Th1 versus Th2 T cell polarization by whole-cell and acellular childhood pertussis vaccines persists upon re-immunization in adolescence and adulthood

Highlights

• Epitope repertoires recognized by aP- and wP-primed donors are similar.

• Magnitude of T cell responses is higher in aP-primed donors.

• IFNγ strongly dominates the T cell response in wP-primed donors, whereas IL-5 is dominant in aP-primed individuals.

• Differential pattern of polarization is maintained even after booster vaccination.

Abstract

The recent increase in cases of whooping cough among teenagers in the US suggests that the acellular Bordetella pertussis vaccine (aP) that became standard in the mid 1990s might be relatively less effective than the whole-bacteria formulation (wP) previously used since the 1950s. To understand this effect, we compared antibody and T cell responses to a booster immunization in subjects who received either the wP or aP vaccine as their initial priming dose in childhood. Antibody responses in wP- and aP-primed donors were similar. Magnitude of T cell responses was higher in aP-primed individuals. Epitope mapping revealed the T cell immunodominance patterns were similar for both vaccines. Further comparison of the ratios of IFNγ and IL-5 revealed that IFNγ strongly dominates the T cell response in wP-primed donors, while IL-5 is dominant in aP primed individuals. Surprisingly, this differential pattern is maintained after booster vaccination, at times from eighteen years to several decades after the original aP/wP priming. These findings suggest that childhood aP versus wP vaccination induces functionally different T cell responses to pertussis that become fixed and are unchanged even upon boosting.

Introduction

Prior to the introduction of effective vaccines, rates of pertussis infection were high in the general US population (more than 150,000 cases per year from 1922 to 1950 [1]. Bordetella pertussis, the causative agent of whooping cough, infects the lungs and upper airways of human hosts. The infection leads to a persistent and paroxysmal cough sometimes associated with post-tussive vomiting and apnea, and is associated with a mortality rate of 1–2%, most frequently among infants younger than 3 months of age [2]. The introduction of the inactivated whole-cell B. pertussis vaccine (wP, containing aluminum salts as adjuvant) in the 1950s led to a dramatic decrease in the number of cases and associated deaths. However, vaccine-related side effects provided a motivation to develop a safer vaccine construct [3]. The subsequently developed acellular vaccine (aP) consists of a mixture of purified antigens from the bacterium, including filamentous hemagluttinin (FHA), pertactin (PRN), pertussis toxin (PT), and fimbrial proteins 2 and 3 (Fim2/3), also formulated with aluminum salts. As the safety profile of aP was much improved, it quickly became the primary vaccine used in infants following approval in the US in the mid to late 1990s and is currently routinely used as booster vaccine for both adults and adolescents [4], [5], [6].

Unexpectedly, 28,660 cases of whooping cough were reported to CDC in 2014, and its incidence is expected to continue to rise by the end of 2015. Interestingly, the highest rates of increase were found in children and adolescents under the age of 20, but not in adults over the age of 20, corresponding to the first cohorts of infants that were aP-vaccinated in the mid 1990s that are now in their teenage years [7]. Accordingly, it is suspected that the switch to aP might be connected to this rise in infections [8], [9], [10].

Both aP and wP induce an increase in antibody titers against all aP components [11], [12], [13], [14]. Adults originally primed with wP remain seropositive for multiple pertussis antigens up to 36 months following vaccination, with antibody titers at greater levels than at pre-vaccination [15]. However, protection against infection persists even after antibody titers have decreased, which suggests that low antibody titers may be protective for the entire inter-boost time, or that protection is due to a cell-mediated component of immunity to B. pertussis [16]. Animal studies suggest that memory CD4 responses to B. pertussis are required for long-lasting immunity to infection, and significant responses in these subsets can be detected after wP vaccination and after infection [17]. In humans, vaccination with aP induces a predominantly Th2 response, which may be associated with less effective protection from disease and infection [18], [19]; however, a reappraisal of the phenotype of T cells induced by vaccination is needed, since most of the seminal work in humans was performed before the appreciation of the substantial plasticity of Th1/Th2 subsets in humans [20], [21], and before the range of additional CD4 T cell subsets was recognized.

Several immunological mechanisms have been proposed to explain the decreasing efficacy of B. pertussis vaccination [8], [9], [10], [22], including the possibility that vaccination with aP induces a qualitatively different T cell response, at the level of which antigens and epitopes are recognized, and/or the phenotypes of T cells recognizing them. Surprisingly, a comparison of immunological parameters observed following aP booster immunization in teenagers originally primed in childhood with aP versus older adults originally primed with wP has not been performed. Characterization of responses to vaccine antigens and epitopes in the two epidemiological cohorts associated with differential disease incidence could inform future mechanistic studies. Accordingly, in the present study we assessed T cell and antibody responses to B. pertussis antigens and defined epitope repertoires using PBMCs from volunteers originally primed with either the wP or aP vaccine.