The effects of differences in pspA alleles and capsular types on the resistance of Streptococcus pneumoniae to killing by apolactoferrin

Highlights

• The strength of binding of apolactoferrin to pneumococci is associated with susceptibility to killing by apolactoferrin (apo-Lf).

• Variation in PspA accounts for most of the variation in the ability of pneumococci to be killed by apo-hLf.

• The ability of PspA to bind apolactoferrin to pneumococci is strongly associated with resistance to killing by apo-hLf.

• Variation in capsule had little affect on susceptibility of pneumococci to apo-hLf. Where it appears to enhance killing by apo-hLf.

Abstract

Pneumococcal surface protein A (PspA) is the only pneumococcal surface protein known to strongly bind lactoferrin on the bacterial surface. In the absence of PspA Streptococcus pneumoniae becomes more susceptible to killing by human apolactoferrin (apo-hLf), the iron-free form of lactoferrin. In the present study we examined diverse strains of S. pneumoniae that differed by 2 logs in their susceptibility to apo-hLf. Among these strains, the amount of apo-hLf that bound to cell surface PspA correlated directly with the resistance of the strain to killing by apo-hLf. Moreover examination of different pspA alleles on shared genetic backgrounds revealed that those PspAs that bound more lactoferrin conferred greater resistance to killing by apo-hLf. The effects of capsule on killing of pneumococci by apo-hLf were generally small, but on one genetic background, however, the lack of capsule was associated with 4-times as much apo-hLf binding and 30-times more resistance to killing by apo-hLf. Overall these finding strongly support the hypothesis that most of the variation in the ability of apo-hLf is dependent on the variation in the binding of apo-hLf to surface PspA and this binding is dependent on variation in PspA as well as variation in capsule which may enhance killing by reducing the binding of apo-hLf to PspA.

Introduction

Streptococcus pneumoniae causes invasive diseases such as otitis media, sepsis, meningitis, and pneumonia in children and the elderly. Every year pneumococcal infections kill an estimated 800,00 children <5 years of age worldwide [1]. Pneumococci are also responsible for 30–50% of all otitis media infections and a significant portion of cases of sinusitis [2], [3], [4]. Pneumococcal pathogenesis is multi-factorial and is the result of an interplay between host immunity and bacterial virulence factors, including the pneumococcal cell-surface virulence proteins and capsular polysaccharide [5], [6].

Pneumococcal capsular polysaccharide and PspA are both virulence factors that reduce complement-dependent phagocytosis [7], [8], [9] and are both generally required for optimal colonization, bacteremia, and sepsis [9], [10], [11], [12], [13]. Although there are over 90 different capsular serotypes [14], [15], less than a third are commonly found in invasive disease in humans [16], [17], [18], [19]. A major role of pneumococcal capsular polysaccharide in invasive infection is the reduction in complement-mediated opsonophagocytosis of pneumococci. Antibodies to capsular polysaccharide have been shown to enhance complement-mediated opsonophagocytosis of S. pneumoniae and more than compensate for the protective effect of the capsule [7]. Low levels of capsule are required for colonization [11], and one of its mechanism of action is to reduce aggregation of the pneumococci in the airway which leads to their clearance [20].

Pneumococcal surface protein A (PspA) is a virulence factor that ranges in size among different strains from 67 to 99 kDa [21], [22]. Despite structural polymorphism, PspAs are immunologically cross-reactive [22], [23]. Although capsule has been reported to largely mask the exposure of pneumococcal surface antigens [9], [24], the exposure of PspA on live pneumococci, as detected by surface fluorescence with monoclonal Ab to PspA, is relatively unaffected by the capsule [25], although in some cases (this manuscript) capsule can reduce the binding of polyclonal antibody to PspA.

PspA has been shown to inhibit complement activation and deposition [26], [27]. Mutants lacking surface expression of PspA are less virulent and are cleared from the blood at a faster rate than their parent strains [9], [28], [29], [30], [31]. In at least some strains of pneumococci, mutants lacking PspA exhibit less colonization than the parent bacteria [12], [32]. PspA is protective against phagocytosis and killing by complement-dependent and complement-independent mechanisms [33]. PspA is also highly immunogenic and immunity to PspA has been shown to protect mice against lung infection, nasal colonization, and sepsis with S. pneumoniae [34], [35], [36], [37], [38] and to enhance complement deposition on pneumococci [9] and killing by phagocytes [39], [40].

PspA is the only pneumococcal surface protein that binds to human lactoferrin. It binds to both conformational forms of human lactoferrin, namely, hololactoferrin (holo-hLF, the iron-saturated form) and apolactoferrin (apo-hLf) the iron free form of human lactoferrin [41], [42]. This observation led to the discovery, using pspA mutants, that PspA can protect against killing of pneumococci by apo-hLf [43]. We also observed that antibody to PspA blocked binding of lactoferrin to PspA and increased the ability of apo-hLf to kill pneumococci [42], [44]. Apo-hLf is found in elevated levels in secretions and in serum during inflammation [41], [42], [43]. Apo-hLf is part of the innate immune system and is thought to help provide protection at the mucosal surfaces against pathogens [45]. When apo-hlf reaches pneumococci its N-terminal end is hydrolysed by a pneumococcal protease and the cationic peptides released are bactericidal for pneumococci [46]. PspA has been found to protect against these apo-hLF peptides [43], [46] and against other cationic peptides (Mirza et al. unpublished). The binding of apo-hlf and it peptides to PspA prevents them from reaching the cell membrane and causing cell death [44]. The susceptibility of pneumococci to killing by apo-hLf was found to vary among different pneumococcal strains [43]. It was not known, however, if this variability in killing by lactoferrin was due to differences in binding of apo-hLf to PspA as a result of structural heterogeneity in PspA or differences in capsular type, which might affect protection against killing by lactoferrin.

In the present study we have examined the association between variability in susceptibility to killing by apolactoferrin and changes in PspA and capsule type. We examined the effects of capsular types and different variants of PspA on resistance to killing. Effects of PspA on live pneumococci were measured by exchanging pspA genes between lactoferrin resistant and susceptible strains. The role of capsule on resistance to killing by apo-hLf was examined using isogenic mutants lacking capsule, or expressing a different capsule.

Our results indicated that some PspAs are more protective against apo-hLf than others and the protection against apo-hLf was strongly associated with the amount of apo-hLf that was bound to the PspA of each pneumococcus. In some cases the removal of capsule resulted in more binding by apo-hLf and more protection against killing by apo-hLf, in other cases removal of capsule had little or no effect on apo-hLf binding or killing by apo-hLf.