Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Dec;19(12):e46065.
doi: 10.15252/embr.201846065. Epub 2018 Oct 18.

The fungal ligand chitin directly binds TLR2 and triggers inflammation dependent on oligomer size

Katharina Fuchs  1 , Yamel Cardona Gloria  1 , Olaf-Oliver Wolz  1 , Franziska Herster  1 , Lokesh Sharma  2 , Carly A Dillen  3 , Christoph Täumer  4 , Sabine Dickhöfer  1 , Zsofia Bittner  1 , Truong-Minh Dang  1 , Anurag Singh  5 , Daniel Haischer  6 , Maria A Schlöffel  6 , Kirsten J Koymans  7 , Tharmila Sanmuganantham  1 , Milena Krach  1 , Thierry Roger  8 , Didier Le Roy  8 , Nadine A Schilling  9 , Felix Frauhammer  10   11 , Lloyd S Miller  3 , Thorsten Nürnberger  6 , Salomé LeibundGut-Landmann  12 , Andrea A Gust  6 , Boris Macek  4 , Martin Frank  13 , Cécile Gouttefangeas  1 , Charles S Dela Cruz  2 , Dominik Hartl  5   14 , Alexander Nr Weber  15
Affiliations

The fungal ligand chitin directly binds TLR2 and triggers inflammation dependent on oligomer size

Katharina Fuchs et al. EMBO Rep. 2018 Dec.

Abstract

Chitin is the second most abundant polysaccharide in nature and linked to fungal infection and asthma. However, bona fide immune receptors directly binding chitin and signaling immune activation and inflammation have not been clearly identified because polymeric crude chitin with unknown purity and molecular composition has been used. By using defined chitin (N-acetyl-glucosamine) oligomers, we here identify six-subunit-long chitin chains as the smallest immunologically active motif and the innate immune receptor Toll-like receptor (TLR2) as a primary fungal chitin sensor on human and murine immune cells. Chitin oligomers directly bind TLR2 with nanomolar affinity, and this fungal TLR2 ligand shows overlapping and distinct signaling outcomes compared to known mycobacterial TLR2 ligands. Unexpectedly, chitin oligomers composed of five or less subunits are inactive, hinting to a size-dependent system of immuno-modulation that appears conserved in plants and humans. Since blocking of the chitin-TLR2 interaction effectively prevents chitin-mediated inflammation in vitro and in vivo, our study highlights the chitin-TLR2 interaction as a potential target for developing novel therapies in chitin-related pathologies and fungal disease.

Keywords: N‐acetyl‐glucosamine; anti‐fungal innate immunity; chitin; inflammation; toll‐like receptor.

PubMed Disclaimer

Figures

Figure 1
Figure 1. Size‐dependent recognition of defined chitin oligomers in humans, mice, and plants
  1. A

    Electron micrograph of a human MoMac engulfing a crude chitin particle (n = 2).

  2. B–E

    (B) Size comparison (to scale) of typical TLR ectodomain and a chitin 7‐mer (seven NAG subunits). Release of IL‐6 and/or TNF release (ELISA) from (C) human MoMacs (n = 6–8), (D) murine BMDMs (n = 6), or (E) THP‐1 cells (n = 3).

  3. F, G

    (F) GFP in vivo fluorescence post‐intradermal injection of chitin oligomers into LysM EGFP/+ C57BL/6 mice (n = 3/group). (G) Representative in vivo imaging results.

  4. H, I

    Leukocyte infiltration in vivo after intratracheal administration assessed by (H) BALF flow cytometry (n = 3–4/group) and (I) histology analysis (n = 2), scale bar = 200 μm, arrows show cellular infiltrate.

  5. J, K

    Arabidopsis thaliana seedling (J) target gene mRNA induction (n = 3) or leaf piece (K) ROS production (0–45 min, n = 6).

Data information: (C, D, F, H, and K) represent data (mean ± SEM) combined from “n” (given in brackets for each panel) technical or biological replicates (human donors, mice, or plant leaves, respectively). In (A, E, G, I, and J), one representative of “n” (given in brackets for each panel) independent experiments is shown (mean ± SD). *P < 0.05 according to Wilcoxon signed rank sum (C, D), one‐way ANOVA (H), or Student's t‐test (E, J).
Figure EV1
Figure EV1. Chitin treatment results in chitinase and target gene induction in mammalian and plant cells, respectively
  1. A

    mRNA induction of CHIT1 relative to TBP and unstimulated conditions in human whole blood treated with large chitin (n = 2).

  2. B

    Chitin azure degradation indicative of chitinase activity in murine BALF from mice previously treated with PBS or house‐dust mite allergen (HDM) in vivo (n = 5). S. griseus recombinant chitinase was used as a positive control.

  3. C

    Primary human MoMacs were stimulated with C10‐15 or large chitin for 24 h and supernatants harvested and incubated with chitin azure (n = 3).

  4. D, E

    Cytokine release induced by chitin is independent of endotoxin. (D) IL‐6 release from stimulated human MoMacs stimulated in the absence or presence of polymyxin B (Pmx B, n = 3), and (E) IL‐6 release from WT or Tlr4 KO BMDMs (n = 3) are shown for one representative batch of C10‐15.

  5. F

    Target gene induction in A. thaliana seedlings. MLO12 and WRKY40 mRNA induction (n = 3).

Data information: (B, C, and E) represent data (mean ± SEM) combined from n biological replicates (mice or donors, each donor represented by one dot). In (A, D, and F), one representative of “n” (given in brackets for each panel) independent experiments is shown (mean ± SD). *P < 0.05 according to one‐sample t‐test (B), Kruskal–Wallis test with Dunn's correction (C), and Student's t‐test (D, E).
Figure EV2
Figure EV2. Chitin oligomers initiate cellular activation in a TLR2‐dependent manner

  1. IL6 or TNF mRNA induction relative to TBP in human MoMacs (n = 3).

  2. Relative CD62L shedding (n = 8, histograms shown for one representative donor) and IL‐8 release (n = 7–13) from human PMNs upon 1‐ or 4‐h stimulation, respectively.

  3. TLR2 mRNA knock‐down efficiency upon Ctrl, MYD88 or TLR2 siRNA transfection in human MoMacs (see Fig. 2c), relative to Ctrl siRNA.

  4. TLR2 immunoblot and flow cytometry in TLR2 CRISPR‐edited THP‐1 cells (n = 1 each). Arrowheads denote endogenous TLR2 (glycosylated).

  5. NF‐κB activity in empty vector (EV) or NOD2, Dectin‐1, or TLR9‐transfected HEK293T cells (n = 2–3).

  6. TNF release (relative to unstimulated) from WT or Clec7a KO (Dectin‐1 deficient) immortalized macrophages (iMacs, n = 2).

Data information: (B and C) represent data (mean ± SEM) combined from 4 to 13 donors (each dot represents one donor). In (A, C, E, and F), one representative of “n” (given in brackets for each panel) independent experiments is shown (mean ± SD). *P < 0.05 according to Wilcoxon signed rank sum (B), one‐way ANOVA with/without Dunnett's correction (A, C), or Student's t‐test (E, F).
Figure 2
Figure 2. TLR2 is the mediator of chitin oligomer‐mediated immune activation
  1. A, B

    TNF released from WT or Myd88 (A) or Tlr2 (B) KO BMDM (total n = 6–9/group combined from three experiments).

  2. C

    IL‐6 released from primary MoMacs (n = 5) treated with non‐targeting (NT), TLR2‐, or MyD88‐specific siRNA relative to NT condition.

  3. D

    CD62L shedding from human PMNs (n = 4–8) with or without isotype control or anti‐TLR2 blocking antibody.

  4. E

    IL‐6 protein released from mock or TLR2 Cas9‐CRISPR‐edited THP‐1 cells (n = 3).

  5. F

    NF‐κB activity in empty vector (EV) or TLR2‐transfected HEK293T cells (n = 3).

  6. G

    MPO release from intradermally injected WT or Tlr2 KO mice (n = 5–15 biopsies each).

  7. H

    BALF IL‐6 and TNF from intratracheally treated WT or Tlr2 KO mice (n = 4–5 mice each).

  8. I

    Relative IL10 mRNA in whole blood of WT or heterozygous TLR2 R753Q carriers (n = 5 each).

Data information: (A–D and G–I) represent data (mean ± SEM) combined from “n” technical or biological replicates (human donors, mice, or samples, respectively). In (E and F), one representative of “n” independent experiments is shown (mean ± SD). *P < 0.05 according to Mann–Whitney U‐test (D, I), one‐sample t‐test (C), Student's t‐test (A, B, E–G), or one‐way ANOVA with Dunnett's multiple comparison (H).
Figure 3
Figure 3. Chitin is a unique fungal TLR2 ligand

  1. NF‐κB activity in TLR2‐transfected HEK293T cells (n = 3).

  2. IL‐8 release and ROS production from primary PMNs (n = 7).

  3. NF‐κB activity in hTLR2 HEK‐Dual cells pre‐treated with TLR1, TLR2, or TLR6 blocking Abs (n = 2).

  4. IRAK4 S186 and T345 phospho‐site containing IRAK4 peptides quantified in proteome‐wide phospho‐screen in THP‐1 cells (n = 1).

  5. Principal component analysis of microarrays conducted on stimulated whole‐blood samples from healthy donors (n = 3–5).

  6. Microarray intensities (excerpt) of chitin‐specific signature genes (n = 5).

  7. Relative cytokine mRNAs in whole‐blood stimulation (n = 5–13).

Data information: (B and E–G) represent data (mean ± SEM) from “n” technical or biological replicates (donors). In (A, C, and D), one representative of “n” independent experiments is shown (mean ± SD). *P < 0.05 according to Wilcoxon signed rank sum (B, G), one‐way ANOVA with Sidak correction (C), and Student's t‐test (A); for details on data acquisition and analysis for panels (D–F) see Appendix.
Figure EV3
Figure EV3. Zymosan contains chitin as a TLR2‐stimulatory component and chitin oligomers induce distinct gene expression profiles

  1. NF‐κB activity in TLR2‐HEK293T cells using zymosan with or without o/n incubation with S. griseus chitinase or chitinase alone. Note: S. griseus chitinase contained TLR2 activity in the absence of an additional stimulant. Incubation of zymosan with this chitinase led to a reduction in this residual level of TLR2 activity.

  2. MALDI‐TOF‐MS analysis of chitinase‐treated zymosan supernatants (n = 2).

  3. Whole‐blood microarray intensities (excerpt) of the 10 genes contributing most strongly to PC1 and PC2, respectively (see Expanded View methods and Code EV1).

  4. Relative cytokine mRNAs in whole‐blood stimulation (n = 5–13).

Data information: In (A and B), one representative of two independent experiments is shown (mean ± SD). (C and D) represent data (mean ± SEM) combined from 3 to 13 donors (each box/dot represents one donor). *P < 0.05 according to one‐way ANOVA with Sidak's correction (A).
Figure 4
Figure 4. Direct chitin TLR2 binding and immune activation can be blocked by SSL3 or short chitin oligomers
  1. A, B

    Flow cytometric quantification (A) or microscale thermophoresis (MST) analysis (B) of Alexa647‐labeled C10‐15 interaction with mTLR2‐Fc protein (n = 4 each).

  2. C, D

    (C) Flow cytometry and (D) fluorescence microscopy of Candida albicans stained with control IgG or mTLR2‐Fc anti‐Fc‐Alexa594, together with ConA‐Alexa488 and CFW (n = 4 each). Scale bar = 5 μm.

  3. E, F

    (E) Docking of chitin 10 (magenta) into the TLR2 (orange) Pam2 (green) lipopeptide binding pocket (close‐up, pdb: 2z7x), (F) overlaid with SSL3 from a SSL3‐inhibited TLR2 complex (pdb: 5d3i).

  4. G

    MST analysis of Alexa647‐labeled C10‐15 and mTLR2‐Fc protein with SSL3 titration (n = 2).

  5. H, I

    (H) NF‐κB activation in TLR2‐transfected HEK293T cells (n = 3) or (I) IL‐8 release from primary PMNs (n = 2) without or with SSL3.

  6. J

    BALF TNF in C57BL/6 mice (n = 5/group) upon C10‐15 administration without or with SSL3.

  7. K

    ROS production between 0 and 45 min post‐chitin application in A. thaliana leaf pieces (n = 6).

  8. L

    MST analysis of Alexa647‐labeled C10‐15 and mTLR2‐Fc Protein with C5 titration (n = 2).

Data information: Quantification sub‐panels (A and C) and panels (K and J) represent data (mean ± SEM in A, C, and J; mean ± SD in K) combined from “n” biological replicates (human donors, mice, or plant leaves, respectively). In A–D, G–I, and L, one representative of “n” independent experiments is shown (for H and I as mean ± SD). *P < 0.05 according to Mann–Whitney U‐test (A), one‐sample t‐test (C), Student's t‐test (H, I), or one‐way ANOVA with Dunnett's multiple comparison (J).
Figure EV4
Figure EV4. TLR2 recognition depends on residues in the hydrophobic channel of the TLR2 ectodomain
  1. A

    Microscale thermophoresis analysis of Alexa647‐labeled C10‐15 and hTLR2 protein (n = 3).

  2. B–D

    Modeling (B), functional NF‐κB activity (C), and expression levels (D) in empty vector (EV) or TLR2‐transfected HEK293T cells (n = 2) for TLR2 WT or point mutants designed to narrow the hydrophobic pocket. In (B), a blue dotted line delineates the hydrophobic pocket, here shown in cross‐section. Black mesh signifies original surface profile in pdb 2z7x, and red mesh the predicted surface of the respective point mutant. In (C), dotted lines signify the levels of NF‐κB activity observed for WT TLR2 for the respective ligand. (D) Expression of TLR2 mutants verified by anti‐Flag immunoblot (n = 1).

  3. E

    Docking grid used for placing of a chitin 5‐mer.

  4. F

    Close‐up on the gatekeeper F349 which restricts access to the cavity in “closed position” (red) but allows access in “open” rotamer position (orange).

  5. G

    Docking poses of a 5‐mer (C5) in the grid specified in e and with F349 in “open” rotamer position.

  6. H

    Potential interactions of chitin 6‐mer (C6) hydrophobic methyl groups with aromatic side chains in the TLR2 hydrophobic cavity.

Data information: In (A, C, and D), one representative of “n” (given in brackets for each panel) independent experiments is shown (mean ± SD). *P < 0.05 according to two‐way ANOVA with Dunnett's correction compared to WT (top sub‐panel, indicated by dotted lines in each sub‐panel) in (C).
Figure EV5
Figure EV5. Size‐dependent immuno‐modulatory effects of oligomeric chitin
  1. A

    BALF IL‐6 in C57BL/6 mice (n = 4–5/group) upon C10‐15 administration without or with SSL3.

  2. B, C

    IL‐6 release from THP‐1 cells (B) or primary MoMacs (C) without or with C5 (n = 3 or 4).

  3. D

    Schematic representation of size‐dependent recognition of chitin oligomers by TLR2. Through the activity of chitinase or physical fragmentation, a range of different sized fragments and oligomers is generated. TLR2 recognition is possible for both particulate chitin as well as diffusible oligomers. As recognition can occur in a window where both activatory potential and solubility are high, distal recognition of fungal pathogens is possible, while the minimum length required for agonism (6 NAG) ensures that mono‐ or disaccharides are prevented from receptor activation. As activating oligomers are cleaved, soluble fragments with antagonistic properties may arise, representing an inherent negative regulatory loop.

Data information: (A and C) represent data (mean ± SEM) combined from “n” biological replicates (human donors or mice, respectively; each dot represents one biological replicate). In (B), one representative of three independent experiments is shown (mean ± SD). *P < 0.05 according to one‐way ANOVA with Dunnett's multiple comparison (A), Student's t‐test (B), or Wilcoxon signed rank sum (C) test.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. To T, Stanojevic S, Moores G, Gershon AS, Bateman ED, Cruz AA, Boulet LP (2012) Global asthma prevalence in adults: findings from the cross‐sectional world health survey. BMC Public Health 12: 204 - PMC - PubMed
    1. Brown GD, Denning DW, Gow NA, Levitz SM, Netea MG, White TC (2012) Hidden killers: human fungal infections. Sci Transl Med 4: 165rv13 - PubMed
    1. Morgulis S (1916) The chemical constitution of chitin. Science 44: 866–867 - PubMed
    1. Lee CG, Da Silva CA, Lee JY, Hartl D, Elias JA (2008) Chitin regulation of immune responses: an old molecule with new roles. Curr Opin Immunol 20: 684–689 - PMC - PubMed
    1. Mack I, Hector A, Ballbach M, Kohlhaufl J, Fuchs KJ, Weber A, Mall MA, Hartl D (2015) The role of chitin, chitinases, and chitinase‐like proteins in pediatric lung diseases. Mol Cell Pediatr 2: 3 - PMC - PubMed

Publication types

MeSH terms