The α‐defensins HNP‐1 and HNP‐2 are dominant self‐peptides presented by HLA class‐II molecules in lesional psoriatic skin

ARTICLE

Auteur(s) : Wolf-Henning Boehncke

Department of Dermatology, Johann Wolfgang Goethe-University, Theodor-Stern-Kai 7, D-60590 Frankfurt am Main, Germany.

Article accepted 09/02/2004

HLA molecules serve as peptide-binding transport and display proteins, evoking effector responses upon recognition by antigen-specific receptors of T-cells. Antigens are usually up to 24 amino acids long and bind to the HLA binding groove in an extended conformation [1]. Whereas HLA class I molecules are present on most cells of the body, class II is expressed primarily on professional antigen-presenting cells (e.g. B-cells). However, cytokine stimulation frequently results in expression of HLA class II molecules also on a wide variety of endothelial and epithelial cells, as is the case of keratinocytes in psoriasis [2]. Peptides bound to HLA class II molecules might be involved in epidermal T-cell activation which in turn seems to be an integral part of the pathogenesis of psoriasis [3, 4]. Therefore, the repertoire of HLA class II bound peptides displayed by the human keratinocyte cell line HaCaT [5] and lesional psoriatic skin was studied.

Material and methods

Cells and skin specimen

2 × 10⁷ HaCaT cells were stimulated for 48 h with recombinant human interferon-γ (1,000 U/ml). In the case of lesional psoriatic skin, a single cell suspension was obtained from a 10 cm² split skin specimen (300 µm) collected from the back of a 43 year old Caucasian woman with extensive plaque-type psoriasis. The patient had not received any systemic or phototherapies for > 4 weeks and no topical therapy except for keratolytic treatment with 10% salicylic acid in vaseline for > 1 week. The split skin specimen was obtained under local anesthesia after written informed consent was obtained.

Peptide isolation

Isolation of HLA class-II molecules was performed as previously described [6]: Briefly, cells were pelleted at 1000 g and lysed with 2% NP-40. The lysate was cleared by ultracentrifugation and submitted to affinity chromatography using the monoclonal antibody L243 (anti-DR, obtained from the American Type Culture Collection [ATCC], Rockville, MD). Elution of the HLA molecules and the bound peptides was performed using trifluoracetic acid/water, pH 2.0. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) showed > 90% purity of the released peptides, which were separated from the HLA molecules by ultracentifugation (Amicon, Danvers, MA) using a 20-kD cutoff membrane (Sartorius, Göttingen, Germany).

Identification of eluted peptides

Further analyses were performed as described previously [7]: Briefly, the peptide pools were separated by capillary reversed phase (RP)-HPLC using an RP-C₁₈ 300 µm X 150 mm column (LC-Packings, Amsterdam, The Netherlands). All collected peaks were assayed for their peptide masses and homogeneity by matrix-assisted laser desorption mass spectrometry (MALDI-MS). Subsequently, microsequencing by N-terminal Edman degradation using an ABI 476A sequencer (Weiterstadt, Germany) was performed.

Nanospray electrospray ionization mass spectrometry (ESI-MS)

For exact mass determination, ESI-MS spectrometry was performed as described previously [7] using a TSQ 7000 mass spectrometer (Finnigasn, La Jolla, Ca).

T-cell activation assays

Freshly isolated PBMCs (2 × 10⁶) were stimulated with 0.1 ng/ml staphylococcal enterotoxins A (SEA) or B (SEB) in the presence of 1-100 µg/ml HNP-1 or -2, respectively. Expression of the T-cell activation marker CD69 was determined by FACS, whereas secretion of interferon-γ, IL-2, TNF-α, IL-4, and IL-5 were determined by means of multiplexed cytometric bead arrays (CBAs).

Results

High-performance liquid chromatography (HPLC) of the material obtained by acid-elution of isolated HLA class II molecules yielded several defined fractions. Figure 1 represents the HPLC profile of the material derived from lesional psoriatic skin. Further analyses of a dominant HPLC fraction by ESI-MS shows 2 different charge states [M + 3H]³⁺ and [M + 4H]⁴⁺ of HNP-1 (1147.3, 861.3) and HNP-2 (1124.4, 843.8) indicating that these peptides have full lengths of 30 and 29 amino acids, respectively (Fig. 1).
Defensins are characterized by three intramolecular disulfide bonds. Assuming that the 3-dimensional structure preserved by these disulfide bonds remains intact, an HLA class II binding motif can be defined comprising isoleucin at position 4 and alanins at positions 6 and 9. Position 1, corresponding to the first binding pocket of HLA class II, is taken by the tyrosine21 which is positioned in line with the three other anchor positions and leads this sequence because of the complex 3-dimensional structure of HNP-1 (Fig. 2).
Occupation of the HLA peptide binding groove with HNP-1 or HNP-2 should have a profound impact on T-cell activation by either conventional antigens or superantigens. For stereometrical reasons, superantigens might no longer be capable of bridging the gap between HLA molecule and T-cell receptor (Fig. 3). To test this hypothesis, the effect of HNP-1 on T-cell stimulation with SEA and SEB was measured. Indeed, a dose-dependent inhibition on the surface expression of the T-cell activation marker CD69 as detected by FACS analyses was observed. This finding was paralleled by reduced secretion of the TH1 cytokines interferon-γ and IL-2 as well as the TH2 cytokines IL-4 and IL-5, and TNF-α, as measured by multiplexed cytometric bead arrays (CBAs). These observations document an inhibitory effect of HNP-1 on superantigen-mediated T-cell activation (Fig. 4). The respective effects were much weaker when HNP-2 was used and reached statistical significance only with regard to CD69 expression (Fig. 4).

Discussion

This report documents that human neutrophil peptides (HNP) 1 and 2 are two dominant self-peptides displayed by HLA class II molecules in lesional psoriatic skin. Since a very thin split-skin specimen from a fully established plaque-type lesion was taken, it is reasonable to assume that the vast majority of cells available for subsequent in vitro analyses were keratinocytes. Moreover, isolation of large amounts of HNP-1 and -2 from interferon-γ stimulated HaCaT cells further indicates that these peptides indeed occupy a substantial portion of HLA class II molecules expressed by keratinocytes.
HNP-1 and -2 belong to a large panel of evolutionary conserved anti-microbial molecules playing a major role in the innate host immunity against bacteria [8]. Based on the connectivity of three disulfide bonds between the six invariant cysteine residues (Fig. 2), α- and β-defensins can be defined. So far, only β-defensins were observed in human skin [9,10], whereas HNP-1 and -2 belonging to the α-defensin subfamily have been localized in neutrophils and Paneth cells of the small intestine. It is noteworthy thattheir masses in this analysis indicate unusual lengths (30 and 29 amino acids, respectively) for HLA class-II associated self-peptides.
Recently, HNP-1 and -2 were found to be dominant HLA-DR associated self-peptides on peripheral blood mononuclear cells of different tumor patients; they can interfere with peptide binding to various HLA-DR alleles [7]. This finding implies direct binding to the HLA peptide binding groove. Assuming that the 3-dimensional structure preserved by their 3 disulfide bonds remains intact, as indicated by the competition kinetics of HNP-1 and -2 [7], a binding motif allowing association with the peptide binding groove of HLA-DR can indeed be defined by molecular modeling.

The presence of large amounts of defensins is expected to not only influence the peptide repertoire presented by HLA molecules due to direct competition for binding, but also to alter superantigen-mediated T lymphocyte activation for stereometrical reasons. Bacterial superantigens are toxins that activate T lymphocytes by connecting the Vβ-chain of their receptors to the outside of an HLA class-II molecule. This mechanism is considered relevant as a trigger factor for autoimmunity in general and psoriasis in particular [11]. Occupation of the HLA peptide binding groove with HNP-1 could make it impossible for superantigens to bridge the gap between HLA molecule and the Vβ-chain of the T-cell receptor. This hypothesis is supported by the dose-dependent inhibition on the surface expression of a T-cell activation marker as well as reduced cytokine secretion.
Thus far, defensin-derived peptides have only been isolated from HLA class-II molecules of cell types expressing them upon stimulation, e.g. in the context of inflammation. Defensins can out-compete antigen [7] and block superantigen-mediated T-cell activation (as demonstrated here), thus exhibiting anti-inflammatory effects. Therefore, it can be envisioned that they may play a role in the down-regulation of inflammation.
Given the role of superantigens as trigger factors in autoimmunity in general and psoriasis in particular, blocking this mode of T-cell activation might also be feasible therapeutically. Thus far, modified superantigens have been effective only against their respective wild-type counterparts, since they prevent superantigen-binding to the respective Vβ-chain by functioning as competitive inhibitors [12]. HNP-1 exhibits a different and unspecific mode of action in as much as it widens the distance between HLA molecule and T-cell receptor Vβ-chain. Its blocking effect should therefore be independent from the Vβ-chain used by the respective superantigen. Given the distinct sets of Vβ-chains used by SEA and SEB for binding, demonstrating inhibitory effects against both of them in vitro might be a first step towards “anti-superantigens” with broader spectra. n

Acknowledgements. The excellent technical assistance of Daniela Dressel and Sandra Diehl is gratefully acknowledged. I am grateful to Falk Ochsendorf for statistical analyses. This project was supported by grant Bo 895/9 of the Deutsche Forschungsgemeinschaft.

References

1. Germain R.N. MHC-dependent antigen processing and peptide presentation: providing ligands for T lymphocyte activation. Cell 1994; 76: 287-99.

2. Gottlieb AB, Lifshitz B, Fu SM, Staiano-Coico L, Wang CY, Carter DM. Expression of HLA-DR molecules by keratinocytes, and presence of Langerhans cells in the dermal infiltrate of active psoriatic plaques. J Exp Med 1986; 164: 1013-28.

3. Baker BS, Swain AF, Griffith CEM, Leonard JN, Fry L, Valdimarsson H. Epidermal T lymphocytes and dendritic cells in chronic plaque psoriasis: the effect of PUVA treatment. Clin Exp Immunol 1985; 61: 526-34.

4. Gottlieb SL, Gilleaudeau P, Johnson R, Estes L, Woodworth TG, Gottlieb AB, Krueger JG. Response of psoriasis to a lymphocyte-selective toxin (DAB389IL-2) suggests a primary immune but not keratinocyte, pathogenetic basis. Nature Med 1995; 1: 442-7.

5. Boukamp P, Petrussevska RT, Breitkreutz D, Hornung J, Markham A, Fusenig NE. Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line. J Cell Biol 1988; 106: 761-71.

6. Halder T, Pawelec G, Kirkin AF, Zeuthen J, Meyer HE, Kalbacher H. Isolation of novel HLA-DR-restricted potential tumor antigens from the melanoma cell line FM3. Cancer Res 1997; 57: 3238.

7. Halder TM, Blggel M, Heinzel S, Pawelec G, Meyer HE, Kalbacher H. Defensins are dominant HLA-DR-associated self-peptides from CD34-peripheral blood mononuclear cells of different tumor patients (plasmacytoma, chronic myeloid leukemia). Blood 2000; 95: 2890-6.

8. Zasloff M. Antimicrobial peptides of multicellular organisms. Nature 2002; 415: 389-95.

9. Ali RS, Falconer A, Ikram M, Bissett CE, Cerio R, Quinn AG. Expression of the peptide antibiotics human β defensin-1 and human β defensin-2 in normal human skin. J Invest Dermatol 2001; 117: 106-11.

10. Harder J, Bartels J, Christophers E, Schröder JM. A peptide antibiotic from human skin. Nature 1997; 387: 861.

11. Boehncke W-H, Dressel D, Zollner TM, Kaufmann R. Pulling the trigger on psoriasis. Nature 1996; 379: 777.

12. Boehncke W-H, Hardt-Weinelt K, Nielsson H, Wolter M, Dohlsten M, Ochsendorf F-R, Kaufmann R, Antonsson P. Antagonistic effects of the staphylococcal enterotoxin A mutant SEAF47A/D227A, on psoriasis in the SCID-hu xenogeneic transplantation model. J Invest Dermatol 2001; 116: 596-601.