Population diversity among Bordetella pertussis isolates, United States, 1935-2009

Abstract

Since the 1980s, pertussis notifications in the United States have been increasing. To determine the types of Bordetella pertussis responsible for these increases, we divided 661 B. pertussis isolates collected in the United States during 1935-2009 into 8 periods related to the introduction of novel vaccines or changes in vaccination schedule. B. pertussis diversity was highest from 1970-1990 (94%) but declined to ≈ 70% after 1991 and has remained constant. During 2006-2009, 81.6% of the strains encoded multilocus sequence type prn2-ptxP3-ptxS1A-fim3B, and 64% were multilocus variable number tandem repeat analysis type 27. US trends were consistent with those seen internationally; emergence and predominance of the fim3B allele was the only molecular characteristic associated with the increase in pertussis notifications. Changes in the vaccine composition and schedule were not the direct selection pressures that resulted in the allele changes present in the current B. pertussis population.

Figure 1

Timeline of pertussis vaccine introduction in the United States and appearance of alleles within the Bordetella pertussis population, 1935–2009. The 8 periods used in this study are indicated at bottom; numbers below indicate number of selected strains during that period (N = 661). wP, whole-cell pertussis vaccine; MLVA 27, multilocus variable number tandem repeat analysis type 27; aP, acellular pertussis vaccine; Tdap, tetanus-diphtheria-aP.

Figure 2

Minimum spanning trees depicting changes within the Bordetella pertussis population, United States, 1935–1996. Multilocus variable number tandem repeat analysis (MLVA) types are represented by circles and are scaled to member count within each panel; multilocus sequence typing (MLST) types are represented by color. A) Periods 1 and 2, 1935–1945 (n = 3) and 1946–1969 (n = 16), respectively. These 2 periods were combined for the generation of the tree. Period 1 (prevaccine era) strains are shown on the left, distantly related to period 2 strains (right side) from the early whole-cell pertussis vaccine (wP) era. B) Period 3, 1970–1990, n = 76. During this period, when wP was in use, a high degree of diversity was identified; 2 predominant MLST types differed by the ptxS1 allele. C) Period 4, 1991–1996, n = 86. During the transition from wP to acellular pertussis vaccine for the 4th and 5th dose of the childhood series, MLVA 27, ptxP3, and prn2 were dominant, and the fim3B allele emerged.

Figure 3

Minimum spanning trees depicting changes within the Bordetella pertussis population, United States, 1997–2009. Multilocus variable number tandem repeat analysis (MLVA) types are represented by circles and are scaled to member count within each panel; multilocus sequence typing (MLST) types are represented by color. A) Period 5, 1997–1999, n = 159, the early years of acellular pertussis vaccine (aP) use. B) Period 6, 2000–2002, n = 98. With aP in use, MLVA 27 with the fim3B allele dominated. C) Period 7, 2003–2005, n = 98. In 2004, during the late aP use period, the novel pertactin allele (prn14) was identified in an isolate from New York. D) Period 8, 2006–2009, n = 125. After the introduction of the aP booster for adolescents and adults, MLST type prn1-ptxP1-ptxS1B-fim3A (previously found in periods 2–4 and 6) reappeared with a new MLVA type.

Figure 4

Frequency (by proportion of all isolates tested) of predominant multilocus variable number tandem repeat analysis (MLVA) types within the Bordetella pertussis population, United States, 1935–2009. MLVA 10 was dominant in period 2 (1946–1969) but decreased through periods 3 (1970–1990) and 4 (1991–1996) while MLVA 18, 27, and 29 emerged. MLVA 27 increased in proportion during period 4 and dominated the population for the rest of the study period; however, the proportion of MLVA 27 has been decreasing since period 6 (2000–2002), allowing for the emergence of other types.

Figure 5

Transitions of frequency (by proportion of all isolates tested) of dominant alleles for each multilocus sequence typing (MLST) type target within the Bordetella pertussis population, United States, 1935–2009. The previous dominant type is denoted by a solid line, with the new dominant type denoted by a dashed line of the same style. The dashed lines of prn2 and ptxP3 overlap with each other and multilocus variable number tandem repeat analysis (MLVA) type 27 (Figure 6), which suggests they arose at approximately the same time and resulted in the new dominant MLVA + MLST profile. The transition from fim3A to fim3B occurred much later than the other transitions.

Figure 6

Comparison of number of pertussis notifications, proportion of vaccine coverage, and proportion of dominant multilocus sequence typing alleles and multilocus variable number tandem repeat analysis (MLVA) type 27 among a random selection of 661 isolates, United States, 1935–2009. Bars indicate case notifications; lines indicate 2-point moving average distributions of frequency for the time periods assigned in this study. Vaccine coverage data were collected for the United States Immunization Survey (USIS, 1962–1985), National Health Interview Survey (NHIS, 1991−1993), and National Immunization Survey (NIS, 1994−2009). No data are available for 1986−1990 because USIS was cancelled (15). The fim3B trend line was temporally and significantly associated with the rate of increase for pertussis notifications.