Two variants among Haemophilus influenzae serotype b strains with distinct bcs4, hcsA and hcsB genes display differences in expression of the polysaccharide capsule

Abstract

Background: Despite nearly complete vaccine coverage, a small number of fully vaccinated children in the Netherlands have experienced invasive disease caused by Haemophilus influenzae serotype b (Hib). This increase started in 2002, nine years after the introduction of nationwide vaccination in the Netherlands. The capsular polysaccharide of Hib is used as a conjugate vaccine to protect against Hib disease. To evaluate the possible rise of escape variants, explaining the increased number of vaccine failures we analyzed the composition of the capsular genes and the expressed polysaccharide of Dutch Hib strains collected before and after the introduction of Hib vaccination.

Results: The DNA sequences of the complete capsular gene clusters of 9 Dutch Hib strains were assessed and two variants, designated type I and type II were found. The two variants displayed considerable sequence divergence in the hcsA and hcsB genes, involved in transport of capsular polysaccharide to the cell surface. Application of hcsA type specific PCRs on 670 Hib strains collected from Dutch patients with invasive Hib disease showed that 5% of the strains collected before 1996 were type II. No endogenous type II Hib strains were isolated after 1995 and all type II strains were isolated from 0-4 year old, non-vaccinated children only. Analysis of a worldwide collection of Hib strains from the pre-vaccination era revealed considerable geographic differences in the distribution of the type I and type II strains with up to 73% of type II strains in the USA. NMR analysis of type I and type II capsule polysaccharides did not reveal structural differences. However, type I strains were shown to produce twice as much surface bound capsular polysaccharide.

Conclusion: Type II strains were only isolated during the pre-vaccination era from young, non-vaccinated individuals and displayed a lower expression of capsular polysaccharide than type I strains. The higher polysaccharide expression may have provided a selective advantage for type I strains resulting in the rapid elimination of type II from the Dutch Hib population after introduction of nationwide Hib vaccination. However, this phenomenon does not explain the increase in the number of Hib vaccine failures in the Netherlands.

Figure 1

Schematic representation of the differences between the type I and type II Hib capsular gene clusters. The top part of the figure shows the genetic organization of the capsular gene cluster of Hib. The figure displays a single copy of the gene cluster with the fused truncated insertion element IS1016 and the incomplete bexA gene on the left hand side. The complete bexA gene and part of the bexB gene, present in the second copy of the gene cluster are depicted at the right hand side. The region containing the bcs4, hcsA and hcsB genes are displayed as a blow up. The vertical lines indicate nucleotide substitutions, closed circles denote amino acid substitutions and Δ indicate deletions.

Figure 2

Minimum spanning trees of 667 strains on which MLVA was performed and of 247 Hib strains analyzed by MLST. Each circle represents a different type. The size of the circle indicates the number of strains with this particular type. The orange area in the circles indicates the portion of type II strains. The numbers in the circles denote the MLST sequence type or MLVA type. Only the numbers of the predominant types and the types carrying capsular genotype II are displayed.

Figure 3

The 1H-NMR spectra of PRP samples obtained from type II and type I Hib strains. NMR profiles of type I Hib strain H1990-0614 and type II Hib strain H1994-0151. The minor absorptions at 3.5–3.65 ppm are caused some residual glycerol.

Figure 4

Expression of the Hib polysaccharide capsule of type I and type II strains as measured by inhibition ELISA. The graph displays the OD values obtained in the ELISA after inhibition by the cell fraction and the culture supernatant obtained from type I and type II strains. Inhibition was performed with 3 different dilutions of the antigens. Each bar denotes the average of the values obtained with 4 different strains and the error bars indicate the standard error. Red bars, inhibition by type I cells; yellow, inhibition by type II cells; dark blue, inhibition by type I culture supernatant; light blue, inhibition by type II culture supernatant. Note that in this graph higher OD values represent lower polysaccharide expression.

Figure 5

SBA with serial dilutions of a high titer serum using type I and type II Hib strains as target cells. The serum sample was diluted over a range of 1:8 (2³) to 1:262,144 (2¹⁸) and mixed with type I or with type II bacteria in the presence of baby rabbit complement. The SBA titer using the type I strain as target is 1:2,048 (2¹¹), whereas the type II target strain yields a titer of 1:4,096 (2¹²). The pro-zone effect, growth in the presence of high concentrations of serum, is more pronounced for the type I strain.

Figure 6

Comparison of capsular structures of type I and type II Hib cells by electron microscopy. Overnight cultures were stained for the presence of polysaccharide capsule, fixed, counterstained and cut into ultrathin sections. All images were made at 67,000× magnification.