Phenotype, ultrastructure, and function of CD1+DR+ epidermal cells that express CD36 (OKM5) in cutaneous T-cell lymphoma

Thrombospondin-1 promotes tumor progression in cutaneous T-cell lymphoma via CD47

CD47 is highly expressed on various hematopoietic malignancies, and enables cancer cells to avoid immunosurveillance. Its ligand, thromobospondin-1 (TSP-1) is a multifunctional protein, and CD47/TSP-1 interactions promote tumor progression in various malignancies. In this study, we investigated roles of TSP-1 and CD47 in cutaneous T-cell lymphoma (CTCL). Flow cytometric analysis and immunohistochemistry showed that CTCL tumor cells and CTCL cell lines (Hut78, HH, and MyLa cells) overexpressed CD47 compared with normal CD4⁺ T cells. Overexpression of CD47 was partially induced by high c-Myc expression in CTCL tumor cells. TSP-1 mRNA expression levels in CTCL lesional skin were higher than those in normal skin and correlated with increased risk of disease-related death. Moreover, TSP-1 was expressed on CTCL tumor cells by immunohistochemistry. Serum soluble TSP-1 levels in patients with Sézary syndrome were significantly elevated. TSP-1 promotes proliferation and survival of CTCL tumor cells, which is inhibited by anti-CD47 neutralizing antibody or CD47 knockdown. Stimulation with TSP-1 also induces cell migration and in vivo growth. These effects were mediated by phosphorylation of ERK1/2 and AKT and expression of survivin. Collectively, our findings prompt a novel therapeutic approach to CTCL based on discovery that CD47/TSP-1 interactions play important roles in progression of CTCL.

Novel human antibody therapeutics: the age of the Umabs

Monoclonal antibodies represent a major and increasingly important category of biotechnology products for the treatment of human diseases. The state-of-the-art of antibody technology has evolved to the point where therapeutic monoclonal antibodies, that are practically indistinguishable from antibodies induced in humans, are routinely generated. We depict how our science-based approach can be used to further improve the efficacy of antibody therapeutics, illustrated by the development of three monoclonal antibodies for various cancer indications: zanolimumab (directed against CD4), ofatumumab (directed against CD20) and zalutumumab (directed against epidermal growth factor receptor).

Clinical efficacy of zanolimumab (HuMax-CD4): two phase 2 studies in refractory cutaneous T-cell lymphoma

The efficacy and safety of zanolimumab in patients with refractory cutaneous T-cell lymphoma (CTCL) have been assessed in two phase 2, multicenter, prospective, open-label, uncontrolled clinical studies. Patients with treatment refractory CD4+ CTCL (mycosis fungoides [MF], n = 38; Sézary syndrome [SS], n = 9) received 17 weekly infusions of zanolimumab (early-stage patients, 280 and 560 mg; advanced-stage patients, 280 and 980 mg). The primary end point was objective response (OR) as assessed by composite assessment of index lesion disease activity score. Secondary end points included physician's global assessment (PGA), time to response, response duration, and time to progression. ORs were recorded for patients in both CTCL types (MF, 13 ORs; SS, 2 ORs). In the high-dose groups (560 and 980 mg dose groups), a response rate of 56% was obtained with a median response of 81 weeks. Adverse events reported most frequently included low-grade infections and eczematous dermatitis. Zanolimumab showed marked clinical efficacy in the treatment of patients with refractory MF, with early onset of response, high response rate, and durable responses. The treatment was well tolerated with no dose-related toxicity other than the targeted depletion of peripheral T cells. A pivotal study has been initiated based on these findings.

Primary ex vivo culture of keratinocytes isolated from hypertrophic scars as a means of biochemical characterization of CD36

The CD36 antigen is a molecule which is ectopically expressed on epidermal keratinocytes of hypertrophic scars and is a good candidate for a marker for a broad range of skin pathologies. Most marker studies have been performed using immunohistochemical techniques on fixed skin sections. Our aim was to investigate the biochemical features of the CD36 expressed in pathological keratinocytes and to find an in vitro model for the study of the regulation of its expression. Here we show how keratinocytes isolated from hypertrophic scars can be cultivated in vitro and employed as a model for the study of these cells. We demonstrated that the antigenic features of the CD36 expressed on keratinocytes of hypertrophic scars are identical to those described for the CD36 expressed by other cell types. The molecule was expressed on the surface of keratinocytes which were non-adherent in vitro. Adherent and proliferating keratinocytes, as well as normal keratinocytes, were CD36 negative both at the surface and intracellularly. The in vitro proliferating cells from hypertrophic scars, but not the normal keratinocytes, showed intracellular expression of CD36 after long-term culture and cell stratification, suggesting a regulated expression of CD36 in pathological keratinocyte differentiation.

Identification of a human dermal macrophage population responsible for constitutive restraint of primary dermal fibroblast proliferation

Primary human dermal cell suspensions prepared from the papillary dermis of keratomed skin strips were used to investigate the effect of indigenous dermal macrophages (HLA-DR+, CD11c+ CD11b+ CD1c- phagolysosome+) upon dermal fibroblast proliferation. Rapid dermal fibroblast expansion was induced upon immunomagnetic bead removal of CD11b+ or CD11c+ cells as well as by removal of more inclusive subsets contained within the DR+ population, but the removal of mast cells, endothelial cells, and CD1c+ dermal Langerhans cells from dermal cell suspensions failed to result in proliferation of the remaining cell subsets. Removal of 1B10+ fibroblasts from macrophage depleted (CD11b-) dermal cell suspensions essentially abrogated the unrestrained proliferation of the CD11b- dermal cells. Flow cytometric cell cycle analysis of cultured macrophage-depleted dermal cells confirmed that the unrestrained proliferating cells contain procollagen I+ as well as procollagen I- dermal fibroblasts. Inhibition of primary fibroblast expansion by adding a supernatant from unfractionated dermal cells suggested that a growth-inhibitory soluble activity of >30,000 kDa dominates the cytokine mixture released by unfractionated fresh dermal cells ex vivo. Inhibitory activity counterbalanced positive fibroblast growth- stimulatory cytokines released by dermal cells because neutralizing antibodies to insulin-like growth factor 1 and interleukin-1 beta resulted in decreased CD11b- dermal cell fibroblast proliferation. These data indicated an important role for dermal macrophages of the DR+ CD11b+ CD11c+ DC1c- phenotype in the normal homeostatic restraint of primary human dermal fibroblast proliferation.

In cutaneous T-cell lymphoma, class II MHC molecules on CD1+ antigen-presenting cells are upregulated in involved compared with uninvolved epidermis

CD1+ antigen-presenting cells in involved epidermis of patients with cutaneous T-cell lymphoma exhibit and enhanced functional capacity to activate autologous CD4+ T cells compared with CD1+ antigen-presenting cells from uninvolved and normal epidermis. Class II major histocompatibility complex molecules are involved in antigen presentation, and their expression on CD1+ Langerhans cells is known to vary. The expression of all three class II (HLA-DR, -DQ, -DP) molecules was therefore determined on CD1+ epidermal cells from both involved and uninvolved epidermis, using flow cytometry. The involved CD1+ epidermal cells exhibited a 1.5-1.6-fold, statistically significant increase in fluorescence intensity after staining of the class II molecules (HLA-DR, -DQ, -DP) compared with CD1+ epidermal cells from uninvolved epidermis. The autologous CD4+ T cells, activation was almost completely blocked by anti-HLA-DR, and partly by anti-HLA-DQ and anti-HLA-DP. In contrast, an antibody against class I, and an irrelevant control antibody, had no blocking effect. In a pokeweed mitogen assay it was demonstrated that autologous CD4+ T cells, activated by involved epidermal cells, demonstrated suppressor activity rather than helper activity. The suppressor activity was dependent on the presence of HLA-DR-positive epidermal cells. Thus, in cutaneous T-cell lymphoma, class II molecules on the individual CD1+ antigen-presenting cell are upregulated in clinically involved compared with uninvolved epidermis, and these molecules are crucially involved in activation of CD4+ T cells.

Retinoic acid upregulates human Langerhans cell antigen presentation and surface expression of HLA-DR and CD11c, a beta 2 integrin critically involved in T-cell activation

Immunomodulatory effects of retinoids may be part of their anti-carcinogenic and anti-inflammatory properties. We studied the in vivo effects of retinoic acid (RA) on antigen-presenting activity of human epidermal Langerhans cells and on accessory cell activity of keratinocytes. Two skin sites from each volunteer were treated in vivo with 0.1% RA or vehicle, respectively, once a day for 4 d. RA-treated epidermal cell (RA-EC) alloantigen presentation to CD4+ T cells in each volunteer tested was consistently greater than that induced by vehicle EC. However, this increased antigen-presenting activity did not lead to autoreactive CD4+ T-lymphocyte proliferation. Elevated unfractionated epidermal antigen-presenting activity of RA-EC was not due to increased keratinocyte major histocompatibility complex (MHC) or intercellular adhesion molecule expression or to other keratinocyte accessory signaling, because incubation of CD1a-fluoroscence-activated cell sorter (FACS)-purified RA-EC inhibited alloantigen presentation, presumably through increased keratinocyte transforming growth factor-beta. By contrast, Langerhans cell function was upregulated; FACS-purified CD1a+ Langerhans cells derived from RA-EC displayed a markedly increased ability, relative to Langerhans cells from vehicle EC, to present alloantigen to T cells. Triple color flow-cytometric analysis of RA-EC and vehicle EC suspensions revealed that RA treatment did not modify the number of DR+ and CD1a+DR+EC, but did result in statistically significant increases in Langerhans cells expression of HLA-DR, CD11c, and CD1c. Another novel finding was that HLA-DR-dependent Langerhans cells antigen-presenting activity in both normal and RA-treated skin was completely blocked by anti-CD11c antibody. Thus, retinoid upregulation of antigen-presenting activity may be due to upregulation of Langerhans cell CD11c, as well as class II MHC. Upregulation of cutaneous immune responsiveness in human skin without autoreactivity has not (to our knowledge) been reported previously, and the Langerhans cell phenotypic and functional state achieved is distinct from previously reported states of Langerhans cell activation.

Distinction of class II MHC+ Langerhans cell-like interstitial dendritic antigen-presenting cells in murine dermis from dermal macrophages

Dermal cells are capable of initiating contact-hypersensitivity responses but the precise identification of the antigen-presenting cell within murine dermis is lacking. Class II major histocompatibility complex (MHC)+ cells with dendritic shape and lacking endothelial factor VIII but expressing the dendritic antigen-presenting cell marker NLDC-145 were observed in the perivascular and interstitial dermis of BALB/c and C3H/HeN skin. The heterogeneous class II MHC+ cells could be divided into two subsets: each was class II MHC+ CD45+ (bone marrow derived) GR-1- (non-neutrophil/macrophage) CD3- (non T), but one subset was CD11b+ (beta 2 integrin) and the other was CD11b-. Ultrastructural examination of class II MHC+ cells revealed the presence of a Langerhans cell-like/indeterminant cell subset with indented nuclei, dendritic morphology, active cytoplasm, and dense intermediate filaments. Phagolysomes and Birbeck granules were not observed in such cells, indicating these were distinct from dermal macrophages and from classical epidermal Langerhans cells, respectively. Cells with a monocyte/macrophage ultrastructural appearance were also noted, likely representing the class II MHC subset expressing CD11b and Ly6c (monocyte/endothelial antigen). Dermal cells in suspension were capable of processing and presenting large protein antigens to antigen-specific T-cell hybridomas; dermal cells also induced the syngeneic mixed lymphocyte reaction. The dermal antigen-presentation activities were totally abrogated by removal of class II MHC+ cells, but not by removal of CD11b+ cells or Ly6c+ cells, indicating that potent antigen-presenting cell activity was restricted to the class II MHC+ CD11b- Ly6c- subset (Langerhans cell-like/indeterminant cells). In conclusion, within a complex array of dermal leukocytes a murine dermal class II MHC+ cell population expressing a Langerhans cell-like/dendritic antigen-presenting cell phenotype and exhibiting potent antigen processing and presenting activity can be identified. The positioning of potent interstitial dendritic antigen-presenting cells at the interface of the vasculature with the dermal interstitium provides rapid access to an antigen-presenting cell as T cells first egress into the skin.

CD1 gene expression in human skin

In human epidermis, expression of CD1a is confined to Langerhans cells (LC), whereas CD1c expression has been observed in dendritic cells of the dermis, as well as the epidermis. In transfected fibroblasts, expression of CD1c at the cell surface appears to exclude expression of either CD1b or CD1a, despite continued transcription of the latter genes. In order to determine whether this mechanism might be operative in human skin, we have compared the expression of CD1a and CD1c on the surface of dermal and epidermal dendritic cells to their expression at the level of mRNA using a combination of dual-label immunofluorescence microscopy, northern blot hybridization, and reverse transcriptase-polymerase chain reaction (RT-PCR). By both immunofluorescence and Northern blotting, CD1c expression was observed in both dermal and epidermal cells, whereas expression of CD1a was confined largely to the epidermis. Moreover, as shown by immunomagnetic bead selection and RT-PCR, CD1a and CD1c were both expressed on epidermal LC, but were absent from other epidermal cell types. These results argue against cell surface exclusion as a mechanism for selective expression of CD1c in human dermis.

Neutrophils, differentiated macrophages, and monocyte/macrophage antigen presenting cells infiltrate murine epidermis after UV injury

We asked whether, as in humans, a population of antigen-presenting macrophages infiltrates the epidermis of ultraviolet (UV)-exposed BALB/c mice. Using three-color flow cytometry on cell suspensions plus in situ immunofluorescence microscopy, the phenotype of normal Langerhans cells was class II major histocompatibility complex (MHC+), CD11b+, NLDC-145+, BM8+ CD45+ and homogeneous. By contrast, in epidermal cells harvested 3 d following UV (UV-EC), there were two subsets of class II MHC+ cells: 1) class II MHChi CD11b+, and 2) class II MHClo CD11b-. Neither expressed the Langerhans cell markers BM8 and NLDC-145. In addition, there were two major populations of class II MHC- CD11b+ cells; half of these expressed the GR-1 neutrophil marker. Langerhans and dendritic epidermal T cells were markedly reduced after UV injury. By electron microscopy, immunomagnetic bead-purified CD11b+ cells in UV-EC were comprised of neutrophils, differentiated macrophages, and mononuclear cells with prominent lysosomes, but no Birbeck granules; the class II MHC+ subset resembled a monocytic cell in between differentiated macrophages and indeterminate dendritic cells. Functionally, immediately following in vivo UV exposure, the allogeneic antigen-presenting cell capacity of UV-EC was reduced to 21 +/- 6% of control epidermal cells (C-EC); by 3 d, antigen-presenting cell activity of UV-EC had recovered to 59 +/- 11% of C-EC, although at this time NLDC-145+ Langerhans cells had reached their lowest number. The recovered antigen-presenting cell activity was critically dependent upon the class II MHChiCD11b+ cells. Sensitization of BALB/c mice through skin that contained these antigen-presenting cells (3 d after UV) resulted in tolerance to dinitrofluorobenzene. By contrast, sensitization through UV-exposed skin immediately after the exposure resulted in unresponsiveness without tolerance, demonstrating temporal association of tolerance with leukocytic infiltration. In summary, murine epidermis responds to an acute UV injury in vivo with an initial abrogation of antigen-presenting activity followed by epidermal infiltration with neutrophils, differentiated macrophages, and monocytic antigen-presenting cells that are distinct from Langerhans cells with regard to expression of Langerhans cell markers and ultrastructure.

Leukemic T cells from patients with cutaneous T-cell lymphoma demonstrate enhanced activation through CDw60, CD2, and CD28 relative to activation through the T-cell antigen receptor complex

Antigen-dependent activation of T cells occurs through the T-cell antigen-receptor complex (TCR/CD3). Antigen-independent T-cell activation may occur through the surface molecules CDw60, CD2, and CD28. We wished to determine whether these antigen-independent T-cell-activation pathways could be involved in proliferation of leukemic T cells from patients with cutaneous T-cell lymphoma (CTCL). Whereas CDw60 was only expressed on 28% +/- 7% (mean +/- SEM) of blood T cells obtained from healthy control subjects (n = 4), CDw60 was expressed on 94% +/- 3% of blood T cells obtained from patients with CTCL (n = 4). Dual color immunofluorescence microscopy of the T-cell infiltrate in involved skin of these patients demonstrated that almost 100% of the T cells expressed CDw60. Not only did T cells in the patients with CTCL express CDw60, but triggering of the T cells with anti-CDw60 resulted in enhanced proliferation relative to anti-TCR/CD3 and mitogenic lectins. Other antigen-independent pathways also appeared highly active in the T cells from patients with CTCL because enhanced proliferation relative to anti-TCR/CD3 or mitogenic lectins was found when anti-CD2 or anti-CD28 plus phorbol ester was used as stimulant. Despite the brisk proliferation induced by anti-CDw60, anti-CD2, or anti-CD28, T cells from the patients did not produce detectable amounts of gamma-interferon. The inability to produce gamma-interferon correlates with our finding of absent (n = 3) or weak (n = 1) intercellular adhesion molecule-1 expression in the lesional keratinocytes in these patients. In conclusion, T cells of patients with CTCL demonstrate elevated expression of a T-cell-independent signaling molecule CDw60 and respond to antigen-independent activating signals.

Interactions of epidermal cells and T cells in inflammatory skin diseases

Multiple cell types and their factors and cytokines are involved in regulating the immune response in inflammatory skin diseases. Stimulatory and inhibitory factors interact to determine whether the immune response is regulated up or down. Normally, stimulatory signals are counteracted by inhibitory signals to prevent an immune reaction from being initiated. However, exogenous antigens and irritants or endogenous factors and altered immunogenic self-peptides can upset this balance. When that occurs, T cells are activated and an inflammatory skin reaction develops. Lymphokines released from such activated T cells can modify the phenotype and function of normal keratinocytes. They can induce the expression of adhesion molecules and receptors involved in antigen presentation. Furthermore, they can also stimulate keratinocyte proliferation. This may be important in development of the hyperplasia seen in inflammatory skin diseases, especially in psoriasis. Cytokines released from the activated keratinocytes can both stimulate and attract T cells to the epidermis and thereby continue the ongoing immune reaction.

Antigen-presenting activity of non-Langerhans epidermal cells in contact hypersensitivity reactions

Despite the critical role of the Langerhans cells in the induction of contact hypersensitivity reactions, non-Langerhans antigen-presenting cells in already sensitized individuals may play a role in the elicitation phase of a contact hypersensitivity reaction. Following epicutaneous challenge with antigens, the number of CD1+DR+ epidermal Langerhans cells increased in a time-dependent way and, concomitantly, CD1-OKM5+DR+ epidermal non-Langerhans cells appeared. In parallel with this, the capacity of epidermal cells to present both alloantigens and auto/nominal antigens increased, and 4 days after initiation of the contact hypersensitivity reactions 33-53% of the epidermal antigen-presenting capacity was due to CD1- non-Langerhans antigen-presenting cells. Thus, contact hypersensitivity skin reactions are accompanied by the appearance of non-Langerhans antigen-presenting cells capable of presenting both alloantigens and auto/nominal antigens.