Arch Dermatol Res (2006) 298:201205 DOI 10.1007/s00403-006-0688-7SHORT COMMUNICATION
Integrin 21 deWciency does not aVect contact hypersensitivity
Manon C. Zweers Lisa Siewe Claudia Wickenhauser Thomas Krieg Axel Roers Beate Eckes
Received: 3 March 2006 / Revised: 7 July 2006 / Accepted: 11 July 2006 / Published online: 5 August 2006 Springer-Verlag 2006
Abstract Collagens in the extracellular matrix arethought to play an important role in regulating inXam-matory responses by aVecting cell adhesion and migra-tion. The contact between collagens and cells isestablished mainly by 11, 21 and 111integrinreceptors. Here, we analyzed the contact hypersensitiv-ity (CHS) reaction in mice that were genetically deW-cient in the collagen receptor 21. Integrin 21 iswidely expressed and has been suggested to play animportant role in mediating inXammatory responses.CHS was induced by applying dinitroXuorobenzene toabdominal skin and challenging with the same reagenton ear skin. Macroscopically and histologically, earswelling in 21-deWcient mice did not diVer from thatin wild-type control mice. Immunohistological detec-tion of inWltrated T lymphocytes, neutrophils and mastcells in inXamed ear skin revealed similar numbers incontrols and integrin 21-deWcient animals. Ourresults suggest that the adhesive functions of integrin21 are dispensable for the CHS response; they maybe compensated for by the collagen receptor 11 orother collagen receptors.
Keywords InXammation T lymphocyte Extracellular matrix Collagen Knock out
The extracellular matrix (ECM) is composed of diVer-ent macromolecules including the families of collagens,laminins, Wbronectin, small and large proteoglycansand many others . Besides giving mechanicalstrength to tissues and serving adhesive functions asscaVold for embedded cells, it provides cues to cellsfacilitating motility and directed migration . ManyECM molecules can modulate speciWc cell functions,such as survival, motility and diVerentiation. Thenature of these responses depends on the cell type andpattern of receptors for ECM molecules expressed bythese cells. Most matrix receptors are transmembraneheterodimers belonging to the family of 1 integrins [2,12]. In skin, collagens are the most abundant type ofECM molecules. They are recognized speciWcally bythree distinct collagen-binding integrins, 11, 21and 111, while no expression of 101 has beendetected in the skin [1, 21]. As collagens are so abun-dant in connective tissues of many organs, the colla-gen-binding receptors have recently received attentionin playing an important role as guidance cues for tumorcells in metastatic progression [9, 18] and in immunereactions [4, 5]. Thus it is thought that once tumor andinXammatory cells have left the vascular system, theytraverse through basement membranes and over longdistances through extravascular spaces utilizing inte-grin-mediated cellcollagen contacts.
Cells of the immune system express 11 and 21to varying levels, depending on the activation state.Monocytes express weak 11 and moderate 21 lev-els with 11 increasing following LPS or IFN- stimu-lation [10, 17]. Neutrophils display 21 uponextravasation, facilitating locomotion through extra-vascular tissue . While naive T cells are devoid ofcollagen integrin receptors, T lymphocytes express
M. C. Zweers L. Siewe T. Krieg A. Roers B. Eckes (&)Department of Dermatology, University of Cologne, Kerpener Str. 62, 50937 Cologne, Germanye-mail: email@example.com
C. WickenhauserDepartment of Pathology, University of Cologne, Cologne, Germany123
202 Arch Dermatol Res (2006) 298:20120511 and 21 some time after activation; hence theywere called very late antigen (VLA) . So far, noexpression of the other collagen-binding integrins,101or 111, has been observed in lymphatic organs, corresponding with their limited expression in car-tilage  or mesenchymal non-muscle tissues .
The function of 11 and 21 in inXammatoryresponses has been addressed in several models usingeither neutralizing antibodies or mice with a geneticdeletion of 11 receptors. These studies have pointedtowards an important role for both 11 and 21 inmediating inXammatory T cell responses . There-fore, we analyzed the T cell-mediated contact hyper-sensitivity (CHS) response using mice with aconstitutive ablation of the integrin 2 subunit due totargeted disruption of the ITGA2 gene .
Integrin 2-deWcient and control mice on a pureC57BL/6 background were sensitized to the contactallergen dinitroXuorobenzene and challenged 6 dayslater on the ear. The inXammatory response was quan-tiWed by daily measurement of ear thickness (Fig. 1).For both experimental groups, the ear-swellingresponse peaked 48 h after challenge and slowlydeclined thereafter. No signiWcant diVerences betweencontrol and 21-null animals were observed at anytime.
As the tissue swelling does not necessarily reXectpotential diVerences in recruitment of individual popu-lations of leukocytes, biopsies of the challenged earswere processed for histological analysis. Immunohisto-chemical staining of T cells using an anti-CD3 antibodyshowed T cell inWltration at 48 h after challenge in earsfrom both wild-type and 2-deWcient animals, while noT cell inWltrate was observed in ears treated with sol-vent (Fig. 2a, b). Quantitative evaluation of T cells didnot reveal signiWcant diVerences between the wild-typeand knock out animals, indicating that lack of 21integrin does not impair T cell inWltration into the chal-lenged skin (0.04 cells/m2 in 21-null vs 0.04 cells/m2 in control).
As integrin 21 is also expressed by other leuko-cytes that contribute to inXammatory responses, 21deWciency may aVect the recruitment of these cell typesin CHS. Therefore, we studied the inWltration of mac-rophages by immunostaining for F4/80 (Fig. 2c, d) andof mast cells (Fig. 2e, f) and neutrophils (Fig. 2g, h)employing chloroacetate esterase and Giemsa staining. However, the numbers of these cells were notaltered in challenged ears of 21 integrin-deWcientmice when compared with control animals.
In addition, we subcutaneously injected 21-nulland control mice with unmethylated CpG dinucleo-tides, as a model of bacterial infection. Unmethylated
CpG dinucleotides are known to induce humoral andcellular responses, which involve T cells as well asmonocytes, macrophages and neutrophils . Alsousing this model, no obvious diVerences in inXamma-tory cell inWltration were observed between 21 inte-grin-deWcient and control mice (data not shown).Furthermore, during the inXammatory phases of cuta-neous wound repair, no diVerence in the number ofinWltrating neutrophils and macrophages was observed.
The Wnding that the deWciency of 21 integrin inour mice had no impact on the CHS reaction or CpG-induced inXammation was surprising because impor-tant roles for integrin-mediated interactions of T cellswith collagen have been demonstrated [3, 5]. Whileapparently not relevant for the activation of nave Tcells in the secondary lymphoid organs, the adhesion tocollagen Wbrils of the ECM in peripheral tissues wasshown to have substantial eVects on the interactions ofdiVerentiated eVector T cells with antigen-presentingcells. By slowing down the migration of the T cell,adhesion to collagen may ensure suYciently long
Fig. 1 No diVerences in ear swelling response in 21-null andcontrol mice. Mice carrying a constitutive deletion of the integrin2 subunit were generated as described . Animals were back-crossed for seven generations onto the C57BL/6 background in aspeciWc pathogen-free facility and all protocols used were ap-proved by the local veterinary authorities. Wild-type and integrin2-deWcient males, 35 months of age derived from heterozygousbreedings, were sensitized by painting 25 l of 0.5% (v/v) dinitro-Xuorobenzene (DNFB, Sigma-Aldrich, Deisenhofen, Germany)in acetone/olive oil (4:1, v/v) onto the shaved belly on day 0. Ani-mals were challenged 6 days later by application of 20 l of 0.2%DNFB to both sides of the right ear, while the left one was left un-treated or received treatment with solvent only. Ear swelling wasmeasured using an engineers precision caliper before challengeand for 4 days thereafter. Results are shown as percentage in-crease in ear thickness SEM. Groups of at least 10 animals pertime point were used (days 1 and 2: wild type n = 19, 2/n = 16; day 3: wild type n = 13, 2/ n = 11; day 4: wild typen = 11, 2/ n = 10). Black bars 2-null mice, white bars controlmice123
Arch Dermatol Res (2006) 298:201205 203Fig. 2 No alterations in inWltration of inXammatory cells in inX-amed ears of 2-null and control mice. Mice were sacriWced 24 or48 h after challenge and ears were either frozen unWxed in opti-mal cutting temperature compound (OCT; Sakura, Torrance,CA, USA) or Wxed for 4 h in 4% paraformaldehyde solution be-fore paraYn embedding. For immunoXuorescence, frozen sec-tions were blocked with normal goat serum (DakoCytomation,Hamburg, Germany) before application of rat-anti-CD3 (Chem-icon, Temecula, CA, USA) for staining of T cells. Alexa 488 nm-conjugated goat anti-rat (Molecular Probes, Eugene, OR, USA)was applied (green) and sections were counterstained with propi-dium iodide (red) and embedded in mounting medium. No diVer-ences in number of T cells per mm2 were observed (a, b). Note
that this antibody also stains dendritic epidermal T cells. Forimmunohistochemistry, sections were incubated with rat anti-mouse F4/80 (MCA497GA, Serotec, Berlin, Germany) for immu-nostaining of macrophages, followed by biotin-conjugated rabbitanti-rat Ig (DakoCytomation). Then, streptavidine horseradishperoxidase was added and aminoethylcarbazol was used as chro-mogenic substrate (DakoCytomation). Macrophage inWltrationinto inXamed ears was similar in 2-null and control mice (c, d).Neutrophil and mast cell inWltration was similar in 2-null andcontrol mice as visualized by naphthol-AS-D-chloroacetate ester-ase (e, f) and Giemsa staining (g, h) according to standard proto-cols . Scale bar = 100 m for ad and 200 m for eh 123
204 Arch Dermatol Res (2006) 298:201205phases of physical contact between the two cells allow-ing eVective T cell activation. In vitro experimentsshowed that activation of eVector T cells by anti-CD3antibodies was signiWcantly enhanced in the presenceof collagen . Integrins 11 and 21, absent fromnave T cells but expressed upon T cell activation, arethe receptors which mediate the interaction of eVectorT cells with collagens. Consistent with the eVects ofcollagen binding on T cell activation described above,inhibition of CHS and delayed type hypersensitivitywas observed upon blocking integrins 11 and 21by speciWc antibodies in vivo . Blocking 21 priorto allergen challenge signiWcantly reduced the earswelling. In addition, the diminished ear swellingresponse in integrin 11 knock out mice could be fur-ther reduced by the administration of anti-2 antibod-ies. The authors suggested that inhibition of CHSresponse was due to reduced leukocyte activation andmigration .
In our mice with a genetic ablation of the 2 integringene, however, we do not Wnd a diminished CHSresponse. The homing of leukocytes to inXamed earswas not altered. This Wnding conXicts with the resultsof studies applying anti-21 antibodies and adds tothe list of results from knock out mice that are at dis-cordance with results from neutralizing antibodies,both aVecting the same protein . In fact, we haveseen that genetic deletion of 21 in our mice elicits aneVect on the vasculature (Zweers and Eckes, unpub-lished observation) that diVers signiWcantly from theeVects of neutralizing anti-21 antibodies described inan earlier study . Our results are in agreement withthe observation that inWltration of polymorphonuclearleukocytes (PMNs) was not altered in a model ofexperimental peritonitis  or in wound healing exper-iments in our 21-null mice (Zweers and Eckes,unpublished observation). Therefore, it must beassumed that the anti-2-blocking antibodies haveundesired activities, which do not involve 21colla-gen interactions.
The most likely explanation for our Wnding ofunaVected CHS in 2 deWciency is that the loss of21 integrin function can be compensated for by11 integrin. Alternatively, T cells might not requirecollagen-binding integrins for migration at all, as ithas been shown that T cells can migrate through col-lagen lattices in an integrin-independent mannerusing transient interactions with collagen Wbers andwithout transforming the focal adhesion complexes. An analysis of T cell function in mice with a deW-ciency for both receptors, the 11 and the 21 inte-grins, should provide answers to these questions inthe future.
Acknowledgments We thank Dr. Ingo Haase for stimulatingdiscussion and Nicole Brggenolte and Silke Kummer for excel-lent technical assistance. This study was supported by the Deut-sche Forschungsgemeinschaft through the SFB 589 at theUniversity of Cologne.
1. Bengtsson T, Aszodi A, Nicolae C, Hunziker EB, Lundgren-Akerlund E, Fassler R (2005) Loss of alpha10beta1 integrinexpression leads to moderate dysfunction of growth platechondrocytes. J Cell Sci 118:929936
2. Brakebusch C, Fassler R (2005) Beta 1 integrin function in vi-vo: adhesion, migration and more. Cancer Metastasis Rev24:403411
3. de Fougerolles AR, Sprague AG, Nickerson-Nutter CL, Chi-Rosso G, Rennert PD, Gardner H, Gotwals PJ, Lobb RR,Koteliansky VE (2000) Regulation of inXammation by colla-gen-binding integrins alpha1beta1 and alpha2beta1 in modelsof hypersensitivity and arthritis. J Clin Invest 105:721729
4. de Fougerolles AR (2003) Integrins in immune and inXam-matory disease. In: Gullberg D (ed) I domains in integrins.Landes Bioscience/Eurekah.com. Georgetown, pp 165184
5. Dustin ML, de Fougerolles AR (2001) Reprogramming Tcells: the role of extracellular matrix in coordination of T cellactivation and migration. Curr Opin Immunol 13:286290
6. Edelson BT, Li Z, Pappan LK, Zutter MM (2004) Mast cell-mediated inXammatory responses require the alpha 2 beta 1integrin. Blood 103:22142220
7. Even-Ram S, Yamada KM (2005) Cell migration in 3D ma-trix. Curr Opin Cell Biol 17:524532
8. Friedl P, Entschladen F, Conrad C, Niggemann B, Zanker KS(1998) CD4+ T lymphocytes migrating in three-dimensionalcollagen lattices lack focal adhesions and utilize beta1 inte-grin-independent strategies for polarization, interaction withcollagen Wbers and locomotion. Eur J Immunol 28:23312343
9. Friedl P, Wolf K (2003) Tumour-cell invasion and migration:diversity and escape mechanisms. Nat Rev Cancer 3:362374
10. Hemler ME (1990) VLA proteins in the integrin family:structures, functions, and their role on leukocytes. Annu RevImmunol 8:365400
11. Holtktter O, Nieswandt B, Smyth N, Muller W, Hafner M,Schulte V, Krieg T, Eckes B (2002) Integrin alpha 2-deWcientmice develop normally, are fertile, but display partially defec-tive platelet interaction with collagen. J Biol Chem277:1078910794
12. Hynes R (2002) Integrins. Bidirectional, allosteric signalingmachines. Cell 110:673687
13. Hynes RO (2002) A reevaluation of integrins as regulators ofangiogenesis. Nat Med 8:918921
14. Kleman JP, Giry-Lozinguez C, van der Rest M (1995) Diver-sity and modularity of extracellular matrix macromolecules atthe gene and protein levels. J Hepatol 22(Suppl):39
15. Krieg AM (2003) CpG motifs: the active ingredient in bacte-rial extracts? Nat Med 9:831835
16. Leder LD (1979) The chloroacetate esterase reaction. A use-ful means of histological diagnosis of hematological disordersfrom paraYn sections of skin. Am J Dermatopathol 1:3942
17. Rubio MA, Sotillos M, Jochems G, Alvarez V, Corbi AL(1995) Monocyte activation: rapid induction of alpha 1/beta 1(VLA-1) integrin expression by lipopolysaccharide and inter-feron-gamma. Eur J Immunol 25:27012705
18. Senger DR, Perruzzi CA, Streit M, Koteliansky VE, de Foug-erolles AR, Detmar M (2002) The alpha(1)beta(1) and123
Arch Dermatol Res (2006) 298:201205 205alpha(2)beta(1) integrins provide critical support for vascularendothelial growth factor signaling, endothelial cell migra-tion, and tumor angiogenesis. Am J Pathol 160:195204
19. Tiger CF, Fougerousse F, Grundstrom G, Velling T, GullbergD (2001) Alpha11beta1 integrin is a receptor for interstitialcollagens involved in cell migration and collagen reorganiza-tion on mesenchymal nonmuscle cells. Dev Biol 237:116129
20. Werr J, Johansson J, Eriksson EE, Hedqvist P, Ruoslahti E,Lindbom L (2000) Integrin alpha(2)beta(1) (VLA-2) is aprincipal receptor used by neutrophils for locomotion inextravascular tissue. Blood 95:18041809
21. White DJ, Puranen S, Johnson MS, Heino J (2004) The colla-gen receptor subfamily of the integrins. Int J Biochem CellBiol 36:14051410123
Integrin a2b1 deWciency does not aVect contact hypersensitivityAbstractReferences
/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 150 /GrayImageDepth -1 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 600 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org?) /PDFXTrapped /False
/Description >>> setdistillerparams> setpagedevice