Novel sialic acid transporter of Haemophilus influenzae

Abstract

Nontypeable Haemophilus influenzae is an opportunistic pathogen and a common cause of otitis media in children and of chronic bronchitis and pneumonia in patients with chronic obstructive pulmonary disease. The lipooligosaccharides, a major component of the outer membrane of H. influenzae, play an important role in microbial virulence and pathogenicity. N-Acetylneuraminic acid (sialic acid) can be incorporated into the lipooligosaccharides as a terminal nonreducing sugar. Although much of the pathway of sialic acid incorporation into lipooligosaccharides is understood, the transporter responsible for N-acetylneuraminic acid uptake in H. influenzae has yet to be characterized. In this paper we demonstrate that this transporter is a novel sugar transporter of the tripartite ATP-independent periplasmic transporter family. In the absence of this transporter, H. influenzae cannot incorporate sialic acid into its lipooligosaccharides, making the organism unable to survive when exposed to human serum and causing reduced viability in biofilm growth.

FIG. 1.

SDS-PAGE and MALDI-TOF-MS of LOS isolated from wild-type NTHi 2019 and NTHi 2019HI1104. (A) SDS-PAGE. Lanes 1 to 4, LOS isolated from wild-type NTHi 2019; lanes 5 to 8, LOS isolated from NTHi 2019HI1104. Bacteria were grown on BHI in the absence (lanes 1, 2, 5, and 6) or presence (lanes 3, 4, 7, and 8) of Neu5Ac. LOS samples in lanes 2, 4, 6, and 8 were treated with neuraminidase prior to loading. LOS from N. gonorrhoeae strain PID2 was used as a molecular weight standard. The LOS was visualized by silver staining. (B and C) MALDI-TOF-MS of O-deacylated LOS from wild-type NTHi 2019 and NTHi 2019HI1104, respectively. Bacteria were grown on media containing supplemental sialic acid. See Table 2 for molecular masses and proposed compositions. Asterisks indicate the addition and number of sialic acid residues; subscripts indicate the number of PEA moieties.

FIG. 2.

Dose dependence of sialic acid incorporation into NTHi 2019 LOS. (A to D) MALDI-TOF-MS of O-LOS of wild-type NTHi 2019 and NTHi 2019nanA in the absence (A and B) and presence (C and D) of sialic acid. See Table 2 for molecular masses and proposed compositions. Asterisks indicate the addition and number of sialic acid residues; subscripts indicate the number of PEA moieties. (E) Incorporation of sialic acid into LOS. MALDI-TOF-MS was carried out on O-LOS isolated from wild-type NTHi 2019 and NTHi 2019nanA grown in media containing increasing amounts of sialic acid. Peaks corresponding to ions of asialylated and sialylated O-LOS glycoforms were integrated and the areas of the peaks summated. The total sialylated O-LOS was then plotted as a percentage of the total O-LOS. □, wild-type NTHi 2019; ▪, 2019nanA.

FIG. 3.

(A) Map of the region in the NTHi genome surrounding siaT (HI0147). (B) Uptake/incorporation of [³H]sialic acid into wild-type NTHi 2019 (⧫), 2019nanA (▪), 2019siaT (▴), and 2019siaTnanA (×). Mid-log-phase bacteria were pelleted and resuspended in RPMI containing a final concentration of 3.3 μM unlabeled sialic acid and 0.7 μM [³H]sialic acid. Samples were removed to a Nuclepore membrane at 10 s and then at 0.5, 1, 3, 5, 10, 15, and 20 min, aspirated, and washed, and the sample were counted in scintillation fluid.

FIG. 4.

(A) SDS-PAGE of LOS isolated from wild-type NTHi 2019 and NTHi 2019siaT. Lanes 1 to 4, LOS isolated from wild-type NTHi 2019; lanes 5 to 8, LOS isolated from the mutant NTHi 2019siaT. Bacteria were grown on BHI in the absence (lanes 1, 2, 5, and 6) or presence (lanes 3, 4, 7, and 8) of Neu5Ac. LOS samples in lanes 2, 4, 6, and 8 were treated with neuraminidase prior to loading. LOS from N. gonorrhoeae strain PID2 was used as a molecular weight standard. The LOS was visualized using silver staining. (B) Western blot of LOS probed with 3F11. Lanes 1, 3, 5, and 7, LOS isolated from wild-type NTHi 2019; lanes 2, 4, 6, and 8, LOS isolated from the mutant NTHi 2019siaT. Bacteria were grown on BHI in the absence (lanes 1 to 4) or presence (lanes 5 to 8) of Neu5Ac. LOS samples in lanes 3, 4, 7, and 8 were treated with neuraminidase prior to loading. LOS from N. gonorrhoeae strain PID2 was used as a molecular weight standard. LOS was probed with monoclonal antibody 3F11, which recognizes a terminal N-acetyllactosamine structure. Antibody binding was detected using peroxidase-labeled goat anti-mouse IgM and a chemiluminescent substrate.

FIG. 5.

Negative-ion MALDI-TOF mass spectra of O-deacylated LOS from wild-type NTHi 2019 and the mutant NTHi 2019siaT. (A) O-LOS from wild-type NTHi 2019 grown on BHI medium without supplemental Neu5Ac; (B) O-LOS from wild-type NTHi 2019 grown on BHI medium supplemented with Neu5Ac; (C) O-LOS from mutant NTHi 2019siaT grown without supplemental Neu5Ac; (D) O-LOS from mutant NTHi 2019siaT grown with supplemental Neu5Ac. See Table 2 for molecular weights and proposed compositions. Asterisks indicate the addition and number of Neu5Ac residues; subscripts indicate the number of PEA moieties.

FIG. 6.

Resistance to serum killing of wild-type NTHi and mutants NTHi 2019nanA, NTHi 2019siaT, and NTHi 2019siaTnanA, respectively. Bacteria were grown on BHI without (A and C) or with (B and D) supplemental Neu5Ac. The scale in all panels is at log₁₀ intervals. (A and B) Bacteria were exposed to normal human serum for 30 min at 37°C from a 20-donor pool of serum from human volunteers with no previous history of serious infections. (C and D) Controls exposed to normal human serum that was heat inactivated at 56°C for 30 min. The ability of the bacteria to grow after treatment with serum was assessed by comparison to growth of untreated bacteria. Serum killing is expressed as log₁₀ change in CFU between treated and untreated bacteria.

FIG. 7.

Biofilm formation by wild-type 2019 and 2019siaT. 2019 and 2019siaT were grown for 2 days in supplemented RPMI diluted 1:10 in PBS, stained with the LIVE/DEAD BacLight bacterial viability kit, and examined by confocal microscopy. Three-dimensional representations of 2019 biofilm (A) and 2019siaT biofilm (B) were rendered using the Volocity imaging program. The wild-type strain (panel A) had a greater proportion of viable cells (green), while a majority of the siaT mutant biofilm (panel B) appeared to be nonviable (red). Squares represent 54 μm.