Staphylococcus aureus throat carriage is associated with ABO-/secretor status

Summary

Objectives

In 30% of carriers, Staphylococcus aureus colonization affects exclusively the pharynx and occurs independently from its presence in the nares. This additional reservoir has implications for S. aureus transmission, infection, and decolonization. Host factors promoting colonization of the throat, however, are unknown.

Methods

We determined pharyngeal and persistent nasal carriage of S. aureus, ABO histo-blood group and ABH secretor status phenotypes in 227 individuals.

Results

Compared to group A/non-secretors, group O/non-secretor individuals were at increased risk of carrying S. aureus in their throat (OR 6.50, 95% confidence interval 1.28–33.03, P = 0.02) and group O/secretor individuals were protected (OR 0.24, 0.07–0.77, P = 0.02). Both associations became moderately stronger after adjusting for persistent S. aureus nasal carriage, which was found to be a risk factor for pharyngeal colonization in the univariable analysis (OR 2.41, 1.35–4.33, p = 0.003). Most simultaneous carriers (72%) had identical S. aureus genotypes in their nose and throat.

Conclusions

These findings are consistent with in vitro studies that proposed a role of histo-blood group antigens as ligands for S. aureus and support their contribution to the observed population variation in nasopharyngeal S. aureus colonization. Based on their tissue specific expression histo-blood group antigens appear to modulate individual S. aureus colonization patterns.

Introduction

The global emergence of antibiotic resistant Staphylococcus aureus as a cause of morbidity in- and outside the health-care setting explains the focus of current research on the evolution and spread of such strains. Colonization of the human skin and mucosal surfaces very likely promotes this process through facilitating the exchange of resistance genes, exposure to antibacterial compounds, selection of resistant clones, and human to human transmission.

While marked progress has been made in clarifying risk factors and determinants of S. aureus nasal carriage1, 2, 3, 4 published research on factors that promote colonization of other body sites is scarce. This neglect seems well justified as it has been shown, that the nose is the primary site of colonization. From there, other body regions are colonized by manual spread and are thus likely to lose colonization after eradication of S. aureus from the nares.5, 6

Pharyngeal carriage of S. aureus, however, was recently shown to occur independently from colonization of the nares. In a large cross-sectional study of 1500 blood donors, 30% of those colonized were found to be exclusive throat and 27% exclusive nasal carriers while 43% were colonized at both sites.⁷ Furthermore, one longitudinal investigation found that about 40% of persistently colonized individuals carried S. aureus at most occasions in the throat only.⁸ These findings suggest that the host-factors involved in colonization of the nose and throat must be, at least partially, distinct.

The clinical and public health importance of throat carriage is supported by a variety of observations. In severely ill patients, colonization of the pharynx and upper respiratory tract has been found to be a risk factor for ventilator associated pneumonia9, 10 and eradication of such colonization to reduce this risk.¹¹ Furthermore, it has been shown, that successful screening and eradication of methicillin resistant S. aureus colonization has to include separate sampling and antibacterial treatment of the throat.12, 13, 14 Finally, it is likely that throat carriage – as shown for nasal carriage¹⁵ – contributes towards the dissemination of bacteria into the environment through increased S. aureus dispersal during upper respiratory infections of viral origin.

It is generally agreed on that adhesion of microorganisms to epithelial surfaces is the first crucial step in establishing colonization.¹⁶ During this process, bacterial adhesins are thought to specifically recognize the epithelial surface.¹⁶ Carbohydrate residues of the glyococalix and glycoconjugates (i.e. glycoproteins and glycolipids) characterize the most outer parts of the epithelial lining and are thus likely to be of importance in establishing this first contact between host and pathogen.17, 18 The large diversity of such oligosaccharide chains together with evidence on their pathogen specificity gave rise to the hypothesis that their presence is a result of a co-evolutionary process balancing the relationship between potentially lethal pathogens and vertebrate hosts.17, 18

One group of glycans showing such striking diversity are histo-blood group antigens (HBGA). In particular those belonging to the ABH-/Lewis system have attracted significant interest as ligands for microorganisms. For instance, adhesion of Helicobacter pylori, Escherichia coli,¹⁹ Streptococcus pneumoniae,²⁰ Salmonella typhimurium,²¹ and Norwalk virus²² to epithelial cells was found to be associated with the phenotypic expression of ABH and Lewis HBGAs.

The quantitative and qualitative presence of ABH-/Lewis HBGAs on epithelia and in body fluids is genetically determined and depends largely on a set of fucosyltransferases (FUT) that add monosaccharide units to precursor molecules in a stepwise manner.²³ One key factor accounting for the variation in expression of these antigens is the α1,2 fucosyltransferase encoded by the FUT2-gene.23, 24 About 80% of Caucasian individuals carry a functional allele of the FUT2- or secretor gene (Se) while the remaining 20% do not.²³ Individuals with the functional variant are generally referred to as secretors. Secretors express fucosylated ABH- and Lewis b (Le^b) antigens in body fluids and on the surface of various tissues. By contrast, non-secretors are not able to synthesize H-antigen – the precursor of Le^b and ABH-antigens – in cells where it is synthesized by the FUT2 enzyme, and are thus phenotypically characterized by the absence of ABH antigens and the presence of Lewis a (Le^a) antigen in fluids and on several epithelial surfaces.²³

Evidence, that the presence of HBGAs on epithelia and in body fluids (e.g. mucus) may also play a role in S. aureus colonization is limited and the few available studies are inconclusive and partially conflicting. One in vitro study showed enhanced binding of S. aureus to buccal epithelial cells of non-secretors compared to secretors and marked inhibition of this binding by pretreatment with anti-Le^a antibodies.²⁵ This binding of S. aureus to Le^a was found to be associated with a 67 kDa adhesin on the S. aureus membrane.²⁶ Although these observations support a specific binding of S. aureus to Le^a, an earlier study failed to establish an association of secretor status with respiratory S. aureus colonization in patients with cystic fibrosis.²⁷ Furthermore, one study in 326 health care workers observed a predominance of blood group A among persistent nasal S. aureus carriers but did not investigate for secretor status.²⁸ Epidemiological studies trying to elucidate the presence of epistatic effects of the ABH and Lewis genes on the individual risk of being colonized with S. aureus are missing.

Such research could be of value in elucidating mechanisms that underlie the observed population variation in S. aureus colonization and their contribution towards individual patterns of colonization. The latter speculation is fostered by the description of substantial differences in AHB Lewis expression between different tissues and body regions.²⁴ In the light of increasing interest in throat carriage as an additional reservoir for S. aureus,7, 8, 12, 29, 30, 31 we hypothesized that ABH/Lewis expression might influence an individual's propensity to be colonized with S. aureus in the nose and throat.