50 55 60 Fig. 10. Rate of development of hypersensitivity reaction. (rom) for 5 guinea pigs aerosolized 3 weeks before skin tests. 65 70 Mean VI 0'< M. KANSASII STRAIN 79 20 CHALLENGE ROUTE SUBCUTANEOUS IN BAYOL OIL ~ ~ 15 ~ fiI 6 H 8 () 10 ~ ~ ~ 5 5 10 15 20 25 30 35 40 TIME (HOURS) 45 50 55 60 "Fig. 11. Rate of deve.lopment of hypersensitivity reaction. (rom) for 5 guinea pigs injected 3 weeks before skin tests. 65 70 Mean U'I ....... M. KANSASII STRAIN 79 CHALLENGE ROUTE SUBCUTANEOUS 20 -~ fI1 ~ 15 ~ til ~ H E-t U ~ 10 rz.. 0 ~ 5 5 10 15 20 25 30 35 40 45 TIME (HOURS) 50 55 60 Fig. 12. Rate of development of hypersensitivity reaction. (mm) for 5 guinea pigs injected 3 weeks before skin tests. 65 70 Mean V1 OJ M. KANSASII STRAIN 79 CHALLENGE ROUTE INTRATESTICULAR ~ ~ 20 ~riI ~ H 15 E-t U ~ r:.. o 10 ~ 5 5 10 15 20 25 30 35 40 45 TIME (HOURS) 50 55 60 Fig. 13. Rate of development of hypersensitivity reaction. (rom) for 5 guinea pigs injected 3 weeks before skin tests. 65 70 Mean U1 \0 M. KANSASII STRAIN 79 CHALLENGE ROUTE INTRACARDIAC -~ ~ 20 ~r:iI - ~ 15 H E-t tJ .( ~ f:) 10 ~ 5 5 10 15 20 25 30 35 40 45 TIME (HOURS) 50 55 60 Fig. 14. Rate of development of hypersensitivity reaction. (rom) for 5 guinea pigs injected 3 weeks before skin tests. 65 70 Mean Q'\ o M. KANSASII STRAIN 79 CHALLENGE ROUTE INTRADERMAL ~ ~ 20 ~ril - Z 15 o H E-t U ~ ~ ~ o 10 ~ 5 5 10 15 20 25 30 35 TIME 40 45 50 55 60 65 70 (HOURS) Fig. 15. Rate of development of hypersensitivity reaction. (rom) for 5 guinea pigs injected 3 weeks before skin tests. Mean 0\ t--' M. KANSASII STRAIN 79 CHALLENGE ROUTE AEROSOL ~ ~ 20 ~ r.q ~ 15 H E-t () I( ~ ~ 10 ~ 5 50 55 60 35 40 45 TIME (HOURS) Fig. 16. Development of skin hypersensitivity reactions. (rom) for 5 guinea pigs injected 3 weeks before skin tests. 5 10 15 20 25 30 65 Mean 70 0'\ N 63 B. Degree of Sensitization i2! Various Inoculation Routes The degree of sensitization varied with the inoculation route. When H37Ra was utilized as the sensitizing organism, maximum reactions were obtained when the organisms were injected either subcutaneously in oil or by intracardiac puncture (Table 1). Cross reactions to PPD-Y and PPD-S were induced by subcutaneous (oil), intradermal, intratesticular, and intracardiac inoculation. Challenge either by mouth or aerosol failed to induce positive reactions (greater than 5 rom) • Injection with ~. kansasii induced skin sensitivity by all routes tested except by mouth (Table 2). Maximum reactions were obtained by subcutaneous injection but no significant difference was observed between this route and intradermal, intracardiac, and intratesticular injections. The use of organisms emulsified in an adjuvant (Bayol oil) increased the induction of cross reactions to PPD-A and PPD-S. In subsequent testing, animals were sensitized by the subcutaneous route. C. Organisms were suspended in Bayol oil. Sensitivitx Patterns Induced Ex~. kansasii and~. tuberculosis The results shown in Tables 1 and 2 indicate that route of inoculation does not greatly influence the patterns of sensitivity. When H37Ra was used as the sensitizing organism, the homologous reaction (PPD-S) was always greatest, the PPD-Y reaction was next, followed by PPD-A. The pattern of sensitivity induced differed from that of H37Ra. by~. kansasii In this case the maximum reactions were to PPD-Y, followed by PPD-A and PPD-S. Figures 17 and 18 show the 64 Table 1 Twenty-four Hour Skin Reactions of Guinea Pigs Injected Challenge Route with~. tuberculosis H37Ra Diameter of Erythema* Sensitins PPD-Y PPD-S PPD-A 9.3 5.5 3.2 Subcutaneous (oil) 15.0 9.3 6.3 Intradermal 13.6 9.4 7.2 Intratesticular 12.6 6.8 6.2 Intracardiac 15.0 9.0 5.0 Aerosol 2.8 2.3 1.3 By mouth 3.3 2.3 2.3 Subcutaneous * Mean reactions for 5 guinea pigs 65 Table 2 Twenty-four Hour Skin Reactions of Guinea Pigs Injected with~. kansasii Diameter of Erythema* Sensitins Challenge Route PPD-Y PPD-S PPD-A Subcutaneous 19.4 7.4 8.6 Subcutaneous (oil) 18.8 11.3 11.8 Intradermal 15.8 7.8 11.0 Intratesticu1ar 16.4 8.2 10.4 Intracardiac 16.0 7.8 10.5 Aerosol 9.8 4.6 6.2 By mouth 3.3 1.3 2.5 * Mean reaction for 5 guinea pigs Fig. 17. Fig. 18. Twenty-four hour s kin r eactio ns of guinea pig injected with H37Ra. Sensitins used include: A=PPD-A, K=PPD-Y T=PPD-S, Sal = Sal ine. Twenty-four hour s k in reactions of guinea pig injected kansasi i. Sensitins used include: A=PPD-A, K=PPD-Y, T=PPD- S, Sal = Sal ine. with~. 67 typical pattern of sensitivity induced by each strain. Inoculations were made subcutaneously. II. STUDIES ON CELLULAR EXTRACTS, CELL WALLS AND CULTURE FILTRATES A. Yields of ~ Week Old Cultures of ~. kansasii and ~. tuberculosis Organisms were grown for 6 weeks in low-form flasks containing 1 liter of Sauton medium. The largest yield of reactive material obtained was the culture filtrate (Table 3). The concentrated filtrate of M. kansasii had a green-grown color and was viscous. The cellular extract had a light green tint with a viscosity of little more than water. The cell wall extract had approximately the same color and viscosity as the cell extract but was more opaque. Cell yields obtained from H37Ra were less than yields obtained from M. kansasii (Table 4). The concentrated culture filtrate was dark brown, viscous and oily. appearance from~. Cell extracts from H37Ra differed in kansasii in that the H37Ra fractions were much more opaque and turbid in appearance. B. Biochemical Analysis of Cellular Extracts, Cell Walls and Culture Filtrates Biochemical analyses were performed on the various fractions to provide some limited chemical characterization of the material prior to electrophoresis. Others (Affronti et al., 1965) have reported certain protein concentrations were required for optimum isolation of tuberculin products. The analyses were performed on 68 Table 3 Fraction Yield from 6 Week Culture of ~. Fraction kansasii in Modified Sauton Medium Before Concentration After Concentration Appearance Filtrate 10 Liters 1 Liter green-brown viscous Cell extract 500 ml 50 ml green-tint clear Cell wall 1 Liter 100 ml green-tint opaque 69 Table 4 Fraction Yield from 6 Week Cultures of M. tuberculosis (H37Ra) in Modified Sauton Medium Fraction Before Concentration After Concentration Appearance Filtrate 10 Liters 1 Liter dark brown viscous Cell extract 260 ml 40 ml straw opaque Cell wall 300 ml 75 ml gray-milky opaque 70 Table 5 Results of Biochemical Analyses of Fractions Obtained From M. kansasii Fraction Carbohydrate* (Anthrone) Nitrogen* (Kjeldahl) Protein* DNA* (Lowry) (Diphenylamine) Filtrate 7.6 3.0 7.5 0.84 Cell extract 3.6 1.7 5.1 0.52 Cell wall 1.1 0.74 3.2 0.36 * Concentration in mg/ml 71 Table 6 Results of Biochemical Analyses of Fractions Obtained from H37Ra Fraction Carbohydrate* (Anthrone) Nitrogen* (Kjeldahl) Protein* DNA* (Lowry) (Diphenylamine) Filtrate 4.3 1.56 4.25 0.51 Cell extract 2.6 0.83 1.41 0.36 Cell wall 0.8 0.33 0.76 0.19 * Concentration in mg/m1 72 the concentrated products. from~. The results of fractions obtained kansasii are found in Table 5. The culture filtrate contained a higher concentration of DNA, nitrogen, and carbohydrate than the cellular fractions. C. The same was true for H37Ra. Polyacrylamide Electrophoresis of Cellular Extracts, Cell Walls and Culture Filtrates. Electrophoretic separation of the 3 fractions involved many trials before definitely reproducible bands were obtained. Adjustments in gel concentrations, time of polymerization and sample concentrations were studied in order to obtain the maximum number of bands with the greatest reproducibility. These investigations resulted in the techniques described previously in the methods section. The cell wall fraction of both M. kansasii and M. tuberculosis never yielded more than 2 very faint bands. These bands, when seen, were located near the spacer gel indicating slow migration. Separation of the culture filtrate material was difficult because of the oily nature of the material. At least 11 bands were distinguished when the spacer gel was used undiluted. The cell extract fraction. contained the best material for electrophoretic separation (Figure 19). At least 10 bands could be found each time a run was performed. The bands were lighter or fainter than those of the filtrate but were also more distinctly separate. Although 10 bands were obtained from the cell extract material, 4 major bands are clearly shown (Figure 19). Because of the wide separation of these 4 major areas, the cell extract was Fig. 19. Culture Ce 11 Filtrate Extract Disc electrophoresis of cell wall, culture filtrate and cell extract material isolated from M. kansasi i. Amidoschwartz protein stain. Fig. 20. Disc electrophoresis of cell extract of H37Ra. Amidoschwartz. Ce 1 1 Wa 1 1 74 chosen to be utilized for the isolation of antigens to be employed for skin tests. The cell extract, culture filtrate and cell wall fractions isolated from H37Ra were also subjected to disc electrophoresis. The results indicated a more complicated pattern of protein constituents. The cell extract (Figure 20) yielded at least 20 bands including bands having electrophoretic mobilities similar to those of M. kansasii cell extract. H37Ra cell wall material was like that of M. kansasii in that only 2 very faint bands were found. Culture filtrate yielded 20-26 bands. The study of bands found in the various fractions of H37Ra will be carried out in future research. III. STUDIES OF BANDS ISOLATED FROM ~ EXTRACTS BY DISC ELECTROPHORESIS TECHNIQUES A. Isolation of Bands The four major bands or areas isolated from the cell extract of M. kansasii were labeled K-l, K-2, K·3 and K-4 (Figure 20). The K-l band moved most rapidly and was extremely close to the bromphenol blue dye front and in most cases could not be distinguished from the dye front. The K-2 area was the most complex area, containing 4 distinct bands. The K-3 area was intermediate between K-2 and K-4 and consisted of one major band and several other faintly staining bands. The K-4 area included at least 2 distinct bands which were located just below (within 10 rom) of the origin of the running gel. 75 Rf values for each of the four major areas were calculated as previously described. Because of the close relationship of band K-l with the bromphenol dye front, this band was given an Rf value of 1.0. The K-l area included the visible dye front and 5 rom of gel on each side of the dye front band. Location of the unstained K-2 area (Rf value 0.69, Table 7) was found by multiplying the distance traveled by the dye front by 0.69 (e.g. if dye front moved 50 rom, band K-2 would be located at 34-35 rom from origin). In order to isolate the area of K-2 the gel was cut 5 rom above the calculated location of the band. The area of K-4 included all the bands within 10 rom of the origin of the running gel and the gel was cut accordingly. B. The remaining gel was considered to be the K-3 area. Chemical Characterization of Bands Nitrogen, DNA, protein and carbohydrate determinations were performed as previously described (see Materials and Methods). The diphenylamine technique for determining the presence of DNA was not sensitive enough to detect the presence of DNA in any of the fractions tested (Table 8). However by Feulgen stain technique a faint red-colored band appeared in the same area as K-I, i.e~, the dye front. Protein analysis by the Lowry technique revealed very small quantities of protein at each band level. Amido-Swartz staining had previously shown (Figures 19-20) the presence of protein at each band. The relationship of protein concentration and skin test potency of each band will be discussed later. at K-4 K-3 K-2 K-l Fig. 21. Disc electrophoresis of cell extract of M. kansasii, showing 4 major bands or areas uti1 ized as skin test agents. Amidoschwartz protein stain. 77 Table 7 Rf Values Calculated for 4 Major Bands Isolated from Cell Extract of M. kansasii Band Rf Value Standard Error 1 1.0 2 0.69 0.02 3 0.51 0.02 4 0.20 0.02 78 Table 8 Results of Biochemical Analysis of Four Major Bands Obtained from Cell Extract of M. kansasii Band Carbohydrate** (Anthrone) Nitrogen** (Kje1dah1) Protein*** DNA (Lowry) (Diphenylamine) 1 12 1.1 0.013 0 2 7 0.9 0.009 0 3 15 0.7 0.012 0 4 30 1.7 0.006 0 * Concentration mg/m1 of eluate ** Concentration ug/m1 of eluate 79 Nitrogen values as determined by the Kjeldahl technique did not show a close correlation with the values obtained for total protein. Treatment of the various band solutions with 10 per cent trichloracetic acid did not lower the total nitrogen value. Carbohydrate content measured as glucose by the Anthrone technique was highest (30 ug/ml) in the K-4 area and next highest in the K-3 area. These data correlated well with the results of PAS staining which indicated the presence of glycoprotein in bands K-4, K-3 and some in K-2. Use,.of the PAS stain also revealed the presence of a band above the spacer gel, i.e., at the interface of sample gel and spacer gel. Preliminary studies had shown no material to be retained in the spacer gel. C. Bio-Assay of ~ Activity Bio-assays (as described under methods) were carried out in order to determine equally effective doses of standard and test preparations. The experimental design required that 3 dilutions of test preparation of approximately equal potency to that of the standard be employed. A pilot study utilizing concentrated test preparations was used to determine approximate relationships of test to a standard antigen. are found in Table 9. The results of this preliminary study These results indicated that concentrated K-3 and K-4 fractions were approximately equal in potency to standard PPD-Y. Bio-assays revealed more quantitatively the relationship to standard PPD-Y. Dilutions of test preparations (K-3, K-4) were made based on the assumption that these agents had 80 Table 9 Pilot Study of Skin Reactions to Determine Approximate Potency of Test Preparations in Normal and M. kansasii Injected Guinea Pigs 24 Hour Skin Test Reactions in mm ~reparation * M. kansasii infected guinea pigs (5) Normal guinea pigs (5) K-1 4.0 1.0 K-2 8.5 2.4 K-3 12.3 2.8 K-4 11.2 3.4 PPD-Y* 10.3 1.4 PPD-Y = 0.0005 Mg/O.l ml (25 TU); obtained from United States Public Health Service. 81 approximately equal potency with that of the standard. Standard test doses were diluted to contain 0.1, 0.32, and 1.0 ug of protein in 0.2 ml, i.e., a low, middle and high dose with the high and low dose different by a full loglO unit. Samples were diluted to corresponding activity levels and labeled as described above, even though actual protein concentration was much lower. A bio-assay was also carried out with a band Y-4 isolated from PPD-Y. This band had an electrophoretic mobility similar to that of K-4. Preliminary studies indicated the potency of Y-4 to be near that of PPD-Y. The data collected from potency tests for K-4, K-3 and Y-4 are shown in Tables 10, 11 and 12 respectively. The results of the computer analysis are summarized in Table 13. The potency of K-4 was found to be only 0.73 of the standard PPD-Y. The bio-assay further revealed highly significant non- parallelism between K-4 and PPD-Y. Significant nonparallelism indicates that a substance or substances other than those associated with the standard is responsible for the reaction. The substances might be considered impurities or other active substances present in the test material. In order for a bio-assay to be valid the test for parallelism must be nonsignificant. However, an assay may fail a test for validity and still provide a contributory estimate of potency (United States Pharmacopeia, 17th Edition). i~e. The lambda (A) value, the index of precision was 0.263. Band K-3 yielded a preparation 2.19 times as potent as PPD-Y 82 Table 10 Bio-assay Data Obtained Using Band K-4 as Skin Test Antigen in M. kansasii Infected Guinea Pigs 24 Hour Skin Test Reactions in nun Pig No. Standard (PPD-Y) Dose* 0.32 1.0 0.1 Test (K-4) Dose** 0.1 0.32 1.0 1 7 11 17 9 13 15 2 6 10 14 5 6 7 3 7 14 17 10 6 16 4 7 13 17 7 13 13 5 6 10 17 5 11 15 6 6 11 23 13 14 15 6.5 11.5 17.5 10.5 13.5 Dose Mean S * ** 8.2 = 2.15 Dose in ug protein/0.2 m1 Dose as related to standard only, not as true value for protein concentration 83 Table 11 Bio-assay Data Obtained Using Band K-3 as Skin Test Antigen in~. kansasii Infected Guinea Pigs 24 Hour Skin Test Reactions in mm Standard (PPD-Y) Dose* 0.32 0.1 1.0 Pig No. Test (K-3) Dose** 0.32 0.1 1.0 1 7 14 15 16 13 25 2 4 11 16 15 19 14 3 7 10 16 11 10 14 4 9 12 17 11 18 17 5 8 17 17 12 16 18 6 5 10 14 4 8 12 6.7 12.3 15.8 11.5 14.0 Dose Mean S * ** 16.7 = 2.75 Dose in ug protein/0.2 m1 Dose as related to standard only, not as true value for protein concentration 84 Table 12 Bio-assay Data Obtained Using Band Y-4 as Skin Test Antigen in~. kansasii Infected Guinea Pigs 24 Hour Skin Test Reactions in nun Pig No. Standard (PPD-Y) Dose* 1.0 0.1 0.32 Test (Y-4) Dose** 0.32 0.1 1.0 1 8 15 16 14 16 24 2 7 14 15 14 15 18 3 8 14 15 17 22 19 4 10 14 19 18 20 22 5 8 8 12 14 13 17 6 7 14 17 13 14 16 8.0 13.2 15.7 15.0 16.7 19.3 Dose Mean S * ** = 1.91 Dose in ug protein/0.2 m1 Dose as related to standard only, not as true value for protein concentration Table 13 Potency of Bands K-4, K-3 and Y-4 Relative to Standard PPD-Y as Determined by Skin Test Reactions in M. kansasii Infected Guinea Pigs Relative Potency Band 95 per cent Confidence Limits Parallelism b s Nonparallel 8.16 2.14 0.263 sib =/\ K-4 0.73 K-3 2.19 1.1 - 4.8 >.05 7.16 2.75 0.384 Y-4 6.12 3.35-14.8 <.05 6.00 1.92 0.320 b = Slope s = Standard of dose-response line deviation 00 V1 86 with nonsignificant parallelism. All other tests for validity were met, e.g., significant regression, no significant differences between preparations and significant slope values. These results indicated close similarity between the substances found in K-3 and PPD-Y. TheAvalue for the bioassay of K~3 was 0.384, a value which would suggest that the above assumption of similarity should be accepted only with the knowledge that there is some question concerning the reliability of the test. Band Y-4 was 6.12 times as potent as PPD-Y. significant at the D5 level was noted. Nonparallelism The nonparallelism of Y-4 was not clear-cut, since all the test responses were higher than the standard reactions and could represent a flattening of the dose-response curve at high dose. The relative potency values of K-4 and K-3 are of further interest because they relate the distribution of skin test active fractions along the gel column. The K-3 band is 3 times as active as band K-4. Statistical analysis revealed no significant differences between skin test sites. guinea pigs. However, there were significant differences between The bioassay design (i.e., random block or Latin square) eliminated these differences. D. Specificity of Bands The specificity of bands K-3 and K-4 was tested by the use of guinea pigs previously injected with H37Ra. The skin test reactions were compared to reactions elicited by standard PPD-S and PPD-Y. 87 Statistical analyses were carried out employing a randomized block design. The results with band K-4 are found in Table 14. Material from band K-4 was not diluted because its potency ratio to PPD-Y was less than 1 (0.73) (Table 13). Although the results indicate a significantly lower response to K-4 material compared with PPD-Y, the difference could be attributed to the difference in potency of the two antigens. Band K-3 was diluted 1:2.19 to a potency assumed to be equal to that of PPD-Y (Table 13). The mean reaction to band K-3 (Table 15) appears to be larger than to PPD-Y. Statistical analysis re- vealed significant differences (.05 level) between K-3 and PPD-Y. E. Antigenicity of Bands K-3 and K-4 The possibility that skin testing with material eluted from bands K-3 and K-4 induced the production of antibodies was studied by use of latex agglutination tests (see Materials and Methods section). Sera from normal guinea pigs, skin tested guinea pigs and ~. kansasii injected guinea pigs were utilized. Normal guinea pigs were skin tested with 0.1 ml of material from either band K-3 or K-4 in 6 sites once a week for three weeks and bled two weeks after the last skin test, i.e., each guinea pig received 1.8 ml of antigen over a period of three weeks. Previous testing did not induce positive reactions. The latex agglutination results indicate that no detectable antibody was produced with either K-3 or K-4. M. kansasii induced titers of 1:512 (Table 16). Injection of living 88 Table 14 Comparison of Reactions Elicited by Band K-4, PPD-Y and PPD-S in H37Ra Injected Guinea Pigs Mean mm of Reaction for 6 Animals 24 Hour Readings Antigen Mean PPD-S 15.3 PPD-Y 9.8 K-4 7.3* Standard Error 0.31 Analysis of Variance Data Described in Table 14 Source of Variation Total Degrees of Freedom Sum of Squares Mean Square 17 266.50 Between Samples 2 200.90 100.45*** Between Blocks 5 59.82 11.96*** 10 5.78 Error * *** 0.58 Significantly different from PPD-Y and PPD-S at .05 level Significantly different at .001 level 89 Table 15 Comparison of Reactions Elicited by Band K-3, PPD-Y and PPD-S in H37Ra Injected Guinea Pigs Mean rom of Reaction for 6 Animals 24 Hour Readings Antigen Mean PPD-S 13.3 PPD-Y 6.8 K-3 9.3* Standard Error 0:71 Analysis of Variance Data Described in Table 15 Source of Variation Degrees of Freedom Sum of Squares 17 214.50 Between Samples 2 128.98 64.49*** Between Blocks 5 54.48 10.89*** 10 31.04 Total Error * *** Mean Square 3.10 Significantly different from PPD-Y and PPD-S at .05 level Significantly different at .001 level 90 Table 16 Results of Latex Agglutination Studies with Sera from Normal, Skin Tested and M_ kansasii Infected Guinea Pigs Sera Source Titer Normal Guinea Pigs o Normal Guinea Pigs After Skin Test with K-3 Material o Normal Guinea Pigs After Skin Test with K-4 Material o M. kansasii Injected Guinea Pigs 1:512 DISCUSSION Hypersensitivity has been classified into two general categories, immediate and delayed. Classically, an immediate reaction is one in which the first overt responses are seen within minutes after injection of antigen. Examples of immediate hypersensitivity include anaphylactic shock and cutaneous reactions of the "wheal and flare" type. These responses are always associated with the presence of humoral antibodies and sensitivity, i.e., the capacity to show response can be passively transferred by means of serum. Delayed hypersensitivity reactions (e.g., the tuberculin reaction) develop more slowly than immediate reactions. It is stated the first macroscopic response is usually seen from 8-12 hours after the antigen injection. The occurrence of delayed hypersensitivity has not been correlated with the presence of humoral antibody and can only be transferred by means of lymphocytic cells or their extracts (Waksman, 1960). The distinction between delayed and immediate reactions based upon time of development should be looked upon critically (Gell, 1959). Studies of the delayed reactions of guinea pigs reported here (Figures 5-16) indicated that reactions may be seen within two hours after antigen injection and were often highly developed 8-10 hours after injection. In addition, delayed reactions in the guinea pig were characterized by maximum reactions at 24-30 hours with a decline in size at 48 and 72 hours. These results 92 corroborate the findings of other investigators (Crowle, 1959; Bauer and Stone, 1961). As has been noted~ immediate reactions are associated with the presence of antibody in the blood whereas delayed reactions are considered to be independent of antibody but dependent upon the presence of reactive substances in lymphocytes. This latter concept (cellular reactivity in delayed hypersensitivity) has recently become controversial because of the demonstration of the presence of globulin specific for mycobacterial antigens in the serum of active and inactive tuberculous patients as well as in those with other mycobacterial diseases. The findings of Parlett and Youmans (1959) have been previously described in another context. For purposes of discussion at this point it is only necessary to note that these workers have found that about 90 per cent of the sera from patients with active tuberculosis react in a serologic (gel diffusion) test. Farr (1965), employing radioactive iodine labelled tuberculin, has found over 40 per cent of tuberculous patients to have high antigen binding substances in their serum. The results with latex agglutination tests, as reported in this thesis (Table 16), indicate the presence of humoral antibody in guinea pigs injected with M. kansasii. These results suggest the possibility that the reactions measured as delayed responses may be combined reactions or may be due entirely to humoral antibody. This latter possibility has been considered at length by 93 Karush and Eisen (1962) who support the view that all delayed hypersensitivity reactions are mediated by low concentrations of high affinity antibody. The rate of development of the delayed guinea pig reactions did not appear to be influenced by the route of inoculation. Sub- cutaneous injection seemed to induce the most specific responses; however, if the organisms were suspended in adjuvant, this apparent specificity was lost. It has been reported that intracardiac or intravenous injections are not effective pathways for the establishment of delayed hypersensitivity (Petroff, et al., 1929; Rich, 1951); however, the work here reported has indicated a high degree of sensitization resulting from intracardiac inoculation. The sensitivity patterns induced by~. kansasii and H37Ra were characteristic for each species, i.e., injection of H37Ra induced a maximal response to PPD-S followed by less response to PPD-Y and even less to PPD-A. M. kansasii induced a maximum response to PPD-Y followed by cross-reactions to PPD-A and PPD-S. Exposure to aerosolization with M. kansasii induced the characteristic sensitivity pattern while exposure to aerosolization with H37Ra failed to induce detectable sensitivity. These results might be attributed to the presence of clumped bacilli in the H37Ra suspension which could prevent deposition of the bacilli in the alveolar spaces or might be due to the lack of invasive capacity of the avirulent organisms reaching the lung surfaces. Presently available tuberculin active substances (e.g. PPD, 94 OT) are mixtures of active materials recovered from culture filtrates. Several workers (Kim et al., 1963; Warfvinge, 1962) have isolated potent, stable mycobacterial antigens from disrupted cells. Studies of cell extract and culture filtrate material by polyacrylamide electrophoresis (Figure 19) have indicated the presence of multiple antigens in each fraction. The cell extract fraction contained the best material for electrophoretic separation; at least 10 bands could be distinguished and four major bands were separated. Warfvinge (1962) reported that endotuberculin, a cellular product obtained by freezing-thawing techniques was more specific than other available skin test agents. Because of this and similar reports (Larsen et al., 1963) and because of easier electrophoretic separation, the cell extract material of M. kansasii was chosen as the source for isolation of more specific skin test antigens. Major problems encountered in isolating band fractions from a polyacrylamide column included: location of the bands without staining, concentration of the band material and removal of toxic acrylamide monomers. Staining the gel columns with a protein dye (amidoschwartz) revealed that each band had a characteristic mobility in relationship to the dye front. runs. Rf values were calculated for many trial The results (Table 7) indicated the reproducibility of the calculated Rf value for each band. always stained. One gel from each run was The stained gel indicated in every case that each major band was located in the proper area. This technique did not 95 exclude the possibility that other bands located near the 4 major ones might overlap from one slice to another. Preliminary testing with bands isolated from the acry1amide column prior to dialysis indicated Some toxic reactions, i.e., they gave positive reactions in normal guinea pigs. Ornstein (1966) suggested that acrylamide polymers were definitely non-toxic, however the monomer form was considered toxic. The large water- insoluble polymer form is held back by standard filtration while dialysis of the filtrate against PVP not only removes the monomer form but serves to concentrate the antigens. There has been some criticism of the use of substances like PVP and Carbowax as dehydrating agents (Crowle, 1966). Carbowax has been noted to cross the dialysis membrane and contaminate the concentrated material; however skin tests in normal guinea pigs with PVP dialyzed (Table 9) showed no indicatiE6uJof inflammatory activity. Comparison of biochemical analysis data (Table 8) with skin test activity (Table 9-12) indicated that the activity might be correlated with a carbohydrate component. This is in agreement with the findings of Chaparas and Baer (1963, 1964) who found tuberculin activity associated with a carbohydrate fraction isolated from the culture filtrate of BCG by means of Sephadex chromatography. Nitrogen assays did not show close correlation with the values obtained for total protein. Treatment of the band solutions with various protein precipitants did not lower the total nitrogen value. In addition, the values obtained for protein (Lowry 96 technique) indicated a low concentration of tuberculoprotein in the isolated band fractions. These results seemed paradoxical since each band fraction was originally identified by means of a protein stain. Later results with polysaccharide stains (PAS) indicated the presence of carbohydrate in the band fractions. It may be postulated that either the isolated band fractions contain carbohydrate contaminated with a small amount of protein or that the active component is a glycoprotein. Further studies to determine the tole of each biochemical constituent are under way. Edwards et al. (1965) have reemphasized the concept that the purpose of a tuberculin-type skin test is to determine whether an individual has been or is infected by a specific microorganism. An ideal skin test would have ideal sensitivity, i.e., be capable of indicating infection while remaining negative when no infection exists. Moreover, the ideal skin test would never confuse infection by one species of organism with another (specificity). These workers have reemphasized that none of the skin tests in present use in the field of mycobacterial disease meet all these requirements. In order to evaluate new skin test antigens, it is necessary to establish their potency in relation to the presently available standard preparations. Other workers (Guld et al., 1965) have determined specificity by comparison of reaction sizes in an adult human population. It was found that one preparation was more specific than some others, including the World Health Organization (WHO) standard. The experimental design and conditions did not 97 allow appropriate controls. A better method of evaluation would determine not only the potency but also the specificity of the antigens in laboratory animals of standard age and with standardized inoculum. sex~ injected If preliminary findings in laboratory animals indicated more specific antigens, then studies should be carried out in man. The bio-assays of bands K-3 and K-4 revealed several interesting facts. PPD~Y The relative potency of K-4 was less than that of while K-3 was more than twice as potent as the PPD-Y standard and three times as potent as K-4. Biochemical analysis indicated that band K-3 contained only one-half as much carbohydrate as K-4 and had twice as much protein. It would seem, then, that tuberculin activity was associated in some way with protein concentration although the concentration was low (0.012 ug/ml). However, the examination of chemical analysis data (Table 8) indicates that band K-l contained the greatest amount of protein and had the least activity of all fractions tested. Significant nonparallelism of the log-dose response curves for the unknown and standard test material might indicate that a substance or substances other than those associated with the standard are responsible for the reaction. Because band K-3 did not show significant nonparallelism it is possible that it contains the same active substances as the standard preparation PPD-Y. The significant values indicating nonparallelism for bands K-3 and Y-4 might indicate several things, e.g., the presence of substances 98 (perhaps acry1amide) which enhance the tuberculin reaction. This phenomenon would be similar to that reported by Be1sey (1966) who found that adjuvant mixed with tuberculin (depot tuberculin) increased the size of the tuberculin reaction in mice. Although it is not apparent from the data presented, there is some evidence that more active material was obtained from the acry1amide columns than was originally placed on them. This might be explained by the depot tuberculin concept mentioned above. Another theory explaining the enhancement effect is the possibility that inhibitory substances have been "stripped" away during the isolation of the active component. It is apparent that the material isolated from the cell extract of M. kansasii is no more specific than presently available PPD-Y, i.e., the band fractions elicit reactions in guinea pigs infected with another species (~. tuberculosis H37Ra). It is of interest that the analysis for parallelism indicated some relationship between the chemical constitution and skin test activity of band K-3 and PPD-Y. The preliminary studies with latex agglutination indicated that the band fractions met the definition of "sensitin" in that the agents elicited delayed hypersensitivity and did not induce the production of circulating antibody. SUMMARY The influence of various inoculation routes on the development and specificity of mycobacterial hypersensitivity reactions was investigated. The results indicated that the rate of development of skin test reactions was not determined by the inoculation route. Skin test reactions were seen within two hours after antigen injection and were often highly developed within 8-10 hours. The reactions were characterized by maximum appearance of inflammation at 24-30 hours followed by a decline in size beginning at 48 and 72 hours. Latex agglutination tests carried out with guinea pig sera 3 weeks after sensitization indicated the presence of circulating antibody at the time skin tests were performed, The sensitivity patterns induced by M. kansasii and M. tuber- culosis, strain H37Ra, were characteristic for each species, i.e., H37Ra induced responses in the order of PPD-S greater than PPD-Y greater than PPD-A. M. kansasii induced responses in the order of PPD-Y greater than PPD-A greater than PPD-S. Cell extracts of M. kansasii obtained by freezing-thawing and ultrasonic disruption were subjected to polyacrylamide (disc) electrophoresis~ At least 10 bands could be distinguished and 4 major bands were separated. Chemical analysis of the material eluted from the 4 major bands indicated that skin test activity was correlated with polysaccharide in association with protein. 100 Skin test bio-assay data were analyzed by computer analysis. Values for parallelism of the log-dose response curves for the unknown and standard indicated that one band (K-3) contained active material of similar chemical constitution to that of PPD-Y. Another band (K-4) , although active, exhibited non-parallelism suggesting the presence of other or different active material. Material isolated from the cell extract of ~. kansasii by disc electrophoresis was not more specific than presently available PPD-Y. However, the studies indicated that the band fractions met the definition of "sensitin" in that the agents were capable of detecting delayed hypersensitivity without inducing the production of antibody. BIBLIOGRAPHY Affronti, L. F. 1959. Purified protein derivatives (PPD) and other antigens prepared from atypical acid-fast bacilli and Nocardia asteroides. Am. Rev. Resp. Dis. 12:284-295. Affronti, L. F., R. C. Parlett, R. A. Cornesky. 1965. Electrophoresis on polyacrylamide gels of protein and polysaccharide fractions from Mycobacterium tuberculosis. Am. Rev. Resp. Dis. 91:1-5. Anderson, R. J. 1932. The chemistry of the lipoids of the tubercle bacilli. Physiol. Rev. ~:166-l89. Anderson, R. J., M. M. Creighton. 1939. The chemistry of the lipids of tubercle bacilli. LVII The mycolic acids of the avian tubercle bacillus wax. J. of BioI. Chem. 129:57-63. Arnason, B. G., B. H. Waksman. 1963. The retest reaction in delayed sensitivity. Lab. Invest. 12:737-747. Asselineau, J., E. Lederer. 1935. Chemical structure and biological activity of mycotic acids. Ciba Foundation Symposium on Experimental Tuberculosis. Little, Brown and Co., Boston, Mass. Austen, K. F., J. H. Humphrey. mechanism of anaphylaxis. 1963. In vitro studies of the Adv. Immunol. 3:1-96. Baer, H., S. D. Chaparas. 1963. The immunology and chemistry of tuberculin. Am. Rev. Resp. Dis. 88:28-32. Baer, H., S. D. Chaparas. 1964. Tuberculin reactivity of a carbohydrate component of unheated BCG culture filtrate. Science 146:245-247. Bauer, J. A., S. H. Stone. 1961. Isologous and homologous lymphoid transplants. I. The transfer of tuberculin hypersensitivity in inbred guinea pigs. J. of Immunol. 86:177-189. Belsey, M. 1966. Tuberculin sensitivity in infected mice. Its detection by depot tuberculin. Am. Rev. Resp. Dis. 93:619-621. Bhadarakom, S. 1963. The transfer of delayed cutaneous reactivity of tuberculin with serum protein fractions. Siriraj Hosp. Gazette 15:597-599. 102 Birkhaug, K. 1939. Effect of I-ascorbic acid on the tuberculin reaction in tuberculous animals. Acta Tuberc. Scand. 13:45-52. Bjerkedal, T. 1960. Degree of tuberculin sensitivity in guinea pigs infected with different doses of virulent tubercle bacilli. Am. J. Hyg. 21:299-313. Bjerkedal, T., H. Nativig. 1962. Comparison of sensitivity to ten different mycobacterial antigens in Norwegian army recruits and BCG-vaccinated guinea pigs. Bull. Intern. Union Against Tuberc. 32:395-397. Bloomfield, A. L., J. G. Mater. 1919. Changes in skin sensitiveness to tuberculin during epidemic influenza. Am. Rev. Tuberc. 3:166-168. Bojalil, L. F., J. Cerbon, A. Trujillo. 1962. Adansonian classification of mycobacteria. J. Gen. Microbiol. 28:333-346. Bovornkitti, S., P. Kangsadal. 1963. Serum transfer of delayed cutaneous hypersensitivity to tuberculin in man. Dis. Chest 43:393-396. Boyd, W. C. 1947. Fundamentals of Immunology. Publishers Inc., New York. Interscience Branch, A., J. R. Cuff. 1930. Tuberculosis. I. Allergic, anaphylactic and immune reactions in guinea pigs following inoculation with heat-killed tubercle bacilli. J. Inf. Dis. 47:151-170. Brody, J. A., R. McAlister. 1964. Depression of tuberculin sensitivity following measles vaccination. Am. Rev. Resp. Dis. 90:607-611. Buhler, V. B., A. Pollak. 1953. Human infection with atypical acid-fast organisms. Report of two cases with pathologic findings. Am. J. Clin. Path. 23:363-374. Burger, M. 1950. Polysaccharides of the mycobacteria. Bacterial polysaccharides their chemica1-afid immunological aspects. Charles C. Thomas Co., Springfield, Ill. Burns, D. T. 1966. Principles of gel filtration or column molecular sieving. Lab. Practice 11:179-183. Campagna, M., H. B. Greenberg. 1964. Epidemiology and clinical course of 41 patients treated for pulmonary disease due to unclassified mycobacteria. Dis. Chest 46:282-288. 103 Carpenter, P. L. 1965. Co., Philadelphia. Immunology and Serology. W. B. Saunders Castelnuovo, G., M. E. Duncan, G. Bellezza. 1964. Mycobacterial antigens: A study of antigens in bacterial extracts and in culture filtrates of different ages. Tubercle 45:246-254. Castelnuovo, G., M. Morellini. 1965. The antigens of mycobacteria and their identification by immunoelectrophoretic analysis. Am. Rev. Resp. Dis. 92:29-33. Chaparas, S. D., H. Baer. 1963. The immunology and 'chemistry of tuberculin. II. Chromatography with sephadex of the nondialyzable tuberculin-activity constituents of BCG culture filtrate. Am. Rev. Resp. Dis. 89:41-47. Chaparas, S. D., H. Baer. 1964. The immunology and chemistry of tuberculin. III. Dialyzable BCG culture filtrate components obtained by sephadex chromatography and fractional dialysis. Am. Rev. Resp. Dis. 90:87-96. Chaparas, S. D., H. Baer. 1965. Lowered delayed skin reactivity to fractions of mycobacterium-bovis (BCG) in guinea pigs sensitized with large doses of BCG. (30504) Proc. Soc. Ex. BioI. and Med. 120:260-262. Chapman, J. S. 1962. Editorial. The present status of the unclassified mycobacteria. Am. J. Med. 11:471-477. Chapman, J. S., H. J. Dewlett, W. E. Potts. 1962. Cutaneous reactions to unclassified mycobacterial antigens. Am. Rev. Resp. Dis. 86:547-552. Chapman, J. S., J. S. Bernard, M. Speight. 1965. Isolation of mycobacteria from raw milk. Am. Rev. Resp. Dis. 21:351-355. Chase, M. W. 1945. The cellular transfer of cutaneous hypersensitivity to tuberculin. Proc. Soc. Ex. BioI. and Med. 59:134-135. Choucroun, N. 1947. Tubercle bacillus antigens. Biological properties of two substances isolated from paraffin oil extract of dead tubercle bacilli. Am. Rev. Tuberc. 56:203-226. Clawson, B. J. 1935. Experiments relative to vaccination against tuberculosis with the Calmette-Guerin bacillus (BCG). Arch. Path. 20:343-368. 104 Coghill, R. D. 1926. The chemical study of bacteria. XII. The albuminglobulin fraction of the tubercle bacillus. J. BioI. Chern. 70:439-447. Craig, J. P., D. L. Wilhelm. 1963. Cutaneous anaphylaxis in the guinea pig and its relative insusceptibility to an antihistamine. J. Immunol. 90~43-51. Crow Ie, A. J. 1958. Immunizing constituents of the tubercle bacillus. Bacteriol. Rev. 22:183-203. Crowle, A. J. 1959. Delayed hypersensitivity in several strains of mice studied with 6 different tests. J. Allergy 30:442-459. Crowle, A. J. 1960. Tuberculin skin reactions in mice hypersensitized by vaccination with living avirulent tubercle bacilli. Am. Rev. Resp. Dis. 81:893-903. Crowle, A. J. 1962. Delayed Hypersensitivity in Health and Disease. Charles C. Thomas Co., Springfield, Illinois. Crowle, A. J. 1963. Tubercle bacillary extracts immunogenic for mice. 5. Specificity of tuberculo-immunity induced by trypsin extracts of tubercle bacilli. Tubercle 44:241-246. Crowle, A. J., C. C. Hu. 1964. Duration of some cutaneous hypersensitivities in mice. J. of Immunol. 93:132-139. Crowle, A. J., C. C. Hu. 1965. Tubercle bacillary extracts immunogenic for mice. 7. Electrophoretic analysis. Tubercle 46:214-223. Crowle, A. J '; Personal communication. Davis, B. J. 1964. Disc electrophoresis. II. Method and application to human serum proteins. New York Acad. of Sci. 121:404-427. Dienes, L., E. W. Shoenheit, L. D. Sheff. 1925. The chemical composition and antigenic properties of the tubercle bacillus. Am. Rev. Tuberc. 11:151-162. Dienes, L. 1929. The antigenic substances of the tubercle bacillus. II. The examination of the antigenic properties. J. Immunol. 12:85-190. Dietz, T. M. 1963. Classification of mycobacteria by aid of skin reactions. Master's Thesis. University of Utah. 105 Dietz, T. M.~ K. W. Maxwell, E.H. Runyon. 1963. Classification of mycobacteria by aid of skin reactions. Bacteriol. Proc. 63:60. Dietz, T., K. Maxwell, E. Runyon, S. Marcus. 1966. Use of latex particles in serology of tuberculosis. Bacteriol. Proc. 66:47. B. Ou 1964" RepeatE:d tuberculin tests at the same site in tuberculin-positive patients. Am. Rev. Resp. Dis. 90:77-86. Duboczy~ Duboczy, B. O. reaction. 1965. Nonspecific early type of tuberculin Am. Rev. Resp. Dis. ~:55-63. Eckman, P. J., J. N. Bickers, A. A. Buechner. 1961. The potential importance of person-to-person contact in the spread of photochromogenic mycobacterial infections. Trans. 20th Res. Con. and Pule Dis. 204-211. Edwards, L. B., C. E. Palmer, P. Z. Edwards. 1955. Further studies of geographic variation in naturally acquired tuberculin sensitivity. Bull. WId. Hlth. Org. 1l:63-83. Edwards, L. B., C. E. Palmer. 1958. Epidemiological studies of tuberculin sensitivity. I. Preliminary results with purified protein. derivatives prepared from atypical acid-fast organisms. Am. J. Hyg. 68:213-232. Edwards, L. B., C. E. Palmer, L. F. Affronti, L. Hopwood, P. Q. Edwards. 1960. Epidemiologic studies of tuberculin sensitivity. II. Response to experimental infection with mycobacteria isolated from human sources. Am. J. Hyg. 21:218-241. L. B., L. Hopwood. 1962. Sensitivity profiles of mycobacterial infection. Bull. Intern. Union against Tuberc. 32:384-394. ~~waLJs, Edwards, L. B. 1963. Current status of the tuberculin test. Ann. New York Acad. of Sci. 106:32-42. Edwards, P. Q., L. B. Edwards. 1960. Am. Rev. Resp. Dis. 81:1-47. Story of the tuberculin test. Engbaek, H. C., A. Jespersen, M. Magnusson. 1964. Danish patients with lung infections caused by~. avium and atypical mycobacteria. Acta Tuberc. et Pneumol. Scand. Supple 56:99-105. 106 Erickson, R., D. W. Smith. 1962. Immunogenicity of defatted mycobacteria in guinea pigs. Am. Rev. Resp. Dis. 85:402-406. Farr, R. S. 1965. Primary interaction with I13l-labeled antigens as an approach to detecting tuberculosis. Am. Rev. Resp. Dis. 92:94-112. Feldman, W. H. 1938. Avian tuberculosis and Wilkins Co., Baltimore, Md. ions. ---------- The Williams Follis, R. H., Jr. 1940. Studies on the cellular response in the early stages of the tuberculin reaction. Bull. Johns Hopkins Hosp. 66:245-252. Francis, J. 1959. The work of the British Royal Commission on Tuberculosis, 1901-1911. Tubercle 40:124-132. Fredrick, J. F. 1964. The origin of the multiple forms of phosphorylase in algae. Ann. New York Acad. of Sci. 121:634-639. Freund, J., K. McDermott. means of adjuvants. 1942. Sensitization to horse serum by Proc. Soc. Ex. BioI. and Med. 49:548-553. Freund, J. 1956. The mode of action of immunologic adjuvant. Adv. Tuberc. Res. 7:130-148. Freund, J., S. H. Stone. 1956. Arthus reaction in the mouse and the rat after intralabial injection of antigens. J. Immunol. 76:138-145. Furcolow, M. L., B. Hewell, W. E. Nelson, C. E. Palmer. 1941. Quantitative studies of the tuberculin reaction. I. Titration of tuberculin sensitivity and its relation to tuberculous infection. Public Health Rep. 56:1082-1100. Gell, P. G. H. 1959. Cellular and humoral aspects of hypersensitive states. H. S. Lawrence Ed., New York, Paul B. Hoeber, Inc., pp. 43-66. Gernex-Rieux, C., M. Gervois, A. Tacquet, C. Voisin, V. Macquet, B. Devulder, M. Lelong, M. Aerts. 1962. Experimental and clinical comparative study of allergy to tuberculin and to an extract of Mycobacterium kansasii. Selected Papers 1:22-34. Glenchur, H., B. E. Fossieck, M. Silverman. 1965. An immediate skin test for the diagnosis of active pulmonary tuberculosis. Am. Rev. Resp. Dis. 92:741-747. Glenchur, H., B. E. Fossieck, Jr., M. Silverman. 1966. A comparison of purified protein derivative and some fractions of Mycobacterium tuberculosis. Am. Rev. Resp. Dis. 93:70-77. 107 Goodwin, T. C., F. F. Schwenthu. 1934. Protective inoculation against tuberculosis in infants by the use of heat-killed human tubercle bacilli. J. Ped. 5:475-483. Gough, G. A. C. 1933. The water-soluble proteins of the tubercle bacillus. Biochem. J. 27:237-242. Gouch, G. A. C. 1934. The purification of tuberculin. Exp. Path. 15:237-242. Brit. J. Gray, D. F., P. A. Jennings. 1955. Allergy in experimental mouse tuberculosis. Am. Rev. Tuberc. 72:171-195. Griffith, A. H., J. Marks, M. Richards. 1963. Low-grade sensitivity to tuberculin in school children. Tubercle 44:135-140. Guld, J., J. Rhodes, E. Sorkin, S. Boyden. 1965. The specificity of tuberculin preparations obtained by chemical fractionation of unheated culture filtrates. Bull. WId. Hlth. Org. 33:385-394. Hammerschlag, A. 1891. II. BAKTERIOLOGISCH-CHEMISCHE UNTERSUCHUNGEN. Uber Tuberkelbacillen. Sentr. Inn. Med. 12:9-18. Hauduroy, P. 1955. Derniers aspects du monde des mycobacteries. Masson and Cie. Paris. Hedgecock, L. W. 1965. Protective and serologically active fractions from phenol extract of Mycobacterium tuberculosis. Am. Rev. Resp. Dis. 92: 923-978. Heidelberger, M., A. E. O. Menzel. 1932. Specific and non-specific cell polysaccharides of human type of tubercle bacillus. Proc. Soc. Exp. BioI. and Med. 29:631-633. Heidelberger, M., A. E. O. Menzel. 1934. Protein fractions of the human strain (H-37) of tubercle bacillus. J. BioI. Chern. 104:655-665. Hjerten, S., S. Jerstedt, A. Tiselius. 1965a. Electrophoretic "particle sieving" in polyacrylamide gels as applied to ribosomes. Anal. Biochem. 2:211-218. Hjerten, S., S. Jerstedt, A. Tiselius. 1965b. Some aspects of the use of "continuous" and "discontinuous" buffer systems in polyacrylamide gel electrophoresis. Anal. Biochem. l:2l9-223. 108 Hofmann, K. R. V. 1894. SUR KENNTNISS DER EIWEISSKORPER IN DEN TUBERKELBACILLEN. Wein. Klin. Wochschr. 7:712-714. Hsu, K. H., D. E. Jenkins, L. R. Soriano. 1963. The tuberculin reaction associated with tuberculous infection. Am. Rev. Resp. Dis. 87:493-503. Hsu, K. H. K., F. Jeu, D. E. Jenkins. 1964. Studies on the specific tuberculin reaction. Tuberculin conversion in tuberculosis contacts. Am. Rev. Resp. Dis. 90:36-47. Jensen, K., R. A. Patnode, H. C. Townsley, M. M. Cummings. 1962 Multiple passive transfer of the delayed type in humans. Am. Rev. Resp. Dis. 85:373-377. Jespersen, A., B. Mackelprang. 1959. Immunity and tuberculin sensitivity in guinea pigs vaccinated with a known number of viable BCG in coarse suspensions and finely dispersed suspensions dispersed by means of ultrasonics. Acta Tuberc. Scand. 12:245-260. Johnston, W. W., D. T. Smith. 1964. Cross reactions with tuberculins on guinea pigs with single and double experimental infections with mycobacteria. Proc. Soc. Exp. BioI. and Med. 116:1024-1028. Jovin, T., A. Chrambach, M. A. Naughton. 1964. An apparatus for preparative temperature-regulated polyacrylamide gel electrophoresis. Anal. Biochem. 2:351-369. Kabot, E. A., N. M. Mayer. 1961. Experimental Immunochemistry. 2nd Ed. Charles C. Thomas, Springfield, Ill. Kahn, M. C., H. Schwarzkopf. 1933. Single cell dissociation of acid-fast bacteria. Mycobacterium of avian tuberculosis; mycobacterium of "rat leprosy." J. Bacteriol. 25:157-200. Karush, F., H. N. Eisen. 1962. Science 136:1032-1039. A theory of delayed hypersensitivity. Kendig, E. L. 1961. Unclassified mycobacteria in children. Correlation of skin tests and gastric cultures. Am. J. Dis. Children 101:749-755. Kendig, E. L. 1963. Incidence of infection and cause of a falsepositive tuberculin reaction. New Eng. J. Med. 268-1001-1002. Kim, H. K., M. Magnusson, M. W. Bentzon. 1963. Tuberculin production. 2. Comparison between amount of tuberculin obtained from the culture filtrate and by extraction of tubercle bacilli. Acta Path. et Microbiol. Scand. 58:501-510. 109 Kim, H. K., M. Magnusson, M. W. Bentzon. 1964. Tuberculin production. 6. Relationship between pH of culture filtrate and yield of tuberculo-protein and tuberculin. Acta Path. et Microbial. Scand. 61:291-302. Kirchheimer, W. F., R. S. Weiser. 1947. The tuberculin reaction: I. Passive transfer of tuberculin sensitivity with cells of tuberculous guinea pigs. Proc. Soc. Exp. BioI. and Med. 66:166-170. Klinenberg, J. R., P. M. Grimley, J. E. Seegmiller. 1965. Destructive polyarthritis due to a photochromogenic mycobacterium. New Eng. J. Med. 272:190-193. Klugh, G. A., P. C. Pratt. 1962. Experimental immunization of guinea pigs with photoch~omogenic acid-fast bacilli. Am. Rev. Resp. Dis. 85:78-83. Kniker, W. T., J. B. LaBorde. 1964. The separation of mycobacterial antigens by ion-exchange chromatography. Am. Rev. Resp. Dis. 89:29-40. Koch, R. 1890. UBER BACTERIOLOGISCHE FORCHUNG. Deutsche Med. Wchnschr. 16:756 (Translated in Lancet l:673, 1890). Koehler, L. H. 1952. Differentiation of carbohydrates by anthrone reaction rate and color intensity. Anal. Chern. 24:1576-1579. Konno, K., K. T. Kurzmann, K. T. Bird, A. Sbarra. 1958. Differentiation of human tubercle bacilli from atypical acid-fast bacilli. I. Niacin production of human tubercle bacilli and atypical acidfast bacilli. Am. Rev. Tuberc. 77:669-674. Krause, A. K. 1916. Studies in immunity to tuberculosis. Experimental studies on the cutaneous reaction to tuberculo-protein. J. Med. Res. 35:1-23. Kubica, G. P., L. P. Gleason. 1960. Studies on the catalase activity of acid-fast bacilli. I. An attempt to sub-group these organisms on the basis of their catalase activities at different temperatures and pH. Am. Rev. Resp. Dis. 81:387-391. Kubin, M., J. Kruml, J. Jelinek. 1962. Experimental generalized granulomatosis in guinea pigs produced by high doses of a photochromogenic strain of mycobacteria. Am. Rev. Resp. Dis. 85:902-909. 110 Kuhne, W. 1893. ERFAHRUNGEN UBER ALBUMOSEN UND PEPTONE. 30:221-253. Z. BioI. Kwapinski, J. B. 1963. Antigenic structure of actinomycetales. VI. Serological relationships between antigenic fractions of actinomyces and nocardia. J. of Bacteriol. 86:179-186. Larsen, C. L., E. Ribi, W. C. Wicht, R. H. List, Goode, G. 1963. Resistance to tuberculosis in mice immunized with BCG disrupted in oil. Nature 198:1214-1215. Lawrence, H. S. 1949. The cellular transfer of cutaneous hypersensitivity to tuberculin in man. Proc. Soc. Exp. BioI. and Med. 71:516-522. Lemaistre, C. 1963. Role of photochromogens in tuberculosis. New York Acad. Sci. 106:62-66. Leon, J., H. I. Hunphrey, M. A. Davis. 1964. photochromogenic mycobacteria for dogs. 90:809-810. Ann. The pathogenicity of Am. Rev. Resp. Dis. Leskowitz, S. 1960. Studies in immunization. II. Immunization with specific precipitates. J. Immunol. 85:56-65. Levene, P. A. 1898. Preliminary communication on the bio-chemistry of the bacillus tuberculosis. Med. Record 54:873-874. Lewis, T. 1927. responses. The blood vessels of the human skin and their Shaw and Sons, Ltd-.- London. ------- Lewis, U. J., M. O. Clark. 1964. Preparative methods for disk electrophoresis with special reference to the isolation of pituitary hormones. Anal. Biochem. 6:303-315. Lind, A. 1965. Immunologic analysis of mycobacterial antigens. Am. Rev. Resp. Dis. 92:54-62. Lowenstein, E., E. P. Pick. 1911. STUDIEN UBER ANTIGENBILDUNG IN ElWEISSFREIEN NAHRMEDIEN. I. BEITRAGE ZUR KENNTNIS DES TUBERKULINS. Biochem. Z. 31:142-152. Lurie, M. B. 1964. Resistance to tuberculosis: experimental studies in native and acquired defensive mechanisms. Harvard University Press. Cambridge, Mass. III Magnusson, M. 1961. Specificity of mycobacterial sensitins. I. Studies in guinea pigs with purified "tuberculin" prepared from mammalian and avian tubercle bacilli, Mycobacterium balnei, and other ac.id-fast bacilli. Am. Rev. Resp. Dis. 83:57-68. M.~ Mo w. Bentzon~ Mo An Bleiker~ H. C. Engbaek~ w. A. Griep, A. Manten. 1961a. Comparative intradermal reactions of tuberculosis patients to purified tuberculin (PPD) and purified sensitin prepared from Mycobacterium ________ Acta Tuberc. Scand. 40 85-102. Magnusson~ Magnusson, M., H. C. Engbaek, W. M. Bentzon. 1961b. Specificity of mycobacterial sensitins. II. Studies in guinea pigs with purified sensitin prepared from unclassified acid-fast bacilli. Am. Rev. Resp. Dis. 83:69-84. Magnusson, M. 1962. Specificity of sensitins. III. Further studies in guinea pigs with sensitin of various species of mycobacterium and nocardia. Am. Rev. Resp. Dis. 86:395-404. Magnusson, M., H. K. Kim, M. W. Bentzon. 1964. Tuberculin production. 5. Relationship between age of culture of Mycobacterium tuberculosis and tuberculin activity of culture filtrates. Acta Path. et Microbiol. Scand. 60:557-570. Martosh, A., V. Eslami, R. J. Atwell. 1962. Simultaneous comparative skin testing of tuberculous patients with antigens derived from M. tuberculosis and certain unclassified mycobacteria. Dis. Chest 41:624-628. Masucci, P., K. L. McAlpine. 1930. Biochemical studies of bacterial derivatives. X. Preparation of human tubercle bacillus protein MA-lOO. Proc. Soc. Exp. BioI. and Med. 27:661-663. Matson, C. F. 1965. Polyacrylamide gel electrophoresis. A simple system using gel columns. Anal. Biochem. 13:294-304. Maxwell, K. W., T. M. Dietz, S. Marcus, G. A. Hill. 1964. Comparative sensitivity of tuberculin tine test and mantoux test in guinea pigs. Am. Rev. Resp. Dis. 89:926-928. McCluskey, R. T., B. Benacerraf, J. W. McCluskey. 1963. Studies on the specificity of the cellular infiltrate in delayed hypersensitivity reactions. J. Immunol. 90:466-477. 112 Meyer, S. N., C. E. Palmer, P. Q. Edwards. 1956. Experimental studies of vaccination, allergy and immunity in tuberculosis. VII. Effect of varying the time interval between vaccination and challenge. Am. J. Hyg. 65:248-263. Middlebrook, G., R. Dubos. 1948. Specific serum agglutination of erythrocytes sensitized with extracts of tubercle bacillio J. Ex. Med. 88:521-528. Mitchell, A., W. Nelson, T. LeBlanc. 1935. Studies in immunity. V. Effect of acute diseases on the reaction of the skin to tuberculin. Am. J. Dis. Child. 49:695-702. Mueller, J. H. 1925. Chemical studies on tuberculin. Acad. Sci. U. S. 11:23-25. Najarian, J. S., J. D. Feldman. delayed hypersensitivity. Proc. Natl. Specificity of passively transferred J. Exp. Med. 118:341-352. O'Grady, F. 1957. Tuberculin sensitization in the mouse. J. Exper. Path. 38:319-325. Brit. Ornstein,L.1964. Disc electrophoresis. I. Background and theory. Ann. New York Acad. Sci. 121:321-349. Ornstein, L. 1966. Personal communication. Pappenheimer, A. M., J. Freund. 1959. Cellular and humoral aspects of the hypersensitive states. H. S. Lawrence Ed., New York. Paul B. Hoeber, Inc. pp. 67-85. Paretzky, M. 1938. Intracutaneous tuberculin test. Tuberc. 38:81-95. Am. Rev. Parlett, R. C., G. P. Youmans. 1958. Antigenic relationships between ninety-eight strains of mycobacteria using gel-diffusion precipitation techniques. Am. Rev. Tuberc. 12:450-461. Pepys, J. 1955. The relationship of nonspecific and specific factors in the tuberculin reaction. Am. Rev. Tuberc. 71:49-73. Pepys, J., R. Augustin, A. B. Paterson. 1959. Common antigenic components of mycobacterial extracts. Tubercle 40:163-172. 113 Petroff, S. A., F. W. Stewart. 1925. Immunological studies in tuberculosis. III. Concerning allergic reactions obtained in animals sensitized with killed tubercle bacilli. J. of Immuno1. 10:677-709. Petroff, S. A., A. Branch, F. B. Jennings, Jr. 1929. Immunological studies in tuberculosis" Vn Resistance of animals sensitized with heat-killed tubercle bacilli to a measured infecting dose. J. of Immuno1. 16:233-257. Pfeiffer, J. H., H. Trunk, J. Leyacker. CHEMISCHEN NATUR DES TUBERKULINS. 24:1115-1117. 1911. ZUR FRANGE DER Wien. K1in. Wochschr. Pfuetze, K. H., L. V. Vo, A. R. Reimann, G. S. Berg, W. Lester. 1965. Photochromogenic mycobacterial pulmonary disease. Am. Rev. of Resp. Dis. 92:470-475. Pharmacopeia of the United States of America. 1965. Mack Publishing Co., Easton, Pa. 17th revision. Phillips, S., J. C. Larkin. 1964. Atypical pulmonary tuberculosis caused by unclassified mycobacteria. Ann. Interu. Med. 60:401-408. Pilcher, J. D. 1930. Diminution in the circulation of the skin: a factor in decreasing the cutaneous tuberculin reaction. Ann. Rev. Tuberc. 21:669-683. Poppius, H., O. Wasz-Htlckert, M. Tuomioja. 1960. A clinical study of the quantitative tuberculin test and of the serological tests for tuberculosis in adult pulmonary tuberculosis. Acta Tuberc. Scand. 39:235-242. Pun. J., L. Lombrozo. 1964. Microelectrophoresis of brain and pineal proteins in polyacrylamide gel. Anal. Biochem. ~:9-20. Racusen, D., N. Ca1vanico. 1964. Preparative electrophoresis on polyacrylamide gel. Anal. Biochem. 7:62-66. Raffel, S. 1946. The relationship of acquired resistance, allergy antibodies and tissue reactivities to the components of the tubercle bacillus. Am. Rev. Tuberc. 54:564-573. Raffel, S. 1948. The components of the tubercle bacillus responsible for the delayed type of "infectious" allergy. J. Inf. Dis. 82:267-293. 114 Raffel, S. 1961. Immunity. Appleton-Century-Crofts Inc. New York. Raymond, S., M. Nakamichi. 1964. Electrophoresis in synthetic gels. II. Effect of gel concentration. Anal. Biochem. 7:225-232. Ribi, E., C. L. Larson, W. List, W. Wicht. 1958. Immunologic significance of the cell wall of mycobacteria Proco SOCo Exp. BioI. and Med. 98:263-265. u Ribi, E., T. Perrine, R. List, W. Brown, G. Goode. 1959. Use of pressure cell to prepare cell walls from mycobacteria. Proc. Soc. Exp. BioI. Med. 100:647-649. Ribi, E., C. L. Larson, W. Wicht, R. List, G. Goode. 1965. Resistance to experimental tuberculosis stimulated by fractions from attenuated tubercle bacilli. (30009) Proc. Soc. Exp. BioI. and Med. 118 926-933. Rich, A. R. 1951. The pathogenesis of tuberculosis. Thomas. Springfield, Ill. Runyon, E. H. 1965. 14:235-287. Pathogenic mycobacteria. Charles C. Adv. Tuberc. Res. Runyon, E. H. 1959. Anonymous mycobacteria in pulmonary disease. Med. Clin. North America 43:273-290. Ruppel, W. G. 1898. ZUR CHEMIE DER TUBERKELBACILLEN. Chern. 26:218-232. I. Z. Physiol. Sabin, F. R., F. R. Miller, C. A. Doan, B. K. Wiseman. 1931. A study of the toxic properties of tuberculoproteins and polysaccharides. J. Exp. Med. 53:51. Schaefer, W. B. 1965. Serologic identification and classification of the atypical mycobacteria by their agglutination. Am. Rev. Resp. Dis. 92:85-93. Seibert, F. B. 1926a. V. The effect of the proteolytic enzymes on tuberculin proteins and the activity of tuberculin. Am. Rev. Tuberc. 13:431-440. Seibert, F. B., E. R. Long. 1926b. IV. Ammonium-sulfate precipitation of the proteins of tuberculin. Am. Rev. Tuberc. 13:408-430. 115 Seibert, F. B. 1926c. The isolation of a crystalline protein with tuberculin activity. Science 63:619-620. Seibert, F. B. 1928. The chemical composition of the active principle of tuberculin. XI. An improved and simplified method for making a standard undenatured tuberculin of any desired strength and a method of chemical assay. J. Biol. Chern. 78:345-362. Seibert, F. B. 1930. The chemical composition of the active principle of tuberculin. XII. Precipitin tests and differentiation of various tuberculin and Timothy-bacillus proteins. Am. Rev. Tuberc. 21:370-382. Seibert, F. B., B. Munday. 1932. XIV. A precipitated purified tuberculin protein suitable for the preparation of a standard tuberculin. Am. Rev. Tuberc. 25:724-737. Seibert, F. B., N. Morley. 1933. The relationship of the tuberculin proteins of different acid-fast bacilli to sensitization as indicated by their reactivity in sensitized animals. J. Immunol. 24:149-156. Seibert, F. B., K. O. Pedersen, A. Tiselius. 1938. Molecular weight,electrochemical and biological properties of tuberculin protein and polysaccharide molecules. J. Exp. Med. 68:413-438. Seibert, F. B., E. Dufour. 1940. Purified protein derivative - its isolation from old tuberculin and fractionation of the residue. A m. Rev. Tuberc. 41:57-70. Seibert, F. B. 1940. Removal of the impurities, nucleic acid and polysaccharide from tuberculin protein. J. Biol. Chern. 113:593-604. Seibert, F. B. 1941a. The chemistry of the proteins of the acidfast bacilli. Bacteriol. Rev. 5:69-95. Seibert, F. B. 1941b. History of the development of purified tuberculin. Am. Rev. Tuberc. 44:1-8. Seibert, F. B., J. T. Glenn. 1941. Tuberculin purified protein derivative. Am. Rev. Tuberc. 44:9-25. Seibert, F. B. 1949. The isolation of three different proteins and two polysaccharides from tuberculin by alcohol fractionation. Their chemical and biological properties. Am. Rev. Tuberc. 59:86-100. 116 Seibert, F. B., M. Stacey, P. W. Kent. 1949. An antigenic polysaccharide, "polysaccharide II", isolated from tuberculin. Biochim. et Biophys. Acta 1:632-640. Seibert, F. B. 1950. Constituents of mycobacteria. Microbiol. 4:35-52. Seibert, F. B. 1953. 13:251-261. Tuberculin proteins. Ann. Rev. Faraday Soc. Discussions Seibert, F. B., E. Soto-Figueroa, E. H. Dufour. 1955. Isolation, identification, and classification of proteins of tuberculin and the tubercle bacillus. Am. Rev. Tuberc. 71:704-721. Seibert, F. B., E. Soto-Figueroa. 1957. Study of tuberculin protein and polysaccharide antigens by gel-diffusion technique. Am. Rev. Tuberc. and Pule Dis. 75:601-607. Singer, E., G. M. J. Rodda. 1961. Non-specific sensitization to old tuberculin: the ubiquity of acid-fast organisms. Tubercle 42:325-332. Sirks, J. L., M. A. Bleiker. 1963. Investigations with a "tuberculin active peptide" Acta Tuber. et Pneumol. Scand. 43:87-98. Smith, D. T., W. W. Johnston. 1964. Single and multiple infections with typical and atypical mycobacteria. Am. Rev. Resp. Dis. 90:899-912. Smith, D. W., J. Robertsen. 1962. Immunogenicity in guinea pigs of lipid fractions of M. tuberculosis. Am. Rev. Resp. Dis. 85:398-401. Smith, D. W., G. B. Fregnan, L. D. Richardson, E. Valdiva. 1964. Induction of acquired resistance in guinea pigs with defatted mycobacterium tuberculosis in guinea pigs. J. Bacteriol. 88:87-92. Someya, S., O. Havashi, Y. Yamamura. 1962. Studies on the antigenicity of tuberculin-active peptide. Am. Rev. Resp. Dis. 86:542-546. Stacey, M. 1955. The proteins of the tubercle bacillus. Ciba Foundation Symposium on Experimental Tuberculosis Bacillus and Host. Little, Brown and Co. Boston, Mass. 117 Steenken, W., W. H. Oatway, S. A. Petroff. 1934. Biological studies of the tubercle bacillus. III. Dissociation and pathogenicity of the Rand S variants of the human tubercle bacillus (H37). J. Expo Med. 60:515-540. Steward, F. C., J. T. Barber. 1964. The use of acrylamide gel electrophoresis in th~ investigation of the soluble proteins of plants. Ann. New York Acad. of Sci. 131:525-531. Stodola, F. H., A. Lesuk, R. J. Anderson. 1938. The chemistry of the lipids of tubercle bacillio LIV. The isolation and properties of mycolic acid. J. BioI. Chern. 126:505-513. Takahashi, Y., T. Onodera, K. Yamamoto. 1961. The behavior of three different kinds of antibodies in tuberculosis: antiprotein, antipolysaccharide, and antiphosphatide. I. Experimental tuberculosis. J. Exp. Med. :555-568. Takahashi, Y. 1962. Specific serum agglutination of kaolin particles sensitized with tubercle phosphatide and its clinical evaluation as serodiagnostic test for tuberculosis. Am. Rev. Resp. Dis. 85:708-719. Takeya, K., Y. Zinnaka, K. Yamaura, T. Toda. 1960. Bacteriophage susceptibility and tuberculin specificity of unclassified mycobacteria. Am. Rev. Resp. Dis. 81:674-681. Takeya, K., K. Nakashima, Y. Tadehara, T. Toda. 1961. Tuberculinspecificity of unclassified mycobacteria isolated in Japan. Jap. J. Microbiol. 2:51-59. Takeya, K., Y. Sinnaka, H. Nakayama, Y. Inoue, Y. Hamada, T. Toda. 1962. Epidemiological study of unclassified mycobacteria infection using purified tuberculin protein. Kyushu J. Med. 13:181-192. Tarshis, M. S., A. W. Frisch. 1952. Chromogenic acid-fast bacilli from human sources. Am. Rev. Tuberc. 65:302-315. Taubenhaus, L. J. 1962. The tuberculin reactions of patients admitted to a surburban New England Hospital. Am. Rev. Resp. Dis. 85:679-701. Tepper, B. S. 1965. Modification of cellular constituents during growth of Mycobacterium phlei. Am. Rev. Resp. Dis. ~:75-82. 118 Toenniessen, E. 1924. DIE SPEZIFISCHE ERKENNUNG UND BEHANDLUNG DER TUBERKULOSE MIT EINEM AUS TUBERKELBAZILLEN GEWONNENEN ElWEISSKORPER (TEBEPROTIN). Deut. Med. Wochschr. 50:629-632. Topley and Wilson. Principles of Bacteriology and Immunity. 1955. 4th Ed. Williams and Wilkins Co. Baltimore, Maryland. Trakatellis, A. C., W. R. Stinebring, A. E. Axelrod. 1963. Studies on systemic reactivity to purified protein derivative (PPD) and endotoxin. I. Systemic reactivity to PPD in pyridoxinedeficient guinea pigs. J. of Immunol. 91:39-45. Tsukamura, M., T. Abo. 1958. Some genetic considerations of INH resistant M. avium. Am. Rev. Tuberc. 77:81-87. Tsukamura, M., S. Tsukamura. 1964a. Different activities of mouse organs in eliminating photochromogenic mycobacteria. Acta Tuberc. et Pneumol. Scand. 44:145-151. Tsukamura, M., S. Tsukamura. 1964b. Relationship between the dose of phenol-killed mycobacteria and appearance of immunity. Jap. J. Tuberc. 1l:55-57. Uhr, J. W., G. Weissman, L. Thomas. 1963. Acute hypervitaminosis A in guinea pigs. II. Effects on delayed-type sensitivity. Proc. Soc. Exp. BioI. and Med. 112:287-291. Waksman, B. H., M. Matoltsy. 1958. Quantitative study of local passive transfer of tuberculin sensitivity with peritoneal exudate cells in the guinea pig. J. Immunol. 81:235-241. Waksman, B. H. 1960. Ciba Foundation Symposium ~Cellular Aspects of Immunity. Eds. Wolstenholme, G. E. W. and O'Connor, M. Little, Brown and Co. Boston, Mass. p. 280. Walker, S. H., L. R. Patron. 1964. Unclassified mycobacterial infections in children. Am. J. Dis. Child. 108:460-469. Warfvinge, L. E. 1962. Studies on tuberculin with special reference to endotuberculin, an unheated fraction extracted by freezing and thawing of living virulent tubercle bacteria. Acta Tuberc. et Pneumol. Scand. 53:1-36 (suppl.) Wayne, L. G. 1962. Two varieties of Mycobacterium kansasii with different clinical significance. Am. Rev. of Resp. Dis. 86:651-656. 119 Weinstein, D., J. F. Douglas. 1964. A note on disc electrophoresis technique. Anal. Biochem. 6:474-475. Weyer, E. M. 1966. Purification of Materials. of Sci. 137:1-402. Ann. New York Acad. Weyl, T. 1891. Sur chemie und toxicologie des tuberkelbacillus. Deut. Med. Wochschr. 17:256-257. Willett, H. P. 1963. Biochemical and biological properties of mycobacterium tuberculosis. Ann. New York Acad. of Sci. 106 16-31. Williams, D. E., R. A. Reisfeld. 1964. Disc electrophoresis in polyacrylamide gels: extension to new conditions of pH and buffer. Ann. New York Acad. of Sci. 121:373-381. Wilson, B. A., R. A. Slepecky. 1962. The lysis of Mycobacterium tuberculosis by microbial enzymes. Am. Rev. Resp. Dis. 86:246-251. Yoneda, M. 1964. A study of two major extracellular protein antigens (a and B) of Mycobacterium tuberculosis. World Health Organization. Meeting of Advisers on Immunogenic Agents in Tuberculosis. Geneva. Youmans, A. S., G. P. Youmans. 1964. Nature of the labile immunogenic substance in the particulate fraction isolated from Mycobacterium tuberculosis. J. Bacteriol. 88:1030-1037. Zinnaka, Y., K. Takeya, H. Nakayama, T. Toda. 1962. Further studies of purified tuberculin protein of unclassified mycobacteria: a modification of purification method.~ 'Kyushu J. Med. Science 13:149-158.