ABSTRACT ANTIGENIC, CHEMICAL, CHROMATOGRAPHIC, AND ELECTROPHORETIC ANALYSES OF UNHEATED CULTURE FILTRATES OF MYCOBACTERIUM BOVIS by Terry J. Dardas The antibody response of rabbits inoculated with heat or beta-propriolactone killed Mycobacterium bovis and/or unheated culture filtrates was studied. Antibody was measured by bacterial agglutination and passive hemagglutin- ation (Middlebrook-Dubos test). The sequential production of mercaptoethano1-sensitive (MES) and mercaptoethanol- resistant (MER) antibody was measured. The relative amount and temporal sequence of synthesis of each type of antibody varied among the experimental groups. Increasing the local- ized dose or concentration of antigen shortened the interval between the detection of MES and MER antibody. When the local antigenic stimulus was decreased (no adjuvant or systemic injection of antigen) or repeated small inoculations were made weekly, MES antibody production was prolonged and the initiation of MER antibody synthesis delayed. Concentrated unheated filtrates from 2, 5, 4, 5, 6, and 7 month cultures of M, bovis were prepared and analyzed chemically, and by chromatography, electrophoresis, and Terry J. Dardas immunodiffusion. The culture filtrates were readily sepa— rated chromatographically into three or more fractions. The chemical and antigenic composition of the fractions varied with the age of the culture filtrate. The amount of high molecular weight protein in the culture filtrates de- creased with longer periods of incubation of the culture. Disc electrophoresis was a very effective means of separating the culture filtrate components. Between 15 and 30 protein components, and 5 to 9 polysaccharide or glyco- protein components were detected in the culture filtrates. All of the protein components had molecular weights of 5,000 or greater. The number of protein bands increased from 16 in two—month-old culture filtrates to 50 in four-month-old culture filtrates. Thereafter, the number decreased; only 15 bands were detected in filtrates from seven-month-old cultures. The number of precipitinogens in the culture filtrates decreased with longer periods of incubation. Sixteen immuno- precipitates were detected in immunograms of two-month-old culture filtrates; only six immunoprecipitates were detected in immunograms of filtrates of cultures incubated for seven months. More precipitinogens were detected in the culture filtrates by immunoelectrophoresis. The number of precipi- tinogens detected by this method varied from 8 to 21. ANTIGENIC, CHEMICAL, CHROMATOGRAPHIC, AND ELECTROPHORETIC ANALYSES OF UNHEATED CULTURE FILTRATES OF MYCOBACTERIUM BOVIS BY Terry Jf/Dardas A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbiology and Public Health 1967 (i\ 0541 out) 0 u -+ C". U‘ 0 ACKNOWLEDGMENTS The author wishes to express his sincere appreciation and thanks to Drs. V. H. and W. L. Mallmann for their con- tinued interest and guidance throughout this investigation. The assistance and suggestions offered by Dr. G. F° Dardas and the other members of the tuberculosis project are also greatly appreciated. A special thanks goes to the authors wife and family who have been the main source of encouragement and patience throughout this investigation. ii TABLE OF CONTENTS Page INTRODUCTION . . . . . . . . . . . . . . . . . . . . 1 LITERATURE REVIEW. . . . . . . . . . . . . . . . . . 3 MATERIALS AND METHODS. . . . . . . . . . . . . . . . 30 Mycobacterium bovis . . . . . . . . . . . . . . 50 Culture filtrates . . . . . . . . . . . . . . . 50 Chemical analyses . . . . . . . . . . . . . . . 51 Production of antisera. . . . . . . . . . . . . 55 Statistical analyses. . . . . . . . . . . . . 58 Ouchterlony double diffusion. . . . . . . . . . 40 Immunoelectrophoresis . . . . . . . . . . . . . 41 Disc electrophoresis. . . . . . . . . . . . . . 43 Dialysis. . . . . . . . . . . . . . . . . . . . 46 Ion exchange column chromatography. . . . . . . 47 Molecular exclusion chromatography. . . . . . . 48 Cellulose acetate membrane electrophoresis. . . 50 RESULTS. . . . . . . . . . . . . . . . . . . . . . . 51 Antibody responses of rabbits to mycobacterial cells and culture filtrates. . . . . . . . 51 Precipitins elicited by alum-precipitated and aluminum chloride precipitated culture filtrate . . . . . . . . . . . . . . . . . 74 Chemical analyses of culture filtrates. . . . . 74 Chromatography of culture filtrates . . . . . . 74 Dialysis of culture filtrates . . . . . . . . . 105 Cellulose acetate membrane electrophoresis of culture filtrates. . . . . . . . . . . . . 105 Disc electrophoresis of culture filtrates . . . 109 Analyses of the culture filtrates and chroma— tographic fractions by immunodiffusion . . 118 Immunoelectrophoresis of culture filtrates. . . 150 DISCUSSION . . . . . . . . . . . . . . . . . . . . . 141 SUMMARY. . . . . . . . . . . . . . . . . . . . . . . 162 LITERATURE CITED . . . . . . . . . . . . . . . . . . 164 iii TABLE 1. 10. LIST OF TABLES Inoculation schedule for rabbits which re- ceived killed cells and/or Mycobacterium bovis culture filtrate. . . . . . . . . . . . Inoculation schedule for rabbits which re— ceived Mycobacterium bovis culture filtrate precipitated with aluminum hydroxide. . . Composition of the stock reagents used for disc electrophoresis. . . . . . . . . . . . Composition of the working solutions used for disc electrophoresis. . . . . . . . . . . . . Mean titers of polysaccharide-specific anti- body produced by rabbits inoculated with killed cells and/or Mycobacterium-bovis cul- ture filtrate . . . . . . . . . . . . . . . . Mean titers of agglutinins produced by rab~ bits inoculated with killed cells and/or Mycobacterium bovis culture filtrate. . . . Mean titers of mercaptoethanol-sensitive and mercaptoethanol-resistant polysaccharide- specific antibody produced by rabbits inocu— lated with killed cells and/or Mycobacterium bovis culture filtrate. . . . . . . . . . . . Mean titers of mercaptoethanol-sensitive and mercaptoethanol-resistant bacterial agglu— tinins produced by rabbits inoculated with killed cells and/or Mycobacterium bovis cul- ture filtrate . . . . . . . . . . . . . . . . Chemical composition of the six unheated, ten fold concentrated Mycobacterium bovis culture filtrates . . . . . . . . . . . . . . . . . . Chemical composition of the fractions obtain- ed by chromatography of Mycobacterium bovis culture filtrate Al on Sephadex G-25. . . . . iv Page 34 59 44 45 52 55 54 55 75 85 LIST OF TABLES - Continued TABLE 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Chemical composition of the fractions ob- tained by chromatography of Mycobacterium bovis culture filtrate B1 on Sephadex G-25. . Chemical composition of the fractions obtain— ed by chromatography of Mycobacterium bovis culture filtrate C1 on Sephadex G-25. . . . . Chemical composition of the fractions obtain- ed by chromatography of Mycobacterium bovis culture filtrate D1 on Sephadex G-25. . . . . Chemical composition of the fractions obtain- ed by chromatography of Mycobacterium bovis culture filtrate E1 on Sephadex G—25. . . . . Chemical composition of the fractions obtain— ed by chromatography of Mycobacterium bovis culture filtrate F1 on Sephadex G-25. . . . . Distribution of 280 mu-absorbing and Folin— reacting material in fractions obtained by chromatography of the six Mycobacterium bovis culture filtrates on Bio-Gels P-100, P—150, and P—200 . . . . . . . . . . . . . . . . . . Distribution of Folin-reacting material in Fraction I obtained by chromatography of the six Mycobacterium bovis culture filtrates on Sephadex 6-25 and Bio-Gels P—100, P—150, and P-200 . . . . . . . . . . . . . . . . . . . . Number of components found in the six Myco~ bacterium bovis culture filtrates by disc electrophoresis, Ouchterlony double diffusion, and immunoelectrophoresis . . . . . . . . . . Rf values of the protein components in the six Mycobacterium bovis culture filtrates separated by disc electrophoresis . . . . . . Precipitinogens demonstrated in the six Myc - bacterium bovis culture filtrates and frac- tions obtained by chromatography on Sephadex G-25 and Bio-Gels P-100, P-150, and P-200 . . Page 84 85 86 87 88 102 105 110 117 119 LIST OF TABLES - Continued TABLE Page 21. Precipitinogens demonstrated in the six Myco- bacterium bovis culture filtrates by Opchterlony double diffusion with antiserum B . . . . . . . . . . . . . . . . . . . . . . 155 vi FIGURE 1. 2. 10. 11. 12. 15. 14. LIST OF FIGURES Hemagglutinins produced by rabbits in Group I (Table 1). . . . . . . . . . . . . . . . . . . Bacterial agglutinins produced by rabbits in Group I (Table 1). . . . . . . . . . . . . . . Hemagglutinins produced by rabbits in Group II (Table 1). . . . . . . . . . . . . . . . . . . Bacterial agglutinins produced by rabbits in Group II (Table 1) . . . . . . . . . . . . . . Hemagglutinins produced by rabbits in Group III (Table 1). . . . . . . . . . . . . . Bacterial agglutinins produced by rabbits in Group III (Table 1). . . . . . . . . . . . . . Hemagglutinins produced by rabbits in Group IV (Table 1). . . . . . . . . . . . . . . . . . . Bacterial agglutinins produced by rabbits in Group IV (Table 1) . . . . . . . . . . . . . . Hemagglutinins produced by rabbits in Group V (Table1)................... Bacterial agglutinins produced by rabbits in Group V (Table 1). . . . . . . . . . . . . . Hemagglutinins produced by rabbits in Group VI (Table 1). . . . . . . . . . . . . . . . . . . Bacterial agglutinins produced by rabbits in Group VI (Table 1) . . . . . . . . . . . . . Hemagglutinins produced by rabbits in Group VII (Table 1). . . . . . . . . . . . . . . . . Bacterial agglutinins produced by rabbits in Group VII (Table 1). . . . . . . . . . . . . . vii Page 56 57 58 59 6O 61 62 65 64 65 66 67 68 69 LIST OF FIGURES - Continued FIGURE 15. 16. 17. 18. 19. 20. 21. 22. 25. 24. 25. 26. Hemagglutinins produced by rabbits in Group VIII (Table 1). . . . . . . . . . . . . . . . Bacterial agglutinins produced by rabbits in Group VIII (Table 1). . . . . . . . . . . . . Ultraviolet absorption Spectrum of M cobac- terium bovis culture filtrate C (Table 9) . . Molecular exclusion chromatography of Myco- bacterium bovis culture filtrate A (Table 9) on Sephadex G-25. . . . . . . . . . 1 . . . Molecular exclusion chromatography of Myco- bacterium bovis culture filtrate B (Table 9) on Sephadex G-25. . . . . . . . . . . . . . . Molecular exclusion chromatography of Myco- bacterium bovis culture filtrate C (Table 9) on Sephadex G-25. . . . . . . . . . . . . . . Molecular exclusion chromatography of Myco- bacterium bovis culture filtrate D (Table 9) on Sephadex G-25. . . . . . . . . . . . . . Molecular exclusion chromatography of Myco- bacterium bovis culture filtrate E (Table 9) on Sephadex G-25. . . . . . . . . . . . . . Molecular exclusion chromatography of Myco— bacterium bovis culture filtrate F (Table 9) on Sephadex G-25. . . . . . . . . . . . . . . Ultraviolet absorption Spectrum of Fraction I (Table 12) obtained by chromatography of Mycobacterium bovis culture filtrate C (Table 9) on Sephadex G-25 . . . . . . . . . . . . Ultraviolet absorption spectrum of Fraction II (Table 12) obtained by chromatography of Mycobacterium bovis culture filtrate C (Table 9) on Sephadex G-25 . . . . . . . . . . . . . Ultraviolet absorption Spectrum of Fraction III(Table 12) obtained by chromatography of Mycobacterium bovis culture filtrate C (Table 9) on Sephadex G-25 . . . . . . . . . . . . . viii Page 70 71 76 77 78 79 80 81 82 91 92 95 LIST OF FIGURES - Continued FIGURE 27. 28. 29. 50. 51. 52. 55. 54. 55. 56. Molecular exclusion chromatography of Myco- bacterium bovis culture filtrate A (Table 9) on Bio-Gels P-100, P-150, and P-200. . . . . . Molecular exclusion chromatography of Myco— bacterium bovis culture filtrate B (Table 9) on Bio—Gels P-100, P-150, and P-200. . . . . . Molecular exclusion chromatography of Myco— bacterium bovis culture filtrate C (Table 9) on Bio-Gels P—100, P—150, and P-200. . . . . . Molecular exclusion chromatography of Myco- bacterium bovis culture filtrate D (Table 9) on Bio-Gels P-100, P-150, and P-200. . . . . . Molecular exclusion chromatography of Myco- bacterium bovis culture filtrate E (Table 9) on Bio-Gels P-100, P—150, and P-200. . . . . . Molecular exclusion chromatography of Myco- bacterium bovis culture filtrate F (Table 9) on Bio-Gels P-100, P-150, and P-200. . . . . . DEAE-cellulose chromatography of Fraction I obtained by molecular exclusion chromatography of Mycobacterium bovis culture filtrate C (Table 9) on Sephadex G-25 . . . . . . . . . Molecular exclusion chromatography of dialyz- able (broken line) and non-dialyzable (solid line) constituents of Mycobacterium bovis culture filtrate C (Table 9) on Sephadex G-25. Immunogram of dialyzable (C) and non-dialyz— able (B) constituents from Mycobacterium bovis culture filtrate C (Table 9) (A). The antiserum was placed in well D . . . . . . . . Cellulose acetate electrophorogram of Fraction I obtained by molecular exclusion chromatog- raphy of Mycobacterium bovis culture filtrate C (Table 9) on Sephadex G-25 . . . . . . . . . ix Page 96 97 98 99 100 101 104 106 107 108 LIST OF FIGURES - Continued FIGURE 57. 58. 59. 40. 41. 42. Schematic representation of the protein (A) and PAS positive (B) components in disc electrophorograms of Mycobacterium bovis culture filtrate A (Table 9) (CF) and Frac— tion I (F-I) obtained by chromatography on Sephadex G-25. The letters S and L refer to the spacer and lower gel areas, respectively. Schematic representation of the protein (A) and PAS positive (B) components in disc electrophorograms of Mycobacterium bovis culture filtrate B (Table 9) (CF) and Frac- tion I (F—I) obtained by chromatography on Sephadex G-25. The letters S and L refer to the Spacer and lower gel areas, respectively. Schematic representation of the protein (A) and PAS positive (B) components in disc electrophorograms of Mycobacterium bovis culture filtrate C (Table 9) (CF) and Frac— tion I (F-I) obtained by chromatography on Sephadex 6-25. The letters S and L refer to the spacer and lower gel areas, respectively. 1 Schematic representation of the protein (A) and PAS positive (B) components in disc electrophorograms of Mycobacterium bovis culture filtrate D (Table 9) (CF) and Frac- tion I (F-I) obtained by chromatography of Sephadex G-25. The letters S and L refer to the Spacer and lower gel areas, respectively. Schematic representation of the protein (A) and PAS positive (B) components in disc electrophorograms of Mycobacterium bovis culture filtrate B (Table 9) (CF) and Frac- tion I (F-I) obtained by chromatography of Sephadex G-25. The letters S and L refer to the spacer and lower gel areas, respectively. Schematic representation of the protein (A) and PAS positive (B) components in disc electrophorograms of Mycobacterium bovis culture filtrate F (Table 9) (CF) and Frac— tion I (F-I) obtained by chromatography on Sephadex G-25. The letters S and L refer to the spacer and lower gel areas, respectively. Page 111 112 115 114 115 116 LIST OF FIGURES - Continued FIGURE Page 45. Ouchterlony double diffusion of Mycobacterium bovis culture filtrate A (Table 9) (A) and Fractions I (B), II (c), and III (D) obtained by chromatography on Sephadex G—25. The anti- serum was placed in the center well. . . . . . 120 44. Ouchterlony double diffusion of Mycobacterium bovis culture filtrate B (Table 9) (A) and Fractions I (B), II (c), and III (D) obtained by chromatography on Sephadex G-25. The anti- serum was placed in the center well. . . . . . 121 45. Ouchterlony double diffusion of Mycobacterium bovis culture filtrate C (Table 9) (A) and Fractions I (B), II (c), and III (D) obtained by chromatography on Sephadex 6—25. The anti— serum was placed in the center well. . . . . . 122 46. Ouchterlony double diffusion of Mycobacterium bovis culture filtrate D (Table 9) (A) and Fractions I (B), II (C), and III (D) obtained by chromatography on Sephadex G-25. The anti- serum was placed in the center well. . . . . . 125 47. Ouchterlony double diffusion of Mycobacterium bovis culture filtrate E (Table 9) (A) and Fractions I (B), II (C), and III (D) obtained by chromatography on Sephadex 6-25. The anti— serum was placed in the center well. . . . . . 124 48. Ouchterlony double diffusion of Mycobacterium bovis culture filtrate F (Table 9) (A) and Fractions I (B), II (C), and III (D) obtained by chromatography on Sephadex 6—25. The anti- serum was placed in the center well. . . . . . 125 49. Ouchterlony double diffusion of Mycobacterium bovis culture filtrate A (Table 9) (A) and Fraction I obtained by chromatography on Sephadex G-25 (B) and Bio-Gels P-100 (C), P-150 (D), and P-200 (E). The antisera were placed in the center wells (F) . . . . . . . . 127 50. Ouchterlony double diffusion of Mycobacterium bovis culture filtrate B (Table 9) (A) and Fraction I obtained by chromatography on Sephadex 6-25 (B), and Bio-Gels P-1OO (C), P-150 (D) and P-200 (E). The antisera were placed in the center wells (F) . . . . . . . . 128 xi LIST OF FIGURES — Continued FIGURE 51. 52. 55. 54. 55. 56. 57. 58. 59. 60. Ouchterlony double diffusion of Mycobacterium bovis culture filtrate C (Table 9) (A) and Fraction I obtained by chromatography on Sephadex G—25 (B) and Bio-Gels P—100 (C), P—150 (D), and P-200 (E). The antisera were placed in the center wells (F). . . . . . . Ouchterlony double diffusion of Mycobacterium bovis culture filtrate D (Table 9) (A) and Fraction I obtained by chromatography on Sephadex G-25 (B) and Bio-Gels P-1OO (C), P-150 (D), and P-2OO (E). The antisera were placed in the center wells (F). . . . . . . . Ouchterlony double diffusion of Mycobacterium bovis culture filtrate E (Table 9) (A) and Fraction I obtained by chromatography on Sephadex 6-25 (B) and Bio—Gels P-100 (C), P-150 (D), and P-200 (E). The antisera were placed in the center wells (F). . . . . . . . Ouchterlony double diffusion of Mycobacterium bovis culture filtrate F (Table 9) (A), and Fraction I obtained by chromatography on Sephadex G-25 (B) and Bio-Gels P-100 (C). The antiserum was placed in the center well . Immunoelectrophorogram of Mycobacterium bovis culture filtrate A (Table 9). . . . . . . . Immunoelectrophorogram of Mycobacterium bovis culture filtrate B (Table 9). . . . . . . . . Immunoelectrophorogram of Mycobacterium bovis culture filtrate C (Table 9): . . . . . . . . Immunoelectrophorogram of Mycobacterium bovis culture filtrate D (Table 9). . . . . . . . . Immunoelectrophorogram of Mycobacterium bovis culture filtrate E (Table 9). . . . . . . . . Immunoelectrophorogram of Mycobacterium bovis culture filtrate F (Table 9). . . . . . . . xii Page 129 150 151 152 155 156 157 158 159 140 INTRODUCTION Mycobacteria have been studied extensively since the discovery of tubercle bacilli by Robert Koch in 1882. The clinical manifestations and pathological characteristics of tuberculosis have been known for many years. Yet little progress has been made in understanding the pathogenesis of the disease or the immunological responses to the causa- tive agent. Many attempts have been made to find specific antigens,antibodies, or other substances in the blood of diagnostic or prognostic value. The tests are not reliable. Diagnosis depends on a combination of the results from tuberculin tests, X-rays, and the presence of tubercle bacilli in clinical Specimens. It must be confirmed by isolation and identification of the causative agent. A basic understanding of the mechanism of antibody production against mycobacterial antigens would greatly supplement our knowledge and understanding of tuberculoimmunity and sensitivity. Perhaps the greatest challenge lies in the isolation of specific antigens or sensitins from mycobacteria. One of the greatest limitations of immunologic tests in tuberculosis is their inability to reliably differentiate between disease and infection or between tuberculosis and tuberculosis-like disease caused by "atypical" acid-fast organisms. Individual antigens or haptens must be isolated from various myco- bacteria and evaluated for differential specificity in serologic tests and in experimentally infected laboratory animals. This is a report of the antibody response of rabbits to mycobacterial antigens. The antibody responses to several different antigen preparations and inoculation procedures were compared. Unheated filtrates of cultures of Mycobacter- ium bovis of different ages were analyzed chemically and by chromatography, disc electrophoresis, and immunodiffusion. HISTOR ICAL REVIEW It is well established that the sera of tuberculous man and animals contains antibodies specific for myco~ bacterial protein, polysaccharide, and lipid (20:127). The occurrence and relative amounts of these types of anti- body are irregular during the course of the disease. Moreover, there is no apparent diagnostic or prognostic correlation between the amount of antibody and the extent, duration, and activity of the disease. Delayed hypersensi- tivity to tuberculoproteins occurs with few exceptions in tuberculous individuals. However, there appears to be little or no correlation between the degree of hypersensitivity and the activity or extent of disease other than that the re- action may be decreased in the terminal stage. The extent of delayed hypersensitivity is not related to the occurrence of serum antibody (86;164;94). As yet, there is no practical method of demonstrating delayed sensitivity in_vitro. The tuberculin test has remained the most reliable test for den tecting tuberculosis. It does not differentiate between active or closed cases, nor the species of mycobacterium causing the sensitivity. The significance of the serologic responses in tuberculo— immunity has been the subject of considerable discussion and controversy for many years (47;127;96). Results from experi— mental animals and clinical observations in humans have led some investigators to conclude that antibodies play an in- significant role (if any) in tuberculoimmunity (96:50:47). According to this view, resistance is primarily a result of the ability of the hypersensitive state to make the intra— cellular environment inhibitory to the growth of the tubercle bacilli. Other investigators (149) believe that antibodies play a decisive role in the acquisition of tuberculoimmunity. Several naturally occurring antimycobacterial serum substances, such as lysozyme, react with tuberculopolysaccharide (149). Antipolysaccharide antibody facilitates the removal of excess tuberculopolysaccharide from the extracellular fluids and allows these inhibitors to act. There is a complex interplay and changing balance of the tuberculopolysaccharides, specific antibodies, and nonspecific serum components. A large portion of the work on the immunology of tubercuw losis has been directed toward the development of serologic tests of diagnostic and prognostic value. Sera from indi- viduals immunized or infected with mycobacteria will agglu— tinate tubercle bacilli (67) and fix complement in the presence of cells or extracts of mycobacteria (152). However, serologic tests based on these reactions are of little diagnostic value because there are many false-negative and flase—positive reactions (20). Both tests have been used to measure the antibody response to the multiple antigens of mycobacterial cells and culture filtrates following immunization of ex— perimental animals (152:41). Because only low titers of precipitins are produced by tuberculous humans, and domestic and laboratory animals, the tube precipitation test is of little diagnostic or prognostic value (20:29:124). The passive hemagglutination test was developed by Middlebrook and Dubos and reportedly measures tuberculopoly- saccharide—Specific antibodies (102:165). Although the test is very sensitive, false-positive reactions occur with sera from normal individuals and patients with other diseases (65:155). Various strains of staphylococci and certain fungi elicited antibody in experimental animals which agglutinated erythrocytes sensitized with tuberculopolysaccharide (79). Hemolytic (105) and anti-globulin (75) modifications have not improved the diagnostic usefulness of the test (68). A passive hemagglutination test Specific for tuberculo- protein was devised by Boyden (17). Tannic acid-treated Sheep red blood cells adsorbed protein but not polysaccharide antigens present in old tuberculin (OT). TubercuIOproteins were also attached to phenolyzed red blood cells by tetra- zotized benzidine (50). Little correlation exists between the results of these tests and the presence of active tuber— culosis (51). Turcotte and co-workers (172) fractionated sera from healthy tuberculin positive individuals and patients with active tuberculosis with DEAE cellulose. The anti-tuberculo- protein antibody in the chromatographic fractions was measured using a passive hemagglutination test. Two types of antibody were readily separated chromatographically. Most of the antibody in the sera from healthy individuals had a high anionic affinity and was serologically inactivated by mercaptoethanol (ME). Sera from tuberculous patients con- tained antibodies of low anionic affinity and were not inactivated by ME. Daniel (59) was unable to confirm these findings. Other passive agglutination tests have employed carriers other than red blood cells such as latex particles coated with CT or PPD-S (45). Approximately 95% of the sera from patients with active tuberculosis caused agglutination whereas 90% of the sera from healthy individuals Sera were negative. When antigen-sensitized bentonite particles were used, two types of agglutinins were found in sera from tuberculous patients; a heat stable (56c, 50 min) antibody and heat labile antibody (52). Heating reduced the titer of sera from tuber— culous patients two to six-fold but not of sera from normal individuals or patients with inactive tuberculosis. Tuberculophosphatide-specific antibodies can be detected in infected or immunized individuals with phosphatide—coated red blood cells or kaolin particles (54;161;162). While anti- protein and anti-polysaccharide antibodies were found in the sera of tuberculous humans and rabbits regardless of the activity or extent of the disease, only anti-phOSphatide antibodies were found in individuals with active disease (164;165). The antibody titer reportedly could be corre- lated with the extent and activity of the disease. Positive tests were obtained with serum from 95% of 1402 patients with active tuberculosis (166). This test was not a re- liable indication of active disease in calves experimentally infected with classical and atypical mycobacteria (128). In 1948, three investigators independently reported the use of diffusion-in-gel techniques for the analyses of complex antigenic mixtures (115;114;51). Since then, many attempts to develop a meaningful serologic test for tubercu- losis have used these techniques. Parlett (118) found precipitins in the sera of 48 of 55 tuberculous patients with a modified immunodiffusion technique. The sera of 58 of 40 tuberculin positive healthy individuals were negative. The number of positive reactions detected varied with the stage of the disease; the percent positive tests were 84.2%, 75.5%, and 57.8% for far advanced, moderately advanced, and minimal tuberculosis, respectively (120). Thurston and Steenken (169) compared the sensitivity of the gel precipi— tation test and the Middlebrook-Dubos test for the detection of antibody in the sera of tuberculous and nontuberculous patients. Antibody was detected in 80% of the sera from tuberculous patients by both tests. Approximately 17% of the sera from nontuberculous patients were positive by both tests. Further refinements of the gel precipitation test such as readdition of reactants and using several antigen dilutions has increased somewhat its sensitivity (120:2;105). This test has not been found to be of value in the serodiagnosis of bovine tuberculosis (95). DeSpite the presence of gross lesions in cattle, very few sera were positive in this test. The Ouchterlony plate technique (115) was not satis- factory for the detection of precipitins in the sera of patients with active tuberculosis (86). Only 17% of 168 patients with active tuberculosis had detectable precipitins. They occurred most frequently in sera from patients with disease of long duration. Similar results were repeated by Glenchur and Kettel (55) and Burrel et al. (22). Forty-four percent of sera from patients with active tuberculosis had precipitins detectable by the tube gel precipitation test; only 57% of the same sera were positive by the plate tech— nique (89). No single serologic test is reliable for the diagnosis of tuberculosis and furthermore, a battery of serologic tests is not reliable (54). Sera from 158 tuberculous patients and 41 tuberculin-negative students were tested for antibodies using the tube double diffusion test, the Middlebrook-Dubos HA test, and the Takahashi kaolin agglutination test. Whereas 85% of the patients sera were positive by one or more tests, only 20% of the sera were positive in all of the tests. Seven percent of the sera from the tuberculin-negative stu— dents were positive by one or more tests. Few investigators have studied the antibody response of experimental animals to tubercle bacilli and their many components. A thorough understanding of the antibody re- sponse to individual mycobacterial antigens would greatly supplement our knowledge and understanding of tuberculo- immunity. Considerable is known about the antibody response of rabbits and other experimental animals to immunization with a number and variety of soluble, cellular and viral antigens (69:15:170). Variations in the dose (175:158) concentration (52), and physical and chemical nature of the antigen, as well as the route of injection (151:84) and im- munization regimen profoundly affect the antibody response. Recent work has focused on the heterogeneity of the immunoglobulins (125) and the temporal sequence of their production following immunization (175). Stelos (157) demonstrated a change in the electrophoretic mobility and sedimentation velocity of hemolysins produced by rabbits immunized with sheep red blood cells. The primary response consisted of the sequential production of two classes of antibodies. The antibodies produced early had a sedimenta- tion value of 19S and migrated with the gamma; globulins lg-M). The antibodies produced later were 78 gammag globulins (lg-G). The appearance of lg-G was accompanied by a decrease or disappearance of lg-M. Bauer and his associ- ates (8:9) inoculated rabbits with a variety of protein and cellular antigens and confirmed the sequential appearance 10 of these two classes of antibody. High molecular weight lg-M lost their precipitating and hemagglutinating activity after reduction with mercaptoethanol (ME): 78 lg—G retained their serologic activity. Macroglobulins were reductively cleaved to two 78 fragments by sulfhydryl compounds (44). The two main classes of antibody could be separated by DEAE column chromatography (10). Early lg-M were bound tightly to the adsorbent whereas lg—G were eluted with the starting buffer. The antibody response of several animal species has sug- gested that viral antigens elicit a different antibody response than soluble protein antigens. Svehag and Mandel (158:159:160) made a thorough study of the kinetics of anti- body production in rabbits immunized with polio virus. Two classes of ME-sensitive lg-M appeared in the circulation 8 to 16 hrs after immunization and persisted for 4 to 5 days. After 56 to 48 hours, two types of lg-G were distinguished by differences in their electrophoretic mobility. The dura— tion of the latent period for the 1g-G could not be shortened by increasing the amount of antigen injected. Small amounts of antigen elicited low levels of lg-M. No lg-G were de- tected. When larger doses were injected, lg-M production was followed by the production of 1g-G. The minimal dose of antigen required to initiate 1g-M synthesis was approximately one fiftieth of that required to initiate lg-G synthesis. 11 Guinea pigs immunized with bacteriophage produced 1g-M which were detectable 20 hours after inoculation (175). Low doses of bacteriophage elicited only 1g-M: larger doses induced the sequential production of both 1g~M and 1g-G. Similar results have been reported by other investigators (21:12). In contrast to the detection of antibody within 24 hrs after the inoculation with viral antigens, antibody to solu— ble protein antigens was not detected for several days to two weeks after injection (9). Both lg-M and ig-G were elicited by soluble proteins, however, the kinetics of pro- duction and the relative amount of each type of immuno- globulin produced varied with the dose of antigen, the route of injection, the chemical and physical prOperties of the protein and several other factors (171). The immune response to cell-attached somatic antigens appears to differ in several respects from that elicited by soluble protein and viral antigens (5:175): both elicited lg-M but less lg-G was elicited by somatic antigens (175:81: 125). lg-G were detected several days after lg-M in rabbits inoculated with soluble and viral antigens (8:9:159) but there was a longer lag period between the detection of lg-M and lg-G in rabbits inoculated with somatic antigens (8:81). The physical state of the antigen as well as the chemical composition can also profoundly influence the nature of the antibody response. Neter and his co-workers (106) studied 12 the antibody response of rabbit to extracts of heat-killed enteric bacteria. Intravenous injection of the soluble antigen resulted in either a minimal antibody response or none at all. In contrast, when the antigens were attached to homologous red blood cells and injected intravenously, a strong antibody response was elicited. Although the antibody response to soluble and particulate protein antigens is Similar, particulate antigens elicitImore 1g-M (10:107:171). Bacterial flagella elicited only lg-M in rats: flagellin (a soluble monomeric preparation of flagella) failed to do so (107). Rabbits produced 10-20 times more 1g-M when soluble thyroglobulin and human gamma globulin were attached to acrylic resin particles than when the same anti- gens were injected in soluble form (171). Many studies of the kinetics of antibody production to protein and somatic antigens have employed the agglutination test which may not be equally sensitive for lg—M and lg-G (11). Rabbit lg-M were 1000 fold more efficient on a molar basis than lg-G as agglutinins for S, typhumurium (129), and 750 times more efficient in the agglutination and hemolysis of human red blood cells (58). From 160 to 180 times as many moles of purified anti-hapten lg-G were required to produce the equivalent amount of hemolysis as produced by one mole of lg-M (110). Both lg—M and 1g-G were detected by radioimmunoelectro- phoreSiS (RIE) 5 to 7 days after rabbits were inoculated with 15 allum-precipitated diphtheria toxoid, human serum albumin, 21nd bovine gamma globulin (55). The authors concluded that tIhe sequential production of these two classes of antibody (was in fact due to differences in sensitivity of the hemag- glubination test for the two immunoglobulins. Rabbits were inoculated with human serum albumin in adjuvant and the serum tested for antibodies by passive hemagglutination, passive cutaneous anaphylaxis (PCA), antigen binding, pre- cipitation and RIE (174). Both lg-M and lg—G were detected six days after immunization: lg—M by passive hemagglutination and RIE and lg-G by antigen binding, RIE, and PCA. The 1g-G were not detectable by hemagglutination until eight days after inoculation. There have been a limited number of studies of the kinetics of production of the major types of immunoglobulins following immunization with mycobacterial antigens. Schoen- berg and his associates (152) reported the sequential appear- ance of complement fixing lg-M and lg-G in the sera of rabbits inoculated intravenously with M, butyricum. Parlett and Chu (121) inoculated rabbits intravenously with viable tubercle bacilli or cell extracts and measured the serum antibody by gel double diffusion after chromatography of the antisera on DEAE cellulose. The precipitins were in the lg-G fraction. Daniel (57:58) has made the most detailed study of antibody responses of rabbits to mycobacteria. Rabbits inoculated with soluble OT or alum-precipitated PPD-S 14 produced only lg-M specific for tuberculopolysaccharide. Heat-killed M, tuberculosis (H57Rv) or OT in incomplete Freund's adjuvant elicited 1g-M which were followed after several days by ig-G. Neither the magnitude nor the duration of the antibody response elicited in rabbits by the inocula- tion of several dosages of viable and heat-killed cells of BCG could be related to the dose or the viability of the antigen (57:58). Whereas lg-M were produced by rabbits which received either viable or heat-killed BCG, only 2 of 8 rabbits which received heat-killed BCG produced lg—G antibody. When M, tuberculosis was injected intravenously a second time five days after an initial injection, only lg—M were detected five days after the last injection (108). The chemical and antigenic compositions of mycobacteria and their culture filtrates have been studied extensively. Since the heated, concentrated culture filtrate known as Old Tuberculin (OT) was used as a chemotherapeutic agent by Koch in 1891 (49). The untoward local, focal, and systemic re- actions caused by OT in tuberculous individuals was recognized subsequently. It became important as a Skin test reagent for the detection of tuberculosis and is still used for Skin tests and as a test antigen in several serologic tests for tubercu— losis. Siebert and Long (92) extensively investigated the chemi- cal and biological activity of various fractions isolated from OT. The Skin—reactive substance in tuberculin was 15 non-dialyzable, precipitated at pH 4.0 with acetic acid and ammonium sulfate, and was destroyed by hydrolysis with hydro- chloric acid or several proteolytic enzymes. They concluded that the active substance(s) was a protein but that OT cone tained many biologically inactive ingredients in addition to the allergenic proteins. Several attempts were made therefore to isolate the proteins free from other biologically inert contaminants by precipitation with ammonium sulfate (155) and trichloroacetic acid (TCA) (154). One of the first tuber- culins prepared in this manner was called synthetic medium old tuberculin (SOTT) (155). The first purified protein derivative (PPD) was made by precipitation of the proteins from heated culture filtrates with TCA (95). The proteins in PPD had molecular weights in the range of 2000 to 4000: somewhat smaller than those in SOTT. Purified protein derivative (PPD) did not elicit precipitins in rabbits or precipitate with antisera (158). Much of the polysaccharide and nucleic acids present in PPD could be removed electro- phoretically (157). In 1941, Seibert and Glenn (159) prepared a PPD by pre- cipitation of the proteins from heated culture filtrates of Mycobacterium tuberculosis with ammonium sulfate. It is called PPD-S and is now the international tuberculin standard of the World Health Organization. Although PPD-S contained some polysaccharide and nucleic acid impurities, it was superior to OT and PPD in that it did not induce tuberculin- type hypersensitivity when used repeatedly in the same 16 individual. The chemical composition and biological activity of PPD-S was more reproducible than PPD (159:145). Although PPD—S was an impressive advancement, physical, chemical, and biological analyses indicated that it was a complex mixture (141:142). Two groups of molecules with different electro- phoretic mobilities and sedimentation velocities were found in PPD-S. Differences in the sedimentation patterns observed during ultracentrifugation of heated culture filtrates and PPD suggested that several monomers had polymerized during precipitation with ammonium sulfate (156). The proteins in unheated culture filtrates had molecular weights in the range of 17,000 to 52,000 (100). Heating reportedly caused the unfolding and elongation of the polypeptide chains without any loss in weight. Prolonged heating resulted in degradation of the proteins to 6,000 molecular weight units with a con- current loss of antigenicity. More recently, the effect of heat on the physical and biological properties of tuberculo- proteins was shown to depend greatly on the initial pH of the solution (75). Because of the denaturative effects of heat on tuberculo— proteins and the loss of antigenic Specificity and biological activity that results, unheated culture filtrates are more promising materials for the isolation of antigens and sensi- tins. Unheated culture filtrates were separated into two fractions by moving boundary electrophoresis (140). One of the fractions was much less antigenic than the other and did 17 not induce tuberculin sensitivity in rabbits. Fourteen fractions were separated by fractional precipitation of un- heated culture filtrates of M, tuberculosis with ammonium sulfate (15:101). Two proteins with different molecular weights were isolated. One of the proteins had a molecular weight of 44,000 and was believed to be identical to a pro- tein isolated earlier by Seibert and co-workers (156). The other protein was not purified. On the basis of differences detected by physicochemical and serologic tests, they con- cluded that only two antigens were present in culture filtrate. In 1949, Seibert (142) treated unheated culture filtrates with ethanol and acetic acid and isolated three proteins (A, B, and C) and two polysaccharides (I and II). They are frequently used for comparisons in other studies because their physical, chemical, and biological prOperties have been studied extensively (145:19). The relative concentrations of the five fractions depended on the strain of M, tuberculosis used, the age of the culture, and the medium on which they were grown. The three proteins differed markedly in sedi- mentation velocity, electrophoretic mobility, chemical compo- sition, antigenic Specificity, and allergenicity in tuberculin~ sensitive animals (145:147). Several antigens were found in each protein fraction by immunodiffusion (148:90). Using the passive hemagglutination test, only four different antigens were found in unheated culture filtrates fractionated by Seibert's method (18). Three of these antigens were found in 18 the corresponding fraction isolated from heated culture filtrates. None of the protein fractions were immunogenic in guinea pigs (54). Polysaccharides I and II differed markedly in physical, chemical, and serologic properties (19). Both failed to elicit an intradermal reaction in tuberculinwsensitive guinea pigs. Polysaccharide II was antigenic in rabbits: polysaccharide I was a hapten. Recently (14) a heteropoly- saccharide was isolated from M, tuberculosis cell walls that was identical to polysaccharide I in chemical, chroma- tographic, and immunological prOpertieS. The authors sug- gested that polysaccharide I found in culture filtrates is released from lipopolysaccharide in the bacterial cell walls during autolysis. Unheated culture filtrates were precipitated with alcohol and zinc acetate (116). The materials that were obtained were better sensitins and more Specific than CT or PPD-S when tested in ECG-vaccinated individuals. A potent and Specific tuberculin was obtained from unheated culture filtrates of M, tuberculosis and M, bovis by precipitation with acid at pH 4.7 (76). The precipitate was dissolved, dialyzed, and reprecipitated several times with ethanol. Sensitins pre- pared by precipitating culture filtrates from classical and atypical mycobacteria with TCA were tested in guinea pigs infected with various mycobacteria (98:99). Although con- siderable cross sensitivity occurred, it was possible to 19 distinguish between infections caused by the two groups on the basis of the degree of sensitivity to the tuberculins. Another group of investigators Skinwtested several human populations with fractions obtained by acid-alcohol precipi- tation of unheated culture filtrates. The supernatant fluid which remained after precipitation of the culture filtrate with acid at pH 4.1 was reprecipitated with several concen— trations of ethanol. The supernatant fluid obtained after precipitation with 76% ethanol was much more Specific than PPD-S for infections caused by M, tuberculosis. Prior to the development of diffusion—in-gel techniques little progress had been made in the antigenic analyses of mycobacteria (17). This method of antigenic analyses has largely replaced the older and much less sensitive serologic techniques. Parlett and Youmans (119) detected six antigens in cell suspensions of mycobacteria by immunodiffusion. Only four antigens were found when unheated culture filtrates were tested (117). Inoue (74), Burtin and Kourilsky (25) and Castelnuovo (25) detected 7, 7 and 10 antigens respectively in unheated culture filtrates. Lind (85:87:88:90) found as many as 17 different antigens in culture filtrates of various mycobacteria. Comparisons of immunodiffusion results are difficult at best. The methods used to obtain mycobacterial antigens and antisera, and the techniques used in immunodiffusion vary markedly. In addition, considerable antigenic differences 20 exist among different species of mycobacteria (117:119:87: 151). Moreover, the antigenic composition of culture filtrates varied among strains and substrains of the same Species (87), and was modified by changes in cultural con- ditions (25). Most attempts to obtain satisfactory separation of the culture filtrates by zone electrophoresis on paper have been unsuccessful (125:42). Two to four bands were poorly separated. Starch gel electrOphoresis separated four pro- teins in TCA precipitates from culture filtrate (45). The amino acid composition of the four isolated fractions did not differ Significantly. A combination of starch electro- phoresis with ammonium sulfate precipitation, and column chromatography yielded two protein antigens from culture filtrates of M, tuberculosis (185:184:185:186). Both anti- gens were readily extractable from the cell surface and accounted for approximately 90% of the antigenic, protein material in the culture medium (185). A very potent hemagglutinating substance was isolated from culture filtrates of M, tuberculosis (4). The super- natant fluid from culture filtrate after TCA precipitation was dialyzed and reprecipitated several times with ethanol. After extraction of the precipitate with chloroform and amyl alcohol, it was dissolved and separated into three fractions by starch block electrophoresis. The active material was a low molecular weight lipOpolysaccharide-peptide of cell wall 21 origin. The purified hapten did not elicit antibody pro- duction in rabbits. The best electrophoretic separation of mycobacterial constituents reported has been by disc electrophoresis (1:151:46). Nineteen protein bands were demonstrated in culture filtrates from human tubercle bacilli (H57Ra) (1). Fractional precipitation of the culture filtrates with am- monium sulfate was performed according to the procedure described by Yoneda (185). All five ammonium sulfate- precipitated fractions contained from 10 to 14 bands when analyzed electrOphoretically. Concentrated culture filtrates from three month old cultures of M, bovis, M, avium, and two strains of Group III of unclassified mycobacteria of bovine origin Contained between 18 and 24 protein components (151). From five to eight polysaccharide components were detected in each culture filtrate. Several protein bands were iso- lated from each culture filtrate by elution from the acrylamide gels after preparative disc electrophoresis. Analyses of the eluates obtained from each culture filtrate was performed by immunodiffusion with four homologous antisera and eleven other antisera. One or more antigens were detected in each culture filtrate which was not found in any of the other cul- ture filtrates. However, re—electrophoresis of the eluates at a different concentration of acrylamide gel separated some eluates into more than one band. An eluate from a Single band was not necessarily a Single antigen. 22 Ion exchange column chromatography has been used with some success to separate the antigens in culture filtrates of mycobacteria. Rhodes (126) separated a fraction obtained by Seibert‘s method (142) into four components on DEAE- cellulose. No Single component was isolated when analyzed by starch gel and paper electrophoresis. Each fraction con- tained several antigens and elicited approximately the same Skin reactivity when tested in tuberculin-Sensitive guinea pigs. AS many as 12 fractions were separated by DEAE-cellulose chromatography of unheated culture filtrates of three strains of M, tuberculosis (78). All of the fractions contained several antigens when analyzed by immunodiffusion and although not separated, there did appear to be Species-specific and strain-Specific antigens. Five fractions were obtained by chromatography of un— heated culture filtrates on carboxymethyl cellulose (90). None were antigenically pure. Rechromatography of the iso- lated peaks failed to improve significantly the antigenic homogeneity of the fractions. Two polysaccharide antigens were separated from culture filtrates of M, tuberculosis by chromatography on Dowex-50 and DEAE—cellulose (179). One of the antigens was pure poly- saccharide; the other antigen contained alanine, glutamic acid, and diaminopimelic acid. Both antigens were probably derived from cell wall lipopolysaccharides by autolysis. 25 Unheated culture filtrates were separated into dialyz- able and non-dialyzable fractions by Baer and Chaparas (6). The non-dialyzable constituents were separated into acetic acid-soluble and acid—insoluble fractions at pH 4.0. The acid-soluble fraction contained polysaccharides primarily and retained its Skin test reactivity after digestion with proteolytic enzymes (7). Several fractions in the non- dialyzable fraction were separated by Sephadex G-50 (26). The concentrated dialyzable fraction contained several pre- cipitinogens and elicited skin reactions in guinea pigs sensitive to tuberculin (26:27). All of the fractions from Sephadex G-25 and G-50 chromatography of the dialysate con- tained several antigens. Since the,yield of biologically active constituents that can be recovered from culture filtrates is usually relatively small, many attempts have been made to obtain soluble ex- tracts from intact or broken cells. These procedures permit the recovery of native or only slightly denatured proteins. Mycobacteria contain readily extractable antigens located at or near the cell surface (19:54). Heidelberger and Menzel (64) extracted killed.M, tuberculosi§_with solutions of in- creasing alkalinity. At least three antigens were detected by serologic analyses of subfractions obtained by sodium sul- fate precipitation of the extracts. A material extracted from broken cells of M, bovis with dilute alkali elicited skin reactions in guinea pigs sensitive 24 to tuberculin (60). The extract was separated into several components by zone electrophoresis. Extracts from different strains varied markedly in their chemical and biological properties. Heckly and Watson (62:65) disrupted tubercle bacilli by Shaking with glass beads and extracted the debris with phos- phate and borate buffers containing ether. The extracts were qualitatively Similar. The borate buffer extract contained mostly protein and little nucleic acid or carbohydrates. Nucleic acid impurities were removed with calcium phosphate and the extracts fractionated by precipitation with ethanol. The most active protein had three times the skin test re- activity of an equivalent amount of PPD-S. Saline and aqueous extracts of three Species of myco- bacterial cells reportedly contained group-specific and strain- specific antigens that could be differentiated serologically (167:168). The antigens were isolated by successive extrac- tion with saline. Aqueous extracts of intact tubercle bacilli closely resembled culture filtrates when analyzed chemically and by electrophoresis (145). Extracts from different strains differed in chemical composition and biological activity. Protein, polysaccharide, and lipid antigens were extract- ed from phenol—killed human tubercle bacilli with dilute buffers (74). Immunodiffusion analyses of the three prepara- tions revealed the presence of 6, 2, and 1 antigen, respectively. 25 Ide and co-workers (70:71) extracted broken mycobacteria with alkaline phOSphate buffers. The extracts closely re- sembled homologous culture filtrates in physicochemical and biological properties. Four fractions were obtained by chromatography of the extracts on DEAE—cellulose (72). The first peak contained largely carbohydrate: the next three peaks were essentially protein. All three of the protein peaks elicited skin reactions in quinea pigs sensitive to tuberculin. All of the fractions contained several precipi- tating antigens (187). Azuma (5) isolated a polysaccharide antigen from an alkaline extract of M, tuberculosis by fractional precipita- tion with ethanol. The antigen was purified by chromatography on Dowex-SO, DEAE cellulose, and Sephadex G-75 and G-200. The active material contained only one precipitating antigen. Paraffin oil extracts of mycobacteria contained two main antigens: a low molecular weight protein that elicited skin reactions in sensitized animals and an antigenic lipopoly- saccharide (28). C00per and Clark (55) prepared a potent tuberculin by allowing mycobacterial cells to autolyze under toluene at 570C. They felt that in this way the denaturative effects of mechanical and physicochemical isolation procedures were avoided. Solid urea was first used to isolate antigens from myco— bacterial cells by Stacey (156). Extracts obtained from 26 M, tuberculosis contained a complex mixture of lipids, nucleic acids, proteins, and polysaccharides. One main com- ponent with the electrophoretic mobility and ultraviolet ad- sorbency of Seibert's C protein was found in urea extracts of human tubercle bacilli (146). The extract was not immuno- genic but elicited skin reactions in tuberculin-sensitive animals and inhibited the migration of leucocytes from tuberculin-sensitive animals in tissue culture. Urea extracts of M, bovis contained five to Six antigens (40). Twenty or more antigens were found in sonic extracts of another lot from the same batch of cells. Yamamura and his co—workers (104:155:109:154) isolated a low molecular weight tuberculin-active peptide (TAP) from heat killed M, tuberculosis by extraction with 0.1 N hydro— chloric acid. After neutralization, the extract was precipi- tated several times with picric acid. The solubilized precipitate was separated into four active fractions by chromatography on DEAE-cellulose. The active fractions did not contain any lipid or carbohydrate. In contrast to PPD-S which evoked sensitivity to tuberculin when injected repeat- edly into guinea pigs, repeated injection of TAP did not sensitize guinea pigs to tuberculin. The inability of TAP to elicit delayed hypersensitivity to tuberculoproteins was attributed to the absence of lipid or polysaccharides. Raffel (125) was unable to sensitize guinea pigs to tubercu- lin by repeated injections of either purified tuberculoprotein 27 or purified waxes from mycobacteria. When they were in— jected together, delayed hypersensitivity to OT and PPD-S could be demonstrated several weeks later. Several investigations have Shown that mycobacteria con- tain several readily extractable immunogenic substances (54; 177). Methanol extracts (176:178) were Slightly less immuno- genic on a weight basis than intact inactivated bacilli. Crowle (56) isolated a water soluble immunogen from acetone killed tubercle bacilli by digestion with trypsin. Selective degradation or removal of peptides, polysaccharides, lipids, and nucleic acids from the extract indicated that the active substance was a polysaccharide. It was probably derived from the cell wall or .cytoplasmic membrane. The purified immuno- gen was five times as effective as intact bacilli on a weight basis and induced immunity of the same degree, longevity, and specificity. Ribosomes and ribonucleic acid isolated from disrupted cells of M, tuberculosis were immunogenic in mice (181:182). Isolated cell walls from various species of myco~ bacteria were as immunogenic as BCG in mice (85). Cell walls isolated from cells disrupted in oil were more immunogenic than those prepared by aqueous disruption. The purified cell walls contained little protein but substantial amounts of lipo- polysaccharides. Rabbits could be sensitized to tuberculin by injection of cell walls but not by protoplasm from dis- rupted tubercle bacilli (82). 28 Extracts of tubercle bacilli broken in a French press were fractionated by gel filtration and chromatography on DEAE-cellulose (55:56:57). All of the fractions contained several antigens and elicited immediate and delayed Skin reactions in tuberculin-sensitive guinea pigs. Dietz (46) used disc electrophoresis to analyze extracts of M, tgberculosis and M, kansasii obtained by repeated freez— ing and thawing and by disruption with glass beads. Four of the ten bands detected were isolated and tested for aller- genicity in tuberculin-sensitive guinea pigs. Two of the eluates elicited skin reactions but lacked Specificity. Neither of the eluates elicited antibody production in animals. Mycobacterial phage lysates have been used as test anti- gens in immunodiffusion test for tuberculosis but no attempt was made to analyze the chemical or antigenic composition of the lysates. Extracts of M, bovis obtained by disintegration of two month old cells with ultrasound contained at least twenty antigens when analyzed by immunoelectrophoresis (40). Variable numbers of antigens were detected in urea, guanidine, triton, desoxycholate, and phOSphate buffer extracts of viable cells. In view of the antigenic and chemical complexity of myco- bacteria, it is perhaps not surprising that difficulty has been encountered in attempts to isolate immunochemically 29 pure antigens or sensitins. Evidence exists that specific- sensitins and antigens are present in culture filtrate and cell extracts. Their isolation and evaluation remains a challenge. MATER IALS AND METHODS Mycobacterium bovis. Culture 510-2 was isolated in 1962 from a gross lesion cow and identified as Mycobacterium bovis by growth and morphologic characteristics, cytochemical tests, virulence for laboratory animals and cattle, and allergenicity for guinea pigs. Culture filtrates. Cultures of M, bovis were grown on the surface of one liter of a modified Proskauer-Beck synthetic medium (180) in diphtheria toxin bottles. The medium contained the following: Asparagin 5.0 gm Monopotassium phosphate 5.0 gm Potassium sulfate 0.5 gm Glycerol 20.0 ml Magnesium citrate 1.5 gm Distilled water g.s.ad. 1000.0 ml The first four ingredients were dissolved in order in 700 m1 of the H20, each being added after the previous in- gredient had completely dissolved. The pH of the solution was adjusted to 7.0 with 40.0% sodium hydroxide, poured into a diphtheria bottle, and autoclaved for 20 minutes at 1210C. To this was added 500 ml of sterile (20 min, 1210c) 0.5% magnesium citrate solution. 50 -- ~n V— 1» mi. :4 by 51 The medium was seeded with fragments of surface growth of M. ngig and incubated 2, 5, 4, 5, 6, and 7 months at 55°C. Bottles with turbid culture fluid beneath the pellicle were discarded. The clear, amber colored culture fluid was drawn off aseptically into 250 ml plastic bottles and centri- fuged three hours at 2000 xg. The supernatant fluid from individual cultures of the same age were pooled and filtered (Berkefeld filter, N grade). The sterile filtrates were poured into sterile 4.5 cm (flat diameter) dialysis tubing (Visking Corporation) and dialyzed two days at 40C against approxi- mately 15 volumes of 0.015 M phosphate buffer pH 7.2. The dialyzed filtrates were concentrated ten fold by pervapora— tion at room temperature. Concentrated, dialyzed, sterile culture filtrate (CF) of the same age was pooled, filtered (Millipore, 0.45 u pore size), diSpensed into 50.0 ml screw capped tubes, and stored at -800C. Chemical analyses. Protein in the CF and chromatographic fractions was measured by the Folin method (77). One ml each of 4.9% sodium potassium tartrate and 2.0% copper sulfate solutions was added to 100.0 ml of 4.0% sodium carbonate solution. Ten ml of the reagent was mixed with one ml of the samples of CF containing 10 to 250 ug of protein and incubated 45 minutes at room temperature. One ml of Folin phenol reagent diluted 1:5 with distilled water was added, mixed immediately, and incubated 15 minutes at room temperature. The optical 52 density (OD) of the solutions at 660 mu was measured with a Beckman D.U. spectrophotometer. The amount of protein in the unknown samples was calculated by comparing the OD of the samples (minus the blank) to those of a standard solution of bovine serum albumin as represented on a standard curve. Carbohydrates were assayed by use of the thymol-sulfuric acid method (150). Seven ml of cold 77% sulfuric acid was placed into 15 ml glass stoppered centrifuge tubes. One ml samples containing 5 to 100 ug of carbohydrate were layered on the sulfuric acid and the tubes placed in an ice bath for 50 minutes. One-tenth ml of a 10% thymol solution in absolute ethanol was carefully layered over the samples and 0.9 m1 of distilled water added. The tubes were stoppered, shaken, and placed in a boiling water bath for twenty minutes. They were then transferred to an ice water bath for five minutes and allowed to stand at room temperature for twenty-five minutes. The OD of the solutions was determined at 500 mu in a Beckman D.U. spectrophotometer. The amount of carbo- hydrate in the samples was calculated by comparing the OD of the samples to those of standard solutions of glucose repre- sented on a standard curve. The amount of nucleic acids present in the CF and chroma- tographic fractions was measured by ultraviolet absorption Spectrophotometry at 280 and 260 mu using a one cm light path. The nucleic acid concentration was calculated by comparing the OD of the samples to those of standard solutions of enolase and nucleic acid represented on a nomogram. 55 The absorbency of ultraviolet light in the range of 190 to 520 mu by culture filtrates and chromatographic fractions was measured in a Beckman D.B. spectrophotometer using a one cm light path. Production of antisera. Several preliminary experiments were performed to de— termine the best way to elicit high titered antisera against mycobacterial antigens. Eight groups of three Dutch rabbits per group were inocu- lated according to the schedule shown in Table 1. Cells of M, bovis strain 510 were grown for three weeks in modified Proskauer Beck media and washed with phosphate buffered saline solution (PBS) pH 7.2 prior to inactivation with heat or beta-propiolactone (Testagar and Company). Cells were heat inactivated at 100°C (moist heat) for thirty minutes. Beta-propiolactone (BPL)—inactivated cells were prepared according to the method of Onyekwere (111). Washed cells were suspended in triple distilled water to a concentration of approximately 10 mg wet weight per ml. The suspension was transferred to a screw capped tube, the pH adjusted to 8.4 with 0.5 M disodiumphosphate and cooled in an ice water bath. A cold solution of BPL was added with constant stirring to a final concentration of 0.4%. The sus- pension was incubated in a 570C water bath for two hours with constant agitation. During this period, the pH of the sus- pension was adjusted to 7.6 at 15 minute intervals by the >mu 08mm Dru co mco_uummc_ um >n smso__oe do to combuswe_ s. _m_s_s_ - om->_m maOcm>mLuc_ I >_m mco_uuo~c_ maomcmuauoam >_xmo3 x_m I _mpucosvomm m_>oa .m.pm___x ocouum_o_aotdmuom I m__mu Ammo m_>os .m.ss___x Dam: - m:8 rm mou_m ucmtmmwmo c_ mco_uommc_ um msoocmu_semm x_m I o_d_u_az: Aoum_ov ucm>awom mum—deouc_ m.pc:mtu I ucm>awv< mnoocmDsunam I own moose—_C arse—so obscure: I muu _ _msacoseom bo->_ +. o.~ o.m “__ou Dam + no ___> «_a_s_sz mom->_ + o.~ o.m m__ss Dam + to __> _siueosesm m>. I o.~ o.m m__ou nae +_ao _> M «:53: em .. o.~ o.m «:8 is + .._o > “_s_uesseom om +_ .o.~ o.m “__su 4am +.ao >_ «_a_a_sz om +. o.~ o.m m__ou cram + to ___ 0:53: om + e.~ o.m m__su m: + ac __ :o_a_s_sz now +. o.~ o Nae . 0.:v020m Munoz _ucm>:flv< Amsvc_muotm Amevm__uu cummuc< lmfiflHmW .mumtu__m «tau—:0 m_>oo E:.touumoou>m.t0\vcm «__ou wo___x vo>~ouot nu_;3 mumnBML tow o_:vorum co_um_:uoc_ ._ o_sss 55 addition of 0.5 M disodium phosphate. The cells were removed after centrifugation, washed four times with sterile dis- tilled water, and suSpended in sterile 0.15 M sodium chloride solution to a concentration of approximately 1.5 mg wet weight per ml. Rabbits which were inoculated with inactivated cells each received a total of approximately nine mg wet weight. All rabbits received two mg of CF protein. The inocula for all the groups except Groups V and VI were prepared by emulsify- ing the antigen suspension in an equal volume of incomplete Freund's adjuvant (Difco Laboratories). Rabbits in Groups I, II, III, V and VII were given six subcutaneous injections at different Sites on the same day (multiple). Six sequential subcutaneous or intravenous injections were given by one in— jection per week to the rabbits in Groups IV and VI, respectively (sequential). Rabbits in Groups VII and VIII each received an initial intravenous injection of one ml of CF followed by multiple or sequential subcutaneous injections. Blood samples were collected from the marginal ear vein of each rabbit prior to inoculation and weekly for eight weeks after inoculation. Sera were decanted, clarified by centrifu— gation, heat inactivated for thirty minutes at 560C, and if to be used in hemagglutination tests, absorbed twice with Sheep red blood cells, and stored at —800C. A portion of each serum was mixed with an equal volume of 0.2 M 2-mercaptoethanol (ME) and incubated twelve hours at room temperature. 56 The treated serums and untreated serums were titrated for antibody immediately after ME treatment. Polysaccharide-specific antibody was measured by a modification (97) of the passive hemagglutination (HA) test devised by Middlebrook and Dubos (102). Sheep red blood cells were collected aseptically into an equal volume of sterile modified Alsevers solution. The cells were removed after centrifugation and washed three times with PBS. Six ml of a 1:15 dilution of mammalian Old Tuberculin (USDA) were added to 0.1 ml of the washed, packed cells and incubated two hours at 570C. The sensitized cells were removed after centrifugation and washed three times with PBS. They were resuspended in PBS to a concentration of 0.5% and used within 24 hours. Serial two-fold dilutions of the serums were made in PBS starting with a 1:10 dilution. Three drOps of the sensi- tized erythrocyte suspension were added to 0.5 ml of the serum dilutions. The tubes were Shaken, incubated for two hours in a 57°C water bath, incubated two hours at room temperature, and overnight at 40C° The reciprocal of the highest serum dilution which caused visible hemagglutination was recorded as the HA titer. The ME-resistant antibody titer was determined by titrating the serums after treatment with ME. The titer of ME-sensitive antibody is the differ- ence between the titers of the serums before and after treatment with ME. 57 Bacterial agglutinins were measured by a tube agglu- tination test-which employed a uniform suSpension of BPL- inactivated M, bgyig. Six month old cells were inactivated with BPL as indicated previously. Aggregated cells were dispersed in a tissue grinder. The turbidity of the sus- pension was adjusted to an OD of 0.2 at 525 mu with PBS. Twenty-five hundredths ml of the cell suspension was added to 0.25 ml of the serially diluted ME-treated and untreated serums. The suSpension was incubated twelve hours at 570C. The presence of antibody was indicated by visible agglutin- ation. The antibody titer was recorded as the highest dilution of the serum which caused visible agglutination. The ME-Sensitive and ME-resistant antibody titers were re— corded as previously described. The efficacy of aluminum chloride and aluminum potassium sulfate (alum) to precipitate CF (5 month incubation, 2.5 mg/ml) precipitinogens was compared. Twenty ml of distilled water and 20.0 ml of 10% alum (66) were added to 16.0 ml of CF. The pH of the solution was adjusted to 6.8 with 5 N sodium hydroxide and the resulting precipitate washed twice with PBS containing 0.01% merthiolate. The washed precipitate was resuspended in 40.0 ml of merthiolated PBS. The aluminum chloride—precipitated inoculum was prepared by adding 2.0 ml of 10.0% aluminum chloride and 2.0 ml of distilled water to a 16.0 ml of CF. The pH of the solution was adjusted to 6.8 with 20.0% sodium hydroxide. - l .1 C» v. I‘.‘ .. . .3 Zn v. 5. C. ‘3 “I .“ a «A «\U N w \ h‘v 58 Two groups of four Dutch Belted rabbits per group were inoculated as Shown in Table 2. Each rabbit received 5 ml (10 mg protein) of aluminum hydroxide-precipitated CF and 1.0 ml (2.5 mg protein) of untreated CF. The serums ob- tained from each rabbit were tested individually for anti— CF precipitins by Ouchterlony immunodiffusion and immuno- electrOphoresis. Inocula were prepared by aluminum chloride precipitation of pooled three and four month incubation CF, and pooled five, Six and seven month incubation CF. Two groups of Dutch rabbits, six rabbits per group, were inoculated as given under A in Table 2. The inoculation schedule was repeated after a one month interim. The serums obtained from each rabbit were tested individually by Ouchterlony immunodiffu- sion for anti-CF precipitins against the homologous CF pool. Satisfactory antisera within each group were pooled, dis- pensed in small tubes, and stored at r8OOC. The antisera are herafter referred to as reference antiserum A (anti-5 and 4 month old CF), and reference antiserum B (anti-5, 6, and 7 month old CF). Statistical analyses. The hemagglutination and agglutination titers of the serums collected weekly from each group of rabbits (Table 1) was converted to the corresponding dilution tube number. The mean of the tube dilution numbers in each group was used for statistical analyses. The variance among the means for 59 table 2. Inoculation schedule of rabbits which received Mycobacterium bovis precipitated with aluminum hydroxide. Amount (ml) Route ‘_gv Operation Al 82 A B l Bled l0.0 10.0 c3 c I . Inoculated 5.0 5.0 Imh lm l4 Inoculated 0 5.0 -- Im 32 . Inoculated l.05 l.0 lp6 lp 36 Bled 25.0 25.0 C C 40 Bled 25.0 25.0 C C lA - Rabbits which received aluminum chloride precipitated culture filtrate (10.0 mg protein per rabbit) 2B - Rabbits which received aluminum potassium sulfate precipitated culture filtrate (l0.0 mg protein per rabbit) 3C - Cardiocentesis lm - Intramuscular 5l.0 - Non-precipitated culture filtrate, 2.5 mg protein per ml 6Ip - Intraperitoneal 40 each week was analyzed by the multiple range test developed by Duncan (48). Significance was determined at the 95% level. Ouchterlony double diffusion. Clean 5 1/4 x 4 inch glass lantern Slides were evenly covered with ten ml of melted 1% agar (Difco) in 0.15 M PBS containing 0.01% merthiolate. The gels were "aged" for at least three hours in a humidified diffusion chamber. Immediately before use, the circular reactant wells were cut and the agar plugs removed. Several different well diameters and interwell diffusion distances were tried. Subsequently, all wells of 6 mm diameter were cut with 6 mm diffusion distance between wells. The protein concentration of the samples was adjusted to approximately 2.5 mg per ml. After the reactants were added to the wells, the Slides were incubated at 280C in a humidified diffusion chamber and observed daily. Replenish- ments of the reactants were made daily for three days with undiluted antiserum and 1:5 dilutions of the antigens. After incubation, the unreacted protein was removed by soaking the gel-covered plates three days in six changes of 0.015 M PBS, pH 7.4. Salts were removed by a 12 hr rinse in dis- tiled water. The plates were overlaid with moist filter paper, dried at 57°C overnight, and stained for ten minutes in a protein triple stain (55). The staining solution contained the following: 41 Thiazine red R 0.1 gm Amidoschwarz 10 B 0.1 gm Light green SF 0.1 gm Acetic acid 2.0 gm Mercuric chloride 0.1 gm Distilled water g.s.ad. 100.0 ml The plates were differentiated in 2% acetic acid and thorough- ly rinsed in tap water. Immunoelectrophoresis. Immunoelectrophoresis was done according to the procedure described by Hirschfeld (66) with modifications. A 2% solu- tion of agar (Difco) in distilled water was poured to 1 cm depth in a flat dish, allowed to solidify, and cut into 1 cm square cubes. The cubes were washed for three days in run- ning tap water and for several days with frequent changes of distilled water, and stored in distilled water at 40C. A barbital buffer system, discontinuous with respect to ionicity, was used: the internal buffer solution (pH 8.6, ionicity 0.09) was used to prepare the gel: the external buffer solution (pH 8.6, ionicity 0.06) was used in the elec- trode vessels of the migration chamber. The buffer solutions contained the following: External buffer Internal buffer Diethylbarbituric acid 1.58 gm 1.66 gm Sodium diethylbarbiturate 8.76 gm 10.51 gm Distilled water g.s.ad. 1000 ml 1000 ml 42 Two parts of the internal buffer solution were mixed with one part of distilled water and heated in flowing steam. An equal volume of melted 2% agar containing 0.02% merthiolate was added. Two and one-half ml of the warm agar solution were Spread evenly over the surface of clean 1 x 5 inch microscope slides. After solidification, the Slides were "aged" at least three hours before used. Immediately before electrophoresis, two circular antigen wells and an antiserum trough were cut in the gels with an LKB gel punch (LKB Instruments, Inc.). The antigen wells were enlarged with a blunt 12 gauge needle and removed. Five ul of the concentrated (approximately 10 mg of protein per ml) CF were placed in the antigen wells and electro- phoresis begun immediately. Electrophoresis was carried out in a Shandon migration chamber (Colab Laboratories, Inc.). The glass microscope slides were supported by a plexiglass plate placed across the bridge supports. Eight slides were used at one time: four sample slides and four blank. The latter were placed on the anode Side of the plate to prevent electrodecantation of the sample Slides. Electrical connection between the slides and the electrode vessels was made with buffer impregnated filter paper strips (Whatman #1). Electrophoresis was carried out at 40C for one and one-half hours at a current of 1.25 mA per Slide (eight volts per cm). Following electrophoresis, the agar plugs in the anti- serum troughs were removed and antiserum added. The slides 45 were incubated in a humidified diffusion chamber for 12 hours with replenishments of antiserum as needed. Incuba— tion was continued for an additional three days at 40C with daily addition of antiserum. The slides were washed, dried, and stained as described under Ouchterlony double diffusion. Disc electrophoresis. The apparatus and technique employed were similar, but Slightly modified, to that described by Ornstein and Davis (112). The composition of the reagents used for preparing the gels is Shown in Tables 5 and 4. The small pore solution was prepared and placed into 0.5 x 6.5 cm glass tubes to a depth of 5.1 cm. It was over- laid carefully with distilled water and allowed to stand 45 minutes. The water was removed and replaced by a layer of large pore solution 1 cm deep. The large pore solution was carefully overlaid with water and photopolymerized for 15 minutes under a fluorescent lamp. The gels were transferred to the upper buffer reservoir. A sample containing approximately 0.25 mg of protein in not less than 0.05 ml and not more than 0.2 ml, was layered on the spacer gel by displacement. Bromphenol marking dye was added to the upper vessel, and a constant current of 5 mA per tube applied at room temperature until the bromphenol blue marking dye had migrated 4.9 cm. The constant current was applied by a Vokam 500 mA DC power supply (Colab Labora- tories, Inc.). T 44 able 3. Composition of the stock reagents used for disc electrophoresis. I T T D A B D R D (A) NHCI 48.0 ml RIS‘ 36.6 gm EMEDZ 0.23 ml H20 3.3.251. I00.0 ml pH 8.93 (C) crylamide 28.0 gm Is‘l 0.74 gm H20 _q._s_._a_g. 100.0 mi (E) IbofIavIn 4.0 mg H20 3.3.29, 100.0 ml (8) INHCI 08.0 ml TRIS 5.98 gm TEMED 0.46 ml ouzo 3.3.91. 100.0 mi pH 6.73 (D) Acrylamide I0.0 gm DHZO 3.3.351. I00.0 ml (F) . Ammonium O.Ih gm persulfate 0:120 _‘g._§._a_c_l_. 100.0 mI 3 I, TRIS - Tris (hydroxymethyl) aminomethane ZTEMED - N, N, N', N', - tetramethylenediamine pH - Adjusted by titrating with INHCI BIS - N, N' - methylenebisacrylamide 45 Table 4. Composition of the working solutions used In disc electrophoresis. Small pore Small pore Large pore Stock Buffer solYtion solution A solutionig solution for reservoirs I part A 4 parts F I part B TRIS2 6.0 gm 2 parts C 2 parts 0 Glycine 28.8 gm l part DHZO I part E DH20 3.3.351. I000 ml 4 parts DH20 pH 8.3 pH 8.9 pH 6.7 'Diluted l:lO with DHZO before use 2TRIS ' trIs (hydroxymethyl) aminomethane () 46 After electrophoresis, the gels were chilled and re- moved from the glass tubes. The proteins in the gels were stained by immersing the gels for thirty minutes in a 0.5% solution of amido black in 5% acetic acid. Excess stain was removed electrophoretically in 5% acetic acid. Polysaccharides in the gels were stained by the periodic acid Schiff method (24). After the gels were removed from the glass tubes, the gels were immersed in 7.5% acetic acid one hour at room temperature. The gels were transferred to a 0.5% periodic acid solution for one hour at 40C. Excess periodic acid was removed from the gels electrophoretically in 7.5% acetic acid. The polysaccharides were stained by placing the gels in cold Schiff reagent and stored in the reagent at 4°C (24). Dialysis. One hundred ml each of CF of three, four, and five month incubation were placed in individual dialysis tubes (Visking Corporation) and dialyzed separately for two days against three changes of 2 L each of distilled water. The retentates were concentrated ten-fold by pervaporation at room tempera- ture and redialyzed eight days against daily changes of 100 ml each of distilled water. The dialysates from the latter were pooled, lyophylized, and reconstituted with ten ml of 0.015 M phOSphate buffer pH 7.2. The retentates and reconsti- tuted dialysates were analyzed by immunodiffusion, chroma- tography on Sephadex G-25, disc electrophoresis, and skin- tests in sensitized rabbits. 47 Ion exchange column chromatography. Dry diethylaminoethyl cellulose powder (DEAE, Schleicher and Schuell Co.) type 20 was thoroughly mixed after settling into a 1 N solution of sodium hydroxide. The mixture was transferred to a Buchner funnel fitted with filter paper and washed repeatedly with the alkali. The filter cake was re- suSpended in a minimal amount of 1 N NaOH. Sufficient 1 N HCl was added to make the mixture strongly acidic. The ad- sorbent was washed immediately with water on a Buchner funnel, resuspended in alkali, washed thoroughly with water again,’ and resuspended in starting buffer solution. Excess "fines‘I were removed by decanting the supernatant fluid after allow- ing the adsorbent to settle one hour in the starting buffer solution. Concentrated phOSphoric acid was added to adjust the pH of the adsorbent to 8.6. The adjusted adsorbent was washed with the starting buffer solution and resuspended to approximately a 2% suspension in the same buffer. Sephadex laboratory columns (1.5 x 50 cm) were packed with the adsorbent until the bed level was Slightly above the desired height. The packed columns were washed for several days with the starting buffer solution. Culture filtrate from a four month old culture of M, Egyi§_was prepared for chromatography by dialysis for 24 hours against 100 volumes of the starting buffer solution (0.005 M TRIS-phosphate pH 8.6). The sample was placed on the column and allowed to sink into the adsorbent. :fi ~HJ I Will ml: the Mm HI 48 The filter pad and Sides of the column were washed with several three m1 portions of the starting buffer. A 5 cm layer of the starting buffer was added to the adsorbent and continuous flow (55 ml per hour) of the eluting fluid started. Proteins were eluted from the adsorbent by continuous gradient elution. A concave salt gradient was produced by a cone-Sphere buffer vessel device. This consisted of a 250 ml Erlenmeyer flask which contained 200 m1 of the limit buffer solution (0.5 M TRIS phosphate pH 5.0 containing 1.6 M sodium chloride) and a 500 ml round bottomed Florence flask which contained 400 of the starting buffer solution. Five ml eluant fractions were collected, and the adsorbency of the eluant at 280 mu was continuously measured and recorded. Molecular exclusion chromatography. Culture filtrates, dialysates, and retentates were frac- tionated chromatographically using medium grade Sephadex G-25 (Pharmacia Fine Chemicals, Inc.) and Bio-Gels P-100, P-150, and P-200 (Bio-Rad Laboratories). The dry Sephadex and Bio-Gel beads were Slowly added to distilled water with constant stirring, and allowed to stand without stirring at least 48 hours before use. Excess "fines" were removed by decanting the supernatant fluid after allowing the hydrated gels to settle one hour in replenished starting buffer solu- tion. 49 Columns used for chromatography with Sephadex G-25 and Bio-Gels were 2.5 x 45 cm and 1.5 x 50 cm, respectively. The inside of the columns were coated by pouring in a warm 1% solution of dimethyl-dichlorosilane in benzene. The solu- tion was removed after several minutes and the benzene evaporated. This procedure was repeated. The gels were de- aerated in a vacuum flask and gradually poured into the columns until the gel column was slightly above the desired bed level. The gel columns were washed for several days with distilled water. Prior to sample application, the void volume of each column was determined using Blue Dextran 2000 (Pharmacia Fine Chemicals, Inc.), a high molecular weight (MW 2,000,000) dextran polymer coupled with a blue chromatophore. Samples were placed in the sample applicator at the top of the column and allowed to settle into the gel. The sample applicator and sides of the column were rinsed with several three ml portions of the eluent (distilled water). A 4 cm layer of eluent was placed on the gel column before continu- ous flow (50 ml per hour) was started. Chromatography was done at room temperature. Three ml fractions were collected, and the absorbency of the eluant at 280 mu was measured and recorded continuously. The contents of selected tubes were pooled, lyophilized (The Virtis Co.) and reconstituted to one-half the original volume. The concentrated fractions obtained from several Du »» :u n.»— Se C\ .nu 50 fractionations were pooled and rechromatographed. The con- tents of appropriate tubes were pooled, lyophylized, recon- stituted to the original sample volume, and stored at -80°C. Designated fractions were analyzed by immunodiffusion, cellulose acetate membrane electrophoresis, disc electrophore- sis, Skin tests in sensitized rabbits, and analyzed for protein, carbohydrate, and nucleic acid. Cellulose acetate membrane electrophoresis. Culture filtrates and rechromatographed fractions from Sephadex G-25 were analyzed by cellulose acetate membrane electrophoresis. Oxoid 2.5 x 12 cm cellulose acetate membranes were supported in a Shandon migration chamber (Colab Labora- tories, Inc.). Current was supplied by a Vokam constant current 500 mA DC power supply. The barbital acetate buffer described by Owen (115) with calcium lactate omitted was em- ployed. The buffer solution, pH 8.6, ionicity 0.07, contained the following: Sodium diethylbarbiturate 5.0 gm _Sodium acetate (anhydrous) 5.25 gm 1 N Hydrochloric acid approximately 54 ml1 Distilled water g.s.ad. 1000 ml .Five ul samples were applied directly over the cathode on each buffer impregnated strip and electrophoresed at a current of 1 mA per strip for two hours at 4°C. The strips were stained overnight in 0.001% nigrosin in 2% acetic acid. J'pH adjusted by titrating with 1 N HCl. RESULTS .Aptibody responses of rabbits to mycobacterial cells and culture filtrates. The antibody titers of sera from rabbits inoculated with (CF in Freund's incomplete adjuvant or with CF+ killed cells of M, bovis are Shown in Tables 5 and 6. Mercaptoethanol- sensitive (MES) and mercaptoethanol-resistant (MER) passive- hemagglutinin and agglutinin titers are given in Tables 7 and 8. Many of the rabbits produced antibodies detectable by both methods one week after inoculation and all but three rabbits by two weeks post-inoculation (Figs- 1 through 16). All of the detectable antibody produced during the first two weeks post-inoculation was MES. Some sera obtained three weeks post-inoculation contained MES antibody: all sera except from Group V rabbits contained MER antibody at four weeks post-inoculation. Group I rabbits produced Significantly more MES bacterial agglutinins for the first two weeks than rabbits in Group II and III (Figs. 2, 4, and 6): there were no significant differ— ences among MES hemagglutinin titers. Detectable bacterial agglutinins and hemagglutinins were not produced by rabbits in Group II until two and three weeks post—inoculation, respectively, and there were not significant differences be- tween the bacterial agglutinin and hemagglutinin titers at any week. 51 .0u0Lu-.$ 0L3u_30 mh>0£ ED~LDDUQDOU>Z L0\DC@ W—.OU UU-~¥ £U-3 UDUI—DUOC- HU-fiDIL >0 VUUTDOLQ >UOD_DCQ Unk~UUQnIUD~L0£UUfim>~OQ to ~0L0u~u E00: I“ 0.00% .>_o>_uoodmoc .0. cu ..u~o m .N ._ .0 amass: co_u:_~o Eaton 0:0 mm 00000 new 0N_m o» ..ouo .0: .0N 52 ._ o_omh cm cream mm wean—:00c_ .asotm sod mu_bomt notch I coaco .o. .o atmu_s N 5.5 m.0 0.x m.m 0.w n.m n.: m.: ___> 0.0 0.0 m.o m.m m.m 0.0 m.m 0.~ __> m.a o.o e.k e.o o.e o.o e.o s.: _s n.m 0.m m.m 0.: 0.: 0.: m.m 0.~ > 5.5 m.m m.m 5.0 0.0 m.m m.: m.0 >- m.m m.m 0.0 m.m m.n 0.5 m.m 0.m ___ ~.m 5.0 o.~ M.“ 0.5 o.o m.¢ o __ m.m 0.x 5.0 0.5 h.o h.@ 0.: m.m _ w m 0 m, : ,mw, M. _ Ndaowm. zo_h<4:uoz. mmhu< mxmmz .oumru__m menu—so m_>ob Ss_touomboo>z LO\ocm m__ou 00...: ru_3 coon—:00c_ mu_boos >0 ouuwoOLQ >0On_ucm u_m_oochuo_rmroumm>_oa mo menu.» coo: .m 0.0mh .>_o>_uoodmot 0. cu ..ouo .m .N ._ .0 Logan: co_u:__u Eaton one no 00000 otm 0~_m OD ..000 0: .0m 55 ._ o_omb cm crocm mm coon—300:. .aaotm sod mu_oomc .0_ .0 mcou_h ours» I 030cc N _ To no. so o.m o... 5m o.m o.~ :; m.0 m.~ m.0 m.m m.m m.m m.o o __> m.: o.m m.m m.m m.: m.: o.m o.m _> o.m o.m 5m Tm m.~ A; To To > og To nm 5: TN o.~ To To 2 m.n o.m m.0 m.m o.m n.m m._ m.0 ___ og e; To Tm Tm o; o o __ og To o.o Tm o.m T... TN o.~ _ m A o m a m (M _ «deals. 222.582. Eta can: m__ou vo___x no.3 boom—:00c_ marrow; >0 powwwwuu_M”_HHMW"mmmm”wwwowwhmowwmnmwwm:Momwmn .0 o_nmh 54 .>.o>.uoodmor .0. 00 ..000 .m .N .. .0 0005:: co.u:..o Eaton 000 mm poooo 00m 0N.m 00 ..000 .0: .0w .0. .0 mcmu.h. >000.0cm Demon.motI.0cmruoODQmosoz I m: >000.ucm o>.0.mcomI.ocm0umODQmoeoz I mm .. 0.0mh c. c300m mm ovum—soon. .aaotm cud mu.00mc 0000» I aaocow o; o o; m; m... k.— ..I. o.... Tm TN T. o... o. A... o m... ___> m.m m.0 0.0 0 0.0 m.0 0.0 m.~ 0.0 m._ 0.. 0.0 0 m.m 0 0.N __> 0.~ m.~ 0.m 0.m 0.. 0.0 m.~ 0.0 0.: m.m .0.m 0.0 0 0.0 0 0.: .> 0.N n.— N.. m.. m.0 N.N 0 0.: 0 0.:. 0 0.: 0 m.m 0 0.~ > m.n m.0 m.0 0.. n.0 m.0 m.m m.. m.: n.. 5.0 0.: 0 m.: 0 m.0 >— 0.m m.0 m.m 0 0.5 0.. 5.0 0.. m.0 .0.. m.m m.m 0 m.m 0 0.m ___ 5.x 0 «.0 0 0.x 0 m.0 5.0 0.5 0 >.m m.~ 0 0.: 0 0 _. «.0 0 >.0 m.0 m.0 m.0 m.m m.. m.: m.~ 0 «.0 n.0 m.m 0 m.m . a m m m m m m m z m x m m m :m mm 03000 0 m s 1% _ 20_h<4:002_ awhm< 02003 A .oumcu..m 0030.30 m.>00 53.0000m0ou>z co\ocm m..uu ou...x 00.3 0000.:ooc. mu.00mc >0 oouavoto >oO0.uco o.m.ouchoo.tm0uomm>.oa ucmum.motI.0cmcuoouamotos 0cm o>.u.mcumI.OcmrquucmocoE mo mcou.u :00: .N 0.0oh 55 .>.0>.uoocm0c .0. OD ..000 .m .N .. .0 00053: co.u3..0 E300m 000 no 00000 000 0~.m 00 ..000 .0: .0N .0. .0 mc0u.h. .. 0.0mh c. czocm mm 0000—3000. .a3otm L00 mu.0omc 0000h I 03000 >000.ucm ucmum.m0cI.Ocmcumoudmocoz I m: >000.ucm 0>.u.mc0mI.0cmcuooudmoc0z I mm N h.0 o m.0 0 0.0 0 n.: m.0 m.m “.0 m.o m.m 0 o.m 0.N ._.> as no no or as M... m... o._ o... n. no 3 o no o __> m... a am a .3 no u... 0.. a... no a m... a as ea ; 0.0 0 m.: m.o m.m m.o 0.N m.. n.o h.. 0 N.— o 5.0 m.o > o; e no a ed no n... no 2 no to m; o no no >_ 2 o a; o 3 e 3 no .3 o e.~ 5. no no no .2 a; o a; e no no a... no 3 .3. no no a o o __ no no fin to .5 no A... o._ n... no o.~ n; no m." o.~ _ a m e m e m a m e m e m e m mm M380 o m I. w. _ 20_h<43002_ xmkm< 02003 A .0umcu..m 0030.30 m.>O0 E3.t0uom0ooxz c0\0cm m..00 00...x 00.3 0000.30Oc. mu.00mc >0 00030000 .0 0.0mh mc.c.03.mmm .m.couom0 acoum.motI.Ocmcuooucmotoe 0cm 0>.u.mc0ma.0cm000oDcmotue mo .mtuu.u c002 .A. 0.0mh. _ 03000 c. no.0om0 >0 00030000 mc.c.u3.mmms0: .. 0030.0 mxmuz 0 h 0 m e m N u . _ _ _ p 3 ,0 3 J/ I. v IN .8 .6 In 5 >000.uc< Damon.m0mu.0cmro0o§mu0ozIf >000.uc< 0>.u.mc0mI.o:mcu0Oudmo00z.fiUI ESLvm DOUMOLUCD‘I I? In I0 0 1|” UBGHON NOI101I0 H0838 .A. 0.0mh. . Q3000 :. no.00m0 >0 00030000 mc.c.u3.mmm .m.000000 .N 0030.0 02003 --00 -I~ MB -In ~0- an r‘N ~I-I “L f 33' W >000.uc< ucmom.m0mu.0cm0u000000002{I >000....c< 0>.u.mcomI.0cmruooudmo002I-I E3000 000000.25... Iv uaawnu N0 um I 0 H0838 8 5 .A. 0.00». ._ 03000 c. mu.00m0 >0 00030000 «0.0.03.00050: .m 0030.0 1 a N 0 0 02003 c . _ . . m m m SI Cl C . 0 I. IN a In >000.uc< ucmum.m0xI.Ocmru0oudmo0ozLur >000.uc< 0>.u.mcomI.0cmsuoouamo00z.uT Io 53000 00000005....- I... I0 a m UiflwflN NOI101I0 H0835 .A. 0.00h. _. 03000 c. mu.00m0 >0 00030000 mc.c.u3.000 .m.0ouomm 02003 Inc-4 .: 0030.0 59 >000 . 0:0 0:3... . 30.. 3005000000002 * >000.uc< 0>.u.mCsmI.Ocmcu0Ou0mu002 1.: E3000 030000.00 I.I 839HON NO I101 I 0 "0835 O 6 .A. 0.0mkv ___ 03000 0. mu.00m0 >0 00030000 mc.c.u3.00080: .0 0030.0 00003 a .. 0 m e n N n n - n - u - p — n Id Ia a a In >000.uc< ucmum.m0mI.ocmru0000mo00z.LGP a >000.0c< 0>.u30004055030300: I.I 0 E3000 00000000: I? Im 0 I0 0 e . 1N UBGNON NOI101I0 NflflBS l 6 .A. 0.0000 __ >000 3.2 0:000 300.. 3000000000000: 1‘: >000 .094 0> n u pmcomn 3000000000000: n'v E300m 00000000: .... _ 0:000 c, 00000 mgmmz 00 >0 00030000 0:, c_u3—mm0 _0_00uu0m ’N 0‘ .0 003mmm UBQHflN N0|lfl1|0 H0835 2 6 .A. 0.00hv >_ 03000 c_ mu_0000 >0 00030000 mc_c_u3_mm050: .m 003m_m a u e mxmmz e . . p m — m m m 0 . n. .0. fl. i In I 3 IN In >000 _ uc< 0:000 0 00¢: 3000000000000: .q >0003c< 0>_u_mc0mn_0000000000000z¢ e 5303 000000005... ., a c In Iu . 'N ? _. UBGHflN N0llfl1l0 H0835 .A. 0.00hv >_ 03009 c_ mu_0000 >0 00030000 mc_cmu3_mm0 _0_00000m mzumz __ 0 0 m 0 'M -N .fi .m 003m.» Inn; Car/9 l1. 635 >000 _ uc< 0:000 _ m0¢u_0:0000300000z la >000~uc< 0>_u_mc0mu_0c00000000000z.nu- E300m 00000000: n°I 1% In uaewnN NO I 1m I (I "0838 .A. 0_nmhv > 0:000 :0 mu_0000 >0 00030000 mc_c_u3_mm080: .m 003m_0 a h c m mxmmzt n . \n p n p u m m t In a t 4 a UN .._. M i. In at >000muc< 0:000_m0mn_o:00000000000z.hur 1A 0 ‘ ‘l I >000Zc< 030 _mc0ml0000000000000z In: 5300.4. 00000000: I.- Im yo . 'N uaawnu NO I101 I (I HOBBS 65 .A. 0.00hv > 03000 0_ mu_0000 >0 00030000 00_0_u3_mm0 _0_00uomm .o. 003000 h c m mxmmz m INu/n- P n W m m .r— 0 >000 _ 004 00000 0 00,0- 30000000000002.‘ >000_uc< 0>_u_mcomu_000000ou00000z*u| .N E300m 000000003... .m .0 . .m um O F” uaewnn Nou1n1lo wnuas \fu-AuaanUCIK Ilv)—U-m:r.va—fvlulx ‘I‘ G tic Cl nl.“ II II II} I II .0. 0.0000 _> 0:000 :0 3mm: m -r m0_0000 >0 00030000 00_0m03_mm0E0: ’M .0. 003000 66 >000 _ 00< 00000 _ 00¢- _00000000000002‘1 >000 _ 00¢ 0> _ 0 _ 0.00m- 30000000000002...- E300m 0000000003.... U39HnN NO I mu 0 HOURS C-oo 9h >000_00< 00000_m0mu_0000000000000z.du >000 _ 00< 0> _ 0 _mc0wu_0000000000000z.u. E300m 000000002... 7 6 .A. 0_00hv _> 03000 0_ 0000000 >0 00030000 mc_0_03_mm0 _0_00000m .m. 003m_0 mxmuz 10 uaawnN NO I 101 I 0 H0835 8 6 .A. 0.00hv __> 03000 00 m0_0000 >0 00030000 mc_0_03_mm080: .m. 003m_0 a o mxmuz HW\\\\\\\\\\AM. . w w m m m - la 1N In >000 _ 00< 0 000 m _ 00¢- 3000000000000; >000_00< 0>_00000m-_0000000000000z&-. 0 5300.0. 000000002... In L. 83800N NO I 1.01 I (I "0835 .A_ 0_00hv _.> 03000 0_ 0000000 >0 00030000 0000_03_mmm _0_00000m .0. 003m_0 mxmmz a . 0 .0. .0 0 r m 0 a . . w t u I“ ..~ Mw - -n 3 >000_00< 00000000m-_000000o000000zucw >000_00< 0>~0_mc0m-_00000000000002fi.. -0 E300m 000 000003.. 10. -0 nu uaawnu N0|ln1|0 unuas O 7 .A. 0.0000 _._> 0300c 0_ 0000000 >0 00030000 0000_03_mm060: .m. 003000 mzmmz a s a m e m N 0 . 0 . . . . . . n :0 a 1N h 0.. . In >000_00< 00000_m0xu_000000000000020AV. . >000_00< 0Z0 _mc0m-_00000000000002¢ e . m. I E300m 00000000; J A? ..m no 00 839NON NO I .101 I 0 "0835 .A. 0.000V .._> 03000 0_ 00—0000 >0 00030000 mc_0_03_mm0 .00000000 .0. 003m_0 ~00 -N «0 I'll} m2wm3 but I’M 'N Inn-O .1 7 >000_00< 00000._m0mn_0000000000000z .dl >000000< 0>_0_mc0mn_0000000000000z.nl. E300m 00000000: .I 0m D 839NON NO I .101 I 0 "0835 72 An early and sustained production of MES hemagglutinins was elicited with CF in incomplete Freund's adjuvant (Group 1). After three weeks an increasing amount of MER hemagglu- tinins were produced (Fig. 1). In contrast, rabbits which received CF+ killed cells (Groups II and III) produced less MES hemagglutinins and for a shorter time (Figs. 3 and 5), and MER antibody was detected one week earlier in these groups and in significantly higher titer. The sequential production of MES and MER bacterial agglutinins was similar in all three groups (Figs. 2, 4, and 6). Hemagglutinins produced by rabbits in Group V were pre- dominately MES (Fig. 9); mercaptoethanol-resistant hemagglu— tinins were not detected until six weeks post-inoculation and only in one serum. The temporal sequence of the produc— tion of bacterial agglutinins was similar to that of hemagglu- tinins. However, a stronger and earlier MER antibody response occurred (Fig. 10). Sera from all of the rabbits in this group contained MES antibody throughout the experiment. There were no significant differences in the temporal sequence or the amount of MES or MER antibodies elicited by CF with heat-killed (Group II) or BPL-killed (Group III) M, boyig (Tables 7 and 8). Antibody was detected one week earlier in the sera from rabbits in Group III. Multiple injections of CF+ BPL-killed g. M (Group III) 'elicited MES and MER hemagglutinins one week earlier than sequential injections of the same antigen in Group IV (Figs. 5 and 7). 73 The amount of MES and MER antibody produced by both groups was not significantly different. Repeated intravenous injections of CF with BPL—killed M, bgyi§_elicited the highest titers of MES antibody (Tables 7 and 8). Mercaptoethanol-resistant bacterial agglutinins and hemagglutins were not detected until three and four weeks respectively, post-inoculation (Figs. 11 and 12). Mercaptoethanol-sensitive hemagglutinins were produced throughout the duration of the experiment; MES agglutinins were not detected after the six week period. The antibody response of rabbits which received sequential or multiple injections of antigens without an intravenous in- jection of CF, Groups III and IV, differed from those which received an initial injection (Groups VII and VII)(Figs. 13 and 15). The peak hemagglutinin titer was produced one week later, and less MER hemagglutinins were produced by Group III rabbits than by Group VII. There were no significant differ- ences between titers of bacterial agglutinins. Rabbits in Group VIII produced more MES hemagglutinins for the first six weeks post-inoculation, less MER hemagglutinins at 4, 5, and 6 weeks post-inoculation, and more bacterial agglutinins for the first four weeks than rabbits in Group IV (Tables 7 and 8). Rabbits in all groups except Group V produced precipitins. Two to four immunOprecipitates were formed with CF. 74 Precipitins elicited by alum-precipitated and aluminum chlorideeprecipitated culture filtrate. From 12 to 15 immunoprecipitates were visible in immuno- electrophorograms of CF with both antisera. No consistent differences were observed in the number or displacement of the lines formed with the antisera obtained from rabbits inoculated with alum-precipitated CF or aluminum chloride- precipitated CF. Chemical analyses of culture filtrates. The concentration of protein, polysaccharide, and nucleic acids, and the ultraviolet absorbancy ratio of the culture filtrates at 280 and 260 mu are shown in Table 9. Culture filtrates B and C contained the greatest amounts of protein, polysaccharide and nucleic acid. The 280/260 absorbancy ratio was approximately the same for all of the culture filtrates. A representative ultraviolet absorption spectrum of one of the culture filtrates (CF-C) is shown in Fig. 17. The peak absorption occurred at approximately 210 and 270 mu. Similar absorption Spectra were obtained for all of the culture filtrates. Chromatography of culture filtrates. The culture filtrates were separated into three or four major 280 mu-absorbing fractions by molecular exclusion chromatography on Sephadex G-25 (Figs. 18, 19, 20, 21, 22, and 23). The chemical composition of the major fractions from the individual culture filtrates are shown in Tables 10,11,12,15,14, and 15. The chromatogram of CF-A contained 75 0000 0_0_03z I <20 _0_0000E mc_00000I_oe>0h I zmhm .0000006 00.00000I0__00 I 200: 32 com 000 30 0mm 00 >000000000 00_0_>000_3 000 00 0.000 I om~\om~m 0o_000 00_000300_ 000 00 000000 I 0m00 E3.00000000>z 000000000m00 0-00 000 .00000003 x_m 000 00 00.0.000500 _00_E0su .m 0_00h 76 .Am c.00hv u m0mL0__m 0030—30 m_>on Eamgw0umnou>z 00 53000000 0000000000 0o_o_>000_= .m_ 003mmm 0050 000200 000: cm m can 3N cum 3% emu 8N AllSNBO 1V3lld0 77 . .0~-0.x00~0000 :0 Am 0.0mhv < 00000__m 0030.30 m_>on E3_0u0umnou>z mo >gamgm00meoLcu co_m3_uxu 00.300.02 .m. 003m_u mumzaz wash 00 00 00 00 00 00 I¢a Ia: nu: 082 GD ALISNEIO 'WfllldO 78 :0 Am c.00hv m 00000__m 0030—30 m_>on E3_Lm0umnou>z mo >camgmo0meoLcu 00—03 as fie 00 ammzzz mash av an .m~-0 x0000000 _uxm 00.300.02 3N w .0. 0030.0 ééé é nu: 082 ('9 MISNHO wauao .m0I0 x0000000 :0 Am o_0mhv u m0mg0__m 0030—30 m_>on E3~Lm0umnou>z mo >camgmo0msogzu co_m3_uxu Lm_3uo_oz .o~ oL3m_u am $52 mag. 00 00 00 0.0 00 W0 s0 79 nm 082 G) MISNBO 1V3lld0 .00I0 x00000om co «m 0.0mpv a m0mg0__m 0030—30 m_>on 53.000umnou>z mo >camLmO0meoLnu co_m3_uxo 00.300.02 ._N 003m_m mumzzz was... 00 0.0 0.0 0.0 0.0 0.0 0.0 .3. . .. 00 I00 .00 O 8 .00 .00 .00 I00 nm 082 Q AllSNBO 1V3lld0 .m0-0 x0000000 :0 Am 0.0mhv u m0mg0__m 0030—30 m_>on E3_Lo0umnou>z mo >Lqmum00meoL£u co~m3_uxu Lm_3uo_oz .NN 003m_m mmmzaz mma. m0 0.0 0.0 an 00 -8 .00 .00 .00 1 8 .00 .00 .00 In: Iad I3 MI 082 9 MISNBO 1V3lld0 82 .m0-0 x00000om :0 Am o_nmhv m m0mL0__m 0030—30 m_>on E3_Lo0uwnou>m.mo >camgmoumeoLco co_m3_uxo Lm_3uu_oz 000232 0030 00 00 gm $0 on 00 i w 4 .00 0000_0 m 032 (P) msuao womo 85 00:30005 002 I zzm 00000000 003 :0.000:. 0:0 :0.:3 :. 053.0> 0:03.0 0:0 0:0 053.0> 0.0> :53.00 0:0 00 0.00m I mmm 35 emu 0:0 35 omm 00 >0:0::0000 00.0.>0:0.3 0:0 00 0.000 I 0030mmN 0:0.000:w .030.>.0:. 0:0 :. 00.00 0.0.03: .0000 0:0 .0 0:00:00 I <2 N0 0:0.000:m .030.>.0:. 0:0 :. .0.:0005 m:.0000:I.05>:0 .0000 0:0 .0 0:00:00 I zmh Nu 0:0.000:m .030.>.0:. 0:0 :. .0.:0005 m:.0000:I:..om .0000 0:0 00 0:00:00 I :00 NJ 0:0.000:m .030.>.0:. 0:0 :. .0.:0005 m:.::00:0I3E owu .0000 0:0 00 0:00:00 I z. m~.o w:.o m.. . o o m.. ___ mm.o mm.o m.mm ..o o.w. w.mm _— o.. mw.o :.m. m.mm o.~m m.:¢ . mum nomm\om~ m0: 53.:0000000Nm,m0 >:00:m000500:0 >0 00:.0000 0:0.000:m 0:0 m0 :0_0.000500 .00.50:u .o. 0.:0h 84 00:30005 00: u :zm 00000000 003 :0.000:0 0:0 :0.:2 :. 053.0> 0:03.0 0:0 0:0 053.0> 0.0> :53.00 0:0 00 0.00: I 0mm 35 com 0:0 35 0mm 00 >0:00:0000 00.0.>0:0.3 0:0 00 0.00: n omm\ow~m 0:0.000:0 .030.>.0:. 0:0 :. 00.00 0.0.03: .0000 0:0 00 0:00:00 I <2 w 0:0.000:0 .030.>.0:. 0:0 :. .0.:0005 0:.0000:I.05>:0 .0000 0:0 00 0:00:00 u 20h Mm 0:0.000:0 .030.>.0:. 0:0 :. .0.:0005 m:.0000:I:..00 .0000 0:0 00 0:00:00 u :00 x: 0:0.000:0 .030.>.0:. 0:0 :. .0.:0005 m:.0:0000 :5 0mm .0000 0:0 00 0:00:00 u z00 53.:0000000>z 00 >:00:mo0050m:0 >0 00:.0000 0:0.000:0 0:0 00 :0.0.000500 .00.50:u ... 0.00h 85 00:30005 00: n zzm 00000000 003 :0.000:0 0:0 :0.:3 :. 053.0> 0:03.0 0:0 0:0 053.0> 0.0> :53.00 0:0 00 0.00: n 00 35 emu 0:0 35 emu 00 >0:0::00:0 00.0.>0:0.3 0:0 00 0.00: n omw\ow~ gm 0:0.000:0 .030.>.0:. 0:0 :. 00.00 0.0.03: .0000 0:0 00 0:00:00 I <2 No 0:0.000:0 .030.>.0:. 0:0 :. .0.:0005 0:.0000:I.05>:0 .0000 0:0 00 0:00:00 I 20h 0m 0:0.000:0 .030.>.0:. 0:0 :. .0.:0005 0:.0000:I:..00 .0000 0:0 00 0:00:00 n 200 N: 0:0.000:0 .030.>.0:. 0:0 :. .0.:0005 m:.::0000 35 omu .0000 0:0 00 0:00:00 n z_ 00.0 00.0 0.. 0 0 0.0. ___ 00.0 00.0 0.00 0.0 0.0 ..00 __ 0. _ 00.0 ..00 0.00 0.00 0.00 . I. ‘0 00 uunn :0 000: 000 0002000 002 00 05: .0 05:... \ 020.000 00 N _ HI... .mNIo X000:00m :0 u 000:0..0 0:30.30 0.>00 53.:0000000>z 00 >:00:mo0050w:0 >0 00:.0000 0:0.000:0 0:0 00 , .:0.0.000500 .00.50:0 .N. 0.:0h 86 00000000 003 :0.000:0 0:0 :0.:3 :. 053.0> 0:03.0 0:0 0:0 053.0> 0.0> :53.00 0:0 00 0.00: a 0mm 35 com 0:0 35 own 00 >0:00:00:0 00.0.>0:0.3 0:0 00 0.00: I omN\om~ m 0:0.000:0 .030.>.0:. 0:0 :. 00.00 0.0.03: .0000 0:0 00 0:00:00 u <2 *0 0:0.000:0 .030.>.0:. 0:0 :. .0.:0005 0:.0000:I.05>:0 .0000 0:0 00 0:00:00 u 20h Na 0:0.000:0 .030.>.0:. 0:0 :. .0.:0005 0:.0000:I:..00 .0000 0:0 00 0:00:00 a :00 Q 0:0.000:0 .030.>.0:. 0:0 :. .0.:0005 m:.::00:0I35 omN .0000 0:0 00 0:00:00 n z.0 :00 0000:30:. 00:30.30 50:0 00:.00nom 00.0 00.0 0.0 0 . 0.0 0.00 _: 00.0 00.0 0.00 0.0 0.0 0.00 : 0. . 00.0 0.00 0.00 0.00 ..mm . 000 0000000 002 00 020.0 .0 0:00 00 0z<0mmr00 08:00:... .mNIo X000:00m :0 o 000:0..0 0:30.30 0.>0: 53.:0000:oomm.00 >:00:m00050::0 >0 00:.0000 0:0.000:0 0:0 00.:0.0.000500 .00_50:o .m. 0.00H 87 00000000 003 :0.000:0 0:0 50.53 c. 053.0> 0:03.0 050 0:0 053.0> 0.0> :53.00 0:0 00 0.00: a mm 35 000 0:0 35 ow0 00 >0c00:0000 00.0.>0:0.3 0:0 00 0.00: n 000\ow0m 0:0.000:0 .030.>.0:. 0:0 c. 00.00 0.0.03: .0000 0:0 00 0:00:00 I <2 00 0:0.000:0 .030.>.0:. 000 c. .0.:0005 mc.0000:n.05>c0 .0000 0:0 00 0:00:00 u zmh Na 0:0.000:0 .030.>.0:. 0:0 c. .0.:0005 mc.0000:nc..ou .0000 0:0 00 0:00:00 n :00 x: 0:0.000:0 .030.>.0:. 0:0 c. .0.:0005 mc.0:0000 35 om0 .0000 0:0 00 0:00:00 zon 53.:0000000>m.00 >cam:mo0050:;0 >0 00c.0000 0:0.000:0 0:0 00 :0.0.000500 .00.50cu .0. 0.000 88 00000000 003 :0.000:0 050 00.03 c. 053.0> 0c03.0 0:0 0c0 053.0> 0.0> :53.00 000 *0 0.00m I wmw 35 000 0:0 35 omN 00 >0:00:0000 00.0.>0:0.3 0:0 00 0.000 a om00ow00 0:0.000:0 .030.>.0:. 000 c. 00.00 0.0.03: .0000 0:0 00 0:00:00 I <2 N0 0:0.000:0 .030.>.0:. 050 c. .0.:0005 mc.0000:n.05>;0 .0000 0:0 00 0:00:00 u Imh gm 0:0.000:0 .030.>.0:. 0:0 2. .0.:0005 mc.0000:n:..om .0000 0:0 00 0:00:00 a 200 N: 0:0.000:0 .030.>.0:. 0:0 2. .0.:0005 mc.n:0000u35 ow0 .0000 0:0 00 0:00:00 n z00 :00 00000302. 00:30.30 50:» 000.0000. 00.0 00.0 0.0 . 0 0.0 0._0 .2 00.0 00.0 0.00 0.0. 0..: 0.00 : 0.. 00.0 0.0. 0.00 0.00 0.00 . 0.; 00000000 002 0 05:0 .05: .0. 0.2000 .0. 003000.00. 0 .m0so X00mcq0m c0 “0.000:0..m 0:30.30 0.>00 amqummmmmmmxm.0o >gam:mo0050m:0 >0 00c.0000 0:0.000:0 0:0 00 c .0.000500 .00.50;o .m. 0.00» 89 Umanmjor 280 mu-absorbing fractions and two minor fractions (FM; 18). Fraction I was eluted in the column void volume (Rf==1.0) and contained 82.0%, 93.9% and 15.9% respectively, cflfthe protein, polysaccharide and nucleic acid in the swung (Table 10). Fraction II had a small shoulder on both sideslof the major peak and was composed of molecules with molecular weights less than 5,000 (Rf = 0.38). The yellow pigment seen in all of the culture filtrates was eluted in this peak. Most of the nucleic acids in the sample (85.5%) were found in fraction II. Only 1.5% of the total 280 mu- absorbing material in the sample was found in fraction III. This fraction contained no polypeptides or oligosaccharides and was almost all low molecular weight polynucleotides. Fraction IV contained 0.5% of the 280 mu-absorbing materials in the sample. Fractions I and II elicited immediate-type skin reactions in sensitized rabbits. Four main fractions were seen in the chromatogram of CF-B (Fig. 19). They contained 49.1%, 36.0%, 12.9%, and 1.8% respectively, of the total 280 mu-absorbing material in the sample (Table 11). The first fraction was eluted in the 00 53.0000000002 . 00 >00000000eot00 >0 00c.0000 Aw. 0.00hv . :0.00000 00 23000000 00.0000000 00.0.>0L0.: .:w 003m.m Aaev :Hozma m><3 0. . .0. . .0 . .0 . .0. . .0 . 0. ALISNBO 1V3|id0 .mNuo x000000m 00 Am 0.00hv 0 00000..» 0030.30 M0>00 E:.00000000>z 00 >0000mo0000000 >0 000.0000 00. 0.00». .. 00.00000 00 63000000 00.0000000 00.0.>000.= .mN 0030.0 a3Ev :hozu0 w><§ . 0mm . £0 . 00.00 0- . £0 . 0m~ . 000 92 AllSNBO TVDILJO 95 .mmno x000000m 00 Am 0.00». 0 00000..m 0030.30 m.>00 E3.00000000>z 00 >0000mo0000000 >0 000.0000 Aw. 0.000. ... 00.00000 00 83000000 00.0000000 00.0.>000.: .00 0030.0 .00. 000200 0002 000 000 000 00.0 0.00 000 00.0 b AilSNBG 1V3lid0 94 Culture filtrate D was separated into three major fractions (Fig. 21). Fraction I was eluted in the column void volume (Rf = 1.0( (Table 15). It contained approxi- mately one-third of the total 280 mu-absorbing material and 82.6%, 92.8%, and 20.9% of the protein, polysaccharide, and nucleic acids, respectively. Fraction II was fairly sym- metrical but contained two other small peaks. The chemical composition of this fraction was similar to that from CF-C. Fraction III contained low molecular weight polypeptides (Rf = 0.25), no oligosaccharides, and a small amount of poly- nucleotides. The chromatogram of CF—E had three fractions (Fig. 22). Fraction I was separated into two subfractions and contained less than one-third of the 280 mu-absorbing material in the sample (Table 14). Fraction II contained more 280 mu- absorbing material, polypeptides, oligosaccharides, and poly- nucleotides than fraction II from CF-D. The chemical compo- sition of fraction III from both culture filtrates was similar. Culture filtrate F was separated into three major frac- tions (Fig. 25). The first fraction was eluted in the column void volume and contained two subfractions. Less protein, polysaccharide, and nucleic acid was found in this fraction than in fraction I from Cf—E (Table 15). More polypeptides and polynucleotides were found in fractions II and III than in the corresponding fractions from CF-E. 95 Chromatograms of the six culture filtrates on Bio-Gels P-100, P-150, and P-200 are shown in Figs. 27, 28, 29, 50, 51, and 52. The distribution of protein and 280 mu—absorbing material in the chromatographic eluants are shown in Table 16. All of the culture filtrates except CF-F were separated into two fractions on all three Bio-Gels. Only one fraction was obtained by chromatography of CF-F on Bio-Gel P-200 (Fig. 52). From 55.1% to 61.9% of the proteins in culture filtrate A-E were eluted in the void volume of P-100 columns (Table 16). Fraction I from Cf-F contained 25.2% of the protein in the sample. The first fraction eluted from P-200 columns contained from 55.5% to 44.8% of the protein in the culture filtrates A-D (Table 16). Twenty-six percent of the protein in CF—E was eluted in the column void volume. All of the protein in CF-F had an Rf value of less than 1.0 when chromatographed on Bio-Gel P-200. The percent distribution of the culture filtrate proteins in fraction I from Sephadex G-25 and Bio-Gels P-100 and P-200 are shown in Table 17. The nondialyzable proteins in fraction I from CF-C ob- tained by rechromatography on Sephadex G-25 were separated into five fractions by gradient elution from DEAE-cellulose (Fig. 55). Approximately 50% of the proteins in the sample were recovered in the chromatographic eluants. Fractions I, II, III, and V were composed of several subfractions with different affinities for the adsorbent. 96 fi— oowum ..u... #4 e I '8. e I ‘1 G 3 mmmzzz man... an m ! .00~-0 0:0 .00.-0 .00_-0 0.00.0.0 00 Am 0.000. < 00000..m 0030.30 m.>00 53.00000000xm.00 >0000m00050000 00.03.0x0 00.300.0z nm 082 G) '0 '0 0 om...n. 1...... I N e I "2 6 I ‘i .e 3 r9... nu: 082 9 '0'0 .mu 0030.0 0.. 0.. .0... '8 :00 um... ...... oo.um .0... .00 ""1 082 0 '0'0 . .oouum ucm .om_-¢ .oo_-m «_ou.o_m :0 Am m_nmkv m mumLu__m 0L:u_:u m_>on E:_Lmuumnouxz mo >zamgm0umeoLzu co_m:_uxm Lm_:uo_oz .mN mgdm_m - xmmzaz wash . a 3 2 3 ... 2. 3 2. 2 m 3. 2 S . m b n I I D p I n W n n I .3 r3 .0 0 . .a. ”a .3@ .3 a m m 7 O o 9 .2 m .3 m . .3. ‘3. I: '3 .3. .3 oouum om_um oo_um mu 082 (9 '0'0 .oo~-¢ ncm .om_-m .oo_-m m_mu.o_m :0 Am m-nmhv u mumgu__m 0L:u_au mw>on E:_L0uumnou>z mo >camLm0umEOgcu co_m=_uxo Lm_:uo_oz .mN ogam_u . _mmmzaz Noah mm ...N m.“ an m mm i ..u .1 m fl mm a .... .8 '3 o o .a mu I2 ‘0 IN: @ z 2 MW MW mm .2 m .8 m .3 .3. um: um: :3 mad C oomum . omTi oo-nm N v nm 082 Q '0'0 99 3N mm, am cowl.— mu 082 Q '0'0 N .oo~-¢ 0cm .om_-¢ .oo_-¢ n.0o-o_m :0 Am 0305 a 303:“— 00330 m_>on Ela..__..._.0.uu_ 0300»: m0 Eamgmoumeozu 53390 00.30.02 .mmmzaz was» a a; uud IN: ind utd um: om_-¢ mu 082 Q '0'0 .om oL=m_m a m 1H: um: um: .qe Yuma 007d nw 082 Q '0'0 100 no oomum nee um: MI 082 Q '0'0 9N mmmzaz mun... m . e— . om-um .oo~-m 0cm .om_-¢ .oo_-¢ n.0u.o_m :0 Am 0.30hv m 0umgu__m 0L:u_:u m_>On E:_L0uumnou>z mo >zamgm0umeoLsu co_m:_ux0 00_:u0_oz I—c we: 1m: "l“ 082 Q '0'0 ._m 003m_u ...N a ..._ w. .a9. 2 uNc In: oo_-m u¢= on: nu: 082 Q '0'0 [ .oo~-m ucm .om_-m .oo_-¢ “.mu.o_a :0 Am 0.30hv m 0u00u__m 0L:u_:u m_>oa E:_L0uumnou>z mo >zamgm0umeoLcu co_m:_ux0 L0_:u0_oz .Nm 00:m_m mmmzaz m2: ...._ a ... _ a m e m a a ..._ m '3 .3 o .o mu .0 ..Nd Q 32 Q 1 z 7.. O 8 8 1 .2 M ‘8 M n n .3. .0... c oowum om-um . 007m :1: i nu: 082 Q '0'0 m 0.000 c. 000.00000 < 00000..» 0030.30 u <0 0co.uu0.0 .000.>.0c. 0;» c. _0.L0u0e mc.~000.uc..00 .0000 0:0 00 0:00000 n :00 00. 000.0000» .0:v_>_vc. 0:0 :0 00.00005 0:00.0000uae 000 0:0 00 0000000 n z00000000EOLcu >0 00:00000 0:0.00000 c. 00.00005 mc_u0000uc__om 0:0 00>on E:_L0uumnou>z x00 onu mo mc.000000uae omw mo co.uaa_0u0_o .0. 0.000 .193 mm 0.000 c. 000.00000 <.00000..0 0030.39 n (m ..-co.00000 c. .0.0000E 0:.00000uc..00 .0000 0:0 00 0000000 a :00 RN 00000..» 0000.00 - 00. 0.0 0.00 0. .0 ... 0.00 0.00 0.00 0 0..... ..00 0.00 0 0. ... 0.00 0.00 0 0.00 0.00 N. .0 0 0.0.. 0. .0 0.00 0.. W 20010. ..00 .0 0:00 001 .0“... 00.0.... 00.-.. 0040.. .oowum .00.0.0 0:0 oo.-0 .00.0.0 .mwno x000000m :0 000000..» 0030.30 0.>00 50.000000000z x.» 000 00 >00000000500cu >0 v0:.00ao . :0.00000 c. .0.0000E m:.00000-¢..00 00 c0.0:0.000.o .0. 0.000 1104 9: oc— .m~no xovmzamm :0 Am o_nmhv u mumgu__m mung—nu m_>on E:_Lmuumnou>z mo >camgmoumeoLzu comma—uxo Lm_:um_oe >3 vmc_muao _ co_uumgu mo >zamgm0umeoLcu mmo_:__muumon E:_Louumnouxr mo muco:u_umcou Amc__ p__0mv m.nm~>_m_uuc0c vcm Aoc__ cuxotnv m.nm~>_m_v mo >camtm0um50tcu comma—uxo Lm_:um.oz .:m mu:m_u mumzaz mash :N :2. I2 0 — ~ m . . . .2 m . . .. N m . A . w . ~ . . ...... m . . . . a . . — . .3 m . — ~ — u n. . . r. m a Q = I06 .I s. E .3 .3 107 .n __mz c. poum_a mm: Eatum_ucm och .A0n Ea.touumnou>z scum mucoau.umcou Amy o.nm~>_m_vnc0c new AuV o.am~>.m.v mo Emtmo::EE_ .mm ou:m_u .108 .munu xmvmcamm :0 Am o.nmpv u oumtu._m otau.au m_>on E:_to.umnou>z mo >gamtm0umeotcu co_m:.uxo cm.auo_oe >n voc_muno _ co_uumtu mo EmLmOLOLQOLuu0_o uumuoum omo_:__ou .mm 0t:m_m 109 No stainable protein bands were detected in fractions II or III in any of the culture filtrates when analyzed electro- phoretically. Disc electrophoresis of culture filtrates. The number of amido black-stained bands in the six cul- ture filtrates were as follows: 16 in CF-A; 25 in CF-B; SO in CF-C; 23 in CF-D; 19 in CF-E; and 15 in CF-F (Table 18); Figs. 57, 58, 39,V4O, 41, and 42. All of the culture fil— trates contained five bands with similar or identical Rf values (Table 19). Twelve other bands with the same mobility were found in three or more of the culture filtrates. No protein bands were found in the spacer gel of any of the disc electrophorograms. There is no satisfactory method for determining the relative amount of proteins in each of the bands. The number of PAS-stained bands in the culture filtrates were as follows: 9 in CF-A; 8 each in CF-B and CF-C; 7 in CF-D; and 5 each in CF-E and CF-F (Table 18). All of the culture filtrates had PAS-stained bands in the Spacer gel, at the spacer gel-lower gel interface, and at the lower end of the lower gel associated with the most anodic protein com- ponent (Figs. 57, 38, 59, 40, 41, and 42). Most of the PAS— stained bands in the lower gel could be correlated with protein components with identical Rf values. Fraction I isolated and rechromatographed on Sephadex G-25 from all of the culture filtrates contained all of the :110 m 3%.. e. 8333.. ... 38:... 33.3 .. <5 m_mmto:aotuum.00caee_ a m_m co_maee.u o.a=ou .eo.tou;u=o - seem mucmcoaeou voc.mumumm_;um v.um u.pomtom u mco.t0u;oao .m_mmto;a05uuo_o om.u >3 moumtu__m menu—:0 m.>om E:.touumnou>z x_m ecu c. peso» mucocoaeou mo Lassa: .m- o-noh 111. .>.o>.uuodmot .mmmtm _mm Luzo— pcm emumam «33 cu Lemme 3 van m mtuuuo_ 03h .muuu xmumnamm co >3dmtm0um50ecu >3 pmcmmu3o «_uuv _ co_uumtm pcm Amuv Am v.3mhv < mumtu__m menu.:u m.>03 E:_tmuum3ou>z mo mEmLmoL03dotuuu.o ommv c. mucmcodeou Amy m>_u.moaum.m>_uumdmmt .mmotm _mm Lm30_ ucm Luumam ecu cu Loewe 3 van m memuum_ och .mmuo xmmeQmm co >3amLmOumeot3u >3 poc_mu3o A.umv _ co_uumtm pcm Auuv Am m.3mhv m meaty—_m menu—nu mm>03 E:_Lmuum3oo>z mo mEmLmoLOLdotuu0_o om_p c. mucmcoaeou Amy m>_u_modum_o>_uuoamot.mmotm _om Luzo— pcm emumqm 033 Cu Lemme 3 pcm m memuuo_ 03h .mmuuo xmpm3aum co >3amtm03meot3u >3 poc.mu3o A_umv _ co_uumtu vcm Amuv Am o_3mhv u mumtu_.m menu—nu m_>03 E:.Lmuum3ou>z wo mEmLmoL03dotuuo.u om_p c. mucmcoasou Amv m>_u_m0dum.0>_300am03 .m0030 .0m 3030— pcm L0umam 033 03 L000; 3 330 m mg0uu0. 03h .munu x0pm3d0m mo >3a03m030eot3u >3 303.0330 A3nuv _ co_uumtu ucm Amuv Am 0.3mhv a 0umtu._m 0t:u_:u m.>03 53.303003ou>m mo m503m0303a03300_0 um33 c. muc0coaeou Amy 0>.u_moaum<¢ p30 “<3 c.0uota 03» mo 30.0033000La03 0.30503um _Iu mu lL H HHH .o: 03:m_u 1:15 .>_0>_3u0am03 .m0030 _0m 3030. 3:0 t0umam 033 03 tmmot 3 330 m 030330_.03h .mwuo X0303Q0m mo >3d03m03meot3u >3 303.0330 A.uuv __co_3umtu 330 «333 Am 0.3mhv m 03033.30 0333.30 m.>03 53.303um3ou>z mo memtmoto3dot3u0_0 om33 c. m3c0coaeou Amy 0>_3_moasw<¢ 3cm A.0>.3003003 .00030 .0m 3030. 330 300030 033 03 30.03 3 330 m 030330. 03» .mmuo x030330m 30 >3303mo305033u >3 303.0330 A.u3. . 30.3003. 330 Aug. Am 0.30h. . 03033... 0333.30 m.>03 53.303003ou>z mo 0503mo3033033u0.0 00.3 3. 0330303500 .3. 0>.3.moaum<3 330 «<. 3.03033 033 .o 30.30330003303 u.302030m .I. .0 l l I I l [ ‘1 .N3 033m.u 1137 Rf values of the protein components in the six M cobacterium bovis cr' culture filtrates. Table 19. M3 d(cm)2 + + + + + + + ++++ ++ + + + + + ++ +++ +++++++ ++ + ++ + ++ ++++ +++ ++++++++++ + ++++++++++++++++++ + ++++++++++ ++++++ +++ ++ ++ + ++ + ++ ++++ + + +++ + ++ + + + ++ + + mm300033333330333333333333330.030 o0000000000000.000000000000000000] mewmfimmwmwmwmwmmmmmmwwnmmmmmwmwwm 000000000 lllllll LLLLLLLLLLLLLhuhuhuhH distance moved by the individual component 3Rf - ratio of the distance moved by the most anodic component and the 9A - culture filtrate.A described in Table 9 2d(om) - distance in centimeters ICF - culture filtrate 118 protein bands detected in the unfractionated culture fil- trates (Figs. 57, 38, 59, 40, 41, and 42). In only one case (CF-E) did purified fraction I contain all of the PAS- stained components found in the unfractionated CF. No protein or PAS-stained components were detected in isolated and rechromatographed fractions II or III from the culture filtrates. The yellow pigment in fraction II mi- grated in close association with the bromphenol blue track- ing dye. Analyses of the culture filtrates and chroma- tographic eluants by immunodiffusion. None of the sera from rabbits used for the production of antibodies specific for CF contained precipitins prior to immunization. The number of immunoprecipitates detected in immunograms of the six culture filtrates with antiserum A plus B was as follows: 16 in CF-A; 15 each in CF-B and CF-C; 15 in CF-D; 8 in CF-E; and 6 in CF-F (Table 18). Fraction I isolated from all the culture filtrates and re- chromatographed on Sephadex G-25 columns contained the same number of precipitinogens as found in the unfractionated 'culture filtrates (Table 20). Some displacement of several individual lines was observed (Figs. 45, 44, 45, 46, 47, and 48). All but two or three of the antigens in each CF- fraction I system formed lines of identity. The lines that did not join were near the antigen well. No precipitinogens were detected in fractions II and III from any of the culture .119 munu xovmcdom co zamtm0umeoLzo >b voc_muno _ co_uomum n Amunuv _umm .m_o_nme.c_ emn_t0mun < oumtu_iu «tau—=u - < manta—_m menu—so n mum «oouuav _ru u- a m a m o -- N m N. __ ~— Aom_-av .-u m m m m. ~_ m_ .Aoo_-av .-u o o o o o o Am~-ov ___-u o o o o o o Amwuuv ..-u m m m. m. m. o. mfimunuv _nu m m m. m. m. e. «do u m ,m u mm. _< co_umummmum uh<¢h4_m manhqau .oounm vcm .om_um .oo_nm m_oono_m new mNno xovmzaom co >ndmum0umeoLcu xn voc_muno mco_uumtm vcm mmumtu__$ menu—:0 m_>on £:_Louumnooxm.x_m ecu c_ voumtumcoeov mcomoc_u_q_ootm .o~ o_nmb 120 ...oz toucou oru c_ noun—a mm: Eatom_ucm och .mNno xovmcaom co >ramtm0umeotco >n uoc_muno «av ___ vcm .Auv __ .Amv _ mco_uumtu vcm A0n E:_Lmuum300xz mo co_m:mw_v o_naou >co~toususo .m¢ ouam_m 121 .__oz Loucoo ogu c_ ovum—a mm: Eatom_ucm och .muuu xovmcaom co >Lamtmoumeotzu >n umc_muno «av ___ wcm .Auv __ .Amv _ mco_uomuu vcm Aon E:_Louomnouxz mo co_m:mm_v m—naou >co_cou;u:o .¢¢ ouam_u <<< :122 ...oz Loucou ecu c_ noon—a mm: Eatom_ucm orb .mNuu xovmcaom co >camtmoumsotzo >3 ooc_muno Aav ___ new .Auv _. .Amv _ mco_uomum vcm Aon e:_touumnoo>z no co_m:mm_v o_b:ov >co_toucuao %/(@\\\ .m: mtam_u 1123 .__oz toucoo ecu c_ ovum—d mm: Eaton—ucm och .mNuu xoumcdom co >zamtm0umsotcu >3 uoc_muno may ___ tam .Auv _. .Amv _ mco_uomtu vcm A0n amitouumnouxz mo co_m:mm_v o_n:ov >co_uounu:o .m: ot:m_m 124 ...o3 toucou ecu c. boom—a mm: Eatom_ucm orb .muuu xovmcaom co >zamtm0umeotco >n no:_muno any ___ vcm .Auv __ .Amv _ mco_uumtu new Aoa E:_Louomnou>z mo :o_m:mm_v 0.3:Op >co_tou;uao ’<< .2 953... ,. ' 1125 ...oz touceo ecu c. boom—a mm: Eacom_ucm ugh .mwuo xovmcaom co >zamtm0umsotru >n poc_muno on ___ vcm .Auv __ .Amv _ mco_uumum vcm A0n E:_touumnou>m mo co_mamm_v o_naov >co_touzuao .m¢ ousm_u 126 filtrates (Table 20). In most cases (CF-A, CF—B, CF-C and CF-D), two to four of the precipitinogens found in the culture filtrates were not found in the corresponding frac- tion I from Bio-Gel P—lOO (Figs. 49, 50, 51, and 52). Fraction I from CF-E and CF-F contained all the precipitino- gens in the corresponding CF (Figs. 55 and 54). Fraction I obtained by chromatography of culture filtrates A-E on Bio- Gel P-1SO contained all but one of the precipitinogens found in the corresponding fraction from Bio-Gel P-lOO (Figs. 49, 50, 51, 52, and 55). From one-quarter to less than one—half of the precipitinogens in the unfractionated culture filtrates were detected in the corresponding fraction I from Bio-Gel P-ZOO (Table 20). Culture filtrate A and ultrasonic extracts (USE) from two-month—old cells of g, boyi§_were antigenically indis- tinguishable when analyzed with antiserum specific for CF or USE. Several other filtrates from three month old cultures of g, boyig reacted poorly or not at all with antiserum A. Antiserum against any of these culture filtrates formed detectable lines with only two or three antigens in CF-B. The number of precipitinogens detected in the six culture filtrates with antiserum B was as follows: 10 each in CF-A, CF-B, CF-C and CF-D; and 6 each in CF-E and CF-F (Table 21). Immunoelectrophoresis of culture filtrates. The number of antigens found in the six culture filtrates by immunoelectrophoresis was as follows: 19 each in CF-A and 127 .Amv m__oz toucou ecu c_ boom—a who: mtom_ucm och .AmV oo~um vcm .on om_ua .Auv oo_um m_ouno_m one any mNIu xovmcaom co >camtmoumsotnu >n voc_mubo _ co_uumtu vcm Aon EsmLouomnooxz mo co_m:mm~v o_n:ou >co_toucuao .m: beam.“ 128 .Amv m__oz Leucou ozu c_ boom—a who: mtomhucm 03h .Amv oouum tam .on om_ua .Auv oo_um m_oo-o_m ucm .Amv mwuu xevmcaom co >3am3m0umeot3u >3 uoc_mu3o _ :o_uumbu vcm Am o_3mhv m mambo—_u beau—so m_>03 EsmLouum3ouxfi wo co_m:mm_v o_3:ou >co_toucu:o .3 2:2... ‘129 .Auv m__oz Loucou o3u c_ boom—a otoz mtom_ucm 03h .Amv oowua new .Aav om_um .Auv oo_um m_oouomm vcm Amy mNIu xoum3aom co >3amtm0umeotcu >3 voc_mu3o _ co_uumuu vcm Am u_3mhv u manta—_L menu—so mu>03 e:_tabumaouxz co co_maee_e o_a=oe >eo_tobeu=o ..m ut=m_u 150 .Auv m__oz Loucoo 033 :_ boom—a otmz mtom_u:m 03h .Auv oownm vcm .onom_ua .Auv oo_u¢ m_mouo_m 3cm Amy mule xoum3aom co >3am3m0umeoLro >3 voc_mu3o _ :o_3umtu vcm Am 0.3MHV a mambo—_w visa—so m_>03 E:_Louum3ooxz mo co_m:mm_v 0.330b >co_tou3u:o .Nm obam_m 131. . ye .Amv m__o3 toucoo oru c_ boom—a otoz meom_u:m 03h .Amv oomua ucm .Aav om_ua .Auv oo_na m_oouo_m one “my mama xoom3aom co >3amtm0umeotcu >3 om:_mu3o . co_uumLm vcm A03 E:_Louom3oo>z mo co_m:mm_v o_3:ov >co_uoucu:o .mm ou:m_u 152 .__03 Loucoo ecu c_ boom—a mm: Eatom.ucm och .AuV oo_na _ouuo.m 3cm Amy mNno xonmzaom co >3am3m0umeotzo >3 ouc.mu3o _ co_uomcm 3cm .A03 E:_Louom3oo>: mo :o_m:mm_v o_3:ov >co_cou3oao V .:m ousmmm 1135 voumrumcoeov m:mm0c_u_a_uotm I m : mountu__m menu—so _m:v_>_vc_ o3u mo mEmtmOc:EE_ c_ wouuoumv moumuma_uota0caee_ $0 to3532 u 3 mum33m3 c_ woumu__o m Eatom_uc< m. m u— o v u 0 ..— m m m o v u m m a _ 3 m c o v u 3 m o. a a . 3 m w o v u 3 m o. u _. _ 3 m e o u u a m e. m H _ a m m o v u 3 m o- < ..m Eatmm_3cm 33.3 co_m:wm_u o_3:ov >co_toucu:o >3 moumtu_mm ot:u_:o m_>03 s:_teuum3ou>aix_m 039 c. toumtumcosov mcom0c~u_a_uoua ..N 0.3mh 154 CF-B; 21 in CF-C; 16 in CF-D; 10 in CF-E; and 8 in CF-F (Table 18). Anodic and cathodic antigens were found in all of the immunoelectrophorograms (Figs. 55, 56, 57, 58, 59, and 60). Many of the lines were very lightly stained with the protein-Specific stain and were difficult to see. None of the lines were stained by lipid or polysaccharide—specific stains. The individual lines in each immunoelectrophorogram differed slightly in their lateral and migrational displace— ment. No attempt was made to identify or locate individual antigens by this method. 1555 .Am o_3mhv < mecca—_w menu—3o mm>03 Esmtouom3oU>z mo Emtmot03aotuoo_00cnee_ .mm al=m_l 1565 .Am o_3whv m mambo—_m mesa—ac m_>03 E:_Louom3ou>z mo EmtmoL03aotuoo_o03:EE_ lllllr. (Ill: Aw ill-I‘ll Iii... . < .mm or:m_a :13] .Am 0.3mkv u mambo—_m menu—so m_>03 E:_Louum3oo>2 mo EmLmoL03aotuuo_00c:EE_ .R 3:3... 15E! .Am o_3mhv a manta—_m otau_:o m~>03 E:_Luuum3oo>z mo EmLmoL03aoLuuo_oOc:EE_ .mm ot=m_e 13$? .Am o_3mkv m mambo—_m menu—:0 m_>03 antouum3oo>z mo EmumotocaoLuoo_oo::EE_ .mm 0.33... 140 .Am o_3mkv m ebony—_m only—:0 m_>03 E:_touom3oo>z mo EmLmoLOLQOLuoo_oo::EE_ .oe at=m_i DISCUSSION Considerable advancements have been made in the past decade in understanding the nature of specific antibodies. Rabbits produce at least two classes of immunoglobulins when inoculated with a variety of antigens (8:9;173). The primary response consists of the sequential production of these two immunoglobulins. The first antibodies to be pro- duced (lg-M) are 195 gammal—globulins which are sensitive to reductive cleavage by 2-mercaptoethanol (ME). This is followed, after several days, by the production of lower molecular weight, 78 gammag-globulins (1g-G) which retain their serologic activity after treatment with ME. Following secondary antigenic stimulation, much higher titers of lg—G are produced shortly after inoculation (9). The temporal sequence and extent of 1g-M production is similar to the primary response. The sequential production of MES and MER hemagglutinins and bacterial agglutinins was observed in rabbits inoculated with soluble and cell-bound antigens of M, boyig. The rela— tive amount and temporal sequence of production of each type of antibody varied among the experimental groups. Although none of the pre-inoculation sera were positive when tested by both serologic tests, it is possible that some of the 141 1.— 142 rabbits used in this investigation had previously been sensi- tized to mycobacterial antigens. Primary sensitization can result from repeated exposure to ubiqutous saprophytic myco- bacteria. Certain antigens are shared by many, if not all mycobacteria and cross reacting antigens can be obtained from apparently diverse sources. Depending on the dosage and frequency of antigenic stimulation, antibody may or may not be elicited. Low doses of antigen elicit only a small amount of 1g-M and no lasting immunologic memory results (175). When larger doses of anti- gen are injected, both 1g-M and 1g-G are produced and a typical anamnestic response can be demonstrated upon rein- jection of the antigen. Thus, some of the sera may have con- tained undetectable amounts of 1g-M or 1g-G prior to inocu- lation. It is also possible that detectable amounts of antibody were present in some of the sera but that neither OT-sensitized red blood cells nor the intact bacterial cells contained homologous antigen(s). This is unlikely, however, because of the antigenic complexity of OT and the frequency of cross agglutination of different mycobacteria (67). It is common practice to tuberculin test experimental animals prior to immunization or collection of sera or cells for ig vitro studies. Tuberculin tests cause an anamnestic antibody response in tuberculous humans (152), cattle and cattle and swine (97). Antibody can also be elicited in normal 3*- 145 tuberculin-negative animals by tuberculin testing. There was a sharp increase in the production of MES hemagglutinins elicited by tuberculin in rabbits sensitized with M, bovis (40). To avoid the possibility of sensitization or the elicitation of an anamestic response in previous sensitized rabbits, the rabbits used in the study were not tuberculin tested prior to use. They were obtained from a colony in which other rabbits are and remain tuberculin—negative. Antibody in the sera of immunized rabbits was detected by bacterial agglutination and the Middlebrook-Dubos passive hemagglutation test. The sequential production of MES and MER antibody was detected by both serologic tests. The mean hemagglutinin titers of all of the groups were consistently higher at each week than the mean bacterial agglutinin titers. These differences can be anticipated due to the greater sensitivity of the passive hemagglutination test, but the differences do not infer that the same antigens are measured. It is most likely that the tests measure different antigen- antibody systems. Culture filtrates contain polysaccharides believed to originate from cell wall lipopolysaccharides dur- ing autolysis (14), and normal sheep red blood cells selec— tively absorb tuberculopolysaccharides. To absorb proteins cells must first be treated, as for example with tannic acid. It is also possible that tuberculopolysaccharides in the medium in which the cells grew absorb to the surface of the mycobacterial cells prior to BPL-inactivation. This is not 144 likely since the cells were treated with various solutions, temperatures and mechanical treatment during inactivation, dispersing and washing. The suspension contains some cell debris but many intact cells were observed when stained with fluorscein—conjugated antibody. The apparent sequential production of MES and MER anti- body may be due to the different sensitivity of the agglutinin test for the two types of immunoglobulins (174). If this is true, much more MER antibody than MES antibody may have to be produced in order to reach detectable concentrations. Further studies will have to be conducted using other serologic tests to determine whether the production of lg-M and 1g-G is simultaneous or sequential. Multiple injections of CF in incomplete Freund's adjuvant (Group I) elicited higher titers of MES bacterial agglutinins for the first two weeks post—inoculation than culture filtrate and heat-killed M, boyi§_in incomplete adjuvant. This may be due to the differences in the relative amounts of effective agglutinogens in cells and CF. The agglutinogens in CF are soluble and readily accessible to antibody producing cells. The same antigens in cells must be liberated prior to process- ing by antibody producing cells. It is also possible that bacterial agglutinins were Specifically sequestered by intact cells in the inocula i vivo. This does not seem probable. The rabbits which received CF with BPL-killed M, bovis (Group III) produced antibody from two to three weeks sooner 145 than rabbits which received CF and heat-killed M, bovis (Group II). It is not known whether these differences are due to variations in the antigen dosage. Although the cells were repeatedly shaken prior to preparation of the inocula, many of the cells were aggregated. It is likely that all of the rabbits did not receive exactly the same amount of cells. It is unlikely that the small difference in the concentra- tions would account for the differences. These results may also be attributed to differences in the nature or extent of antigen inactivation by heat and BPL. Perhaps heat-killed M, bovis contains less effective antigen than BPL-killed M, bovis or this may reflect only at BPL-inactivated cells used to test agglutinins. It is important to realize however, that the serologic tests used probably measure several dif- ferent antigen-antibody systems. Consequently, neither test is likely to reveal meaningful information about the alter- ation of antigenic specificity in complex mixtures of antigens. The magnitude and duration of MES hemagglutinin produc- tion was lower in rabbits inoculated with inactivated cells plus CF (Groups II and III) than in rabbits which received only CF (Group I) or inactivated M, bovis. Moreover, earlier and higher titers of MER hemagglutinins were produced by the rabbits in Groups II and III. Sequential injection of CF plus BPL-killed cells (Group IV) prolonged the production of MES and delayed the onset of MER hemagglutinin production when compared to multiple injections of the same inoculum 146 (Group III). Rabbits given six simultaneous subcutaneous injections of CF without adjuvant (Group V) produced only MES hemagglutinins until six weeks post-inoculation. A similar pattern of bacterial agglutinins production was observed. An initial intravenous injection of CF followed by multiple injections of CF with BPL-killed M,bovis (Group VIII) extended the duration of MES hemagglutinin production longer than rabbits which did not receive initial intravenous injection (Group III). More and earlier MER hemagglutinins were produced by rabbits in Group III. These results may be best understood by relating the mode and route of injection to the localized dose or concentration of the antigen. The interval between the appearance of MES antibody and the first appearance of MER antibody could be shortened by concentration of the antigen or localizing the dose, that is, injecting in an area and/or with a substance so that the antigen is retained longer at the site of injection. 0n the other hand, when the localized antigenic stimulus was de- creased by not using adjuvant, systemic injections, or re— peated small injections at weekly intervals, MES antibody production was prolonged and the initiation of MER antibody synthesis delayed. How can this be rationalized with the theory that lg-G acts as an inhibitor of lg-M? Is the pro- longed production of 1g-M only an indirect effect and could it be more correctly stated that intravenous injection sup- pressed 1g-G production, or at least, increase the time 147 required for localized antigen to elicit 1g-G antibodies? The addition of heat or BPL-killed.M, bovis (Groups II and III) increased both the localized dose and concentration of antigen in contrast to the inoculation of CF or cells alone. Similarly, rabbits which received multiple inocu- lations of CF plus killed M, bovis received a greater initial antigenic stimulus. Although rabbits in Groups III and VII received the same amount of antigen, the localized dose of antigen was less in these rabbits which received an intra- venous inoculation (Group VII). The longest interval between the appearance of MES and MER antibody (7 weeks) occurred in rabbits given CF with inactivated M, bovis without adjuvant. None of the rabbits in the group had palpable granulomae or produced detectable precipitins. Importance of adjuvant most likely lies in the retention of antigen in the tissue and the associated inflammatory reSponse which thereby potentiates the production of antibody. The preferential evocation of MES antibody by the intra— venous inoculation of particulate antigens is well-known (171). Repeated intravenous injections of CF plus BPL-killed M, bovis elicited high titers of MES antibody that persisted throughout the duration of the experiment. The production of MES antibodies was also potentiated by an initial intra- venous injection of CF (Group VII and VIII). It was of considerable importance to determine the best way of eliciting large amounts of antibodies Specific for 148 CF antigens. Aluminum hydroxide-precipitated CF elicited a much better precipitin reSponse than either killed M, bovis or inactivated cells plus CF incorporated in Freund's ad- juvant. These results are not comparable, however, due to the differences in the antigen dose and route of inoculation. Washing the alum hydroxide-precipitate did not noticeably reduce the efficacy of the precipitate to elicit precipitins. The biopolymers found in mycobacterial culture filtrates are derived from bacterial cells primarily as a result of autolysis. Materials may also be liberated into the culture fluid during cell growth or may be extracted from the cell envelopes. Autolysis is a complex chemical process involving a state of dysequilibrium between cell wall biosynthesis and catabolism. Autolysis of some cells occurs during every phase of the growth cycle of most bacteria (80). However, more occurs at the end of the stationary phase when nutrient exhaustion or the accumulation of toxic substances halts the growth of the culture. During autolysis, cytoplasmic materials are released from cells as the result of enzymatic attack on cell walls. This process occurs gradually so that in the case of mycobacteria cultures, considerable time may be required before the culture media contains suitable con- centrations of antigens. Thus, there is ample time for considerable denaturation and degradation of the macromole- cules by protolytic enzymes, peptidases, carbohydrases, and nucleases. The moderate temperature of incubation (55C) 149 and aqueous environment facilitate enzymatic hydrolysis of macromolecules (61). All of the culture filtrates of M, bovis contained pro— teins, polysaccharides, and nucleic acids in various states of polymerization and denaturation. The relative concen- trations of these macromolecules varied considerably in the culture filtrates. The chemical and antigenic composition of the filtrates varied with the length of incubation of the culture. There was no correlation between the age of the culture and the protein, polysaccharide or nucleic acid concentration. However, CF-B and CF-C contained the highest concentrations of these constituents. It is possible that this is in part due to variations in the amount of growth that occurred in the cultures. No actual measurements of the number of cells in the individual cultures were made, so it cannot be stated with certainty that this is not the case. However, a thin confluent layer of growth formed on the surface of the culture media within one month after inocu— lation and large lots of filtrates were pooled. Moreover, the filtrates were all prepared in the same manner. The constituents in the culture filtrates were readily separated into three or more fractions by molecular exclusion column chromatography on Sephadex G-25. In this technique, molecules are separated solely on the basis of molecular size. The approximate molecular weight range of a solute can be estimated by relating the volume of eluent required to 150 elute the solute to the void volume of the column. Molecules that have Rf values of 1.0 are eluted in the column void volume and have molecular weights equal to or greater than the exclusion value of the gel. Thus, molecular exclusion chromatography is a rapid and convenient means of detecting changes in the state of polymerization of the three major classes of biomacromolecules in culture filtrates of differ- ent ages. The amount of protein in the chromatographic eluants was measured chemically by the Folin method and spectrophoto- metrically at 280 mu. Although the results of both tests were in fairly good agreement for fraction I of culture filtrates A through D, they were considerably different for fraction I of culture filtrates E and F. The reasons for these discrepancies is unknown. Both test measure the amino acids tyrosine and tryptophane. Although nucleic acids have a maximal absorbancy at 260 mu, they also absorb 280 mu ultraviolet light. The partial absorption of 280 and 260 mu ultraviolet light by nucleic acids and proteins accounts for the maximal absorbancy at 270 mu in the ultraviolet absorp- tion Spectra of the culture filtrates. This may also account for the discrepancies in the Spectrophotometric determina— tions of protein in the chromatographic eluants. The percent distribution of proteins of molecular weight 5,000, 100,000 and 200,000 or greater varied with the age of the culture filtrate. From 80% to 90% of the proteins in 151 culture filtrates A through E had molecular weights equal to or greater than 5,000. Only 50% to 60% of the proteins in the same culture filtrates were greater than 100,000 molecular weight. Sixty percent of the proteins in CF-F had molecular weights greater than 5,000; 25.2% had molecular weights greater than 100,000. From 55% to 45% of the proteins in culture filtrates A, B, and C had molecular weights greater than 200,000. Only 26% of the proteins in CF-E were greater than 200,000 molecular weight. None of the proteins in CF-F had molecular weights in this range. Thus, considerable degradation of the large mycobacterial proteins occurs in the culture fluid during incubation. Most, if not all, bacteria contain peptidases that are released into the culture medium during autolysis (150). These enzymes hydrolyze small proteins to low molecular weight polypeptides and amino acids. The Folin reacting materials in fraction II obtained by chromatography of the culture fil- trates on Sephadex G-25 had molecular weights considerably less than 5,000 (Rf = 0.56 to 0.40). They are probably poly- peptides. It is difficult to determine whether these poly- peptides represent products of degradation of larger proteins or arise from polypeptide pools within viable cells. Twice as much polypeptide was found in fraction II from CF-F as in the same fraction from CF-A. It is most likely that the majority of these polypeptides were derived from larger proteins by enzymatic hydrolysis. 152 The presence of low molecular weight polypeptides in fraction III is further indication of degradation of larger proteins. Polypeptides were not detected in fraction III from culture filtrates A, B, or C. However, increasing amounts were found in this fraction obtained from culture filtrates D, E, and F. Most of the chemical tests for the measurement of pro- tein are based on the determination of aromatic amino acids, nitrogen, the peptide bond, or absorption of 280 mu ultra- violet light. None of these test, however, give any indication of the size or molecular weight of the proteins being measured. The dividing line between the size requirement for proteins and polypeptides has been arbitrarily set at 10,000. This is also near the lower limit for antigenicity of proteins, although some lower molecular weight "proteins" are antigenic (16). These factors must be considered when mixtures such as CF are used to elicit antibody. This is exemplified by the fact that only 60% of the Folin-reacting material in CF-F had molecular weights of 5,000 or greater. The presence of 280 mu and 260 mu-absorbing material in the culture filtrates is indicative of autolysis. The per- cent distribution of these materials in Chromatograms of the Six culture filtrates varied depending on the age of the culture. The chromatogram of CF-A contained very little fraction III. However, approximately one-third or more of the total 280 mu—absorbing materials in culture filtrates 155 D, E, and F were found in this fraction. Increasing amounts of 280 mu-absorbing material were found in fraction I of culture filtrates A, B, and C. Thereafter, the relative amount of 280 mu-absorbing material in fraction I of culture- filtrates D, E, and F decreased. Moreover, fraction I in the older culture filtrates (D, E, and F) was split into several subfractions. These changes reflect the greater amount of autolysis and degradation of macromolecules that occurred in the older cultures. Further indication of the extent of autolysis and de- gradation that occurred in the cultures was provided by the distribution of nucleic acids in the three fractions. Most of the nucleic acids in the culture filtrates were found in fraction II. Fraction I of CF-C and CF-D contained greater amounts of nucleic acid than the same peak from any of the other filtrates. This suggests that considerable autolysis occurred at this time. Culture filtrate C also contained high concentration of proteins and polysaccharides. The culture filtrates contained variable amounts of dialyzable and nondialyzable materials. The dialysate con- tained only trace amounts of fraction I but most of fractions II and III found in the culture filtrates. This provides further evidence for the low molecular weight of these ma- terials. The dialysate elicited immediate—type skin reactions in sensitized rabbits but no precipitinogens were detected by Ouchterlony double diffusion or immunoelectrophoresis. 154 This result differs from that reported by Chaparas and Baer (27). They found several precipitinogens in the di- alyzable fraction from BCG culture filtrates. Only trace amounts of several protein components were seen in disc electrophorograms of the dialysates. The nondialyzable fractions contained all of the precipitinogens and protein components found in the homologous culture filtrate. These results indicate that dialysis does not appreciably reduce the number of antigens in CF. This is a good way of remov- ing low molecular weight materials from CF that interfere with protein determinations. Moreover, the dialysate con- tains very little protein but may be a good source of sensitin materials. The proteins in culture filtrates A, B, and C were not well separated by cellulose acetate membrane electrophoresis. Considerable trailing occurred and only three to four bands were discernable. However, from five to six bands were discernable in electrophorograms of purified fraction I from the same culture filtrates. The bands were fairly well separated. There were no qualitative differences in the pro— tein patterns obtained. The improved separation and resolu- tion obtained with these fractions is most likely due to the absence of low molecular weight materials. This may explain the failure of other investigators to obtain satisfactory separation of the proteins in culture filtrates by this method (125). Elution of these bands from large cellulose 155 acetate strips may be useful as a primary step in the iso- lation of mycobacterial antigens. Disc electrophoresis was a very effective means of separating the components in the six culture filtrates. Moreover, the technique is simple, rapid, and the patterns obtained are highly reproducible. The number of amido black- stained (protein) and PAS-stained (polysaccharide or glyco- protein) components detected in disc electrophorograms varied with the length of the incubation of the culture. The number of protein bands increased from 16 in CF-A to 50 in CF-C. Thereafter the number decreased; only 15 protein bands were detected in CF-F. All of the filtrates contained five protein bands with Similar or identical electrophoretic mobility. Seven other bands with the same Rf value were found in three or more of the culture filtrates. Culture filtrate C contained all but two of the protein bands found in the other filtrates. It appears, therefore, that new components did not appear in the culture medium after the fourth month of incubation. Rather as the length of incuba- tion continued, there was a progressive degradation of the proteins already present. The number of polysaccharide or glycoprotein components in the culture filtrates decreased with longer periods of incubation. The disc electrophorograms of all the filtrates had two polysaccharide bands in the Spacer gel. There was no consistent pattern in the presence or absence of these glyc0proteins in culture filtrates of different ages. 156 All of the protein and polysaccharide components de- tected in the disc electrophorograms had molecular weights of 5,000 or greater. This is indicated by the fact that the disc electrophorograms of fraction I from Sephadex G-25 contained all the bands seen in the corresponding CF. No bands were seen in disc electrOphorograms of fractions II or III. This affords further evidence for the non-protein nature (polypeptide) of the Folin-reacting material in these fractions. Immunodiffusion is a very useful means of analyzing the precipitinogenic content of the culture filtrates. However, the results obtained by this technique vary greatly depending on the way the test is performed and the quality of the antiserum employed. Repeated inoculation of CF in incomplete Freund's adjuvant was not an effective means of eliciting anti-CF precipitins in rabbits. Prolonged immuni— zation in this manner may be more effective, however, the Specificity of the antisera would probably be decreased. Only six antigens were found in concentrated culture filtrates when analyzed with antisera elicited by this method. The addition of killed M, boyi§_to culture filtrates from the same species of mycobacterium did not improve the precipi- tinogenicity of the inocula. Immunograms of filtrates from three month cultures of M, ngig had only three lines when analyzed with antisera elicited in this manner. In the present investigation, only four lines were detected in 157 immunograms of filtrates of three month old cultures of M. bovis. Preliminary studies showed that multiple intramuscular injections of aluminum hydroxide-precipitated CF elicited good precipitating antisera in rabbits. From Six to eight lines were detected in immunograms of culture filtrates of M, bovis with these antisera. However, by carefully select- ing the optimal conditions for analyses, as many as 16 lines could be detected. Results of disc electrophoresis and molecular exclusion chromatography indicate that CF may con- tain as many as 50 different protein components with molecu— lar weight of 5,000 or greater. Most of the precipitinogens in CF have molecular weights greater than 150,000. It is likely that most if not all, of these proteins may elicit precipitins under suitable conditions. From 19 to 21 lines were detected in immunoelectrOphorograms of culture filtrates A, B, and C. These results point out the antigenic and chemical complexity of mycobacterial culture filtrates. It is clear that earlier investigations have detected relatively few of the antigens in CF. The methods on which reports of fewer numbers of antigen have been based should be seriously questioned. Immunodiffusion was an effective means of enumerating the precipitinogens in culture filtrates of different ages. There was a progressive reduction in the number of precipi- tinogens in culture filtrates of increasing age. Sixteen 158 lines were detected in immunograms of CF-A; immunograms of CF-F had only six lines. These results were confirmed by immunoelectrophoresis. Fewer precipitinogens were detected in all but one of the filtrates (CF-F) when analyzed with antiserum B than with antisera A. This is most likely due to the fact that younger filtrates contain antigens that are absent or present only in low concentrations in older filtrates. The detection of more antigens in culture filtrates D and E with antiserum A than with the homologous antiserum B is not so readily explained. Many (perhaps most) of the proteins present in the older culture filtrates (D, E, and F) have undergone some degree of denaturation due to the prolonged exposure to the relatively warm temperatures in an aqueous environment. Denaturation occurs by unfolding and/or refolding of the poly- peptide chains resulting in a new tertiary structure. The extent of denaturation and alteration of antigenic specificity that results depends on the extent of chain rearrangement that occurs. It is also probable that smaller proteins are eliminated from the rabbit more readily which reduces the relative or effective concentration of the protein. Denatured proteins elicit precipitins that may or may not react with the native protein. Therefore, it is possible that antiserum againSt the native proteins (antiserum A) will react with the denatured form whereas the converse will not occur. Some support for this interpretation is provided by results 159 of immunodiffusion analyses of different batches of M, bovis culture filtrates of the same age with homologous and hetero- logous antisera. Whereas strong reactions occurred with homologous syStemS (8 to 10 lines) only two or three lines were detected,with heterologous antisera. One would expect that if the proteins were in the native undenatured state (no change in antigenic Specificity) stronger cross reactions would occur. Antisera specific for ultrasonic extracts of M, bovis (40) reacted equally well with CF-A. The sonic extract presumably contained native antigens. Moreover CF-A and ultrasonic extracts from two month old M, bovis were indistinguishable when analyzed with antiserum A. These re- sults indicate that filtrates from two month old cultures contain antigens in a native or only Slightly denatured state. Continued incubation of the culture may lead to considerable denaturation of the proteins and the acquisition of new anti- genic Specificity. Immunoelectrophoresis was superior to Ouchterlony double diffusion for the enumeration of the precipitinogens in the culture filtrates. In every case, more precipitinogens were detected by immunoelectrophoresis than by the Ouchterlony method. The number of lines detected in immunoelectrophoro- grams were from 21 in CF-C to 8 CF-F. The possibility of superimposition (masking) of precipitates after immunoelectro- phoresis is considerably less than after Ouchterlony double diffusion since the antigens are first separated by electro- phoresis. Two or more antigens can form superimposed 160 precipitates only if their electrophoretic mobility, dif- fusion coefficient, and reactant ratios are nearly identical. The probability of the simultaneous occurrence of all of these factors is very low. Despite the demonstrated sensi- tivity of this technique, several antigens in each CF were undoubtedly not detected because of masking of immunoprecipi- tates. A great need exists for purified Specific antigens and sensitins from mycobacteria that could be used for the diag- nosis of tuberculosis and identification of the causative agent. Although much progress has been made, most investiga- tions have failed to obtain immunochemically pure prepara- tions. Moreover, almost nothing is known about the role of individual mycobacterial antigens in tuberculoimmunity or hypersensitivity. It is obvious that purified antigens must be used in order to draw meaningful conclusions regarding their respective role in tuberculosis. Disc electrophoresis hold much promise as a means of isolating individual antigens (151). However, the application of this technique is limited to proteins that have dissimilar electrophoretic mobility. Unfortunately, due to the great number of CF proteins with Similar mobilities, the usefulness of this technique is limited to only a few proteins. A pre- liminary procedure is needed to separate the proteins in CF based on properties other than electric charge. Proteins with dissimilar mobility could then be readily isolated by 161 preparative disc electrophoresis and studied in experimen- tally infected animals. The results of molecular exclusion chromatography on Bio Gel P~200 suggest that this technique may be useful in this regard. Fraction I from CF-A contained only 6 of the 16 precipitinogens detected in the unfraction- ated CF. Moreover all of these antigens have molecular weights of 200,000 or greater and are free of lower molecular weight constituents. The technique is rapid, easily per- formed, and can be used on a preparative scale. S UMMARY Antibodies elicited in rabbits by concentrated unheated culture filtrates and heat or betaprone-killed cells of M, bovis were measured. Bacterial agglutinins and passive hemagglutinins (tuberculopolysaccharide specific) were ti- trated before and after treatment with 2-mercaptoethanol. The variance among the mean weekly antibody titers were analyzed statistically. Mercaptoethanol sensitive (MES) and mercaptoethanol resistant (MER) antibodies were produced sequentially but relative amounts and temporal sequences were altered by the state of the antigen and the inoculation Schedule. Injecting the total amount of antigen initially shortened the interval after the production of MES antibody and the production of MER antibody. Repeated small inocula— tions at weekly intervals, no adjuvant, or intravenous inocu- lations of antigens prolonged MES antibody production and delayed production of MER antibody. The constituents in unheated concentrated culture fil- trates from 2, 5, 4, 5, 6, and 7 month-old-cultures of M, bovis were readily separated into 5 or more fractions by molecular exclusion chromatography on Sephadex G-25. The first fraction contained all of the protein and precipitino- gens, and most of the polysaccharides. Fractions II and III contained predominately polynucleotides, polypeptides, and 162 165 oligosaccharides. The percent distribution of proteins of molecular weight of 5,000, 100,000, and 200,000 or greater varied with the length of incubation of the culture. The older culture filtrates contained less high molecular weight protein than the younger culture filtrates. None of the precipitinogens in culture filtrate were dialyzable. The components in the dialysate were largely materials from fractions II and III, only trace amounts of fraction I were found. By disc electrophoresis, between 15 and 50 amido black-staining components and five to nine PAS-staining components in the culture filtrates were separated. All of the amido black-staining components had molecular weights greater than 5,000. The number of com- ponents detected varied with the length of the incubation of the culture. There were 16 protein bands in CF-A, 50 in CF-C, and 15 bands in CF-F. Sixteen immunoprecipitates were detected in immunograms of CF-A; Six immunoprecipitates in CF-F. In every case, more precipitinogens were detected in the culture filtrates by immunoelectrophoresis than by the Ouchterlony method. The number varied from 21 in CF-C to 8 in CF—F by immuno- electrophoresis. 6. L I TERATURE C ITED Affronti, L. F., R. C. Parlett, and R. A. Cornesry. 1965. Electrophoresis on polyacrylamide gels of protein and polysaccharide fractions from Mycobacterium tuberculosis. Am. Rev. Resp. Dis. 91:1-5. Alshabkhoun, A., P. T. Chapman, M. F. White, and A. DeGroat. 1960. A study of the double-diffusion gel precipitation test in tuberculous patients. With Special reference to technical problems. Am. Rev. Resp. Dis. 81:704-708. Altemeier, W. A., J. B. Robbins, and R. T. Smith. 1966. Quantitative studies of the immunoglobulin sequence in the response of the rabbit to a somatic antigen. J. Exptl. Med. 124,445-460. Aoki, Y., K. Nagai, and T. Watanoke. 1961. Preparation of a highly potent hemagglutinating antigen of Mycobacterium tuberculosis. Am. Rev. ReSp. Dis. 84,907-910. . Azuma, I., H. Kimura, T. Nunaka, and Y. Yamamura. 1967. Isolation and purification of anaphylactically active polysaccharides from human tubercle bacilli. J. Bact. 25:770-771. Baer, H., and S. D. Chaparas. 1965. The immunology and chemistry of tuberculin. I. The isolation of dialyzable and non-dialyzable tuberculin-active components from un- heated BCG culture filtrates. Am. Rev. Resp. Dis. 88:28-52. Baer, H. and S. D. Chaparas. 1964. Tuberculin reactivity of a carbohydrate component of unheated BCG culture filtrate. Science 146:245-247. Bauer, D. C., and A. B. Stavitsky. 1961. On the different molecular forms of antibody synthesized by rabbits during the early response to a Single injection of protein and cellular antigens. Proc. Natl. Acad. Sci. 41:1667-1670. Bauer, D. C., M. J. Mathies, and A. B. Stavitsky. 1965. Sequence of synthesis of gamma, macroglobulin and gammag globulin antibodies during primary and secondary responses to proteins, Salmonella antigens, and phage. J. Exptl. Med. 111,889-907. 164 10. 11. 12. 15. 14. 15. 16. 17. 18. 19. 20. 21. 165 Benedict, A. A., R. J. Brown, and R. Ayengar. 1962. Physical prOperties of antibody to bovine serum albumin as demonstrated by hemagglutination. J. Exptl. Med. 115:195-208. Benedict, A. A. 1965. Sensitivity of passive hema- gglutination assay of 7S and 198 antibody in primary rabbit antibovine serum albumin sera. Nature 206:1568- 1572. Berlin, B. B. 1965. The nature of antibodies in mice injected with influenza virus vaccine. Proc. Soc. Exptl. Biol. N. Y. 115:1015-1017. Bevilacqua, E. B. and J. R. McCarter. 1948. The pro- teins in unheated culture filtrates of human tubercle bacilli. I. Fractionation and determination of physical chemical properties. J. Exptl. Med. 815229-244. Birnbaum, S. E. and L. F. Affronti. 1967. Immunochemi- cal characterization of the polysaccharides of Myco- bacterium tuberculosis and Mycobacterium ransasii. Bact. Proceed. New York, pp. 101. Boyd, W. C. 1966. Fundamentals of Immunology. 4th ed. Interscience Publishers, New York. Borek, F., Y. Stupp, and M. Sela. 1967. Formation and isolation of rabbit antibody to a synthetic antigen of low molecular weight. J. Immunol. ‘§§:759-744. Boyden, S. V. 1951. The adsorption of porteins on erythrocytes treated with tannic acid and subsequent hemagglutination by antiprotein sera. J. Exptl. Med. 955107-120. Boyden, S. V. and E. Sorkin. 1955. A study of antigens active in the tannic acid hemagglutination test present in filtrates of cultures of Mycobacterium tuberculosis. J. Immunol. 15:15-27. Boyden, S. V. and E. Sorkin. 1956. Antigens of Myco- bacterium tuberculosis. Adv. Tuberc. Res. 1:17-51. Boyden, S. V. 1958. The immunological response of antigens of the tubercle bacillus. Some experimental aspects. Progr. allergy. 5:149-214. Bradley, S. G. and D. W. Watson. 1965. Production of neutralizing antibody by mice injected with actinophage. J. Immunol. 99:782-787. 22. 25. 24. 25. 26. 27. 28. 29. 50. 51. 52. 166 Burrel, R. G., M. Rheins, and J, M. Birkeland. 1956. Tuberculosis antibodies demonstrated by agar diffusion. II. Further characterization of these antibodies by their distribution in human tuberculosis sera. Am. Rev. Tuberc. 14:259—244. Burtin, P. and R. Kourilsky. 1959. Reserches immuno- logiques sur les antigenes et haptenes solubles du bacille tuberculeux. I. Denombrement et identification des constituants precipitont avec les immunoserums specifiques. Ann. Inst. Pasteur. _912148-170. Canalco. 1965. Canal Industrial Corporation, Bethesda, Maryland. Castelnuovo, G., M. E. Duncan, and O. Bellezza, 1964. Mycobacterial antigens. A study of antigens in bacterial extracts and in culture filtrates of different ages. Tubercle. 45:246-254. Chaparas, S. D. and H. Baer. 1964. The immunology and chemistry of tuberculin. II. Chromatography with Sephadex of the non-dialyzable tuberculin-active constituents of BCG culture filtrate. Am. Rev. Resp. Dis. 89:41-48. Chaparas, S. D. and 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. 99:87-96. Choucroun, N. 1947. Tubercle bacillus antigens. Biological properties of two polysaccharides itolated from paraffin oil extracts of dead tubercle bacilli. Am. Rev. Tuberc. 56,205-226. Choucroun, N. 1949. Precipitin test for carbohydrate antibodies in human tuberculosis. Am. Rev. Tuberc. .59:710-712. Cole, L. R. and V. R. Farrell. 1955. A method for coupl~ ing protein antigens to erythrocytes. I. Description of the method. J. Exptl. Med. 102:651-645. Cole, L. R., J. J. Matloff, and V. R. Farrell. 1955. A method for coupling protein antigens to erythrocytes. II. Use of the method in the diagnosis of tuberculosis. J. Exptl. Med. 192,647-655. Cornesky, R. A. 1967. The bentonite flocculation test: an acid in the diagnosis of tuberculosis. Medical Ab- stracts, National Tuberculosis Association, Pittsburg, pp. 22. 55. 54. 55. 56. 57. 58. 59. 40. 41. 42. 45. 44. 45. 167 Corper, H. J. and C. Clark. 1946. Autolytic tuberculin. Its properties and the significance of its mode of form- ation. Am. Rev. Tuberc. 54:401-412. Crowle, A. J. 1958. Immunizing constituents of the tubercle bacillus. Bact. Rev. 22:185-197. Crowle, A. J. 1961. Immunodiffusion. Academic Press, New York. Crowle, A. J. 1962. Tubercle bacillary extracts immuno- genic for mice. III. Chemical degradation studies on the immunogen extracted from tubercle bacilli by trypsin digestion. Tubercle 45:178-184. Daniel, T. M. 1966. Observations on the antibody response of rabbits to mycobacterial antigens. J. Immunol., 95:100-108. Daniel, T. M. 1967. Molecular characterization of the antibody response in rabbits to intravenously injected BCG. Am. Rev. Resp. Dis. 95:262-269. Daniel, T. M. and G. L. Brawn. 1967. Immunoglobulin response to tuberculopolysaccharide in patients with tuberculosis. Medical Abstracts, National Tuberculosis Association, Pittsburg, pp. 21. Dardas, G. F. 1967. Immunologic responses of rabbits to Mycobacterium bovis: Chemical, electrophoretic, and antigenic analyses of bacillary exttacts. Thesis, Michigan State University. Dardas, G. F., T. J. Dardas, and V. H. Mallmann. 1967. Immunologic responses evoked by mycobacterial components: antibody responses of rabbits to cells and culture fil- trates of Mycobacterium bovis. Am. Rev. Resp. Dis. In Press. Dekker, T. P. 1960. Paper chromatography and paper electrophoresis of tuberculoproteins. Nature. 188:1191- 1192. Desbordes, J., F. Percheron, M. Foyet, and J. Blamet. 1960. Contribution a la connaissance des prateines des mycobacterus. Ann. Inst. Pasteur. 98:55-50. Deutsch, H. F. and J. I. Morton. 1957. Dissociation of human serum macroglobulins. Science. 125:600-605. Dietz, T., K. Maxwell, and E. Rungon. 1966. Use of latex particles in serology of tuberculosis. Bact. Proceedings. Los Angeles, pp. 47. 46. 47. 48. 49. 50. 51. 52. 55. 54. 55. 56. 57. 58. 168 Dietz, T. M., M. J. Smith, and E. H. Runyon. 1967. Mycobacterial species characterized by electrophoretic protein analyses. Medical Abstracts, National Tuberculo- sis Association, Pittsburg, pp. 15. Dubos, R. J. 1964. Acquired immunity to tuberculosis. Am. Rev. Resp. Dis. 99:505-515. Duncan, D. B. 1955. Multiple range and multiple F-tests. Biometrics. 11,1-42. Edwards, P. Q. and L. B. Edwards. 1960. Story of the tuberculin test. From an epidemiologic vieWpoint. Am. Rev. Resp. Dis. 81:1-47. Elberg, S. S. 1960. Cellular immunity. Bact. Rev. 24:67-95. Elek, S. D. 1948. The recognition of toxigenic bacterial strains ig_vitro. Brit. Med. J. 1,495-497. Farr, R. S. and F. J. Dixon. 1960. The effect of antigen concentration on the initiation of detectable antibody synthesis in rabbits.' J. Immunol. 85:250-257. Freeman, M. J. and A. B. Stanitsky. 1965. Radio-immuno- electrophoretic study of rabbit anti-protein antibodies during the primary response. J. Immunol. 95:981-990. Froman, S., R. Burge, M. Gedebou, and M. J. Pickett. 1966. The merits of applying a battery of tests for the serology of tuberculosis. Medical Abstracts, National Tuberculosis Association, San Francisco, pp. 65. Glenchur, H. and L. J. Kettel. 1965. A study of the agar double-diffusion test in human tuberculosis. Am. Rev. Resp. Dis. 91:86-96. Glenchur, H., E. Fosseick, and M. Silverman. 1965. An immediate skin test for the diagnosis of active pulmonary tuberculosis. Am. Rev. Resp. Dis. 925741-748. Glenchur, H., E. Fossieck, and M. Silverman. 1966. A comparison of a purified protein derivative of some fractions of Mycobacterium tuberculosis. Am. Rev. ReSp. Dis. 95370-77. Greenbury, C. L., D. H. Moore, and L. A. C. Nunn. 1965. Reactions of 7S and 198 components of immune rabbit antisera with human group A and AB red blood cells. Immunol. 6:412-418. 59. 60. 61. 62. 65. 64. 65. 66. 67. 68. 69. 70. 169 Grey, H. M. 1964. Studies on changes in the quality of rabbit-bovine serum albumin antibodies following immunization. Immunol. .l:82-88. Gronwall, A. 1944. An attempt of physicochemical analysis of protein antigens isolated from the tubercle bacillus. Honorary volume for the Svedberg, 1944, pp. 540. Haurowitz, F. 1965. The Chemistry and Function of Proteins. Academic Press, New York. Heckly, R. J. and D. W. Watson. 1950. Extraction of proteins and other constituents from the tubercle bacillus, with some observations on their chemical and biological properties. An. Rev. Tuberc. 61:798-808. Heckly, R. J. and D. W. Watson. 1951. Extraction and fractionation of water-soluble components from tubercle bacilli. Am. Rev. Tuberc. 64:602-619. Heidelberger, M. and A. E. O. Menzel. 1954. Protein fractions of the human strain (H57) of the tubercle bacillus. J. Biol. Chem. 104:655. Hinson, K. F., A. R. Jones, and J. A. Chamberlin. 1952. The Middlebrook-Dubos reaction in relation to the clinical aspects of pulmonary tuberculosis. Brit. J. Tuberc. 46:60-68. Hirschfeld, J. 1960. Immunoelectrophoresis-proceedure and application to the study of group specific variations in sera. Scientific Tools. 1:18-25. Hobby, G. L., W. B. Redmond, E. H. Runyon, W. B. Schaffer, L. G. Wayne, and R. H. Wickelhausen. 1967. A study on plumonary disease associated with mycobacteria other than mycobacterium tuberculosis. Identification and character- ization of the mycobacteria. XVIII. A report of the Veterans Administration--Armed Forces COOperative Study. Am. Rev. Resp. Dis. 95:954-971. Hollander, A. G., M. F. Frobisher Jr., and K. Kolish. 1955. Clinical evaluation of the hemogglutination reaction. Am. Rev. Tuberc. [61:497-502. Humphrey, J. H. and R. G. White. 1965. Immunology for Students of Medicine. 2nd ed., F. A. Davis Company, Philadelphia. Ide, Y., M. Takanami, H. Nemoto, H. Yugi, and H. Halakeyama. 1961. Studies on tuberculin active substance. I. Prepara- tion of tuberculin active substance from tubercle bacilli bodies. Bull. Natl. Inst. Anim. Hlth. 42:91-97. 170 71. Ide, Y., M. Takanami, H. Nemoto, H. Yugi, and H. Halokeyama. 1961. Studies on tuberculin active substance. II. Bio- logical studies on fractions extracted from tubercle bacilli bodies. Bull. Natl. Inst. Anim. Hlth. 42:99-107. 72. Ide, Y., H. Yugi, and M. Takanami. 1961. Studies on tuberculin active substance. III. Fractionation of phos- phate buffer extracts of tubercle bacilli. Natl. Inst. Anim. Hlth. Quart. 1:155-141. 75. Ide, Y., H. Yugi, H. Halakeyama, and H. Nemoto. 1965. Effect of heat-treatment on a tuberculin active protein. Natl. Inst. Anim. Hlth. Quart. 5:44-48. 74. Inoue, T. 1957. Agar diffusion precipitation with tubercular antigens of somatic origin. Ann. Tuberc. 1565-76. 75. Jones, A. R. and K. F. W. Hinson. 1955. Observations on the serology of pulmonary tuberculosis. Am. Rev. Tuberc. §§;759-745. 76. Jones. M. A., A. B. Larsen, T. H. Vardaman, and L. A. Baisden. 1955. Studies on the fractionation of tubercu- lin and johnin. I. Fractionation of tuberculin proteins. Am. Rev. Tuberc. §§;425-458. 77. Kabat, E. A. and M. M. Mayer. 1961. Experimental Immuno— chemistry. 2nd ed., Charles C. Thomas Publishers, Springfield, Ill. 78. Knicker, W. T. and J. B. LaBorde. 1964. The separation of mycobacterial antigens by ion-exchange chromatography. Am. Rev. ReSp. Dis. 89,29-40. 79. Lagercrantz, R. 1956. Immunological studies of myco- bacteria, I. The ability of various bacteria and fungi to induce production of antibody against erythrocyte- adsorbable antigens from the tubercle bacillus. Acta Path. et Microbiol. Scand. .§§:416-424. 80. Lamanna, C. and M. F. Mallette. 1965. Basic Bacteriology. 5rd ed., Williams and Wilkins Co., Baltimore. 81. Landy, M., R. P. Sanderson, and A. L. Kackson. 1965. Humoral and cellular aspects of the immune response to the somatic antigens of Salmonella enteritidis. J. Exptl. Med. 122:485-504. 82. Larson, C. L., E. Ribi, W. C. Wicht, and R. List. 1961. Skin reactions produced in rabbits by cell walls and proto- plasm of Mycobacterium tuberculosis and Mycobacterium butyricum. Am. Rev. Resp. Dis. _§§:184-195. 85. 84. 85. 86. 87. 88. 89. 90. 91. 171 Larson, C., E. Ribi, W. Wicht, R. List, and G. Goode. 1965. Resistance to tuberculosis in mice immunized with BCG disrupted in oil. J. Bact. 198:1214-1215. Leskowitz, S. and B. H. Waksman. 1960. Studies in im- munization. The effect of route of injection of BSA in Freund's adjuvant on the production of circulating anti- body and delayed hypersensitivity. J. Immunol. 84:58-72. Lind, A. 1959. Serological studies of mycobacteria by means of the diffusion-in-gel technique. I. Preliminary investigations. Int. Arch. Allergy. 14:264-278. Lind, A. 1960. Serological studies of mycobacteria by means of the diffusion-in-gel technique. II. Investi- gations of the precipitins formed after injection with Mycobacterium tuberculosis and the relation to the anti- bodies demonstrated by the Middlebrook-Dubos hemagglu- tination test. Int. Arch. Allergy. 16:556-551. Lind, A. 1960. Serological studies of mycobacteria by means of the diffusion-in-gel technique. III. A difference in precipitinogenic content found in substrains of BCG Int. Arch. Allergy. .4131'9- Lind, A. 1960. Serological studies of mycobacteria by means of the diffusion-in-gel technique. IV. The pre- cipitinogenic relationships between different species of mycobacteria with Special reference to Mycobacterium tuberculgsis, M. phlie, M. smegmatis, and M. auium. Int. Arch. Allergy. 11:500-522. Lind, A. 1961. Serological studies of mycobacteria by means of the diffusion-in-gel technique. V. A compari- son between the plate and the tube techniques for the demonstration of precipitins in sera from tuberculous patients. Int. Arch. Allergy. .1§:505-517. Lind, A. 1961. Serological studies of mycobacteria by means of the diffusion-in-gel technique. VI. Physical, chemical, and immunological characteristics of precipi- tinogenic factors in an antigenic preparation from a culture filtrate of virulent strain of Mycobacterium tuberculosis. Int. Arch. Allergy. 1851-45. Lind, A. 1961. Serological studies of mycobacteria by means of diffusion-in-gel techniques. VII. Investi- gations of precipitinogens demonstrated in various preparations of mycobacterial origin. Int. Arch. Allergy. 18:1-15. 92. 95. 94. 95. 96. 97. 98. 99. 100. 101. 102. 172 Long, E. R. and F. B. Seibert. 1926. The chemical composition of the active principle of tuberculin. VII. The evidence that the active principle is a protein. Am. Rev. Tuberc. 88:448-455. Long, E. R., F. B. Seibert, and J. D. Aronson. 1955. A standard tuberculin (purified protein derivative) for uniformity in diagnosis and epidemiology. Tubercle 89:755-759. Long, E. R. 1958. The Chemistry and Chemotherapy of Tuberculosis. Williams and Wilkins Company, Baltimore. Long, K. R. and F. H. Top, 1964. The detection of anti- body against tubercle bacilli in the serum of tuberculin- positive cattle by an agar double diffusion precipitation technique. Am. Rev. Resp. Dis. 88:49-54. Lurie, M. B. 1964. Resistance to tuberculosis: Experi- mental Studies in Native and Acquired Defense Mechanisms. Harvard University Press, Cambridge, Mass. Mallmann, V. H., P. Robinson, and M. D. McGanin. 1964. An anamnestic-like reaction elicited by tuberculin. Am. J. Vet. Rech. 88:695-698. Magnusson, M. 1961. Specificity of mycobacterial sensitins. I. Studies in guinea pigs with purified tuberculin pre- pared from mammalian and avian tubercle bacilli. Am. Rev. Resp. Dis. 88:57-68. Magnusson, M. and M. W. Bentzon. 1961. Specificity of mycobacterial sensitins. II. Studies in guinea pigs with purified sensitins prepared from unclassified acid- fast bacilli. Am. Rev. ReSp. Dis. 88:69-84. McCarter, J. R. and D. W. Watson. 1942. The relationship of antigenicity, physical-chemical properties, and poly- saccharide content of tuberculins to their intracutaneous activity. J. Immunol. 88:85-98. McCarter, J. R. and E. B. Bevilacqua. 1948. The proteins in unheated culture filtrates of human tubercle bacilli. II. Determination of serological properties. J. Exptl. Med. 81:245-258. Middlebrook, G. A. and R. J. Dubos. 1948. Specific serum agglutination of erythrocytes sensitized with extracts of tubercle bacilli. J. Exptl. Med. 88:521-528. 105. 104. 105. 106. 107. 108. 109. 110. 111. 112. 115. 175 Middlebrook, G. A. 1950. A hemolytic modification of the hemagglutination test for antibodies against tubercle bacillus antigens. J. Clin. Invest. .88z1480-1486. Morisawa, S., A. Tonaka, K. Shokima and Y. Yamamura. 1960. Studies on tuberculin active peptide. I. The isolation, crystallization and prOperties of tuberculin active peptide from tubercle bacilli. Biochim. Biophys. Acta. 88:252-258. Morton, W. and H. J. Dodge. 1965. Modifications of the agar double-diffusion precipitation test for tuberculosis. Am. Rev. Resp. Dis. 88:264-266. Neter, E., H. Y. Whong, T. Suzuki, and E. A. Garzynski. 1964. Differences in antibody response of rabbits to intravenously injected soluble and cell-attached entero- bacterial antigens. Immunol. .l:657-664. Nossal, G. J. V., G. L. Ada, and C. M. Austin. 1964. Antigens in immunity. II. Immunogenic properties of flagella "polymerized flagellin" and flagellin in the primary response. Australian J. Exptl. Biol. Med. Sci. .48:285-295. Oconnor, J. E. and L. S. Snyder. 1967. Antibody pro- duction to Mycobacterium tuberculosis. Bacterial. Pro- ceed., Los Angeles, pp. 62. Okada, Y., S. Morisowa, K. Shajima, M. Kilayama, and Y. Yamamura. 1965. Improved method for the isolation of tuberculin active peptide. J. Biochem. 84:484-490. Onoue, K., N. Tanigaki, Y. Yagi, and D. Pressman. 1965. Ig-M and Ig-6 anti-hapten antibody; hemolytic, hemagglu- tinoling, and precipitating activity. Proc. Soc. Expt. Biol. (N. Y.) 888:540-545. Onyekwere, O. O. 1962. The tuberculoimmunogenicity for mice and guinea pigs of Mycobacterium bovis inacivated with beta-propiolactone compared to that of BCG, phenolin-activated cells and Youmans extract. Thesis, Michigan State University. Ornstein, L. and B. J. Davis. 1961. Disc electro- phoresis. Distillation Products Industries, Rochester, New York. Ouchterlony, O. 1948. $2 vitro method for testing the toxin-producing capacity of diphtheria. Acta. Pathol. Microbiol. Scand. 88:516-524. 114. 115. 116. 117. 118. 119. 120. 121. 122. 125. 125. 124. 174 Oudin, J. 1948. L'analyse immunochemique qualitative methode par diffusion des antigens ous sein de l'immuneserum precipitant gelose. Ann. Inst. Pasteur. 18:50-51. Owen, J. A. 1958. Paper electrOphoresiS of proteins and protein-bound substances in clinical investigations. Adv. in Clin. Chem. Igz257-500. Pangborn, M. C. and K. Birkhoug. 1954. A purified tuberculin fraction from heated cultures in testing BCG-vaccinated subjects. Am. Rev. Tuberc. 88:500-505. Parlett, R. C. and G. P. Youmans. 1956. Antigenic relationships between mycobacteria as determined by agar diffusion techniques. Am. Rev. Tuberc. 18:657-649. Parlett, R. C., G. P. Youmans, C. Rehr, and W. Lester. 1958. The detection of antibodies in the serum of tuberculous patients by an agar double diffusion precipi- tation technique. Am. Rev. Tuberc. 11:462-472. Parlett, R. C. and G. P. Youmans. 1959. Antigenic relationships between 98 strains of mycobacteria using gel diffusion precipitation techniques. Am. Rev. Tuberc. 11:450-461. Parlett, R. C. and C. A. Rehr. 1959. Further studies on gel diffusion tests in tuberculosis. I. Methods for standardization of the antigen and detection of small amounts of antibody. Am. Rev. Resp. Dis. .88: 886-894. Parlett, R. C. and Y. M. Chu. 1966. Observations on circulating antibody reSponseS to mycobacterial stimuli. Am. J. Clin. Pathol. 48:552-561. Pike, R. M. and M. L. Schultze. 1964. Production of 7S and 19S antibody to the somatic antigens of Salmonella typhosa in rabbits. Proc. Soc. Exptl. Biol. Med. 188: 829-855. Pike, R. M. 1967. Antibody heterogeneity and serologi- cal reactions. Bact. Rev. 88:157-174. Raffel, S. 1948. The components of the tubercle bacillus responsible for the delayed type of infectious allergy. J. Infect. Dis. 88:267-295. Rheins, M. S., R. G. Burrell, and J. M. Birkeland. 1956. Tuberculous antibodies demonstrated by agar-diffusion. I. Specificity and incidence of agar diffusing anti- bodies in rabbit sera. Am. Rev. Tuberc. .14:229-258. 125. 126. 127. 128. 129. 150. 151. 152. 155. 154. 155. 156. 175 Rhodes, J. M. and E. Sorkin. 1954. Paper electrophoretic separation of tuberculin constituents. Experimentia 10:427. Rhodes, J. M. 1961. Chromatography of tuberculoproteins on DEAE cellulose. Int. Arch. Allergy. 88:257-270. Rich, A. R. 1951. The Pathogenesis of Tuberculosis. Blackwell Scientific Publications, Oxford. Robinson, P. J. 1964. Tuberculopolysaccharide and tuberculophosphatide passive agglutination with bovine and porcine serums. Thesis, Michigan State University. Robbins, J. B., K. Kenney, and E. Suter. 1965. The iso- lation and biological activities of rabbit gamma M and gamma G anti-Salmonella typhemuriun antibody. J. Exptl. Med. 888:585-. Rogers, H. J. 1961. The dissimilation of high molecular weight substance in I. C. Gunsalus and R. Y. Stamer (ed.), The Bacteria, Vol. II. Academic Press, New York. Roszman, I. R. 1966. Disc electrophoretic and immuno— diffusion studies of mycobacterial culture filtrates. Thesis, Michigan State University. Schoenberg. M. D., J. C. Rupp, and R. D. Moore. 1964. Antibody response to mycobacterial antigens. Brit. J. Exptl. Pathol. 88:12-20. Seibert, F. B. and B. Munday. 1951. The chemical compo- sition of the active principle of tuberculin. XIV. Analysis of the colloidal components of tuberculin with Special reference to the relation of protein and carbo- hydrate. Am. Rev. Tuberc. .§§325‘4O° Seibert, F. B. and B. Munday. 1952. The chemical compo— sition of the active principle of tuberculin. XV. A precipitated purified protein suitable for the preparation of a standard tuberculin. Am. Rev. Tuberc. 88:724-757. Seibert, F. B. 1954. The isolation and prOperties of the purified protein derivative of tuberculin . Am. Rev. Tuberc. 88:715-720. Seibert, F. B., K. O. Pedersen, and A. Tiselius. 1958. Molecular weight, electrochemical, and biological proper- ties of tuberculin protein and polysaccharide molecules. J. Exptl. Med. .882415-458. 157. 158. 159. 140. 141. 142. 145. 144. 145. 146. 147. 148. 149. 150. 176 Seibert, F. B. 1940. Removal of the impurities, nucleic acid and polysaccharide from tuberculin protein. J. Biol. Chem. 155:592-601. Seibert, F. B 1941. The chemistry of the proteins of the acid-fast bacilli. Bact. Rev. 8:69-95. Seibert, F. B. and J. T. Glenn. 1941. Tuberculin purified protein derivative. Preparation and analysis of a large quantity for standardization. Am. Rev. Tuberc. 88:9-25. Seibert, F. B. and J. W. Nelson. 1945. Proteins of tuberculin. J. Am. Chem. Soc. 88:272-278. Seibert, F. B. 1944. The chemistry of tuberculin. Chem. Rev. 88:107-127. Seibert, F. B. 1949. Isolation of three different proteins and two polysaccharides from tuberculin by alcohol fractionation. Their chemical and biological properties. Am. Rev. Tuberc. _88:86-101. Seibert, F. B. 1950. Constituents of mycobacteria. Ann. Rev. Microbiol. .8:55-52. Seibert, F. B. 1950. Progress in the chemistry of tuberculin. Adv. Tuberc. Res. 8:1-29. Seibert, F. B., C. Crumb, and C. H. DuFour. 1951. Antigenic differences in two tuberculin protein frac- tions. J. Infect. Dis. .88:252-258. Seibert, F. B. and A. M. Fabrizio. 1952. The isolation and chemistry of a protein fraction of Mycobacterium tuberculos8§ and its ability to sensitize cells. Am. Rev. Tuberc. .§§=314'554- Seibert, F. B. 1955. Tuberculin proteins. Faraday Society Discussions, pp. 251-261. Seibert, F. B. and E. Sote-Figueroa. 1957. Studying tuberculin protein and polysaccharide antigens by gel diffusion techniques. Am. Rev. Tuberc. .18:601-607. Seibert, F. B. 1960. The interplay of an immune sub- stance with tuberculopolysaccharide and its antibody in tuberculosis. J. Inject. Dis. 105:52-60. Shetlar, M. R. and Y. F. Masters. 1957. Use of thymol sulfuric acid reaction for determination of carbohydrate in biological material. Anal. Chem. 88:402-410. 151. 152. 155. 154. 155. 156. 157. 158. 159. 160. 161. 177 Shulman, S., L. Hubler, and E. Wilebsky. 1964. Antibody response to immunization by different routes. Science. 145:815-817. Smith, D. T. and N. B. Scott. 1950. Clinical interpre- tation of the Middlebrook-Dubos hemogglutination test. Am. Rev. Tuberc. .88:121-127. Someya, S., O. Hayashi, and Y. Yamamura. 1962. Studies on the antigenicity of tuberculin-active peptide. Amf Rev. Resp. Dis. .88:542-546. Someya, S., K. Kayama, N. Asami, T. Kalaoka, and Y. Yama- mura. 1966. Studies on the tuberculin reactivity of the tuberculin-active peptide. Jap. J. Microbiol. 88: 221- 227. Spanberg, J. 1959. A clinical study of the hemagglu- intation reaction in tuberculosis. Acta Tuberc. Scand. 26:1-96. Stacey, M., P. W. Kent, and E. Nassau. 1951. Poly- saccharide complexes isolated from Mycobacterium tubercu- losis. Biochim. Biophys. Acta. 1:146—152. Stelos, P. 1958. Comparitive study of rabbit hemolysins to various antigens. I. Hemolysins to beef red blood cells. J. Infect. Dis. 102:105-115. Svehag, S. E. and B. Mandel. 1965. The formation and prOperties of polio virus-neutralizing antibody. I. 195 and 78 antibody formation: differences in kinetics and antigen dose requirement for induction. J. Exptl. Med. 888:1-19. Svehag, S. E. 1964. The formation and properties of polio virus-neutralizing antibody. II. 198 and 73 antibody formation: differences in antigen dose require— ment for sustained synthesis, anamnesis, and sensitivity to x-irradiation. J. Exptl. Med. 888:21-59. Svehag, S. E. 1964. The formation and properties of polio virus-neutralizing antibody. III. Sequential changes in electrophoretic mobility of 19S and 7S anti- bodies synthesized by rabbits after a single virus injection. J. Exptl. Med. 888:225-240. Takahaski, Y. and K. Ono. 1961. Study on the passive hemagglutination reaction by the phosphatide of Myco- bacterium tuberculosis. I. The reaction and its Specificity. Am. Rev. Resp. Dis. .882581-585. 162. 165. 164. 165. 166. 167. 168. 169. 170. 178 Takahashi, Y. and K. Ono. 1961. Study on the passive hemagglutination reaction by the phosphatide of Mycobacterium tuberculosis. II. Conditions of the reaction. Am. Rev. Resp. Dis. 88:586-595. Takahashi, Y., S. Fujita, and A. Sasaki. 1961. The Specificity of the passive hemagglutination methods used in the serology of tuberculosis. J. Exptl. Med. 888: 1141-1154. Takahashi, Y., T. Onadera, and K. Yamamoto. 1961. The behavior of three different kinds of antibodies in tuberculosis: antiprotein, antipolysaccharide, and anti- phosphatide. I. Experimental tuberculosis. J. Exptl. Med. 888:555-567. Takahashi, Y., K. Mochizuki, and Y. Nagayama. 1961. The behavior of three different kinds of antibodies in tuberculosis: antiprotein, antipolysaccharide, and anti- phosphatide. II. Human tuberculosis. J. Exptl. Med. 888:569-579. Takahashi, Y. 1962. Specific serum agglutination of koalin particles sensitized with tubercle bacillus phosphatide and its clinical evaluation as a serodiag- nostic test for tuberculosis. Am. Rev. Resp. Dis. .88:708-719. Thurston, J. R., M. S. Rheins, and T. Huziwara. 1956. Serologic investigations of mycobacterial antigens. I. The specificities of aqueous and saline extracts of BCG, Mygobacterium phlei, and H57RN. Am. Rev. Tuberc. 18:565-570. Thurston, J. R. and M. S. Rheins. 1956. Serological investigations of mycobacterial antigens. II. Specifici- ties of antigens obtained by successive aqueous extracts of BCG and Mycobacterium phlei. Am. Rev. Tuberc. 18:565-570. Thurston, J. R. and W. Steenken Jr. 1960. Comparison of gel precipitation and sensitized erythrocyte techniques for the detection of antibodies in the sera of tuberculous and non-tuberculous patients. Am. Rev. Resp. Dis. 88: 695-705. Wilson, G. S. and A. A. Miles. 1957. Topley and Wilson's Principles of Bacteriology and Immunity. 4th ed. Williams and Wilkins, Co., Baltimore. 179 171. Torrigiani, G. and I. M. Raitt. 1965. The enhancement of 195 antibody production by particulate antigen. J. Exptl. Med. 122:181-195. 172. Turcotte, R., S. O. Freedman, and A. H. Sehon. 1965. A new hemagglutination proceedure for the diagnosis of active tuberculosis. Am. Rev. Resp. Dis. 88:725-728. 175. Uhr, J. W. and M. S. Finkelstein. 1967. The kinetics of antibody formation. Progr. Allergy. 88:57-85. 174. Wei, M. M. and A. B. Stavitsky. 1967. Molecular forms of rabbit antibody synthesized during the primary response to human albumin. Immunol. “88:451-444. 175. weidanz, W. P., A. L. Jackson, and M. Landy. 1964. Some aspects of the antibody response of rabbits to immunization with enterobacterial somatic antigens. Prac. Soc. Exptl. Biol. Med. 888:852-857. 176. Weiss, D. W. and R. J. Dubos. 1955. Antituberculosis immunity induced in mice by vaccination with killed tubercle bacilli or with a soluble bacillary extract. J. Exptl. Med. 888:515-550. 177. Weiss, D. W. 1959. Vaccination against tuberculosis with nonliving vaccines. I. The historical background. Am. Rev. Resp. Dis. 88:540—558. 178. Williams, C. A. and R. J. Dubos. 1959. Studies on fractions of methanol extracts of tubercle bacilli. I. Fractions which increase resistance to infection. J. Exptl. Med. 888:981-1004. 179. Yamamura, Y., Y. Orada, S. Nagamatsu, and T. Imada. 1965. The isolation from tuberculin of two different polysaccharides having anaphylactic activity. Am. Rev. Resp. Dis. 88:859-845. 180. Youmans, G. P. and A. G. Karlson. 1947. Streptomycin sensitivity of tubercle bacilli. Studies on recently isolated tubercle bacilli and the development of re- sistance to streptomycin 8g_vivo. Am. Rev. Tuberc. .88:529-555. 181. Youmans, A. S., G. P. Youmans, and I. Millman. 1957. Immunogenicity of particles isolated from Mycobacterium tuberculosis. Proc. Soc. Exptl. Biol. N. Y. .88:762-768. 182. 185. 184. 185. 186. 187. 180 Youmans, A. S. and G. P. Youmans. 1964. Nature of the labile immunogenic substance in the particulate fraction isolated from Mycobacterium tuberculosis. J. Bact. 88:1050-1057. Yoneda, M. and Y. Fukui. 1961. Extracellular proteins of tubercle bacilli. I. Zone electrophoretic purifi- cation of the proteins of a virulent strain (H57RN) of Mygobacterium tuberculosis. Bilken J. 8:25-40. Yoneda, M. and Y. Fukui. 1961. Extracellular proteins of tubercle bacilli. II. Some serological properties of purified extracellular proteins of a virulent strain (H57RN) of Mycobacterium tuberculosis. Bilken J. 4:121-150. Yoneda, M. and Y. Fukui. 1961. Extracellular proteins of tubercle bacilli. III. Column chromatbgraphical purification of the proteins of a virulent strain (H27RN) of Mycobacterium tuberculosis. Bilken J. 8:187-196. Yoneda, M. and Y. Fukui. 1965. Extracellular proteins of tubercle bacilli. V. Distribution of alpha and beta antigens in various mycobacteria. Bilken J. 8:201-225. Yugi, H., H. Halakeyama, H. Nemato, and Y. Ide. 1962. Studies on tuberculin active substance. IV. Biological studies on phOSphate buffer extracts from tubercle bacilli. Natl. Inst. Anim. Hlth. Quart. 8:140-147.