- Email: [email protected]
Typing of Streptococcus pneumoniae: past, present, and future
Typing of Streptococcus pneumoniae: Past, Present, and Future Jørgen Henrichsen, MD
Early works leading to the detection of the pneumococcus and eventually to the appreciation that isolates differed in agglutination and that antisera differed in their capacity to protect against pneumococcal infection in the mouse protection test are reviewed. Studies by researchers from Europe, South Africa, and the United States over nearly five decades led to the introduction of serum therapy. Rapid typing methods thus became very important, and type-specific serum therapy generated a dramatic decrease in the number of deaths from pneumococcal pneumonia. Just before the introduction of sulfonamides and, shortly thereafter, penicillin, the use of horse sera was replaced by the use of rabbit sera for a number of reasons. The present methods of typing comprise the capsular reaction test, latexand coagglutination, and capillary precipitation, to name the most important; these use a large variety of antisera. Newer methods include the use of DNA probes and DNA sequence-based subtyping. Am J Med. 1999;107(1A):50S–54S. © 1999 by Excerpta Medica, Inc.
From the World Health Collaborating Center for Reference and Research on Pneumococci, Statens Serum Institut, Copenhagen, Denmark. Requests for reprints should be addressed to Jo¨rgen Henrichsen, MD, World Health Collaborating Center for Reference and Research on Pneumococci, Statens Serum Institut, Artillerivej 5, DK-2300 Copenhagen S, Denmark. 50S © 1999 by Excerpta Medica, Inc. All rights reserved.
T
he complete history of typing of Streptococcus pneumoniae seems never to have been told, although the monographs of Heffron1 and White,2 published in 1939 and 1938, respectively, and reprinted in 1979, cover it in detail until 1938. However, World War II disrupted communication among researchers on both sides of the Atlantic, and the appearance of sulfonamides and, a few years later, penicillin diminished interest in serum therapy, which eventually vanished. As a consequence, interest in typing of pneumococci also waned. This review connects current knowledge of typing, typing methods, typing nomenclature, and various aspects of serum therapy with those described in the Heffron and White monographs. Additionally, newer typing and subtyping methods, and those that loom in the future, will be discussed briefly.
THE BEGINNING In 1880, Pasteur in France and Sternberg in the United States independently isolated, cultured, and described the pneumococcus. Sternberg inoculated his own saliva into rabbits and Pasteur inoculated rabbits with the saliva of a child who had died of rabies.1,2 Pasteur published his findings in 1881, several months before Sternberg did. In 1897, using rabbit antipneumococcic serum, French researchers Bezanc¸on and Griffon found differences in agglutination reactions between strains of pneumococci and concluded that several races of pneumococci existed. They found concordant results from studying human cases of pneumonia.2 In 1910 in Germany, Neufeld and Ha¨ndel obtained high-potency monovalent sera by immunizing rabbits, donkeys, and horses with a collection of strains isolated from a series of pneumonia patients. With these sera, the investigators showed that the results obtained by mouse protection experiments agreed with those of the agglutination reaction. Neufeld and Ha¨ndel were the first to recommend that sera be developed for all types for the purpose of serum treatment of pneumonia.2 In 1913, working with their own strains as well as with strains originally isolated by Neufeld and Ha¨ndel, Dochez and Gillespie of the Rockefeller Institute divided pneumococci into four groups (I–IV; group IV was a heterogeneous collection of strains).2 They obtained their results using protection and agglutination methods. That same year, Lister studied opsonization of pneumococci recently isolated from native laborers in South Africa by 0002-9343/99/$20.00 PII S0002-9343(99)00100-X
A Symposium: Typing of Streptococcus pneumoniae/Henrichsen
sera of pneumonia patients. His results led to the division of his cultures into five groups (A–E). The typing system of Dochez and Gillespie was compared with Lister’s classification system. From 1921 to 1932, Cooper et al3 in New York managed to establish 32 types, chiefly using monovalent rabbit and horse sera as the agglutination method. Their work and that of many others led to the introduction of serum therapy for pneumococcal pneumonia.
TYPING METHODS Early on, researchers using various immune sera realized that different isolates of pneumococci could be divided into different races, groups, or types through agglutination and mouse protection methods. Little by little it became clear, as noted in White’s monograph, that “Whatever benefit is to be derived from the use of antipneumococcic serum depends on the rapid and accurate determination of the type of pneumococcus causing the infection.”2 The question thus became one of devising ways of shortening the time elapsing between the collection of the specimen and the identification of the serologic type of the infecting pneumococcus.2 Among the most important methods White2 described were the following: ● ●
●
●
●
Mouse protection test (1910). Culture agglutination. In 1918, Avery inoculated washed sputum in broth containing 1% glucose; after 5 hours of incubation, the culture was used as an agglutinating antigen with three types of sera. Urine precipitation test. This was based on the 1917 finding by Dochez and Avery that pneumococci in vivo as well as in vitro produce a soluble-specific substance that is easily detected with an appropriate immune serum. Stained slide microscopic agglutination test. Sabin introduced this method in 1929. Sputum is injected into the peritoneal cavity of a mouse, and a few hours later, a small amount of peritoneal exudate is withdrawn and mixed with diagnostic sera of the various types in separate drops on the same slide. The drops are then smeared and the slide fixed, stained, and examined under the microscope. Quellung reaction. First described by Neufeld and Etinger-Tulczynska in 1933, this is based on the Neufeld reaction described in 1902 and is now also known as the capsular reaction test. Then and now, this is probably the fastest, most specific, and easiest of all methods. It continues to be carried out essentially as described by Austrian.4
SERUM THERAPY Heffron1 and White2 both described the production of antipneumococcal therapeutic antisera in detail, includ-
ing methods of immunization, methods of concentrating serum, potency tests, in vitro tests, safety tests, and treatment of lobar pneumococcal pneumonia itself, through 1938. By 1886 Fra¨nkel had already observed that rabbits which recovered from an experimental pneumococcal infection were resistant to subsequent inoculation with the same organism.2 In 1891, the Klemperer brothers demonstrated that the young of actively immunized mother rabbits usually acquired protection and, furthermore, that the serum of actively immunized rabbits, when injected into the bloodstream of normal rabbits, protected the latter from an otherwise fatal infection due to their inoculation with pneumococci 24 hours later.1 In 1910, Neufeld and Ha¨ndel outlined suitable means of producing specific serums of high potency and developed measures for their standardization.1 At the hospital of the Rockefeller Institute in 1913, Cole et al1 treated cases of lobar pneumonia caused by types I or II pneumococci with specific immune horse serum. However, treatment with immune horse serum at times generated unimpressive results, particularly during World War I. In Boston in 1924, Felton was more successful, using immune horse sera he had concentrated by one of his own methods.1 Good results from the use of concentrated immune horse serum in suitable cases were also reported in 1928 by Cecil and by Bullowa et al, in 1929 by Bullowa, and in 1930 by Finland, among others.1 In 1934, Barnes and White offered specific experimental evidence that serum of much higher specific antibody content could be obtained from rabbits than from horses. Horsfall, Goodner, MacLeod, and Harris in 1937 and Bullowa the same year subsequently reported favorable results in the treatment of a number of cases of pneumococcus pneumonia with specific immune rabbit serum.1 Heffron added that the available evidence suggested antiserum produced in the rabbit may have certain advantages over that produced in the horse.1 In 1938, hemolytic reactions that sometimes led to death were reported in occasional patients receiving antipneumococcic type XIV horse sera.5 This was ascribed to antigenic similarity between the pneumococcal type 14 polysaccharide and the ABO blood group antigens.5,6 Therefore, just before World War II, there were multiple reasons to switch from therapeutic sera produced in horses to sera produced in rabbits: all rabbits respond to immunization, whereas not all horses do, and rabbits do so much sooner than do horses. Specific antibody concentrations obtained are higher in rabbits than in horses. Rabbit antibodies show much less cross-reactivity than do horse antibodies. The rabbit immunoglobulin (Ig) molecule is smaller than that of the horse. Rabbit typespecific antibody provides better opsonophagocytosis. Rabbit anti–type 14 antisera do not lead to occasional life-threatening hemolytic reactions.
July 26, 1999
THE AMERICAN JOURNAL OF MEDICINE威
Volume 107 (1A)
51S
A Symposium: Typing of Streptococcus pneumoniae/Henrichsen
Incidentally, the rabbit type-specific pneumococcal antisera were used to treat pneumococcal lobar pneumonia in Denmark for the first time in 1937; this was the only country in Europe that ever undertook serum therapy. Dosages were 100 –200 mL to 400 mL, given sequentially. Due to the war and to the appearance of sulfonamides and then, in the mid-1940s, of penicillin, larger studies of the use of rabbit sera were never carried out, and their use slowly diminished. In Denmark, however, these sera were used until the early 1950s.
PNEUMOCOCCAL TYPE NOMENCLATURE As mentioned earlier, between 1921 and 1932, Cooper et al6 described 32 types, which they numbered consecutively. In 1940, Kauffmann et al7 described several new types and proposed gathering closely related types into groups, using as an example group 7 (consisting of types 7, 7A, 7B, and 7C). In 1941, Walter et al8 described 17 new types and, following the nomenclature proposed by the Danes, designated certain of the types as subtypes. In 1944, Bernice Eddy of the Division of Biologics Control, National Institutes of Health, wrote: “Numbers have been proposed to designate every known pneumococcic type without consideration of cross-reactions with other types.”9 In a 1947 letter to Dr. St. John-Brooks, secretary-general of the International Association of Microbiologists, Annabel W. Walter of the New York City Department of Health observed, “Your letter of December 11 concerning nomenclature of pneumococcus types came as somewhat of a surprise. I agree that a single scheme of nomenclature is desirable and I shall be glad to serve on the committee with you and Doctors Kauffmann, Mørch and Eddy. The use of sulphonamides and penicillin for treating pneumonia and the reduction in laboratory staff during the war resulted in discontinuance of our studies of pneumococcus type separation.” Correspondence with Dr. Kauffmann had, of course, not been possible from the fall of Denmark to the end of the war (copy of letter in author’s possession). In 1954, Kauffmann and Lund (alias Mørch) suggested that their nomenclature10 should receive international recognition. In the 1950s and 1960s, a number of international committee meetings were held and the issue of nomenclature debated, but no conclusion was reached. In 1974 the International Subcommittee on the Taxonomy of Streptococci met in Prague before its activities came to a temporary halt. Still no decision was reached. In 1970, Lund proposed adding the letter F (for first) to the first type of each group, with the exception of groups 6 and 9.11 When S. pneumoniae type 47A was described in 1972,12 the number of known capsular types totaled 83.13 Finally, at a 1980 meeting of what was then known as 52S July 26, 1999 THE AMERICAN JOURNAL OF MEDICINE威
the Bureau of Biologics of the US Food and Drug Administration (FDA) in which FDA officials, vaccine manufacturers, representatives from the Centers for Disease Control, the World Health Organization, and others participated, Henrichsen represented the sole producer of reference typing antisera and commercially available diagnostic typing antisera. He insisted on continuing to label diagnostic pneumococcal typing antisera according to the Danish nomenclature. Reasons included the fact that nomenclature of types clearly was not subject to the rules and recommendations of The International Code of Nomenclature of Bacteria, that the Centers for Disease Control had already adopted the Danish type nomenclature, and that the experience of working with two systems of nomenclature over the last 5 years had shown that doing so was very confusing if not, at times, impossible. In 1985, Austrian et al14 described type 16A, and in 1995, Henrichsen described an additional six types, bringing the numbers of known types to a total of 90.15
TYPING METHODS TODAY Available Typing Antisera Omniserum, nine pooled sera labeled A through I, 46type or group antisera, and the so-called factor sera, their production, and their use have all been described in detail elsewhere.13 Today, they cover the whole range of 90 types, and factor sera have become freely available. An additional group of five pooled sera, labeled P through T, have become available as Pneumotest (Statens Serum Institut, Copenhagen, Denmark) together with pools A through F plus H. Using the pools in a chessboard fashion, this test allows typing of nearly 90% of all invasive disease isolates.16 Additionally, a number of monospecific antisera that can replace some factor sera have been raised in rabbits after induction of immunologic tolerance to heavily cross-reactive types by intravenous injection of high doses of purified capsular polysaccharides of these types.17 Finally, monoclonal pneumococcal antisera are being produced in a number of laboratories.18 –20 Typing Methods The capsular reaction test has been described in detail elsewhere.3,13,21 When carried out by experienced personnel, it is fast and easily interpretable. A pertinent argument in favor of using this test is that with it the whole range of diagnostic pneumococcal typing antisera has been produced using stepwise control procedures. The test also measures the final titers. As a consequence, crossreactions that are not detected by the capsular reaction test may appear if the diagnostic sera are used in more sensitive methods, as is sometimes the case (e.g., in countercurrent immunoelectrophoresis22), which is therefore not recommended. Agglutination tests using either protein-A–rich staphylococci or latex particles coated with
Volume 107 (1A)
A Symposium: Typing of Streptococcus pneumoniae/Henrichsen
pneumococcal antisera have been used successfully.23,24 Finally, the Centers for Disease Control and Prevention are currently typing by means of capillary precipitation, simply using liquid culture supernatants and typing antisera.
9. 10.
11.
FUTURE TYPING METHODS Pneumococci, especially penicillin-resistant isolates, are being characterized by techniques beyond typing, that is, being characterized by their penicillin-binding protein genomic profiles and cloning of the genes in Escherichia coli.25,26 Likewise, polymerase chain reaction– based typing by means of analysis of capsular genes is coming into use.27,28 Restriction fragment length polymorphism analysis has been known for identification of infrasubspecific taxa29 and also holds promise for subtyping of pneumococci.30,31 A quotation from Bowden seems appropriate as a closing thought: “There seems to be little doubt that serological techniques will be used in identification of bacteria for several years to come. However, DNA probes can perform the same role as antisera in identification and may be used in preference to serology in some cases in the future. Serology will retain value in determining antigenic similarities between bacteria and in demonstrating those antigens in microorganisms that are significant in the diagnosis of disease; neither of these goals can be achieved by DNA probes.”32 Because it remains to be seen whether the use of the new conjugate vaccine will affect the prevalence of disease causing types, continued type surveillance will be of major importance.
12.
13.
14.
15. 16. 17.
18.
19.
20.
21. 22.
REFERENCES 1. Heffron R. Pneumonia; with Special Reference to Pneumococcus Lobar Pneumonia (1939). [Reprinted, Harvard University Press, 1979.] 2. White B. The Biology of Pneumococcus (1938). [Reprinted, Harvard University Press, 1979.] 3. Cooper G, Rosenstein C, Walter A, Peizer L. The further separation of types among the pneumococci hitherto included in group 14 and the development of antisera for these types. J Exp Med. 1932;55:531–554. 4. Austrian R.The Quellung reaction: a neglected microbiologic technique. Mount Sinai Med J. 1976;43:699 –709. 5. Finland M, Curnen EC. Agglutinins for human erythrocytes in type XIV anti-pneumococcic horse serums. Science. 1938;87:417– 418. 6. Beeson PB, Goebel WF. The immunological relationship of the capsular polysaccharide of type XIV pneumococcus to the blood group A specific substance. J Exp Med. 1939;70: 239 –247. 7. Kauffmann F, Mørch E, Schmith K. On the serology of the pneumococcus group. J Immunol. 1940;39:397– 426. 8. Walter A, Guevin BH, Beattie, et al. Extension of the separation of types among the pneumococci: description of 17 types in addition to types 1 to 32 (Cooper) with recommen-
23.
24.
25.
26.
27.
July 26, 1999
dations for terminology of all types reported through 1940. J Immunol. 1941;41:279 –294. Eddy BE. Nomenclature of pneumococcic types. US Public Health Rep. 1944;59:449 – 451. Kauffmann F, Lund E. Memorandum on the nomenclature of the pneumococcus group. Int Bull Bacteriol Nomencl Taxon. 1954;4:125–128. Lund E. On the nomenclature of the pneumococcal types. Int J Syst Bacteriol. 1970;20:321–323. Lund E, Munksgaard A, Steward SM. A new pneumococcus type. Type 47A. Acta Pathol Microbiol Scand Sect B. 1972;80:497–500. Lund E, Henrichsen J. Laboratory diagnosis, serology and epidemiology of Streptococcus pneumoniae. In: Bergan T, Norris JR, eds. Methods in Microbiology. London: Academic Press, 1978:241–262. Austrian R, Boettger C, Dole M, et al. Streptococcus pneumoniae type 16A, a hitherto undescribed pneumococcal type. J Clin Microbiol. 1985;22:127–128. Henrichsen J. Six newly recognized types of Streptococcus pneumoniae. J Clin Microbiol. 1995;33:2759 –2762. So¨rensen UBS. Typing of pneumococci by using 12 pooled antisera. J Clin Microbiol. 1993;31:2097–2100. Henrichsen J, Robbins JB. Production of monovalent antisera by induction of immunological tolerance for capsular typing of Streptococcus pneumoniae. FEMS Microbiol Lett. 1992;94:89 –94. So¨rensen UBS, Agger R, Bennedsen J, Henrichsen J. Phoshorylcholine determinants in six pneumococcal capsular polysaccharides detected by monoclonal antibody. Infect Immun. 1984;43:876 – 878. Kolberg J, Aaberge IS, Jantzen E, et al. Murine monoclonal antibodies against pneumococcal capsular polysaccharide types 4, 8, 23F and 19A/19F. APMIS. 1992;100:91–94. Kolberg J, Høiby EA, Jantzen E. Detection of the phosphorylcholine epitope in streptococci, Haemophilus and pathogenic Neisseriae by immunoblotting. Microbiol Pathog. 1997;22:321–329. Henrichsen J. The pneumococcal typing system and pneumococcal surveillance. J Infect. 1979;1(suppl 2):31–37. Henrichsen J, Berntson E, Kaijser B. Comparison of counterimmunoelectrophoresis and the capsular reaction test for typing of pneumococci. J Clin Microbiol. 1980;11:589 – 592. Kronvall G. A rapid slide-agglutination method for typing pneumococci by means of specific antibody adsorbed to protein A containing staphylococci. J Med Microbiol. 1973; 6:187–190. Dajani AS. Agglutination tests for the diagnosis of meningitis. In: Coonrod JD, Kunz LJ, Ferraro MJ, eds. The Direct Detection of Microorganisms in Clinical Samples. London: Academic Press, 1983:143–157. Sibold C, Henrichsen, J, Ko¨nig A, et al. Mosaic pbp genes of major clones of penicillin-resistant Streptococcus pneumoniae have evolved from pbp genes of a penicillin-sensitive Streptococcus oralis. Mol Microbiol. 1994;12:1013– 1023. Laible G, Lurz R, Keck W, et al. Penicillin-binding protein 2x of Streptococcus pneumoniae: cloning and overproduction of the PBP 2x gene in E. coli. Eur J Biochem. 1992;207: 943–949. Watson DA, Kapur V, Musher DM, et al. Identification, cloning and sequencing of DNA essential for encapsulation of Streptococcus pneumoniae. Curr Microbiol. 1995;31: 251–259. THE AMERICAN JOURNAL OF MEDICINE威
Volume 107 (1A)
53S
A Symposium: Typing of Streptococcus pneumoniae/Henrichsen 28. Garcia E, Lopez R. Molecular biology of the capsular genes of Streptococcus pneumoniae. FEMS Microbiol Lett. 1997; 149:1–10. 29. Grimont F, Grimont PAD. Ribosomal ribonucleic acid gene restriction patterns as potential taxonomic tools. Ann Inst Pasteur/Microbiol. 1986:137B:165–175. 30. Martin B, Humbert O, Ca´mara M, et al. A highly conserved repeated DNA element located in the chromosome of
54S July 26, 1999 THE AMERICAN JOURNAL OF MEDICINE威
Streptococcus pneumoniae. Nucleic Acids Res. 1992;20: 3479 –3483. 31. Zhou L, Hui FM, Morrison DA. Characterization of IS 1167, a new insertion sequence in Streptococcus pneumoniae. Plasmid. 1995;33:127–138. 32. Bowden GHW. Serological identification. In: Goodfellow M, O’Donnell AG, eds. Handbook of New Bacterial Systematics. London: Academic Press, 1993:429 – 462.
Volume 107 (1A)
Early works leading to the detection of the pneumococcus and eventually to the appreciation that isolates differed in agglutination and that antisera differed in their capacity to protect against pneumococcal infection in the mouse protection test are reviewed. Studies by researchers from Europe, South Africa, and the United States over nearly five decades led to the introduction of serum therapy. Rapid typing methods thus became very important, and type-specific serum therapy generated a dramatic decrease in the number of deaths from pneumococcal pneumonia. Just before the introduction of sulfonamides and, shortly thereafter, penicillin, the use of horse sera was replaced by the use of rabbit sera for a number of reasons. The present methods of typing comprise the capsular reaction test, latexand coagglutination, and capillary precipitation, to name the most important; these use a large variety of antisera. Newer methods include the use of DNA probes and DNA sequence-based subtyping. Am J Med. 1999;107(1A):50S–54S. © 1999 by Excerpta Medica, Inc.
From the World Health Collaborating Center for Reference and Research on Pneumococci, Statens Serum Institut, Copenhagen, Denmark. Requests for reprints should be addressed to Jo¨rgen Henrichsen, MD, World Health Collaborating Center for Reference and Research on Pneumococci, Statens Serum Institut, Artillerivej 5, DK-2300 Copenhagen S, Denmark. 50S © 1999 by Excerpta Medica, Inc. All rights reserved.
T
he complete history of typing of Streptococcus pneumoniae seems never to have been told, although the monographs of Heffron1 and White,2 published in 1939 and 1938, respectively, and reprinted in 1979, cover it in detail until 1938. However, World War II disrupted communication among researchers on both sides of the Atlantic, and the appearance of sulfonamides and, a few years later, penicillin diminished interest in serum therapy, which eventually vanished. As a consequence, interest in typing of pneumococci also waned. This review connects current knowledge of typing, typing methods, typing nomenclature, and various aspects of serum therapy with those described in the Heffron and White monographs. Additionally, newer typing and subtyping methods, and those that loom in the future, will be discussed briefly.
THE BEGINNING In 1880, Pasteur in France and Sternberg in the United States independently isolated, cultured, and described the pneumococcus. Sternberg inoculated his own saliva into rabbits and Pasteur inoculated rabbits with the saliva of a child who had died of rabies.1,2 Pasteur published his findings in 1881, several months before Sternberg did. In 1897, using rabbit antipneumococcic serum, French researchers Bezanc¸on and Griffon found differences in agglutination reactions between strains of pneumococci and concluded that several races of pneumococci existed. They found concordant results from studying human cases of pneumonia.2 In 1910 in Germany, Neufeld and Ha¨ndel obtained high-potency monovalent sera by immunizing rabbits, donkeys, and horses with a collection of strains isolated from a series of pneumonia patients. With these sera, the investigators showed that the results obtained by mouse protection experiments agreed with those of the agglutination reaction. Neufeld and Ha¨ndel were the first to recommend that sera be developed for all types for the purpose of serum treatment of pneumonia.2 In 1913, working with their own strains as well as with strains originally isolated by Neufeld and Ha¨ndel, Dochez and Gillespie of the Rockefeller Institute divided pneumococci into four groups (I–IV; group IV was a heterogeneous collection of strains).2 They obtained their results using protection and agglutination methods. That same year, Lister studied opsonization of pneumococci recently isolated from native laborers in South Africa by 0002-9343/99/$20.00 PII S0002-9343(99)00100-X
A Symposium: Typing of Streptococcus pneumoniae/Henrichsen
sera of pneumonia patients. His results led to the division of his cultures into five groups (A–E). The typing system of Dochez and Gillespie was compared with Lister’s classification system. From 1921 to 1932, Cooper et al3 in New York managed to establish 32 types, chiefly using monovalent rabbit and horse sera as the agglutination method. Their work and that of many others led to the introduction of serum therapy for pneumococcal pneumonia.
TYPING METHODS Early on, researchers using various immune sera realized that different isolates of pneumococci could be divided into different races, groups, or types through agglutination and mouse protection methods. Little by little it became clear, as noted in White’s monograph, that “Whatever benefit is to be derived from the use of antipneumococcic serum depends on the rapid and accurate determination of the type of pneumococcus causing the infection.”2 The question thus became one of devising ways of shortening the time elapsing between the collection of the specimen and the identification of the serologic type of the infecting pneumococcus.2 Among the most important methods White2 described were the following: ● ●
●
●
●
Mouse protection test (1910). Culture agglutination. In 1918, Avery inoculated washed sputum in broth containing 1% glucose; after 5 hours of incubation, the culture was used as an agglutinating antigen with three types of sera. Urine precipitation test. This was based on the 1917 finding by Dochez and Avery that pneumococci in vivo as well as in vitro produce a soluble-specific substance that is easily detected with an appropriate immune serum. Stained slide microscopic agglutination test. Sabin introduced this method in 1929. Sputum is injected into the peritoneal cavity of a mouse, and a few hours later, a small amount of peritoneal exudate is withdrawn and mixed with diagnostic sera of the various types in separate drops on the same slide. The drops are then smeared and the slide fixed, stained, and examined under the microscope. Quellung reaction. First described by Neufeld and Etinger-Tulczynska in 1933, this is based on the Neufeld reaction described in 1902 and is now also known as the capsular reaction test. Then and now, this is probably the fastest, most specific, and easiest of all methods. It continues to be carried out essentially as described by Austrian.4
SERUM THERAPY Heffron1 and White2 both described the production of antipneumococcal therapeutic antisera in detail, includ-
ing methods of immunization, methods of concentrating serum, potency tests, in vitro tests, safety tests, and treatment of lobar pneumococcal pneumonia itself, through 1938. By 1886 Fra¨nkel had already observed that rabbits which recovered from an experimental pneumococcal infection were resistant to subsequent inoculation with the same organism.2 In 1891, the Klemperer brothers demonstrated that the young of actively immunized mother rabbits usually acquired protection and, furthermore, that the serum of actively immunized rabbits, when injected into the bloodstream of normal rabbits, protected the latter from an otherwise fatal infection due to their inoculation with pneumococci 24 hours later.1 In 1910, Neufeld and Ha¨ndel outlined suitable means of producing specific serums of high potency and developed measures for their standardization.1 At the hospital of the Rockefeller Institute in 1913, Cole et al1 treated cases of lobar pneumonia caused by types I or II pneumococci with specific immune horse serum. However, treatment with immune horse serum at times generated unimpressive results, particularly during World War I. In Boston in 1924, Felton was more successful, using immune horse sera he had concentrated by one of his own methods.1 Good results from the use of concentrated immune horse serum in suitable cases were also reported in 1928 by Cecil and by Bullowa et al, in 1929 by Bullowa, and in 1930 by Finland, among others.1 In 1934, Barnes and White offered specific experimental evidence that serum of much higher specific antibody content could be obtained from rabbits than from horses. Horsfall, Goodner, MacLeod, and Harris in 1937 and Bullowa the same year subsequently reported favorable results in the treatment of a number of cases of pneumococcus pneumonia with specific immune rabbit serum.1 Heffron added that the available evidence suggested antiserum produced in the rabbit may have certain advantages over that produced in the horse.1 In 1938, hemolytic reactions that sometimes led to death were reported in occasional patients receiving antipneumococcic type XIV horse sera.5 This was ascribed to antigenic similarity between the pneumococcal type 14 polysaccharide and the ABO blood group antigens.5,6 Therefore, just before World War II, there were multiple reasons to switch from therapeutic sera produced in horses to sera produced in rabbits: all rabbits respond to immunization, whereas not all horses do, and rabbits do so much sooner than do horses. Specific antibody concentrations obtained are higher in rabbits than in horses. Rabbit antibodies show much less cross-reactivity than do horse antibodies. The rabbit immunoglobulin (Ig) molecule is smaller than that of the horse. Rabbit typespecific antibody provides better opsonophagocytosis. Rabbit anti–type 14 antisera do not lead to occasional life-threatening hemolytic reactions.
July 26, 1999
THE AMERICAN JOURNAL OF MEDICINE威
Volume 107 (1A)
51S
A Symposium: Typing of Streptococcus pneumoniae/Henrichsen
Incidentally, the rabbit type-specific pneumococcal antisera were used to treat pneumococcal lobar pneumonia in Denmark for the first time in 1937; this was the only country in Europe that ever undertook serum therapy. Dosages were 100 –200 mL to 400 mL, given sequentially. Due to the war and to the appearance of sulfonamides and then, in the mid-1940s, of penicillin, larger studies of the use of rabbit sera were never carried out, and their use slowly diminished. In Denmark, however, these sera were used until the early 1950s.
PNEUMOCOCCAL TYPE NOMENCLATURE As mentioned earlier, between 1921 and 1932, Cooper et al6 described 32 types, which they numbered consecutively. In 1940, Kauffmann et al7 described several new types and proposed gathering closely related types into groups, using as an example group 7 (consisting of types 7, 7A, 7B, and 7C). In 1941, Walter et al8 described 17 new types and, following the nomenclature proposed by the Danes, designated certain of the types as subtypes. In 1944, Bernice Eddy of the Division of Biologics Control, National Institutes of Health, wrote: “Numbers have been proposed to designate every known pneumococcic type without consideration of cross-reactions with other types.”9 In a 1947 letter to Dr. St. John-Brooks, secretary-general of the International Association of Microbiologists, Annabel W. Walter of the New York City Department of Health observed, “Your letter of December 11 concerning nomenclature of pneumococcus types came as somewhat of a surprise. I agree that a single scheme of nomenclature is desirable and I shall be glad to serve on the committee with you and Doctors Kauffmann, Mørch and Eddy. The use of sulphonamides and penicillin for treating pneumonia and the reduction in laboratory staff during the war resulted in discontinuance of our studies of pneumococcus type separation.” Correspondence with Dr. Kauffmann had, of course, not been possible from the fall of Denmark to the end of the war (copy of letter in author’s possession). In 1954, Kauffmann and Lund (alias Mørch) suggested that their nomenclature10 should receive international recognition. In the 1950s and 1960s, a number of international committee meetings were held and the issue of nomenclature debated, but no conclusion was reached. In 1974 the International Subcommittee on the Taxonomy of Streptococci met in Prague before its activities came to a temporary halt. Still no decision was reached. In 1970, Lund proposed adding the letter F (for first) to the first type of each group, with the exception of groups 6 and 9.11 When S. pneumoniae type 47A was described in 1972,12 the number of known capsular types totaled 83.13 Finally, at a 1980 meeting of what was then known as 52S July 26, 1999 THE AMERICAN JOURNAL OF MEDICINE威
the Bureau of Biologics of the US Food and Drug Administration (FDA) in which FDA officials, vaccine manufacturers, representatives from the Centers for Disease Control, the World Health Organization, and others participated, Henrichsen represented the sole producer of reference typing antisera and commercially available diagnostic typing antisera. He insisted on continuing to label diagnostic pneumococcal typing antisera according to the Danish nomenclature. Reasons included the fact that nomenclature of types clearly was not subject to the rules and recommendations of The International Code of Nomenclature of Bacteria, that the Centers for Disease Control had already adopted the Danish type nomenclature, and that the experience of working with two systems of nomenclature over the last 5 years had shown that doing so was very confusing if not, at times, impossible. In 1985, Austrian et al14 described type 16A, and in 1995, Henrichsen described an additional six types, bringing the numbers of known types to a total of 90.15
TYPING METHODS TODAY Available Typing Antisera Omniserum, nine pooled sera labeled A through I, 46type or group antisera, and the so-called factor sera, their production, and their use have all been described in detail elsewhere.13 Today, they cover the whole range of 90 types, and factor sera have become freely available. An additional group of five pooled sera, labeled P through T, have become available as Pneumotest (Statens Serum Institut, Copenhagen, Denmark) together with pools A through F plus H. Using the pools in a chessboard fashion, this test allows typing of nearly 90% of all invasive disease isolates.16 Additionally, a number of monospecific antisera that can replace some factor sera have been raised in rabbits after induction of immunologic tolerance to heavily cross-reactive types by intravenous injection of high doses of purified capsular polysaccharides of these types.17 Finally, monoclonal pneumococcal antisera are being produced in a number of laboratories.18 –20 Typing Methods The capsular reaction test has been described in detail elsewhere.3,13,21 When carried out by experienced personnel, it is fast and easily interpretable. A pertinent argument in favor of using this test is that with it the whole range of diagnostic pneumococcal typing antisera has been produced using stepwise control procedures. The test also measures the final titers. As a consequence, crossreactions that are not detected by the capsular reaction test may appear if the diagnostic sera are used in more sensitive methods, as is sometimes the case (e.g., in countercurrent immunoelectrophoresis22), which is therefore not recommended. Agglutination tests using either protein-A–rich staphylococci or latex particles coated with
Volume 107 (1A)
A Symposium: Typing of Streptococcus pneumoniae/Henrichsen
pneumococcal antisera have been used successfully.23,24 Finally, the Centers for Disease Control and Prevention are currently typing by means of capillary precipitation, simply using liquid culture supernatants and typing antisera.
9. 10.
11.
FUTURE TYPING METHODS Pneumococci, especially penicillin-resistant isolates, are being characterized by techniques beyond typing, that is, being characterized by their penicillin-binding protein genomic profiles and cloning of the genes in Escherichia coli.25,26 Likewise, polymerase chain reaction– based typing by means of analysis of capsular genes is coming into use.27,28 Restriction fragment length polymorphism analysis has been known for identification of infrasubspecific taxa29 and also holds promise for subtyping of pneumococci.30,31 A quotation from Bowden seems appropriate as a closing thought: “There seems to be little doubt that serological techniques will be used in identification of bacteria for several years to come. However, DNA probes can perform the same role as antisera in identification and may be used in preference to serology in some cases in the future. Serology will retain value in determining antigenic similarities between bacteria and in demonstrating those antigens in microorganisms that are significant in the diagnosis of disease; neither of these goals can be achieved by DNA probes.”32 Because it remains to be seen whether the use of the new conjugate vaccine will affect the prevalence of disease causing types, continued type surveillance will be of major importance.
12.
13.
14.
15. 16. 17.
18.
19.
20.
21. 22.
REFERENCES 1. Heffron R. Pneumonia; with Special Reference to Pneumococcus Lobar Pneumonia (1939). [Reprinted, Harvard University Press, 1979.] 2. White B. The Biology of Pneumococcus (1938). [Reprinted, Harvard University Press, 1979.] 3. Cooper G, Rosenstein C, Walter A, Peizer L. The further separation of types among the pneumococci hitherto included in group 14 and the development of antisera for these types. J Exp Med. 1932;55:531–554. 4. Austrian R.The Quellung reaction: a neglected microbiologic technique. Mount Sinai Med J. 1976;43:699 –709. 5. Finland M, Curnen EC. Agglutinins for human erythrocytes in type XIV anti-pneumococcic horse serums. Science. 1938;87:417– 418. 6. Beeson PB, Goebel WF. The immunological relationship of the capsular polysaccharide of type XIV pneumococcus to the blood group A specific substance. J Exp Med. 1939;70: 239 –247. 7. Kauffmann F, Mørch E, Schmith K. On the serology of the pneumococcus group. J Immunol. 1940;39:397– 426. 8. Walter A, Guevin BH, Beattie, et al. Extension of the separation of types among the pneumococci: description of 17 types in addition to types 1 to 32 (Cooper) with recommen-
23.
24.
25.
26.
27.
July 26, 1999
dations for terminology of all types reported through 1940. J Immunol. 1941;41:279 –294. Eddy BE. Nomenclature of pneumococcic types. US Public Health Rep. 1944;59:449 – 451. Kauffmann F, Lund E. Memorandum on the nomenclature of the pneumococcus group. Int Bull Bacteriol Nomencl Taxon. 1954;4:125–128. Lund E. On the nomenclature of the pneumococcal types. Int J Syst Bacteriol. 1970;20:321–323. Lund E, Munksgaard A, Steward SM. A new pneumococcus type. Type 47A. Acta Pathol Microbiol Scand Sect B. 1972;80:497–500. Lund E, Henrichsen J. Laboratory diagnosis, serology and epidemiology of Streptococcus pneumoniae. In: Bergan T, Norris JR, eds. Methods in Microbiology. London: Academic Press, 1978:241–262. Austrian R, Boettger C, Dole M, et al. Streptococcus pneumoniae type 16A, a hitherto undescribed pneumococcal type. J Clin Microbiol. 1985;22:127–128. Henrichsen J. Six newly recognized types of Streptococcus pneumoniae. J Clin Microbiol. 1995;33:2759 –2762. So¨rensen UBS. Typing of pneumococci by using 12 pooled antisera. J Clin Microbiol. 1993;31:2097–2100. Henrichsen J, Robbins JB. Production of monovalent antisera by induction of immunological tolerance for capsular typing of Streptococcus pneumoniae. FEMS Microbiol Lett. 1992;94:89 –94. So¨rensen UBS, Agger R, Bennedsen J, Henrichsen J. Phoshorylcholine determinants in six pneumococcal capsular polysaccharides detected by monoclonal antibody. Infect Immun. 1984;43:876 – 878. Kolberg J, Aaberge IS, Jantzen E, et al. Murine monoclonal antibodies against pneumococcal capsular polysaccharide types 4, 8, 23F and 19A/19F. APMIS. 1992;100:91–94. Kolberg J, Høiby EA, Jantzen E. Detection of the phosphorylcholine epitope in streptococci, Haemophilus and pathogenic Neisseriae by immunoblotting. Microbiol Pathog. 1997;22:321–329. Henrichsen J. The pneumococcal typing system and pneumococcal surveillance. J Infect. 1979;1(suppl 2):31–37. Henrichsen J, Berntson E, Kaijser B. Comparison of counterimmunoelectrophoresis and the capsular reaction test for typing of pneumococci. J Clin Microbiol. 1980;11:589 – 592. Kronvall G. A rapid slide-agglutination method for typing pneumococci by means of specific antibody adsorbed to protein A containing staphylococci. J Med Microbiol. 1973; 6:187–190. Dajani AS. Agglutination tests for the diagnosis of meningitis. In: Coonrod JD, Kunz LJ, Ferraro MJ, eds. The Direct Detection of Microorganisms in Clinical Samples. London: Academic Press, 1983:143–157. Sibold C, Henrichsen, J, Ko¨nig A, et al. Mosaic pbp genes of major clones of penicillin-resistant Streptococcus pneumoniae have evolved from pbp genes of a penicillin-sensitive Streptococcus oralis. Mol Microbiol. 1994;12:1013– 1023. Laible G, Lurz R, Keck W, et al. Penicillin-binding protein 2x of Streptococcus pneumoniae: cloning and overproduction of the PBP 2x gene in E. coli. Eur J Biochem. 1992;207: 943–949. Watson DA, Kapur V, Musher DM, et al. Identification, cloning and sequencing of DNA essential for encapsulation of Streptococcus pneumoniae. Curr Microbiol. 1995;31: 251–259. THE AMERICAN JOURNAL OF MEDICINE威
Volume 107 (1A)
53S
A Symposium: Typing of Streptococcus pneumoniae/Henrichsen 28. Garcia E, Lopez R. Molecular biology of the capsular genes of Streptococcus pneumoniae. FEMS Microbiol Lett. 1997; 149:1–10. 29. Grimont F, Grimont PAD. Ribosomal ribonucleic acid gene restriction patterns as potential taxonomic tools. Ann Inst Pasteur/Microbiol. 1986:137B:165–175. 30. Martin B, Humbert O, Ca´mara M, et al. A highly conserved repeated DNA element located in the chromosome of
54S July 26, 1999 THE AMERICAN JOURNAL OF MEDICINE威
Streptococcus pneumoniae. Nucleic Acids Res. 1992;20: 3479 –3483. 31. Zhou L, Hui FM, Morrison DA. Characterization of IS 1167, a new insertion sequence in Streptococcus pneumoniae. Plasmid. 1995;33:127–138. 32. Bowden GHW. Serological identification. In: Goodfellow M, O’Donnell AG, eds. Handbook of New Bacterial Systematics. London: Academic Press, 1993:429 – 462.
Volume 107 (1A)