Indian Journal of Medical Microbiology Home 

Year : 2001  |  Volume : 19  |  Issue : 4  |  Page : 197--200

Cost-effective method of serotyping streptococcus pneumoniae using staphylococcal co-agglutination

B Rajalakshmi, R Kanungo 
 Department of Microbiology, JIPMER, Pondicherry - 605 006, India

Correspondence Address:
R Kanungo
Department of Microbiology, JIPMER, Pondicherry - 605 006


Typing of Streptococcus pneumoniae to determine the serotype prevalence has paved the way for polyvalent vaccines to prevent invasive pneumococcal infection. Variation of serotype prevalence in different geographical areas necessitates typing of strains from these areas for effective vaccine protection. High cost of antisera very often is a hindering factor in undertaking this exercise. We have tried to evaluate typing by co-agglutination to reduce cost. Clinical isolates of S.pneumoniae from Pondicherry and surrounding Tamil Nadu were serotyped using antisera coated Staphylococcus aureus Cowan I strain and compared with standard quellung reaction. There was hundred percent correlation. By this method we could determine the serotypes causing invasive infections in this area. A commercially available Pneumotest kit was used as source of type specific antisera. Serotype 1 was found to be the major isolate (20.1%) by both the tests. Twenty-four isolates (13%) belonged to the nonvaccine types. Rest of the isolates was made up by serotypes 6, 5, 19, 23 and 12. Co-agglutination method was found to be a simple rapid and economical technique. Ten milliliters of the reagent could be made, using 0.1 ml of standard antisera. Shelf life was found to be six months at 40C.

How to cite this article:
Rajalakshmi B, Kanungo R. Cost-effective method of serotyping streptococcus pneumoniae using staphylococcal co-agglutination.Indian J Med Microbiol 2001;19:197-200

How to cite this URL:
Rajalakshmi B, Kanungo R. Cost-effective method of serotyping streptococcus pneumoniae using staphylococcal co-agglutination. Indian J Med Microbiol [serial online] 2001 [cited 2020 Jul 2 ];19:197-200
Available from:

Full Text

Streptococcus pneumoniae is the predominant cause of morbidity and mortality worldwide. It is a common etiological agent of invasive infections like meningitis, pneumonia and septicaemia. After the success of Haemophilus influenzae type b vaccine, S. pneumoniae has become the major cause of bacterial meningitis now accounting for an estimated 2705 cases annually in the United States[1] and even larger number in the developing countries.[1] It is also an important pathogen causing bacterial pneumonia in the developed and developing countries in both children and adults. Of the 90 serotypes identified so far only few types are responsible for most invasive infections in any geographical area.[2] Distribution of pneumococcal serotypes varies according to the region. The currently employed 23-valent-pneumococcal vaccine incorporates 90% of the types causing invasive infections.

Although typing of S.pneumoniae based on their capsular nature was an age-old technique, it was neglected after the advent of antibiotics. Presently typing of strains causing invasive infection has become essential as a strategy for vaccine prevention, in the light of the emergence of multi drug resistant S. pneumoniae occurring worldwide. The present typing methods comprise of the capsular reaction test, latex and co-agglutination and capillary precipitation.[3],[4],[5] Newer methods of typing include the use of DNA probes and DNA sequence-based subtyping.[3]

Currently Omniserum, a nine pooled pneumococcal antisera labeled A through I, 46-types of group antisera, factor sera covering the whole range of 90 types are available. In addition, a group of five pooled sera, labeled P through T with pools A through F and H are available as a Pneumotest kit (Statens Serum Institut, Copenhagen, Denmark). Using these pools, typing of most of serotypes included in the presently available 23-valent pneumococcal vaccine is possible by chess-board system. Monoclonal pneumococcal antisera are also being developed.[6]

Although a variety of typing methods are available in the Western countries its availability with logistic problems of transportation, lack of distributing agencies and above all high cost are some of the problems faced in the developing countries. Agglutination tests using protein-A rich Staphylococci (Cowan I strain) have been used successfully for typing of pneumococci.[4],[7] This method has not been exploited for use in economically backward nations.

This study was conducted to determine the serotype prevalence in Pondicherry and surrounding area of Tamil Nadu, using the co-agglutination technique for typing the clinical isolates of S.pneumoniae from various infections, with an objective of providing a costeffective method of determining type prevalence.

 Materials and Methods

Pneumococcal isolates from specimens like blood, CSF, pleural fluid, ascitic fluid, peritoneal fluid and other sites were identified according to the standard methods.[8] They were stocked in skimmed milk at -700C until further testing. Prior to typing, the stock cultures were thawed and subcultured on trypticase soy agar with 5% sheep blood and incubated in 5% CO2 incubator overnight.

Pneumococcal antisera: Pneumotest kit containing antisera included in the 23 valent pneumococcal vaccine obtained from Statens Serum Institut, Copenhagen, Denmark, was used as source for antisera.

Co-agglutination reagent: The method followed was as described by Lalitha MK et al.[4] Staphylococcus aureus Cowan I strain (kindly donated by Lalitha MK, Christian Medical College, Vellore) were grown in 500 mL Todd-Hewitt broth (Hi-media) overnight at 370C in shaker water bath. Cells were washed five times with 0.03 M Phosphate Buffered Saline (PBS. pH 7.3). After washing, these cells were suspended in 0.5% Formaldehyde PBS and left at room temperature for 3 hours with intermittent shaking. Cells were then washed again with PBS. The suspension was heated in a water bath at 560C for 30 minutes to remove the foul odour. Finally a 10% Cowan I suspension was made for sensitization with different pools of pneumococcal antisera.

Sensitization was carried out as follows. Each of 0.1 mL of different pool antisera were mixed with 1 mL of 10% Cowan I cell suspension. The suspension was kept at room temperature for 30 minutes for sensitization and centrifuged. The cells were then suspended to 1% in PBS containing 0.1% sodium azide and stored at 40C.

Co-agglutination test

Colonies of overnight culture of S. pneumoniae were suspended in saline (opacity of the suspension was adjusted to McFarland standard 1). 25 ml of the pneumococcal suspension was mixed with 25 ml of the sensitized Co-A reagent on a clean glass slide. After mixing for 3 minutes the test was considered positive if there was clumping with clearing of the background. From the 12 pools tested, pools that showed positive results were noted and used for the identification of the serotype according to the chart provided by the manufacturer of the pneumotest kit [Table:1].

Standard method followed was as per the instructions of the manufacturers. 1-4 ml of a pneumococcal suspension from overnight culture was placed on a clean glass slide. Equal amount of antiserum was added and mixed thoroughly with the droplet on the slide. A coverslip was placed on the mixture and examined under a microscope using an oil immersion lens, (x1000 total magnification) after 10-15 minutes. Observation of well-defined swollen bacteria was taken as a positive reaction. Lyophilized type 1 pneumococcus and its homologous type serum 1 provided by the manufacturers were used as positive control.


Except a single strain, which showed autoagglutination all the clinical isolates could be typed using Co-agglutination and capsular swelling reaction. Out of 184 strains of S.pneumoniae isolated, 24 (13%) isolates belonged to non-vaccine type. Serotype 1 contributed 20.1% (37/184). This was followed by serotype 6 which accounted for 11.4% (21/184) of the total isolates. The other common serotypes of S. pneumoniae isolated were serotype 5, 19, 23, 12, 7, 18, 3, 14 and 15 [Table:2]. Predominance of serotype 1 was seen in the strains isolated from the CSF [Table:3]. Among isolates from bacteremic pneumococcal pneumonia, serotype5 was predominant. The non-vaccine types of S.pneumoniae were isolated from cases of pneumonia as well as other invasive infections. None of the 184 strains isolated belonged to serotype 2. The results were found to have 100% correlation with those obtained by the standard capsular swelling method.


A total of 184 strains of S.pneumoniae could be typed by the Co-agglutination technique following a chess-board system using 12 pools of pneumococcal antisera. The results had 100% agreement with the standard Quellung reaction. Co-agglutination technique was found to be cost effective when compared to Quellung test, which required large quantity of the specific antisera. However, not all strains could be placed in the types not included in the Pneumotest Kit. By using this some strains were found to belong to non-vaccine types, which were not incorporated in the Pneumotest kit. The other drawback of chess-board system using pooled antisera was the inability of subtyping of strains, which belonged to any particular serotype. This was also borne out by Lalitha MK et al.[4] On the other hand, advantage of this method was that using only 12 pools of different antisera, around 21 vaccine serotypes could be identified. This method was found to be easy and quick as it could be done by a one-step procedure of referring to the chart provided by the manufacturers.[9] Time required to perform Quellung test was more, as the slide had to be kept aside after the addition of antisera for the halo to become visible. A good microscope with x1000 total magnification was also a prerequisite for the test. In contrast, the Co-agglutination test was found to be simple, less time consuming and did not require a microscope. The reagent had a shelf life of six months at 40C. Colony suspension from plates could be directly tested and naked eye observations made within minutes.

Cost of typing by Co-agglutination was calculated and found to be Rs.11.60 per test. On the other hand using the Pneumotest kit, obtained from Statens Serum Institut, cost per test was found to be Rs.187.00, an increase by 16 fold.

It is to be noted that Co-agglutination test is a simple, rapid and cost-effective technique, which can be employed for serotyping the strains of S.pneumoniae routinely in a diagnostic microbiology laboratory. This method can be implemented in a developing country where cost is a limiting factor in serotyping of pneumococcal isolates.

Typing of all significant clinical isolates from different geographical region will help in building a data bank pertinent to that area. This data will in turn provide useful information for the formulation of pneumococcal vaccines. Serotypes that are predominant in a particular area can therefore be included in the pneumococcal vaccines to prevent invasive infections. Predominance of serotype 1 among our isolates was similar to the report of multicentric IBIS (Invasive Bacterial Surveillance Study), and an earlier study from Chandigarh.[10],[11] Serotypes 6, 19, 7, 5, 15, 14, 4, 16 and 18 were the types encountered by these groups. Another study from Madurai reported serotypes 9, 10, 11, 14, 15, 19, 23 and 33 in infants colonized with S.pneumoniae.[12]

Co-agglutination was found suitable for serotyping clinical isolates of S.pneumoniae in the diagnostic laboratory. Chess-board system with the help of 12 pooled sera could detect 21 serotypes, which are included in the currently available 23-valent pneumococcal vaccine.


Authors wish to thank Dr. M.K. Lalitha, Professor,Department of Microbiology, Christian Medical College Hospital, Vellore for imparting training to B. Rajalakshmi and cross checking some of the strains.


1Schuchat A, Deaver-Robinson K, Wenger JD. Bacterial meningitis in the United States in 1995. N Engl J Med 1997; 337: 970-976.
2Klein JO. The epidemiology of pneumococcal disease in infants and children. Rev Infect Dis 1981; 3: 246-253.
3Henrichsen J. Typing of Streptococcus pneumoniae: past, present and future. Am J Med 1999; 107 (1A): 505-508.
4Lalitha MK, Pai R, John TJ, Thomas K, Jesudason MV, Brahmadathan KN, Sridharan G, Steinhoff MC. Serotyping of Streptococcus pneumoniae by agglutination assays: a cost-effective technique for developing countries. Bull WHO 1996; 74(4): 387-390.
5Singhal A, Lalitha MK, John TJ, Thomas K, Raghupathy P, Jacob S, Steinhoff MC. Modified Latex agglutination test for rapid detection of Streptococcus pneumoniae and Haemophilus influenzae in cerebrospinal fluid and direct serotyping of Streptococcus pneumoniae. Eur J Clin Microbiol Infect Dis 1996; 15: 472-477.
6Kolberg 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.
7Kronvall 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.
88. Duguid JP, Ross PW. Streptococcus pneumoniae (Pneumococcus) in: Collee JG, Duguid JP, Fraser AG and Marmion BP (eds) Mackie and MeCartney Practical Medical Microbiology, 13th edn. Longman Group (FE) Ltd, New York, 1989; 327-334.
9Lalitha MK, Thomas K, Satish KR, Steinhoff MC, and the IBIS study group. Serotyping of Streptococcus pneumoniae by Coagglutination with 12 pooled antisera. J Clin Microbiol 1999; 37(1): 263-265.
10Thomas K, Lalitha MK, Steinhoff MC and IBIS Study Group. Invasive Bacterial Infection Surveillance. Ind J Med Microbiol 1999; 17(2): 66-71.
11Ayyagari A, Kumar L, Agarwal KC, Ganju S, Sharma P, Pancholi V. Prevalence of serotypes of Streptococcus pneumoniae in patients in and around Chandigarh. Indian J Med Res 1984; 79: 174-177.
12Coles CL, Kanungo R, Rahmathullah L, Thulasiraj RD, Kalz J, Santhosham M, Tielsch JM. Pneumococcal nasopharyngeal colonization in young South Indian infants. Pediatr Infect Dis J 2001; 20(3): 289-295.