Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2010 Aug 2;207(8):1637-46.
doi: 10.1084/jem.20100575. Epub 2010 Jul 12.

Novel Mechanism for the Generation of Human Xeno-Autoantibodies Against the Nonhuman Sialic Acid N-glycolylneuraminic Acid

Affiliations
Free PMC article

Novel Mechanism for the Generation of Human Xeno-Autoantibodies Against the Nonhuman Sialic Acid N-glycolylneuraminic Acid

Rachel E Taylor et al. J Exp Med. .
Free PMC article

Abstract

The nonhuman sialic acid N-glycolylneuraminic acid (Neu5Gc) is metabolically incorporated into human tissues from certain mammalian-derived foods, and this occurs in the face of an anti-Neu5Gc "xeno-autoantibody" response. Given evidence that this process contributes to chronic inflammation in some diseases, it is important to understand when and how these antibodies are generated in humans. We show here that human anti-Neu5Gc antibodies appear during infancy and correlate with weaning and exposure to dietary Neu5Gc. However, dietary Neu5Gc alone cannot elicit anti-Neu5Gc antibodies in mice with a humanlike Neu5Gc deficiency. Other postnatally appearing anti-carbohydrate antibodies are likely induced by bacteria expressing these epitopes; however, no microbe is known to synthesize Neu5Gc. Here, we show that trace exogenous Neu5Gc can be incorporated into cell surface lipooligosaccharides (LOS) of nontypeable Haemophilus influenzae (NTHi), a human-specific commensal/pathogen. Indeed, infant anti-Neu5Gc antibodies appear coincident with antibodies against NTHi. Furthermore, NTHi that express Neu5Gc-containing LOS induce anti-Neu5Gc antibodies in Neu5Gc-deficient mice, without added adjuvant. Finally, Neu5Gc from baby food is taken up and expressed by NTHi. As the flora residing in the nasopharynx of infants can be in contact with ingested food, we propose a novel model for how NTHi and dietary Neu5Gc cooperate to generate anti-Neu5Gc antibodies in humans.

Figures

Figure 1.
Figure 1.
Anti-Neu5Gc antibodies in human infants. (A and B) Levels of anti-Neu5Gc IgM (A) and IgG (B) antibodies in infant sera (n = 15, filled circles) from birth (cord), 3, 6, at 12 mo (for each infant) and adult sera from the pregnant mothers (n = 9, open squares) were measured by ELISA against Neu5Gcα2–6Lacβ–HSA. Each data point represents the mean of triplicate values from one individual, quantified according to an IgM or IgG standard curve. Horizontal lines represent mean values for each group. (C) Neu5Gcα2–6Lacβ–HSA coated on an ELISA plate was treated with mild periodate (periodate) or inactivated periodate (mock periodate), and then analyzed for binding by IgG antibodies from infants at 12 mo of age (n = 15). Bars represent mean absorbance values at OD490 ± SEM. Statistical analysis was performed using an unpaired two-tailed Student’s t test. (D) Infant sera (same infants as in A and B) were analyzed by ELISA for IgM antibodies against α-Gal-PAA (dashed line). For comparison, anti-α-Gal IgM levels were plotted with anti-Neu5Gc IgM levels (solid line; same data as in Fig. 1 A). Values represent mean IgM levels ± SEM.
Figure 2.
Figure 2.
Dietary Neu5Gc does not elicit anti-Neu5Gc antibodies. (A and B) WT (open bars, n = 8) and Cmah−/− (filled bars, n = 8) were fed ∼1 mg/kg/day Neu5Gc (from normal chow) after weaning (4–6 wk total), immunized with human RBC ghosts (WT, n = 4; Cmah−/−, n = 4), or immunized with chimpanzee RBC ghosts (WT, n = 4; Cmah−/−, n = 4). Sera were analyzed by ELISA for IgM (A; note the broken y axis) and IgG (B) antibodies against Neu5Gcα-PAA and shown as mean absorbance values at OD405 ± SEM. Statistical analysis was performed using an unpaired two-tailed Student’s t test. n.s, not significant, *, P < 0.05; **, P < 0.01. Data are representative of greater than three independent experiments.
Figure 3.
Figure 3.
NTHi can efficiently take up and incorporate Neu5Gc. (A) Dose-dependent uptake and expression of NTHi (2019) grown in a Sia-free defined media with 0.1 µM–1 mM Neu5Gc. Neu5Gc was detected by flow cytometry analysis using a chicken anti-Neu5Gc IgY antibody. (B) 2019 grown in 1 mM Neu5Gc were treated with sialidase (sialidase) or heat-inactivated sialidase (mock sialidase) and probed with a chicken anti-Neu5Gc IgY antibody in a whole-cell ELISA. Data are representative of greater than three independent experiments and show the mean of triplicate absorbance values at OD405. Error bars represent SD.
Figure 4.
Figure 4.
Anti-NTHi antibodies in human infants. Infant sera (same infants analyzed in Fig. 1; n = 15) were analyzed by whole-cell ELISA for IgM antibodies against NTHi strain 2019 grown in Sia-free media (filled circles). Adult sera (n = 9) obtained from the pregnant mothers of the infants in this study were analyzed in parallel with infant sera for IgM anti-NTHi antibodies (open squares). Each circle or square represents the mean of triplicate values from one individual, quantified according to an IgM standard curve. Horizontal lines represent mean values for each group. Statistical analysis was performed using an unpaired two-tailed Student’s t test.
Figure 5.
Figure 5.
Neu5Gc expressed on NTHi induces anti-Neu5Gc antibodies in Cmah−/− mice. (A and B) Cmah−/− mice were injected intraperitoneally with heat-killed NTHi (without adjuvant) which had been grown in sialic-acid free media (Sia-free NTHi, n = 16) or with sialic-acid free media with 1 mM Neu5Gc (Gc-NTHi, n = 17). All mice were injected a total of three times at 2-wk intervals. Sera collected after the third injection were analyzed by ELISA for IgM (A) and IgG (B) antibodies against Neu5Gcα-PAA. The highest and lowest value from each group was removed before graphing. Horizontal lines represent the mean values. Statistical analysis was performed using an unpaired one-tailed Student’s t test. (C) Pooled mouse serum from mice injected intraperitoneally with heat-killed NTHi grown in sialic-acid free media with 1 mM Neu5Gc was analyzed in an ELISA for levels of anti-Neu5Gc IgG antibodies against Neu5Gcα2–3Galβ1–4Glcβ-HSA (black bar), Neu5Gcα2–6Galβ1–4Glcβ-HSA (gray bar), Neu5Gcα2–3Galβ1–4GlcNAcβ-HSA (white hashed bar), and Neu5Gcα2-6Galβ1-4GlcNAc-HSA (white bar). Values represent mean IgG levels, quantified according to a mouse IgG standard curve. Error bars represent SD. For comparison, serum from 12-mo-old infants (n = 15) was analyzed for levels of anti-Neu5Gc antibodies against Neu5Gcα2–3Galβ1–4Glcβ-HSA (black bar), Neu5Gcα2–6Galβ1–4Glcβ-HSA (gray bar; same data as shown in Fig. 1 B). Values represent mean IgG levels, quantified according to a human IgG standard curve. Error bars represent SEM. Data are representative of greater than three independent experiments.
Figure 6.
Figure 6.
NTHi uptake and expression of Neu5Gc from baby food and its recognition by purified human anti-Neu5Gc antibodies. (A) WT NTHi strain 2019 (WT, solid line) and Sia transporter mutant strain of 2019 (SiaT Mutant, dashed line) were grown in Sia-free media with or without the addition of commercially available baby foods and analyzed by flow cytometry for Neu5Gc staining using the chicken anti-Neu5Gc IgY antibody. Secondary antibody for WT and SiaT Mutant (not depicted) and IgY control for SiaT Mutant (not depicted) all showed similar shifts compared with the IgY control for WT (solid gray). (B) Anti-Neu5Gc IgG antibodies purified from serum from a single individual (S34) was used to probe NTHi in a whole-cell ELISA. NTHi was grown in Sia-free media (Sia-Free NTHi) or Sia-free media with 1 mM Neu5Gc (Gc-NTHi), and binding of human antibodies is shown the mean of triplicate values at OD405. Error bars represent SD. Statistical analysis was performed using an unpaired two-tailed Student’s t test. ***, P < 0.0001. Data are representative of two (A) and three (B) independent experiments.

Similar articles

See all similar articles

Cited by 50 articles

See all "Cited by" articles

References

    1. Allen S., Zaleski A., Johnston J.W., Gibson B.W., Apicella M.A. 2005. Novel sialic acid transporter of Haemophilus influenzae. Infect. Immun. 73:5291–5300 10.1128/IAI.73.9.5291-5300.2005 - DOI - PMC - PubMed
    1. Anderson P., Johnston R.B.J., Jr., Smith D.H. 1972. Human serum activities against Hemophilus influenzae, type b. J. Clin. Invest. 51:31–38 10.1172/JCI106793 - DOI - PMC - PubMed
    1. Angata T., Varki A. 2002. Chemical diversity in the sialic acids and related alpha-keto acids: an evolutionary perspective. Chem. Rev. 102:439–469 10.1021/cr000407m - DOI - PubMed
    1. Bardor M., Nguyen D.H., Diaz S., Varki A. 2005. Mechanism of uptake and incorporation of the non-human sialic acid N-glycolylneuraminic acid into human cells. J. Biol. Chem. 280:4228–4237 10.1074/jbc.M412040200 - DOI - PubMed
    1. Bouchet V., Hood D.W., Li J., Brisson J.R., Randle G.A., Martin A., Li Z., Goldstein R., Schweda E.K., Pelton S.I., et al. 2003. Host-derived sialic acid is incorporated into Haemophilus influenzae lipopolysaccharide and is a major virulence factor in experimental otitis media. Proc. Natl. Acad. Sci. USA. 100:8898–8903 10.1073/pnas.1432026100 - DOI - PMC - PubMed

Publication types

MeSH terms

Feedback