Competition among Streptococcus pneumoniae for intranasal colonization in a mouse model M. Lipsitch a, b, *, J.K. Dykes a , S.E. Johnson a , E.W. Ades a , J. King c , D.E. Briles c , G.M. Carlone a a Division of Bacterial and Mycotic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, GA 30333, USA b Department of Biology, Emory University, Atlanta, GA 30322, USA c Department of Microbiology, University of Alabama at Birmingham, Birmingham, AL 35294, USA Received 29 November 1999; accepted 24 January 2000 Abstract Widespread use of conjugate vaccines against Streptococcus pneumoniae, by reducing carriage of S. pneumoniae serotypes included in the vaccine, may result in an increase in nasopharyngeal carriage of Ð and disease from Ð nonvaccine serotypes of the same species. Mathematical models predict that the extent of such replacement will depend positively on the degree to which carriage of vaccine-type S. pneumoniae inhibits acquisition of nonvaccine-type pneumococci, and may depend negatively on the inhibition of vaccine-type pneumococci by nonvaccine-type pneumococci. We used a mouse model of intranasal carriage of pneumococci to test whether such inhibition occurs between dierent pneumococcal strains. Mice carrying a streptomycin- resistant derivative of S. pneumoniae BG9163 (serotype 6B) as a resident strain showed reduced levels of colonization when challenged intranasally by optochin-resistant derivatives of the same strain and of a serotype 23F pneumococcus, BG8826. Inhibition could be overcome by increasing the dose of the challenge strain. Carriage of optochin-resistant BG9163 did not inhibit acquisition of the streptomycin-resistant variant. Colonization by a challenge strain did not signi®cantly aect the level of colonization with the resident strain. These results provide evidence that is consistent with several hitherto untested assumptions of mathematical models of serotype replacement and suggest that a biological mechanism exists that could account for serotype replacement that is observed in clinical trials. The ®ndings provide a basis for further studies of in vivo interactions between strains of S. pneumoniae. Published by Elsevier Science Ltd. Keywords: Streptococcus pneumoniae; Serotype replacement; Nasopharyngeal carriage; Mathematical models; In vivo models 1. Introduction and background Conjugate vaccines against Streptococcus pneumo- niae are currently in clinical trials [1±5]. These vaccines immunize against 7, 9, of the 90 known pneumococcal serotypes. Preliminary results from these ongoing trials in infants indicate that these vaccines are highly pro- tective against invasive disease [1] and oer signi®cant protection against pneumonia [1] and otitis media [1,4], from the serotypes included in the vaccine. In ad- dition, the vaccine oers partial protection against nasopharyngeal carriage of the included serotypes [5± 7]. Because nasopharyngeal carriers of pneumococci are the major source for transmission of these bacteria, conjugate vaccines are expected to reduce transmission and prevalence of vaccine-type pneumococci. This re- duction in pneumococcal transmission should help to reduce the incidence of disease from vaccine-type pneumococci among unvaccinated individuals in com- munities where vaccination is used, via herd immunity Vaccine 18 (2000) 2895±2901 0264-410X/00/$ - see front matter Published by Elsevier Science Ltd. PII: S0264-410X(00)00046-3 www.elsevier.com/locate/vaccine * Corresponding author: Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA. Tel.: +001-617-432-4559; fax: +001-617-566-7805. E-mail address: mark_lipsitch@harvard.edu (M. Lipsitch).