Differential Expression of Activation Epitopes of β1 Integrins inPsoriasis and Normal Skin

Pablo F. Penas, Manuel Gomez,* Guadalupe F. Buezo, Luis Rios, Marıa Yanez-Mo,* Carlos Cabanas,†Francisco Sanchez-Madrid,* and Amaro Garcıa-DıezDepartments of Dermatology and * Immunology, Hospital Universitario de la Princesa, Universidad Autonoma de Madrid, Spain; †Department of Biochemistryand Molecular Biology, Facultad de Medicina, Universidad Complutense de Madrid, Spain

In the epidermis, β1 integrin expression is normallyconfined to the basal layer; however, suprabasal expres-sion of β1 integrins in keratinocytes has been found inpsoriasis, and it has been suggested that it could be apathogenic factor of the disease. We have investigatedherein the functional state of β1 integrins of humankeratinocytes in normal skin and psoriasis. The expressionof β1-activation-reporter epitopes was monitored withtwo monoclonal antibodies, HUTS-21 and MG5A7, thatrecognize epitopes whose expression parallels functionalactivity of β1 integrins and correlates with the ligandbinding activity of these heterodimeric glycoproteins.We have found that keratinocytes express activationepitopes of β1 integrins, and that these epitopes can be

In the epidermis, β1 integrin expression is normally confinedto the basal layer; however, there are several situations in whichsuprabasal, differentiating keratinocytes express β1 integrins.This has been described during wound healing (Hertle et al,1992; Jones et al, 1993), and in several cutaneous diseases,

including psoriasis (Horrocks et al, 1991; Ralfkiaer et al, 1991; Hertleet al, 1992; Pellegrini et al, 1992; Giannelli et al, 1994), eczema, andlichen planus (Ralfkiaer et al, 1991).

Human keratinocytes express several β1 integrins that include: α2β1,a collagen receptor; α3β1, a laminin and collagen receptor; and α5β1,a receptor for fibronectin (Guo et al, 1991; Watt and Hertle, 1994).In keratinocytes, β1 integrins not only mediate cell adhesion andmigration, but also regulate stratification and the initiation of terminaldifferentiation (Watt and Hertle, 1994), and it is a marker of stem cells(Jones and Watt, 1993). It has been shown that suprabasal integrinexpression is not a direct consequence of inflammation because it isnot induced by intradermal injections of cytokines (Hertle et al, 1995),and it has been suggested that suprabasal expression of integrins couldbe a pathogenic factor in psoriasis (Carroll et al, 1995).

Integrin function is regulated at two different levels: selectiveexpression (Guo et al, 1991; Kim et al, 1992) and modulation of itsbinding properties (Schwartz et al, 1995; Sanchez-Mateos et al, 1996).Although most studies have focused on the regulation of expression,recently new antibodies that recognize the activated state of integrins

Manuscript received August 27, 1997; revised February 24, 1998; acceptedfor publication March 10, 1998.

Reprint requests to: Dr. Pablo F. Penas, Servicio de Dermatologıa, HospitalUniversitario de la Princesa, Diego de Leon 62, 28006 Madrid, Spain.

Abbreviation: NHK, normal human keratinocytes.

0022-202X/98/$10.50 · Copyright © 1998 by The Society for Investigative Dermatology, Inc.

19

modulated by manganese. The expression of activationepitopes of β1 integrins was related to an enhancedadhesion to fibronectin and collagen. Immunohisto-chemical studies of normal and psoriatic skin withHUTS-21 and other monoclonal antibodies indicatethat, although there is suprabasal expression of β1integrins in psoriasis, these molecules seem to be in aninactive state. Moreover, most β1 integrins in lateral andapical surfaces of basal keratinocytes of psoriasis arealso in a nonactive conformation, implying a decreaseof activity compared with normal skin, in which activeβ1 integrins are distributed all over the basal keratino-cytes. Key words: activation-dependent epitope/HUTS-21/keratinocyte. J Invest Dermatol 111:19–24, 1998

allow the study of the modulation of their binding properties (Lenteret al, 1993; Miyake et al, 1994; Bazzoni et al, 1995; Mould et al, 1995;Yednock et al, 1995). For the β1 family the monoclonal antibody(MoAb) HUTS-21 has been generated and characterized. The epitoperecognized by this MoAb is contained within the sequence 355–425of the β1 chain. Its expression parallels functional activity of VLAintegrins and correlates with the ligand binding activity of theseheterodimeric glycoproteins (Luque et al, 1996).

In order to assess the functional state of human keratinocytes innormal and diseased skin, we have studied herein the expression andregulation of β1 activation-reporter epitopes in cultured keratinocytesin normal skin and psoriasis.

MATERIALS AND METHODS

Keratinocytes Normal human keratinocytes (NHK) were cultured, with fewmodifications, as previously described (Jones and Watt, 1993; Parenteau, 1994).In brief, epidermal strips were isolated from the skin of cosmetic surgery using1.2 U dispase per ml (Boehringer, Mannheim, Germany) for 24 h at 4°C.After incubation of epidermal strips in 0.05% trypsin/0.02% ethylenediaminetetraacetic acid for 25 min at room temperature, keratinocytes were seededonto plastic dishes at 40,000 cells per cm2. The culture medium consisted ofone part Ham’s F-12 medium and three parts Dulbecco’s modification of Eagle’smedium (DMEM, GIBCO, Scotland) supplemented with 10% fetal calf serum,1.8 3 10–4 M adenine (Sigma, St. Louis, MO), 0.5 µg hydrocortisone per ml(Sigma), 5 µg insulin per ml (Boehringer), 10 ng epidermal growth factor perml (ICN Biochemicals, Irvine, CA), 5 µg transferrin per ml (Sigma), and 10 ngcholera toxin per ml (ICN Biochemicals). The medium was replaced everyother day and cells were passed when almost confluent. Cultures were usedbetween the first and fourth passages.

Antibodies MoAb used in this study directed to the common β1 integrinchain have previously been described: the activating TS2/16 (CD 29) (Hemleret al, 1984), the activation-reporter HUTS-21 (Luque et al, 1996), and the

20 PENAS ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

adhesion-blocking Lia1/2 and VJ1/14 (Campanero et al, 1992). MG5A7 is anew anti-β1 MoAb generated by immunizing mice with human umbilical veinendothelial cells, which recognizes an epitope on β1 integrins that appears uponactivation of β1 integrins with different stimulatory agents, like Mn21 cations.The P3X63 mouse myeloma protein (IgG1, kappa) was used as negative control.The HP2/9 anti-CD44 MoAb has been previously described (de la Heraet al, 1989).

Flow cytometry Keratinocytes were detached with 0.05% trypsin and 0.02%ethylenediamine tetraacetic acid, counted, and aliquoted into separate tubes(100,000 cells per tube). Cells were washed in DMEM or HEPES/NaCl buffer(20 mM HEPES, 150 mM NaCl, 2 mg D-glucose per ml, pH 7.4) andresuspended in DMEM or HEPES/NaCl buffer containing the appropriateconcentrations of divalent cations or stimuli. The corresponding MoAb wasadded and cells were incubated for 15 min at 37°C. After this incubation, cellswere washed in phosphate buffered saline (PBS) and incubated with 100 µl of1:100 dilution of fluorescein-isothiocyanate-conjugated sheep anti-mouse IgGsecondary antibody (DAKO) for 20 min at 4°C. Cells were washed andresuspended in PBS with 10 µg propidium iodide per ml and then analyzed ina Becton Dickinson FACScan flow cytometer. Cell analysis were gated bothon forward- and size-scatter intensities and propidium iodide fluorescence todiscard dead cells.

Immunofluorescence digital confocal microscopy Keratinocytes wereplated in complete medium on glass coverslips and grown until almost confluent.Cells were fixed in 3.7% paraformaldehyde in PBS. Coverslips were thenincubated with the primary antibody for 30 min at 37°C, washed, and incubatedwith the fluorescein-isothiocyanate-conjugated sheep anti-mouse IgG secondaryantibody (DAKO) at 1:100 for 30 min at 37°C. Specimens were examinedwith an MRC 1024 Bio-Rad confocal system (Bio-Rad Richmond, CA)mounted on a Zeiss Axiovert 135 microscope (Zeiss, Oberkochen, Germany)equipped with a 363, 1.4 NA planapochromat objective. Images were storedon magnetoptical disks.

Cell attachment assays Cell adhesion assays were essentially performed aspreviously described (Luque et al, 1996). Briefly, 96 well flat-bottomed plates(Titertek, ICN Biochemicals) were coated with the β1 integrin ligands type 1collagen (ICN Biochemicals) or fibronectin (Sigma) at the indicated concentra-tions by overnight incubation at 4°C. After saturation of free plastic sites ofwells with 1% boiled bovine serum albumin in PBS for 1 h at room temperature,plates were washed three times with PBS and once with 200 µl of HEPES/NaCl buffer. Fifty microliters of HEPES/NaCl buffer containing 2.5 µg ofpurified MoAb and/or 2 mM MnCl2 were then added to the wells; 2.5 3 105

keratinocytes in 50 µl of HEPES/NaCl buffer were added to each well andallowed to sediment onto the bottom of the wells for 10 min at 4°C. Plateswere then transferred to a CO2 incubator and incubated for 30 min at 37°C.The percentage of cells that remained adhered after three gentle washes of thewells with 200 µl of warm HEPES/NaCl buffer was calculated by measuringthe absorbency of the wells at 540 nm after fixation and staining with 0.5%crystal violet in 20% methanol.

Immunohistochemistry of cutaneous biopsies Eight patients withpsoriasis were involved in the study after obtaining an informed consentapproved by the Ethical Committee of the Hospital Universitario de la Princesa.Normal skin from excisional surgery from six patients served as controls. Abiopsy specimen was obtained from representative areas on extremities underlocal anesthesia with 2% mepivacaine hydrochloride. All biopsy specimens weresnap frozen embedded in OCT (Miles, Naperville, IL) in liquid nitrogen andserially sectioned using a rotary cryostat microtome. Cryostat-cut unfixedsections were mounted on glass slides, covered with acetone for 10 min, driedat room temperature and stored at –70°C.

Immunohistochemical staining of psoriatic and normal skin was performedin parallel on frozen sections using a two-step immunoperoxidase technique asdescribed (Fernandez-Herrera et al, 1989). The sections were incubated in ahumidity chamber with the primary monoclonal antibody for 45 min. Slideswere rinsed in 1 M Tris-HCl buffer pH 7.6, for 10 min and then coveredwith rabbit-anti-mouse immunoglobulin conjugated to horseradish peroxidase(DAKO). After two washes in Tris buffer, enzyme staining was developed for10 min with diaminobenzidine and H2O2. The staining reaction was stoppedby washing the slides in Tris buffer. Finally, they were counterstained withhematoxylin for 1 min, rinsed in tap water, dehydrated, and mounted.

Sections of skin and keratinocytes in culture were examined by twoindependent observers. The extent and intensity of staining for each MoAbwere recorded and photographs were taken.

RESULTS

Expression of activation-reporter epitopes of β1 integrin inkeratinocytes We explored whether the expression of the activation-

Figure 1. Influence of manganese on the expression of activationepitopes of β1 integrins in NHK. (a) Flow cytometry profiles of theinfluence of Mn21 10 mM in the expression of HUTS-21. (b) Expression ofactivation epitopes HUTS-21 and MG5A7 correlated to the concentrationof Mn21 in the assay medium. NHK were resuspended in DMEM containingthe appropriate concentrations of Mn21. The corresponding MoAb was addedand cells were incubated at 37°C. The data are given as the percentage offluorescence intensity with respect to basal expression. Each point is the mean6 SEM. The number of experiments is given in parentheses.

dependent conformational epitopes of β1 were regulated in NHK.The expression of β1 was monitored by using TS2/16 and Lia1/2MoAb, which recognize constitutively expressed epitopes, and HUTS-21 and MG5A7 MoAb, specific for activation-related epitopes. WhenNHK were incubated at 4°C, the activation β1 epitopes were notdetected, whereas a high expression of TS2/16 and Lia1/2 epitopeswas observed. Incubation at 37°C showed a significant expression ofthe activation β1 epitopes, whereas TS2/16 and Lia1/2 expression wasnot altered. Furthermore, the expression of β1-activation epitopes wasalso determined on keratinocytes under conditions of divalent cationsdescribed to change β1 integrins to a high affinity state. Mn21 hasbeen described as one of the most important activating cations of β1integrins (Luque et al, 1996). We found an enhanced expression of theactivation epitopes detected by HUTS-21 and MG5A7 (Fig 1a). Theinduced expression of the activation epitopes correlated with theconcentration of Mn21 used (Fig 1b).

Although Mn21 has been described to enhance adhesion of Tlymphoblasts to fibronectin and type I collagen (Luque et al, 1996),we performed adhesion studies to explore if the increased expressionof HUTS-21 induced by Mn21 was correlated by changes in theadhesion of keratinocytes through β1 integrins (Fig 2). We found thatboth TS2/16 and Mn21 greatly enhance the adhesion of NHK tofibronectin (0.5–1 µg per ml) and type I collagen (0.25–0.5 µg per

VOL. 111, NO. 1 JULY 1998 β1-INTEGRIN ACTIVITY IN PSORIASIS 21

Figure 2. Manganese stimulates the adhesion of keratinocytes tofibronectin and type I collagen through β1 integrins. The dependency ofthis adhesion on VLA integrin function is demonstrated by the blockade withthe inhibitory anti-β1 MoAb VJ1/14. Incubation with the activating anti-β1TS2/16 MoAb was used as a positive control. Control with lymphoblasts wascarried out in parallel (data not shown). NHK were seeded onto plastic wellscoated with fibronectin at 0.5 µg per ml and type I collagen at 0.25 µg per mlcontaining 2.5 µg of purified MoAb and/or 2 mM MnCl2. One representativeexperiment performed in triplicate is shown. The data are given as the percentageof absorbency with respect to control.

ml). We also found that the MoAb VJ1/14, a very potent inhibitory anti-β1 MoAb, almost reversed the enhanced adhesion induced by Mn21.

Localization of activated β1 integrins on cultured keratino-cytes Immunofluorescence digital confocal microscopy studies ofcultured NHK showed different patterns of expression with the anti-β1 antibodies studied. Both TS2/16 and Lia1/2 showed fluorescencein all four optical sections taken at a pitch of 0.5 µm indicatingexpression all over the cell membrane, with an enhancement of thefluorescence in the intercellular surfaces (Fig 3a–d).

In contrast, activated epitopes on β1 integrins assessed by HUTS-21 and MG5A7 MoAb showed little intercellular staining. Conspicuouslinear arrangement of fluorescence in the basal surface of the cell wasfound in most keratinocytes with both antibodies (Fig 3e, f ). Nostaining was found in other optical sections of the cell above the basalsurface. The negative control P3X63 did not show any relevant staining.

Expression of activated β1 integrins in normal skin andpsoriasis We found that in normal skin, β1 integrins, evaluated byTS2/16 and Lia1/2, were present only in the basal cell layer ofkeratinocytes, and these cells were stained throughout the cell withconspicuous staining in the membrane (Fig 4a, b). Dermal endotheliumshowed a strong reactivity similar to the basal cell layer.

HUTS 21 and MG5A7 antibodies stained only the basal cell layerof normal skin, and keratinocytes were stained all over the cell, bothat the membrane and at the cytoplasm (Figs 4c, d, 5a, b). Capillariesand other endothelia showed similar staining to that found on theepidermis.

In psoriatic skin, there was a strong expression of β1 integrins inthe basal cell layer, predominantly on the membrane. A consistentstaining could be found in keratinocytes of suprabasal layers withTS2/16 and Lia1/2 antibodies but only in the tips of dermal papillae(Fig 4e, f ). The endothelium of capillary vessels showed a similarstaining to that found in the basal cell layer.

Activation β1 epitopes were detected consistently only in the dermo–epidermal junction, with scant or no staining in the rest of the cell(Figs 4g, h, 5d, e). Endothelium showed a stronger staining, whichwas similar to what we found in normal skin.

As a control for staining intensity we included the CD44 hyaluronatereceptor (Fig 5c, f ). A comparable CD44 staining pattern was foundin both normal and psoriatic skin.

DISCUSSION

Integrins are important molecules in cell–cell and cell–extracellularmatrix interactions and regulate cell functions and behavior in many

Figure 3. Localized expression to the basal surface of the cell ofactivation epitopes of β1 integrins in cultured keratinocytes.Immunofluorescence digital confocal microscopy of cells with TS2/16 (a–d)and HUTS 21 (e, f). For each field, four optical sections from the cell–matrixcontact area to the top were obtained at a pitch of 0.5 µm. Only the first twooptical sections of HUTS 21 are shown due to the absence of fluorescence inthe others.

cell types (Schwartz et al, 1995). Cellular adhesive properties areregulated through the selective expression of integrins (Guo et al, 1991)as well as by the modulation of the binding properties of thesereceptors. The avidity of integrins for their ligands is not constant andis dynamically regulated (Sanchez-Mateos et al, 1996). The existenceof different conformations of β1 integrins in vivo has been inferredfrom both cell adhesion studies with keratinocytes and the observationsrecently described on the expression of activated β1 integrins in vitro(Kim and Yamada, 1997); however, a direct assessment of the expressionof activated β1 in normal and diseased skin has not been explored sofar. The use of the HUTS-21 and MG5A7 MoAb, which detectactivation-dependent epitopes of β1 integrins (Luque et al, 1996), hasfacilitated the study of activated β1 expression both in vivo and in vitro.These MoAb recognize epitopes on a regulatory region of the commonβ1 subunit of VLA integrins whose expression parallels functionalactivation and ligand binding by these adhesion receptors (Luque et al,1996; Gomez et al, 1997).

We have found that keratinocytes express activation epitopes of β1integrins, we have shown that these epitopes can be modulated bymanganese, an inducer of adhesion of keratinocytes (Hotchin et al,1993) and a potent inducer of activation epitopes in β1 integrins(Luque et al, 1996), and we have shown that these epitopes are relatedto adhesion to fibronectin and type I collagen. These data are inagreement with previous reported observations with T lymphocytes(Luque et al, 1996). In cultured NHK, our results showed that, whereas

22 PENAS ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Figure 4. Different pattern of expression of β1 integrins in normal and psoriatic skin. Immunohistochemical study of the expression in normal skin (a–d)and psoriasis (e–h) of TS2/16 (a, e), Lia1/2 (b, f), HUTS-21 (c, g), and MG5A7 (d, h). Suprabasal expression of TS2/16 and Lia1/2 is found in psoriasis whereasHUTS 21 and MG5A7 only stain the dermo–epidermal junction. Scale bar: 50 µm; inserts, 20 µm.

β1 integrins can be found all over the keratinocyte membrane,activation epitopes appear mainly in a linear arrangement on the basalsurface of cells where focal adhesions are formed (Kim and Yamada,1997). This distribution could be the reflection of the interactions withextracellular matrix ligands, similar to what has been described inhuman lymphocytes (Sanchez-Mateos et al, 1996; Gomez et al, 1997).Under physiologic conditions, HUTS-21 detects a ligand inducedbinding site epitope on β1 (Gomez et al, 1997). In addition, stimulatingagents acting from without the cell, such as Mn21 and activating TS2/16 MoAb, can impose a similar conformation to that induced by ligandoccupancy.

Expression of β1 integrin has been related to the proliferative stateof keratinocytes (Jones and Watt, 1993; Bata-Csorgo et al, 1995).Parallel to accelerated proliferation, the β1 integrin expression becamevery bright in vitro. As the proliferation slowed down on confluentcultures, the intensity diminished (Bata-Csorgo et al, 1995). Moreover,β1 integrins are downregulated when keratinocytes initiate terminaldifferentiation (Jones and Watt, 1993; Hodivala and Watt, 1994).Nevertheless, keratinocytes in vitro have a high proliferative rate(Rheinwald and Green, 1977) similar to psoriatic epidermis, andexpress β1 integrins together with several markers of differentiation(peanut agglutinin, involucrin, K1/K10 keratins) (Hodivala and Watt,1994; Bata-Csorgo et al, 1995). Transgenic mice with suprabasalexpression of integrins have shown features reminiscent of chronic

plaque psoriasis, suggesting that the relationship between hyper-proliferation and suprabasal β1 integrin expression could be causal(Carroll et al, 1995). Nevertheless, the functional state of these β1integrins has not been formally addressed.

Our immunohistochemical studies with MoAb that recognizeactivated β1 integrins indicate that, although there is suprabasalexpression of β1 integrins in psoriasis, these molecules seem to be inan inactive state. Moreover, most β1 integrins in lateral and apicalsurfaces of basal keratinocytes are also in a nonactive conformation,implying a decrease of activity compared with normal skin, in whichactive β1 integrins are distributed all over the basal keratinocytes. Ourresults with cultured NHK, showing intense β1 fluorescence on cell–cell contacts and the apical surface of keratinocytes but unrelated tothe expression of HUTS-21 or MG5A7, suggest a similar pattern ofexpression to that found in psoriatic skin. Probably, the psoriatic-likebehavior of NHK in culture could explain the discrepancy in thestaining patterns.

It has been proposed that the presence of integrins in the differentiat-ing cell layers stimulates proliferation in the basal layer, perhaps bysignaling that there is a deficit in the size of the differentiatedcompartment that should be replenished through increased proliferation(Carroll et al, 1995). Our results show that the expression of β1 is notcorrelated with its functional state, thus indicating that downregulationof the β1 functional activity occurs in psoriasis. Suprabasal expression

VOL. 111, NO. 1 JULY 1998 β1-INTEGRIN ACTIVITY IN PSORIASIS 23

Figure 5. Downregulation of β1 integrin activity in psoriasis, compared with normal skin. Immunohistochemical study of the expression in normal skin(a, b) and psoriasis (d, e) of HUTS-21. Anti-CD44 (HP2/9) MoAb, used as staining control, showed a similar intensity on both normal (c) and psoriatic (f) skin.Scale bar: (a, b, d, e) 20 µm; (c, f) 50 µm.

of inactive β1 might be related to the hyperproliferative state ofkeratinocytes. β1 integrins mediate keratinocyte adhesion to extra-cellular matrix components (Watt and Hertle, 1994), and its down-regulation may be one of the first steps in the differentiation processthat leads keratinocytes to the spinous layer (Adams and Watt, 1990).The mechanisms leading to the elimination of β1 from the membrane(Dalton et al, 1995; Hotchin et al, 1995) could not be so efficient todestroy the protein in the short time it takes keratinocytes to migratefrom the basal to the suprabasal layers. Expression of β1 integrins inlateral and apical surfaces of NHK in culture and in skin biopsies havesuggested its implication in cell–cell interactions (Carter et al, 1990;Larjava et al, 1990; Watt and Hertle, 1994). Although studies incultured NHK with other anti-β1 activation reporter MoAb (Kim andYamada, 1997) has shown that β1 integrins in cell–cell contact are notin the ligand-occupied conformation, as we found with HUTS 21 andMG5A7, our results in normal skin imply that these integrins are, atleast partially, in an activated state. A decrease of activation epitopesexpression in psoriasis, a disease of keratinocyte hyperproliferation,suggests that activated β1 integrins on lateral and apical surfaces of thecell might be signalling inhibition of proliferation.

This work was supported by grants from the Fondo de Investigacion Sanitaria 95/0196and 96/2009 and by a fellowship from the Fondo de Investigacion Sanitaria (to G. F. B.)

REFERENCES

Adams JC, Watt FM: Changes in keratinocyte adhesion during terminal differentiation:reduction in fibronectin binding precedes α5β1 integrin loss from the cell surface.Cell 63:425–435, 1990

Bata-Csorgo Z, Hammerberg C, Voorhees JJ, Cooper KD: Kinetics and regulation ofhuman keratinocyte stem cell growth in short-term primary ex vivo culture. J ClinInvest 95:317–327, 1995

Bazzoni J, Shih D-T, Buck CA, Hemler ME: Monoclonal antibody 9EG7 defines a novelβ1 integrin epitope induced by soluble ligand and manganese, but inhibited bycalcium. J Biol Chem 270:25570–25577, 1995

Campanero MR, Arroyo AG, Pulido R, et al: Functional role of α2/β1 and α4/β1integrins in leukocyte intercellular adhesion induced through the common β1subunit. Eur J Immunol 22:3111–3119, 1992

Carroll JM, Romero MR, Watt FM: Suprabasal integrin expression in the epidermis oftransgenic mice results in developmental defects and a phenotype resembling psoriasis.Cell 83:957–968, 1995

Carter WG, Wayner EA, Bouchard TS, Kaur P: The role of integrins α2β1 and α3β1 incell-cell and cell-substrate adhesion of human epidermal cells. J Cell Biol 110:1387–1404, 1990

Dalton SL, Scharf E, Briesewitz R, Marcantonio EE, Assoian RK: Cell adhesion toextracellular matrix regulates the life cycle of integrins. Mol Biol Cell 6:1781–1791, 1995

Fernandez-Herrera J, Sanchez-Madrid F, Garcıa Dıez A: Differential expression of the 4F2activation antigen on human follicular epithelium in hair cycle. J Invest Dermatol92:247–250, 1989

Giannelli G, Savoia P, Schivaldi O, Lospalluti M, de Luca M, Marchissio PC, QuarantaV: Psoriatic lesions in patients with chronic liver disease are distinct from psoriasisvulgaris lesions, as judged on basis of integrin adhesion receptors. Hepatology 20:56–65, 1994

Gomez M, Luque A, del Pozo MA, Hogg N, Sanchez-Madrid F, Cabanas C: Functionalrelevance during lymphocyte migration and cellular localization of activated β1integrins. Eur J Immunol 27:8–16, 1997

Guo M, Kim LT, Akiyama SK, Gralnick HR, Yamada KM, Grinnell F: Altered processingof integrin receptors during keratinocyte activation. Exp Cell Res 195:315–322, 1991

Hemler ME, Sanchez-Madrid F, Flotte TJ, et al: Glycoproteins of 210 000 and 130 000m.w. on activated T cells: cell distribution and antigenic relation to components onresting cells and T cell lines. J Immunol 132:3011–3018, 1984

de la Hera A, Acevedo A, Marston W, Sanchez-Madrid F: Function of CD44 (Pgp-1)homing receptor in human T cell precursors. Int Immunol 1:598–604, 1989

Hertle MD, Kubler M-D, Leigh IM, Watt FM: Aberrant integrin expression duringepidermal wound healing and in psoriatic epidermis. J Clin Invest 89:1892–1901, 1992

24 PENAS ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Hertle MD, Jones PH, Groves RW, Hudson DL, Watt FM: Integrin expression by humanepidermal keratinocytes can be modulated by interferon-γ, transforming growthfactor-β, tumor necrosis factor-α and culture on a dermal equivalent. J Invest Dermatol104:260–265, 1995

Hodivala KJ, Watt FM: Evidence that cadherins play a role in the downregulation ofintegrin expression that occurs during keratinocyte terminal differentiation. J CellBiol 124:589–600, 1994

Horrocks C, Dunca JI, Oliver AM, Thomson AW: Adhesion molecule expression inpsoriatic skin lesions and the influence of cyclosporin A. Clin Exp Immunol 84:157–162, 1991

Hotchin NA, Kovach NL, Watt FM: Functional down-regulation of α5β1 integrin inkeratinocytes is reversible but commitment to terminal differentiation is not. J CellSci 106:1131–1138, 1993

Hotchin NA, Gandarillas A, Watt FM: Regulation of cell surface β1 integrin levels duringkeratinocyte terminal differentiation. J Cell Biol 128:1209–1219, 1995

Jones J, Sugiyama M, Watt FM, Speight PM: Integrin expression in normal, hyperplastic,dysplastic and malignant oral epithelium. J Pathol 169:235–243, 1993

Jones PH, Watt FM: Separation of human epidermal stem cells from transit amplifyingcells on the basis of differences in integrin function and expression. Cell 73:713–724, 1993

Kim LT, Ishihara S, Lee CC, Akiyama SK, Yamada KM, Grinnell F: Altered glycosylationand cell surface expression of beta 1 integrin receptors during keratinocyte activation.J Cell Sci 103:743–753, 1992

Kim LT, Yamada KM: Evidence that β1 integrins in keratinocyte cell-cell junctions arenot in the ligand-occupied conformation. J Invest Dermatol 108:876–880, 1997

Larjava H, Peltonen J, Akiyama SK, Yamada SS, Gralnick HR, Uitto J, Yamada KM:Novel function for β1 integrins in keratinocyte cell–cell interactions. J Cell Biol110:803–815, 1990

Lenter M, Uhlig H, Hamann A, Jeno P, Imhof B, Vestweber D: A monoclonal antibodyagainst an activation epitope on mouse integrin chain β1 blocks adhesion of

lymphocytes to the endothelial integrin α6β1. Proc Natl Acad Sci USA 90:9051–9055, 1993

Luque A, Gomez M, Puzon W, Takada Y, Sanchez-Madrid F, Cabanas C: Activatedconformations of VLA integrins detected by a group of antibodies (HUTS) specificfor a novel regulatory region (355–425) of the common β1 chain. J Biol Chem271:11067–11075, 1996

Miyake K, Yamashita Y, Kimoto M: A calcium or manganese-dependent epitope on theintegrin β1 chain recognized by a unique MoAb. Int Immunol 6:1221–1226, 1994

Mould AP, Garratt AN, Askari JA, Akiyama SK, Humphries MJ: Identification of a novelanti-integrin monoclonal antibody that recognises a ligand-induced binding siteepitope on the β1 subunit. FEBS Lett 363:118–122, 1995

Parenteau N: Skin equivalents. In: Leigh IM, Watt FM (ed.). Keratinocyte Methods.Cambridge: Cambridge University Press, Cambridge, 1994, pp. 45–55

Pellegrini G, de Luca M, Orecchia G, et al: Expression, topography and function ofintegrins are severy altered in KC from involved and uninvolved psoriatic skin. JClin Invest 89:1783–1795, 1992

Ralfkiaer E, Thomsen K, Vejlsgaard GL: Expression of a cell adhesion protein (VLA β) innormal and disease skin. Br J Dermatol 124:527–532, 1991

Rheinwald JG, Green H: Epidermal growth factor and the multiplication of culturedhuman epidermal keratinocytes. Nature 265:421–424, 1977

Sanchez-Mateos P, Cabanas C, Sanchez-Madrid F: Regulation of integrin function. SemCancer Biol 7:99–109, 1996

Schwartz MA, Schaller MD, Ginsberg MH: Integrins: emerging paradigms of signaltransduction. Ann Rev Cell Dev Biol 11:549–599, 1995

Watt FM, Hertle MD: Keratinocyte integrins. In: Leigh IM, Lane EB, Watt FM (ed.).The Keratinocyte Handbook. Cambridge University Press, Cambridge, 1994, pp.153–164

Yednock TA, Cannon C, Vandevert C, et al: α4β1 integrin-dependent cell adhesion isregulated by a low affinity receptor pool that is conformationally responsive toligand. J Biol Chem 270:28740–28750, 1995