Advertisement

Cutaneous Lymphocyte Antigen Is a Potential Therapeutic Target in Cutaneous T-Cell Lymphoma

Open AccessPublished:July 15, 2022DOI:https://doi.org/10.1016/j.jid.2022.06.016
      Cutaneous T-cell lymphoma (CTCL) such as Sézary syndrome or mycosis fungoides corresponds to an abnormal infiltration of T lymphocytes in the skin. CTCL cells have a heterogeneous phenotype and express cell adhesion molecules such as cutaneous lymphocyte antigen (CLA) supporting skin homing. The use of a mAb (HECA-452) against CLA significantly decreased transendothelial migration and survival of CTCL cells from patient samples and My-La cell line. The decrease of CLA expression by inhibition of its maturation enzyme, ST3 β-galactoside α-2,3-sialyltransferase 4, also impaired CTCL cell migration, proliferation, and survival. We confirmed in vivo that treatment with anti-CLA mAb decreased the tumorigenicity as well as dissemination of CTCL cells in different tissues compared with the control group. Our findings provide evidence of the involvement of CLA in CTCL cell migration and survival, supporting that CLA inhibition could represent an actionable therapy in patients with CTCL.

      Abbreviations:

      CLA (cutaneous lymphocyte antigen), CTCL (cutaneous T-cell lymphoma), MF (mycosis fungoides), SC (Sézary cell), shRNA (short hairpin RNA), shST3Gal-4 (ST3 β-galactoside α-2,3-sialyltransferase 4‒targeted short hairpin RNA), SS (Sézary syndrome), ST3Gal-4 (ST3 β-galactoside α-2,3-sialyltransferase 4)

      Introduction

      Cutaneous T-cell lymphoma (CTCL) corresponds to an abnormal infiltration of T lymphocytes in the skin. The most common is mycosis fungoides (MF), which is characterized by an indolent evolution with patches and plaques on the skin and epidermotropism of atypical and clonal T cells. In contrast, Sézary syndrome (SS) is less frequent but more aggressive and is characterized by an erythroderma associated with pruritus and a clonal expansion of CD4+ T cells circulating in the blood (
      • Olsen E.
      • Vonderheid E.
      • Pimpinelli N.
      • Willemze R.
      • Kim Y.
      • Knobler R.
      • et al.
      Revisions to the staging and classification of mycosis fungoides and Sezary syndrome: a proposal of the International Society for Cutaneous Lymphomas (ISCL) and the cutaneous lymphoma task force of the European Organization of Research and Treatment of Cancer (EORTC).
      ). Patients with SS have a poorer prognosis than those with MF, with a 5-year survival rate of approximately 24‒40% (
      • Fink-Puches R.
      • Zenahlik P.
      • Bäck B.
      • Smolle J.
      • Kerl H.
      • Cerroni L.
      Primary cutaneous lymphomas: applicability of current classification schemes (European Organization for Research and Treatment of Cancer, World Health Organization) based on clinicopathologic features observed in a large group of patients.
      ;
      • Wilcox R.A.
      Cutaneous T-cell lymphoma: 2016 update on diagnosis, risk-stratification, and management.
      ). Treatments are not curative, especially in advanced stages, except in cases eligible for allogeneic stem cell transplantation (
      • de Masson A.
      • Beylot-Barry M.
      • Bouaziz J.D.
      • Peffault de Latour R.P.
      • Aubin F.
      • Garciaz S.
      • et al.
      Allogeneic stem cell transplantation for advanced cutaneous T-cell lymphomas: a study from the French Society of Bone Marrow Transplantation and French Study Group on Cutaneous Lymphomas.
      ).
      Recent studies have focused on the immunophenotypic characterization of Sézary cells (SCs) using flow cytometry. Although a central memory phenotype of SC had been suggested by
      • Campbell J.J.
      • Clark R.A.
      • Watanabe R.
      • Kupper T.S.
      Sezary syndrome and mycosis fungoides arise from distinct T-cell subsets: a biologic rationale for their distinct clinical behaviors.
      , it seems to be more heterogeneous with cells exhibiting the phenotype of resident memory, effector memory, central memory, or naive T lymphocytes (
      • Poglio S.
      • Prochazkova-Carlotti M.
      • Cherrier F.
      • Gros A.
      • Laharanne E.
      • Pham-Ledard A.
      • et al.
      Xenograft and cell culture models of Sézary syndrome reveal cell of origin diversity and subclonal heterogeneity.
      ;
      • Roelens M.
      • Delord M.
      • Ram-Wolff C.
      • Marie-Cardine A.
      • Alberdi A.
      • Maki G.
      • et al.
      Circulating and skin-derived Sézary cells: clonal but with phenotypic plasticity.
      ).
      Such studies have however identified that blood clonal T cells express a combination of molecules (cutaneous lymphocyte antigen [CLA] and CCR4), which could allow them to migrate to the skin (
      • Campbell J.J.
      • Clark R.A.
      • Watanabe R.
      • Kupper T.S.
      Sezary syndrome and mycosis fungoides arise from distinct T-cell subsets: a biologic rationale for their distinct clinical behaviors.
      ;
      • Kamarashev J.
      • Burg G.
      • Kempf W.
      • Hess Schmid M.H.
      • Dummer R.
      Comparative analysis of histological and immunohistological features in mycosis fungoides and Sezary syndrome.
      ). On the basis of such observations, an antibody directed against the CCR4 chemoreceptor (mogamulizumab) was tested on CTCL cell lines before the phases II and III clinical trials, and it is Food and Drug Administration approved since 2018 for treatment of adult patients with relapsed or refractory MF or SS after at least one previous systemic therapy (
      • Chang D.K.
      • Sui J.
      • Geng S.
      • Muvaffak A.
      • Bai M.
      • Fuhlbrigge R.C.
      • et al.
      Humanization of an anti-CCR4 antibody that kills cutaneous T-cell lymphoma cells and abrogates suppression by T-regulatory cells.
      ;
      • Kim Y.H.
      • Bagot M.
      • Pinter-Brown L.
      • Rook A.H.
      • Porcu P.
      • Horwitz S.M.
      • et al.
      Mogamulizumab versus vorinostat in previously treated cutaneous T-cell lymphoma (MAVORIC): an international, open-label, randomised, controlled phase 3 trial.
      ;
      • Kasamon Y.L.
      • Chen H.
      • de Claro R.A.
      • Nie L.
      • Ye J.
      • Blumenthal G.M.
      • et al.
      FDA approval summary: Mogamulizumab-kpkc for mycosis fungoides and Sézary syndrome.
      ;
      • Ramelyte E.
      • Dummer R.
      • Guenova E.
      Investigative drugs for the treatment of cutaneous T-cell lymphomas (CTCL): an update.
      ).
      Alternatively, it has been shown in MF and SS skin biopsies that skin-infiltrating lymphocytes express CLA in the vast majority of cases (
      • Kamarashev J.
      • Burg G.
      • Kempf W.
      • Hess Schmid M.H.
      • Dummer R.
      Comparative analysis of histological and immunohistological features in mycosis fungoides and Sezary syndrome.
      ;
      • Picker L.J.
      • Michie S.A.
      • Rott L.S.
      • Butcher E.C.
      A unique phenotype of skin-associated lymphocytes in humans. Preferential expression of the HECA-452 epitope by benign and malignant T cells at cutaneous sites.
      ). Although its expression by CTCL cells has been reported, functional data on the role of the CLA molecule in CTCLs are still missing. In healthy donors, CLA is absent on the surface of naïve T cells but is expressed by 10‒25% of circulating memory T cells (CD45RO+) (
      • Berg E.L.
      • Yoshino T.
      • Rott L.S.
      • Robinson M.K.
      • Warnock R.A.
      • Kishimoto T.K.
      • et al.
      The cutaneous lymphocyte antigen is a skin lymphocyte homing receptor for the vascular lectin endothelial cell-leukocyte adhesion molecule 1.
      ). T cells expressing CLA interact with the E-selectin, an adhesion molecule expressed by activated endothelial cells (
      • Fuhlbrigge R.C.
      • Kieffer J.D.
      • Armerding D.
      • Kupper T.S.
      Cutaneous lymphocyte antigen is a specialized form of PSGL-1 expressed on skin-homing T cells.
      ). Such interaction is necessary for the skin-homing properties of T cells. CLA shares a common core protein with PSGL-1. The two proteins are encoded by the same gene, SELPLG, and differ, as a result of post-translational modification (
      • Fuhlbrigge R.C.
      • Kieffer J.D.
      • Armerding D.
      • Kupper T.S.
      Cutaneous lymphocyte antigen is a specialized form of PSGL-1 expressed on skin-homing T cells.
      ;
      • Maverakis E.
      • Kim K.
      • Shimoda M.
      • Gershwin M.E.
      • Patel F.
      • Wilken R.
      • et al.
      Glycans in the immune system and the Altered Glycan Theory of Autoimmunity: a critical review.
      ). In myeloid leukocytes, the enzyme ST3 β-galactoside α-2,3-sialyltransferase 4 (ST3Gal-4) allows Sialyl Lewisx biosynthesis that is required for CLA maturation and then specific CLA/E-selectin interaction (
      • Mondal N.
      • Buffone A.
      • Stolfa G.
      • Antonopoulos A.
      • Lau J.T.Y.
      • Haslam S.M.
      • et al.
      ST3Gal-4 is the primary sialyltransferase regulating the synthesis of E-, P-, and L-selectin ligands on human myeloid leukocytes.
      ).
      More recently, CLA was selected to be specifically targeted by a light-sensitive molecule on MF cells (
      • Silic-Benussi M.
      • Saponeri A.
      • Michelotto A.
      • Russo I.
      • Colombo A.
      • Pelizzo M.G.
      • et al.
      Near infrared photoimmunotherapy targeting the cutaneous lymphocyte antigen for mycosis fungoides.
      ). Therefore, we decided to investigate the expression of CLA in both patients’ circulating SCs and MF cell line. In this study, we evaluated CLA functions on migration and survival of patient SCs and My-La cell line both in vitro and in vivo after targeting CLA by a specific mAb or inhibition of the CLA maturation enzyme, ST3Gal-4.

      Results

      CLA is expressed in the majority of SS blood samples

      A total of 36 patients with SS (stage B2 at diagnosis) were included in our study. PBMC containing circulating SCs were isolated. The patients’ characteristics such as sex, age, and TNMB (tumor, node, metastasis, blood) stage at the time of analysis are available in Supplementary Table S1. For all samples, we evaluated by flow cytometry (FACS) the percentage of CLA+ T cells in the SC clonal (TCRVβ+CD3+) or total CD4+CD3+ population for patients with an indeterminate TCRVβ (Figure 1a). Such percentage of CLA+ SC was heterogeneous among patients (Figure 1b). A total 34 of the 36 patient samples exhibited more than 10% of CLA+ cells, and 16 exhibited more than 50% of CLA+ cells in the tumor clonal population (Supplementary Table S1). The quantification using HALO software of the CLA+ area on the total biopsy surface was heterogeneous (range = 0.04‒2.45%) among 12 patients (Figure 1c and Supplementary Figure S1a). The coimmunostaining of CLA and CD3 on the patients' skin biopsies revealed that approximately 80% of CD3+ cells residing in the skin were CLA+ compared with 83.7% on normal skin biopsies (Supplementary Figures S1b and c). For patients whose skin and blood were matched, no correlation was found between the proportion of CLA+ area in skin-resident and circulating CLA+CD3+ cells (Figure 1d).
      Figure thumbnail gr1
      Figure 1Proportions of CLA+ T cells in SS blood and skin samples are heterogeneous between patients. (a) Immunophenotyping of tumor cells. PBMCs prepared from the samples of 36 patients with SS were analyzed by flow cytometry to determine the percentage of CLA+ T cells in the clonal tumor population (TCRVβ2+CD3+). TCRVβ variant specific to each patient was previously determined using the TCRVβ Repertory Kit (Beckman Coulter, Brea, CA). An example of gating strategy for patient 5 is shown. (b) The percentage of CLA+ tumor T cells is variable in blood SS samples. The horizontal bar represents the median of CLA+ cells inside the tumor population. (c) IHC showing CLA expression in the skin biopsies of patients with SS. Example of anti-CLA staining on skin sections of three patients with SS. Scale bars = 100 and 30 μm. (d) Percentages of CLA+ cells in the blood and CLA+ area in the skin are not correlated. These percentages were compared using simple linear regression test. CLA, cutaneous lymphocyte antigen; FSC, forward scatter; IHC, immunohistochemistry; SS, Sézary syndrome; SSC, side scatter.

      Targeting CLA inhibits CTCL cell transendothelial migration

      The involvement of CLA in CTCL cell migration was investigated by targeting CLA with a mAb in a transendothelial migration assay using a Transwell system coated with human microvascular cells from the dermis or human umbilical vein endothelial cells. Three hours after inducing migration with SDF-1 and CCL19/21 chemokines, the percentages of migrating My-La cells and SCs treated with an anti-CLA mAb were significantly lower than that of the isotype control antibody group (n = 3 My-La, 4.33 vs. 20.23%, P = 0.05, n = 3 for patients with SS, 23.37 vs. 40.28 %, P < 0.001) (Figure 2a). In parallel, we also increased the expression of CLA ligand (E-selectin) on endothelial cell surface by mimicking inflamed tissue by lipopolysaccharide exposure (Supplementary Figure S2a). My-La cell migration was also significantly increased after E-selectin induction and impaired after anti-CLA mAb treatment (Supplementary Figure S2b). Interestingly, in an inflammatory context, anti-CLA antibody has a higher effect on migration inhibition than on steady state condition (Supplementary Figure S2b). These data show that CLA is involved in CTCL migration process through dermal vascular cells.
      Figure thumbnail gr2
      Figure 2Anti-CLA mAb inhibits CTCL cell transendothelial migration and alters cell survival. (a) Anti-CLA mAb treatment decreases CTCL migration. My-La cell line or cells of patients with SS were incubated with an anti-CLA mAb (15 or 30 μg/ml HECA-452) or isotype control in a Transwell system coated with HMVEC-D cells for 3 hours. The histogram represents the percentage of migration of My-La cells (n = 3) or cells of patients with SS (n = 3). (b) Anti-CLA mAb increases CTCL cell death. Percentage of My-La (n = 3) or cell death of patients with SS after anti-CLA mAb or isotype control treatment for 3 or 3 and 72 hours, respectively. Statistical analysis was performed using a Mann‒Whitney test. ∗P < 0.05 and ∗∗∗P < 0.0001. Data are expressed as mean ± SEM. CLA, cutaneous lymphocyte antigen; CTCL, cutaneous T-cell lymphoma; HMVEC-D, human microvascular cells from the dermis; SS, Sézary syndrome.

      Targeting CLA alters CTCL cell survival

      To investigate whether CLA is involved in CTCL cell survival, we performed an apoptosis‒necrosis assay on the My-La cell line and on fresh CLA+ SCs from patients after treatment with anti-CLA mAb. Three and 72 hours after treatment, the percentage of total dead cells significantly increased when cells were treated with the anti-CLA mAb in comparison with treatment with the isotype control antibody (n = 3 My-La cells for 3 hours, 65.9 vs. 22.9%, P < 0.0001; n = 7 patients for 3 hours, 25.1 vs. 5.2%, P < 0.0001; n = 3 patients for 72 hours, 65.5 vs. 38.8%, P < 0.0001) (Figure 2b). Moreover, the percentage of CLA+ cells and their expression levels on the cell surface were significantly correlated with the efficacy of the antibody on tumor cell death (R2 = 0.7498, P = 0.0117; R2 = 0.6579, P = 0.0268) (Supplementary Figure S3). This also supported that the effects of the anti-CLA antibody were specific and dependent on CLA expression, as also shown by the absence of an increase in cell death when CLA-negative cell lines were treated with the anti-CLA antibody (Supplementary Figure S4a). In addition, treatment of CD3+CD4+ cells from healthy donors with the anti-CLA mAb did not increase cell death, suggesting no effect on normal CD4+ T cells even if 30% of the cells expressed CLA (Supplementary Figure S4b and S4c). This could be explained by a lower expression of CLA as observed on the surface of healthy donor CD4+ compared with CTCL cells (Supplementary Table S1, Supplementary Table S2). These data support that CLA expression is involved in both CTCL proliferation and survival.

      Impaired CLA maturation reduces transendothelial migration and survival of CTCL cells

      The CLA molecule shares a common core with PSGL-1 and differs by post-translational modification mediated by ST3Gal-4 (Figure 3a, left). To confirm whether the decrease of CLA levels could affect cell survival, we silenced ST3Gal-4, the maturation enzyme of CLA, using two lentiviruses encoding short hairpin RNA (shRNA) targeting ST3Gal-4 mRNA on exon 5 or exon 8 and GFP as a reporter gene (Supplementary Figures S5). My-La cells transduced with ST3Gal-4‒targeted shRNA (shST3Gal-4) shST3Gal-4 ex5 or shST3Gal-4 ex8 had decreased ST3Gal-4 mRNA levels compared with those transduced with control shRNA (Supplementary Figure S6a). As expected, in these cells, the expression of PSGL-1 was not decreased by ST3Gal-4 silencing (Supplementary Figure S6b). Sorted GFP+ My-La cells transduced with ST3Gal-4 shRNA vectors exhibited a high decrease of CLA expression than those transduced with control shRNA (mean fluorescent intensity = 20,091 vs. 3,398, n = 3) (Figure 3a, right). This was associated with a significant decrease in transendothelial migration capacity of CTCL cells (n = 3 My-La cells, 15.8 for control-targeted shRNA vs. 7.6 for shex5 and 2.5% for shex8, P < 0.001) (Figure 3b).
      Figure thumbnail gr3
      Figure 3Decreasing cell surface CLA expression by ST3Gal-4 gene silencing impairs CTCL cell transendothelial migration and cell survival. (a) ST3Gal-4 shRNAs decrease CLA surface expression. Schematic representation of CLA and PSGL-1 molecules using Biorender.com (left). Dot plots (right) represent the transduction rate of My-La cells with a lentivirus encoding a GFP reporter and shRNAs targeting ST3Gal-4 or a nonsilencing control. Histograms show the decrease of CLA expression on My-La cell surface after ST3Gal-4 silencing. (b, c, d) A decrease in CLA cell surface expression inhibits (b) transendothelial migration and (c) cell number and increases (d) cell death. (b) shControl or shST3Gal-4‒transduced My-La cells were plated for 3 hours in Transwell coated with human microvascular cells. The histogram represents the percentage of My-La cell migration. (c, d) Cell count and death percentage in each condition after 72 h. Statistical analysis was performed using a Mann‒Whitney test. ∗∗P < 0.001 and ∗∗∗P < 0.0001. Data are presented as mean ± SEM. n = 3. CLA, cutaneous lymphocyte antigen; CTCL, cutaneous T-cell lymphoma; h, hour; shControl, control-targeted short hairpin RNA; shRNA, short hairpin RNA; shST3Gal-4, ST3 β-galactoside α-2,3-sialyltransferase 4‒targeted short hairpin RNA; ST3Gal-4, ST3 β-galactoside α-2,3-sialyltransferase 4.
      The decrease of CLA expression at the CTCL cell surface was also associated with significantly decreased cell numbers after 72 hours of culture (Figure 3c). Both decreased proliferation in shST3Gal-4 ex8 condition (Supplementary Figure S6c) and increased cell death were observed with shST3Gal-4 ex5 and shST3Gal-4 ex8 compared with the control-targeted shRNA condition (n = 3, 37.8 and 49.7 vs. 20.7%, P < 0.001) (Figure 3d). These data support that the presence of CLA on the cell surface is involved in CTCL transendothelial migration, proliferation, and survival.

      Targeting CLA mediates cell apoptosis

      To understand the mechanisms of cell death induced by the anti-CLA mAb, caspases-3/7, caspase-8, and caspase-9 activities and mitochondria permeability were measured in My-La cells 30 minutes and 3 hours after treatment. At these two time points, the anti-CLA antibody induced an increase in the caspase 3/7 activity compared with isotype control antibody (Figure 4a). Caspase-8 and caspase-9 activities were also both increased by the anti-CLA mAb, showing that the cell death is due to a global activation of apoptotic pathways (Figure 4a). Tetramethylrhodamine methyl ester staining was also reduced after treatment with the anti-CLA mAb, showing a loss of the mitochondrial membrane potential and suggesting that at least part of My-La cell apoptosis is due to the induction of the intrinsic pathway (38.12%, P < 0.001) (Figure 4b). Altogether, the anti-CLA antibody induced rapid activation of the different apoptotic pathways in My-La cells.
      Figure thumbnail gr4
      Figure 4Anti-CLA mAb activates apoptotic pathways. (a) Anti-CLA mAb increases caspase 3/7, 8, and 9 activities. My-La cells were incubated with anti-CLA antibody mAb (HECA-452) or isotype control for 30 min or 3 hours. The percentage of apoptotic cells (caspase activity positive) and necrotic cells (Hoechst+, caspase activity negative) were studied by flow cytometry (left). Quantification of cells with caspase 3/7, 8, and 9 activities after 30 min or 3 hours of treatment (right). Statistical analysis was performed using a Mann‒Whitney test. ∗P < 0.05 and ∗∗∗P < 0.0001. (b) Anti-CLA mAb increases mitochondria permeability. Mitochondrial membrane potential was measured by flow cytometry using TMRM dye on My-La cell line after 3 hours of incubation with an anti-CLA antibody (HECA-452) or isotype. Statistical analysis was performed using a Mann‒Whitney test. ∗∗P < 0.001. Data are presented as mean ± SD. CLA, cutaneous lymphocyte antigen; min, minute; TMRM, tetramethylrhodamine methyl ester.

      Anti-CLA treatment impairs the capacity of tumorigenesis of My-La cells in vivo

      My-La cells were previously treated for 2 hours with the anti-CLA mAb or with the corresponding isotype control antibody and then injected intrahepatically in immunodeficient mice. After 2 weeks of clinical monitoring, the mice were killed (Figure 5a).
      Figure thumbnail gr5
      Figure 5Anti-CLA mAb pretreatment decreases the survival and dissemination of CTCL cells in vivo. (a) Experimental strategy. Schematic representation of the experimental strategy in vivo created using Biorender.com. (b) Pictures of livers from mice treated with anti-CLA mAb (HECA-452) or its isotype control. We can assess the invasion at the site of injection (n = 5 for each group). (c) Decreased percentage of HLA-ABC+ cells in the peritumoral liver section and the kidney in the anti-CLA group. Example of IHC using anti‒HLA-ABC on peritumoral liver and kidney sections of mice injected with cells treated with anti-CLA mAb or isotype control and quantification of HLA-ABC+ cells area percentages. ∗P < 0.05. Scale bars = 500 μm. CLA, cutaneous lymphocyte antigen; CTCL, cutaneous T-cell lymphoma; IHC, immunohistochemistry.
      Macroscopic analysis showed a decrease in tumor size at the site of injection in the anti-CLA antibody group compared with that in the control group (Figure 5b). The percentage of HLA-ABC+ cells analyzed by immunohistochemistry on peritumoral liver and kidney sections was significantly lower in the anti-CLA group than in the control group (Figure 5c). In both groups, clusters of My-La cells were observed at portal spaces and mainly in the peritubular areas of the renal cortex without predominant infiltration of the glomeruli or juxtaglomerular apparatus (Supplementary Figure S7). The anti-CLA treatment on My-La cells led to a decrease in both tumorigenesis and the spreading of My-La cells in different tissues.

      Discussion

      The CLA molecule was found expressed on CTCL cell surface on skin biopsies a long time ago (
      • Kamarashev J.
      • Burg G.
      • Kempf W.
      • Hess Schmid M.H.
      • Dummer R.
      Comparative analysis of histological and immunohistological features in mycosis fungoides and Sezary syndrome.
      ). However, our work brings evidence of the biological role of CLA in skin homing and tumorigenesis properties of CTCL cells. It provides evidence that CLA is involved in transendothelial migration of CTCL cells from the samples of patients with SS and a T-MF cell line, as already shown for normal memory T cells (
      • Clark R.A.
      • Chong B.
      • Mirchandani N.
      • Brinster N.K.
      • Yamanaka K.
      • Dowgiert R.K.
      • et al.
      The vast majority of CLA+ T cells are resident in normal skin.
      ). Surprisingly, the targeting of the CLA molecule or decreasing its expression on the cell surface impairs in vitro survival of CTCL cells but not of normal CD4+ T cells. This loss of survival occurs through apoptotic signal resulting in caspase 3/7, 8, and 9 activation. Using a mice xenograft model (
      • Andrique L.
      • Poglio S.
      • Prochazkova-Carlotti M.
      • Kadin M.E.
      • Giese A.
      • Idrissi Y.
      • et al.
      Intrahepatic xenograft of cutaneous T-cell lymphoma cell lines: A useful model for rapid biological and therapeutic evaluation.
      ), we observed that targeting the CLA molecule decreases CTCL tumorigenesis and spreading in vivo.
      As expected, we showed a wide interpatient heterogeneity in CLA expression on SCs in blood and skin biopsies (
      • Campbell J.J.
      • Clark R.A.
      • Watanabe R.
      • Kupper T.S.
      Sezary syndrome and mycosis fungoides arise from distinct T-cell subsets: a biologic rationale for their distinct clinical behaviors.
      ;
      • Kamarashev J.
      • Burg G.
      • Kempf W.
      • Hess Schmid M.H.
      • Dummer R.
      Comparative analysis of histological and immunohistological features in mycosis fungoides and Sezary syndrome.
      ;
      • Sokolowska-Wojdylo M.
      • Wenzel J.
      • Gaffal E.
      • Lenz J.
      • Speuser P.
      • Erdmann S.
      • et al.
      Circulating clonal CLA(+) and CD4(+) T cells in Sezary syndrome express the skin-homing chemokine receptors CCR4 and CCR10 as well as the lymph node-homing chemokine receptor CCR7.
      ). The absence of correlation between the number of circulating and skin-resident CD3+CLA+ cells supports that CLA expression may be heterogeneous and/or plastic between blood and skin compartments. Interestingly, a recent single-cell study showed that malignant T cells from the skin exhibit higher expression of genes related to T-cell activation status and cell cycle progression than those from blood cells (
      • Herrera A.
      • Cheng A.
      • Mimitou E.P.
      • Seffens A.
      • George D.D.
      • Bar-Natan M.
      • et al.
      Multimodal single-cell analysis of cutaneous T cell lymphoma reveals distinct sub-clonal tissue-dependent signatures.
      ). Another study also revealed a higher proliferative index in the skin-derived T cells in SS than those from the blood, explained by a higher mTOR pathway activation in skin-derived SCs through the tumor skin microenvironment (
      • Cristofoletti C.
      • Bresin A.
      • Picozza M.
      • Picchio M.C.
      • Monzo F.
      • Helmer Citterich M.
      • et al.
      Blood and skin-derived Sezary cells: differences in proliferation-index, activation of PI3K/AKT/mTORC1 pathway and its prognostic relevance.
      ). Whether the difference in the proliferative status of these two SC compartments may account for the difference in CLA expression level remains to be investigated.
      CLA is a molecule involved in memory T-cell homing into the skin. In the literature, the effect of the blocking potential of anti-CLA mAb (HECA-452 clone) on immune cells notably was not well-established (
      • De Boer O.J.
      • Horst E.
      • Pals S.T.
      • Bos J.D.
      • Das P.K.
      Functional evidence that the HECA-452 antigen is involved in the adhesion of human neutrophils and lymphocytes to tumour necrosis factor-alpha-stimulated endothelial cells.
      ;
      • Kummitha C.M.
      • Shirure V.S.
      • Delgadillo L.F.
      • Deosarkar S.P.
      • Tees D.F.J.
      • Burdick M.M.
      • et al.
      HECA-452 is a non-function blocking antibody for isolated sialyl Lewis x adhesion to endothelial expressed E-selectin under flow conditions.
      ). In this study, we show that in a transwell system coating with human vascular cells, HECA-452 treatment blocks the CTCL transendothelial migration. This decrease could be both explained by the cell death induced by treatment within a 3-hour time and by the destabilization of E-selectin‒CLA interaction. Indeed, we have also shown that overexpression of E-selectin on microvascular cell surface significantly increases CTCL cell transendothelial migration.
      The use of shRNAs in silencing the maturation enzyme ST3Gal-4 decreased CLA cell surface levels and had the same effect as using the HECA-452 antibody on cell migration. This suggests that the effect of the anti-CLA antibody would be due to blocking functions.
      ST3Gal-4 involvement in cancer cell migration processes was previously described. However, except for stem/stromal cell biology, its role in cell survival remains not well-established (
      • Guerrero P.E.
      • Miró L.
      • Wong B.S.
      • Massaguer A.
      • Martínez-Bosch N.
      • Llorens R.
      • et al.
      Knockdown of α2,3-sialyltransferases impairs pancreatic cancer cell migration, invasion and E-selectin-dependent adhesion.
      ;
      • Templeton K.
      • Ramos M.
      • Rose J.
      • Le B.
      • Zhou Q.
      • Cressman A.
      • et al.
      Mesenchymal stromal cells regulate sialylations of N-glycans, affecting cell migration and survival.
      ). Our results suggest that targeting ST3Gal-4 sialyl transferase in the context of CTCL could represent another strategy to induce tumor cell death. However, alternative therapeutic drugs are necessary because current inhibitors of sialyltransferase are not ST3Gal-4 specific and may have therefore important side effects (
      • Templeton K.
      • Ramos M.
      • Rose J.
      • Le B.
      • Zhou Q.
      • Cressman A.
      • et al.
      Mesenchymal stromal cells regulate sialylations of N-glycans, affecting cell migration and survival.
      ). These unexpected effects include an increase in tumor cell survival as observed in human multiple myeloma cells (
      • Natoni A.
      • Farrell M.L.
      • Harris S.
      • Falank C.
      • Kirkham-McCarthy L.
      • Macauley M.S.
      • et al.
      Sialyltransferase inhibition leads to inhibition of tumor cell interactions with E-selectin, VCAM1, and MADCAM1, and improves survival in a human multiple myeloma mouse model.
      ).
      So far, the impact of HECA-452 on CTCL cell survival has not been reported except after the combination of the mAb to a photoabsorber dye, the hydrophilic phtalocyanine IRdye 700DX (IR700), actionable by near-infrared light (
      • Silic-Benussi M.
      • Saponeri A.
      • Michelotto A.
      • Russo I.
      • Colombo A.
      • Pelizzo M.G.
      • et al.
      Near infrared photoimmunotherapy targeting the cutaneous lymphocyte antigen for mycosis fungoides.
      ). No direct effect of HECA-452 on CTCL survival was observed, which might be due to the use of a lower concentration of antibody than in our study. Interestingly, we showed that CLA targeting induced CTCL cell death with no effect on normal CD4+ T-cells survival, although they partially express CLA. Indeed, a lower level of CLA expression in normal cells than in the My-La cell line and patients' SCs could explain the specific effects on tumor cells.
      CLA is a molecule from the PSGL-1 family that differs by post-translational modifications and the addition of sialyl Lewisx motifs. PSGL-1 is described in the literature as a potential inhibitor of T-cell function by dampening TCR signaling through extracellular signal‒regulated kinase and protein kinase B dephosphorylation (
      • Tinoco R.
      • Otero D.C.
      • Takahashi A.A.
      • Bradley L.M.
      PSGL-1: a new player in the immune checkpoint landscape.
      ). Our data suggest that CLA may have an opposite effect in CTCL tumor cells because CLA expression protects them from apoptosis. Indeed, the process of T-cell survival may differ between CTCL and normal T cells. At basal levels and after stimulation, the activation of TCR pathways is reduced in CTCL cells compared with that in healthy donor cells (
      • Fargnoli M.C.
      • Edelson R.L.
      • Berger C.L.
      • Chimenti S.
      • Couture C.
      • Mustelin T.
      • et al.
      Diminished TCR signaling in cutaneous T cell lymphoma is associated with decreased activities of Zap70, Syk and membrane-associated CSK.
      ;
      • Klemke C.D.
      • Brenner D.
      • Weiss E.M.
      • Schmidt M.
      • Leverkus M.
      • Gülow K.
      • et al.
      Lack of T-cell receptor–induced signaling is crucial for CD95 ligand up-regulation and protects cutaneous T-cell lymphoma cells from activation-induced cell death.
      ). This phenomenon also confers resistance of CTCL cells to activation-induced cell death due to the downregulation of CD95 (
      • Klemke C.D.
      • Brenner D.
      • Weiss E.M.
      • Schmidt M.
      • Leverkus M.
      • Gülow K.
      • et al.
      Lack of T-cell receptor–induced signaling is crucial for CD95 ligand up-regulation and protects cutaneous T-cell lymphoma cells from activation-induced cell death.
      ). This differential activation or differentiation status of normal CD4+ and CTCL cells could account for a differential response to HECA-452 treatment.
      In addition, modulation of activation-induced cell death regulators was not observed after HECA-452 treatment of CTCL cells (data not shown). In contrast, anti-CLA antibody induced apoptotic pathways in CLA+ CTCL cells through an in vitro rapid activation (within 30 minutes) of caspases 3/7, 8, and 9 effectors compared with that of other potential or current therapeutic agents such as curbitacin, bexarotene, or cobomarsen (
      • Brouwer I.J.
      • Out-Luiting J.J.
      • Vermeer M.H.
      • Tensen C.P.
      Cucurbitacin E and I target the JAK/ STAT pathway and induce apoptosis in Sézary cells.
      ;
      • Seto A.G.
      • Beatty X.
      • Lynch J.M.
      • Hermreck M.
      • Tetzlaff M.
      • Duvic M.
      • et al.
      Cobomarsen, an oligonucleotide inhibitor of miR-155, co-ordinately regulates multiple survival pathways to reduce cellular proliferation and survival in cutaneous T-cell lymphoma.
      ). How HECA-452 exerts its effects did not require complement fixation because it induces cell death in the absence of serum (data not shown) and needs to be clarified.
      In this study, we also showed an in vivo biological role of CLA in T-cell tumorigenesis by pretreating the cells with mAbs and injecting them into a mouse model (
      • Andrique L.
      • Poglio S.
      • Prochazkova-Carlotti M.
      • Kadin M.E.
      • Giese A.
      • Idrissi Y.
      • et al.
      Intrahepatic xenograft of cutaneous T-cell lymphoma cell lines: A useful model for rapid biological and therapeutic evaluation.
      ). CLA targeting by HECA-452 treatment decreased tumor development at the injection site, probably through the induction of cell death and a decrease in the ability of CTCL cells to generate a tumor in vivo.
      This work evidenced the specific roles of CLA in CTCL cell survival, proliferation, and migration. In addition, we also showed that CLA could be targeted either directly using a mAb on fresh patient SCs and a T-MF cell line or by silencing its maturation enzyme. Anti-CLA treatment impaired both CTCL cell migration and survival by inducing apoptosis that may depend on the level of CLA expression on CTCL cells and/or the activation status of CTCL cells, sparing normal CD4+ blood T cells from a side effect. CLA may therefore represent a valuable therapeutic target in the treatment of CTCL. Indeed, for a portion of patients whose tumor cells are CLA+, a drug targeting CLA could be proposed in combination with current therapies such as anti-CCR4.

      Materials and Methods

      Sézary patient cells

      Blood samples from 36 patients diagnosed with SS (T4NxMxB2 stage) were obtained from the Onco-Dermatology Department of the Bordeaux University Hospital (Bordeaux, France). Patient characteristics at the time of analysis are provided in Supplementary Table S1. Patients gave written informed consent in accordance with the declaration of Helsinki and national ethic rules. The French Ministry of Higher Education, Research, and Innovation has approved the manipulations of the samples of patients with SS (DC-2008-412). PBMC were isolated using a Pancoll centrifugation technique (PAN-Biotech, Aidenbach, Germany). Tumor (TCRVβ+CD3+) cells were purified by FACS from PBMC as described below and then characterized in the different experiments.

      Patient cell cultures

      SCs from patients purified by flow cytometry cell sorting as described below were cultured as previously described (
      • Armstrong F.
      • Brunet de la Grange P.
      • Gerby B.
      • Rouyez M.C.
      • Calvo J.
      • Fontenay M.
      • et al.
      NOTCH is a key regulator of human T-cell acute leukemia initiating cell activity.
      ;
      • Poglio S.
      • Prochazkova-Carlotti M.
      • Cherrier F.
      • Gros A.
      • Laharanne E.
      • Pham-Ledard A.
      • et al.
      Xenograft and cell culture models of Sézary syndrome reveal cell of origin diversity and subclonal heterogeneity.
      ). Briefly, SCs were cultivated in a T-cell lymphoma medium that is composed of α-MEM medium, 10% fetal bovine serum, 10% human AB serum (Jacques Boy Biotechnology Institute, Reims, France), 50 ng/ml human stem cell factor, 20 ng/ml human Fms-like tyrosine kinase-3, 20 ng/ml IL-7, 20 nM insulin, and 1% penicillin/streptomycin. SCs were held at 37 °C, in an atmosphere with 5% carbon dioxide, in an incubator.

      Flow cytometry analyses

      TCRVβ variant characterization

      The clonal population was characterized by the dominant TCRVβ variant using the TCR Vβ Repertory Kit (Beckman Coulter, Brea, CA). Staining and controls were made according to the provider’s instructions with anti‒CD8-PE-Cy7 (RPA-T8), anti‒CD4-BV421 (RPA-T4), and anti‒CD3-APC-H7 (SK7) antibodies.

      Quantification of CLA expression

      SCs were pretreated with fragment constant blocking reagent (human BD fragment constant Block, BD Biosciences, Franklin Lakes, New Jersey). SCs were then stained with anti‒CLA-APC (HECA-452), anti‒TCRVβ-PE (dominant clone), and anti‒CD3-APC-H7 (SK7) or CD4-APC-H7/BV421 (RPA-T4). The percentage of CLA+ cells was quantified by flow cytometry (FACS) among TCRVβ+CD3+/CD4+ cells or CD4+CD3+ cells for patients with an indeterminate TCRVβ.
      All FACS analyses were performed with a Canto II cytometer (BD Biosciences, Le Pont-de-Claix, France) at the UB’Facsility platform (TBM Core, Bordeaux, France). BD FACSDiva software (BD Biosciences) was used for data interpretation.

      Sézary and My-La cell sorting

      SCs were previously stained with anti‒TCRVβ-PE, anti‒CD8-PE-Cy7, anti‒CD4-BV421, and anti‒CD3-APC-H7 antibodies, as mentioned earlier. The clonal cell population (TCRVβ+) was sorted using BDFACSAria (BD Biosciences). TCRVβ+ cells were selected among CD3+, CD4+, and CD8‒ cells, and the purity was evaluated. The transduced My-La cells were sorted by selecting the transduced GFP+ cells.

      Transendothelial migration assays

      A total of 5 and 8 μm-porous transwells were used for the SC and the My-La cell line, respectively (ThinCert, Greiner Bio-one, Chon Buri, Thailand). Transwells were coated with 0.2% gelatin solution for 1 hour at 37 °C, and 1.5 × 105 human microvascular cells from the dermis or human umbilical vein endothelial cells were seeded into transwells. After 2 days, chemoattractants were added in the lower chambers: SDF-1 (100 ng/ml) of CCL19/21 (100 ng/ml). Then, 5 × 105 sorted Sézary or My-La cells were pretreated for 30 minutes with 30 or 15 μg/ml anti-CLA mAb, respectively, (HECA-452, BioLegend, San Diego, CA) or its isotype control (RTK2118) or not treated (control condition) and seeded in the upper chamber. After 3 hours of incubation at 37 °C, cells in the lower chamber were counted, and the percentage of migrating cells was calculated.

      Ethics statement

      Animal experiments were performed in level 2 animal facilities (Bordeaux University) after approval by the French Ministry of Higher Education, Research and Innovation and with the agreement of the local Ethics Committee on Animal Experiments CEEA50 of Bordeaux (national agreement number APAFIS#29572-2021020517035043 v4).

      Intrahepatic mouse model

      A total of 1 × 106 cells were intrahepatically xenografted in male NSG immunodeficient mice (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ) aged 8‒12 weeks under 2.5% isoflurane anesthesia (Belmont, Piramal Healthcare, Northumberland, United Kingdom), as previously described (
      • Andrique L.
      • Poglio S.
      • Prochazkova-Carlotti M.
      • Kadin M.E.
      • Giese A.
      • Idrissi Y.
      • et al.
      Intrahepatic xenograft of cutaneous T-cell lymphoma cell lines: A useful model for rapid biological and therapeutic evaluation.
      ). Mice were clinically monitored for 2 weeks with global surveillance of animal health. Two weeks after engraftment, mice were killed, body and liver weights were measured, and liver/body ratio was then determined. Liver and kidneys were fixed in 4% formaldehyde (Merck Millipore, VWR International S.A.S, Fontenay-sous-Bois, France) for immunohistochemistry.

      Immunohistochemistry

      Immunohistochemistry was performed as described previously (
      • Andrique L.
      • Poglio S.
      • Prochazkova-Carlotti M.
      • Kadin M.E.
      • Giese A.
      • Idrissi Y.
      • et al.
      Intrahepatic xenograft of cutaneous T-cell lymphoma cell lines: A useful model for rapid biological and therapeutic evaluation.
      ) and in the Supplementary Materials and Methods.

      Statistical analysis

      The statistical analysis was performed using the Prism software (GraphPad Software, La Jolla, CA). The unpaired Mann‒Whitney test was used to analyze migration and survival assays with Spearman correlation test. Significance was defined as ∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001.

      Data availability statement

      No datasets were generated or analyzed during this study.

      Conflicts of Interest

      The authors state no conflict of interest.

      Acknowledgments

      The authors would like to thank the team of the Onco-Dermatology and Pathology Tumor Biobank Departments (Bordeaux University Hospital) for helping us in providing fresh and fixed patient samples and also the patients who agreed with informed consent to the research. Mouse experiments were greatly facilitated by Benoît Rousseau and Julien Izotte from the A2 Animal facility at Bordeaux University. We thank the vectorology platform Vect'UB for providing lentiviral particles and for technical support and CNRS UMS3427, INSERM US005, Université de Bordeaux. We thank Atika Zouine and Vincent Pitard for technical assistance at the Flow cytometry facility, CNRS UMS 3427, INSERM US005, Université de Bordeaux. This work was supported by Inserm , Bordeaux University and CHU Bordeaux institutions and grants from the Ligue Régionale Contre le Cancer (Comités de Gironde and Dordogne) and the Société Française de Dermatologie.

      Author Contributions

      Conceptualization: SPo, MPC, SPe; Formal Analysis: SPo, SPe, MPC, FC, JV, YI, LAM, ER; Funding Acquisition: SPo; Investigation: SPo, SPe, MPC, FC, JV, YI, LAM, ER; Methodology: SPo, MPC, DC; Supervision: SPo; Writing - Original Draft Preparation: SPo, SPe; Writing - Review and Editing: MPC, MBB, APL, JPM

      Supplementary Materials

      Cutaneous T-cell lymphoma cell line culture

      My-La cell line derived from a T-MF (kindly gifted by Kaltof, Aarhus University, Aarhus, Denmark) (
      • Kaltoft K.
      • Bisballe S.
      • Dyrberg T.
      • Boel E.
      • Rasmussen P.B.
      • Thestrup-Pedersen K.
      Establishment of two continuous T-cell strains from a single plaque of a patient with mycosis fungoides.
      ). Cells were cultured in RPMI medium (with 10% fetal bovine serum and 1% penicillin/streptomycin) at 37 °C, in an atmosphere with 5% carbon dioxide, in an incubator.

      Endothelial cell culture

      We used human microvascular cells derived from the dermis and human umbilical vein endothelial cells derived from the umbilical vein (Lonza, Bale, Switzerland). Cells were cultured in EBM-2 (Lonza) at 37 °C, in an atmosphere with 5% carbon dioxide, in an incubator. All the compounds used to maintain the cells were those recommended by the cell lines provider (Lonza).

      Transduction of cells

      pLKO.1-U6-GFP vectors encoding a GFP reporter and a control-targeted short hairpin RNA or short hairpin RNA targeting exon 5 (ST3 β-galactoside α-2,3-sialyltransferase 4‒targeted short hairpin RNA [shST3Gal-4] ex5) or exon 8 (shST3Gal-4 ex8) of ST3 β-galactoside α-2,3-sialyltransferase 4 gene ST3Gal4 transcripts were produced in our laboratory (Supplementary Figure S5). The lentiviral production was done by the Vect’UB platform (TBM Core, INSERM UMS 005, Bordeaux, France). The My-La line was transduced at a Multiplicity of Infection of 10.

      RT-qPCR

      Total RNA from control-targeted short hairpin RNA‒, shST3Gal-4 ex5‒, and shST3Gal-4 ex8‒transduced cells was extracted using the Direct-zol RNA MiniPrep (ZYMO Research, Irvine, CA) as recommended by the manufacturer. A total of 200 ng of RNA were used to generate cDNA using the SuperScript II reverse transcriptase kit (Invitrogen, Waltham, MA). qPCR on the ST3Gal4 gene was performed using Takyon No Rox SYBR MasterMix dttP Blue (Eurogentec, Seraing, France) with the following primers: 5′-CAAGACGCCATCTGCTTACGA-3′ forward and 5′-GCACCTGAGGCTCTGGATGTT-3′ reverse. The Stratagene Mx3005P system (Agilent Technologies, Santa Clara, CA) was used to perform the RT-qPCR. Gene expression was normalized to the expression of the housekeeping gene TBP (forward primer: 5'-CACGAACCACGGCACTGATT-3', reverse primer:5'-TTTTCTTGCTGCCAGTCTGGA-3'). Relative gene expression was calculated using the 2-ΔΔCt according to the method described by
      • Livak K.J.
      • Schmittgen T.D.
      Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method.
      .

      Proliferation assays

      Proliferation assays were performed using CarboxyFluorescein Succinimidyl Ester (Cell Trace CFSE Proliferation kit, Fisher Scientific, Illkirch, France)-labeled on My-La cell line according to the manufacturer instructions. My-La control-targeted short hairpin RNA‒, shST3Gal-4 ex5‒, or shST3Gal-4 ex8‒transduced cells were cultured in their medium for 3 days, and acquisition was performed with a Canto II cytometer (BD Biosciences, Franklin Lakes, New Jersey) at the UB’Facsility platform (TransBiomed Core facility, Bordeaux University, Bordeaux, France). FlowJo software (BD Biosciences, Le Pont de Claix, France) was used for data interpretation.

      Apoptosis‒necrosis assay

      Apoptosis‒necrosis assay was performed on Sézary cells, My-La cell line, or healthy donors' CD4+ T cells. A total of 1 × 105 cells were cultured in 24-well plate in their medium and treated with an anti‒cutaneous lymphocyte antigen mAb or its isotype control (30 μg/ml) for 3 or 72 hours with a treatment every 24 hours. Data were obtained from the mean of replicates for each condition. The percentage of total dead cells was evaluated by FACS after staining with Annexin V-APC and Hoechst as recommended by the manufacturer. Annexin V‒ Hoechst+ cells were considered necrotic cells, early apoptotic cells were Annexin V+ and Hoechst‒, and late apoptotic cells were Annexin V+ and Hoechst+. Apoptosis was assessed by measuring the activities of caspases-3/7, caspase-8, and caspase-9 by FACS using the FAM-FLICA detection kit according to the manufacturer’s instructions (Bio-Rad Laboratories, Marnes-la-Coquette, France). Apoptosis was also detected by the loss of mitochondrial membrane potential using tetramethylrhodamine methyl ester dye. After treatment, cells were incubated with 200 nM tetramethylrhodamine methyl ester for 30 minutes at 37 °C, and the fluorescence was measured by flow cytometry as described earlier. The loss of the fluorescence corresponds to an increase in mitochondria permeability because tetramethylrhodamine methyl ester is a fluorescent lipophilic cation able to accumulate in active mitochondria with intact membrane potentials.

      Immunohistochemistry

      Mayer’s hematoxylin labeling, human leukocyte antigen (HLA-ABC, 1:100 of antibody 70328; Abcam, Paris, France), and cutaneous lymphocyte antigen (HECA-452, 1:100 of antibody; Biolegend, San Diego, CA) immunohistochemistry were performed on mice organ or human Sézary syndrome skin biopsy sections (3-μm thick of formalin-fixed, paraffin-embedded organs). A secondary horseradish peroxidase‒labeled ImmPRESS anti-mouse kit (Vector Laboratories, Burlingame, CA) and the Liquid DAB+ Substrate Chromogen System (Dako, Glostrup, Denmark) were used to reveal HLA-ABC or CLA‒positive cells. Slides were analyzed using Panoramic Scan (3DHISTECH, Budapest, Hungary) and Panoramic Viewer software, version 1.15.4 (3DHISTECH,) or using microscope LEICA DMR coupled with a camera NIKON DS-FI2 and NIS BR imaging software, version 4.0 (Nikon, Champigny sur Marne, France). Quantitative evaluation of HLA-ABC or CLA labeling was performed using ImageJ (National Institutes of Health, Bethesda, MD) image analysis software, version 1.48 (Java), or HALO image analysis software (Indica Labs, Albuquerque, New Mexico).

      Immunofluorescence

      Immunofluorescence was performed on human Sézary syndrome skin biopsies or control normal skin sections (3-μm thick of formalin-fixed, paraffin-embedded organs). After blocking, cells were incubated with a rabbit anti-CD3 antibody (polyclonal 1:200, A0452, Agilent) and rat anti‒cutaneous lymphocyte antibody (HECA-452, 1:50 of Antibody; BioLegend) and with a fluorescent-labeled secondary anti-rabbit Alexa 594 labeled (goat, 1:400, A32740, Molecular Probes, Fischer Scientific, Waltham, MA) and anti-rat Alexa 488‒labeled (goat, 1:400, A11006, Molecular Probes, Fischer Scientific, Waltham, MA). The coverslips were mounted on glass slides using an antifade mounting medium with DAPI (VECTASHIELD, Vector Laboratories). Images and scans were done using a microscope ZEISS Axiolmager Z2 coupled with MetaSystems imaging software or Metafer scanning system (MetaSystems, Altlussheim, Germany). The quantitative evaluation of cutaneous lymphocyte antibody-positive area labeling/total tissue area was performed using HALO image analysis software (Indica Labs).
      Figure thumbnail fx1
      Supplementary Figure S1CLA+ T cells in patients with SS and normal skin biopsies. (a) IHC showing the expression of CLA molecule in skin biopsies of patients with SS. Example of IHC using anti-CLA mAb (HECA-452) on skin sections of three other patients with SS. Scale bar = 50 μm. (b) Immunofluorescence showing the expression of CLA molecule by CD3+ cells in SS and normal skin biopsies. Examples of CLA (green), CD3 (red), and DAPI (blue) stainings. Scale bars = 20 μm. (c) Quantification of the percentage of CLA+ cells in CD3+ population using HALO software. CLA, cutaneous lymphocyte antigen; IHC, immunohistochemistry; SS, Sézary syndrome.
      Figure thumbnail fx2
      Supplementary Figure S2Anti-CLA mAb inhibits CTCL cells transendothelial migration. (a) E-selectin expression. E-selectin expression on HUVECs treated or not treated with LPS for 8 hours. (b) Percentage of migration of CTCL cells treated with anti-CLA mAb or isotype control. My-La cells were incubated with an anti-CLA mAb (15 μg/ml HECA-452) or isotype control in a Transwell system coated with HUVECs pretreated or not with LPS for 8 hours. Three hours after induction of migration by addition of SDF1α (100 ng/ml) and/or CCL19/21 (100 ng/ml) in the lower chamber, migrating cells were harvested and counted. The histogram represents the percentage of migration of My-La cells treated with anti-CLA mAb or isotype control. Statistical analysis was performed using a Mann‒Whitney test. ∗∗P < 0.001 and ∗∗∗P < 0.0001. CLA, cutaneous lymphocyte antigen; CTCL, cutaneous T-cell lymphoma; HUVEC, human umbilical vein endothelial cell; LPS, lipopolysaccharide.
      Figure thumbnail fx3
      Supplementary Figure S3Anti-CLA mAb alters the survival of CTCL cells. (a‒g). Percentage of CLA+ cells and evaluation of cell death of CTCL cells treated with anti-CLA or isotype control. SCs from patients were incubated with an anti-CLA mAb (HECA-452) or isotype control for 3 h. The percentage of apoptotic cells (Annexin V+) and necrotic cells (Hoechst+, Annexin V‒) were determined by flow cytometry. (h, i) Percentage of cells expressing CLA and its expression intensity are correlated with cell death. The RFI used for this analysis is calculated on the day of the treatment experiment. Correlation between the percentage of CLA+ cells or RFI of CLA and cell death induced after treatment with anti-CLA mAb (HECA-452). Statistical analysis was performed using a Mann‒Whitney test. ∗P < 0.05. The correlations were analyzed by linear regression test using Graph Pad Prism software. CLA, cutaneous lymphocyte antigen; CTCL, cutaneous T-cell lymphoma; h, hour; RFI, relative fluorescence intensity; SC, Sézary cell.
      Figure thumbnail fx4
      Supplementary Figure S4Anti-CLA mAb has no effect on CLA-negative CTCL cell line survival and healthy donor cells. (a, b) Percentage of cell death of (a) CTCL cell lines or (b) three healthy donors CD3+CD4+ cells treated with anti-CLA mAb or isotype control. Cells were incubated with an anti-CLA mAb (HECA-452) or isotype control for 3 or 72 h with a treatment every 24 h. The percentage of apoptotic cells (Annexin V+) and necrotic cells (Hoechst+, Annexin V‒) were studied by flow cytometry (left). Quantification of My-La cell death (percentage of apoptotic and necrotic cells) after 3 hours of treatment (right). (c) PBMCs prepared from healthy donors were analyzed by flow cytometry to determine the percentage of CD3+CD4+CLA+ T cells. Statistical analysis was performed using a Mann‒Whitney test. ∗P < 0.05. CLA, cutaneous lymphocyte antigen; CTCL, cutaneous T-cell lymphoma; h, hour.
      Figure thumbnail fx5
      Supplementary Figure S5Lentiviral vector construct. (a) pLKO.1-U6-GFP vectors. pLKO.1-U6-GFP vectors are encoding a GFP reporter and a shControl or shex5 or shex8 of shST3Gal-4 transcripts. (b) Sequences of sense-strand oligonucleotides used for cloning of ST3Gal-4 ex5 and ST3Gal-4 ex8 shRNA vectors as well as forward and reverse primers used for RT-qPCR analysis of ST3Gal-4 transcript levels. LTR, long terminal repeat; shControl, control-targeted short hairpin RNA; shex5, short hairpin RNAs targeting exon 5; shex8, short hairpin RNAs targeting exon 8; shRNA, short hairpin RNA; shST3Gal-4, ST3 β-galactoside α-2,3-sialyltransferase 4‒targeted short hairpin RNA.
      Figure thumbnail fx6
      Supplementary Figure S6Decreasing cell surface CLA expression by ST3Gal-4 gene silencing impairs cell proliferation and has no effect on PSGL-1 expression. (a) Relative mRNA expression of ST3Gal-4. Evaluation of the relative mRNA expression of ST3Gal-4 in My-La cells transduced with the lentivirus encoding a control shRNA or shRNA targeting exon 5 or exon 8 of ST3Gal-4 gene transcripts. (b) PSGL-1 expression of CTCL cells after ST3Gal-4 inhibition. Evaluation of the PSGL-1 expression after transduction with lentivirus encoding a control shRNA or shRNA targeting exon 5 or exon 8 of ST3Gal-4 transcripts. (c) Effect of CLA inhibition by ST3Gal-4 gene silencing on proliferation was evaluated on My-La cells CFSE+. The intensity of CFSE staining was evaluated by FACS after 3 days in culture. Statistical analysis was performed using a Mann‒Whitney test. ∗P < 0.05. CFSE, carboxyfluorescein succinimidyl ester; CLA, cutaneous lymphocyte antigen; shRNA, short hairpin RNA; ST3Gal-4, ST3 β-galactoside α-2,3-sialyltransferase 4.
      Figure thumbnail fx7
      Supplementary Figure S7Liver and hepatic involvements after injection of pretreated My-La cells in NSG mice. Example of IHC using anti‒HLA-ABC antibody (brown areas) on peritumoral liver and kidney sections of mice injected with cells treated with anti-CLA mAb or isotype control. Black arrows show a sinusoid of livers. Red arrows show the kidney glomeruli. Scale bars on the left column = 100 μm. Scale bars on the right column = 50 μm. CLA, cutaneous lymphocyte antigen; IHC, immunohistochemistry.
      Supplementary Table S1Patients’ Characteristics and Percentage of CLA Expression on Patients’ Circulating SS Cells
      PatientsSexAge at SS Diagnosis (y)Stage at AnalysisPercentage of Circulating CLA+ Cells among SS Cell CompartmentRFI
      Patient 1M45T4N3M1B25572
      Patient 2F51T4N3M0B2860
      Patient 3M63T4N0M0B22959
      Patient 4M66T4N3M0B23554
      Patient 5F81T4N0M0B298340
      Patient 6F81T4N3M0B20155
      Patient 7F80T4N0M0B297295
      Patient 8F73T4N0M0B21245
      Patient 9M64T4N0M0B25799
      Patient 10F60T4N1M0B25227
      Patient 11F77T4N0M0B22888
      Patient 12F57T4N0M0B24621
      Patient 13F79T4N0M0B26831
      Patient 14M51T4N0M0B23628
      Patient 15F89T4N0M0B285587
      Patient 16M82T4N1M0B290440
      Patient 17M57T4N0M0B27758
      Patient 18F77T2N0M0B191216
      Patient 19F64T4N0M0B11328
      Patient 20F81T4N0M0B11455
      Patient 21F70T2N0M0B25336
      Patient 22F87T4N0M0B23952
      Patient 23F63T4N0M0B21842
      Patient 25M64T4N3M0B25356
      Patient 27M59T4N3M0B23219
      Patient 28F87T2bN0M0B21352
      Patient 29F79T1N0M0B22047
      Patient 30F93T4N1M0B23636
      Patient 32F60T4N3M0B253176
      Patient 33M66T4NXM0B23043
      Patient 34M64T4N3M0B23685
      Patient 35M29T4N3M0B253248
      Patient 36M67T4N1M0B251119
      Patient 38F61T4N0M0B121555
      Patient 39M62T4N0M0B245631
      Patient 40F93T2aN0M0B298362
      Median4259
      Mean145145
      Abbreviations: CLA, cutaneous lymphocyte antigen; F, female; M, male; RFI, relative fluorescence intensity; SS, Sézary syndrome.
      Supplementary Table S2My-La Cell Line and CD3+CD4+ Healthy Donors Cells Percentage of CLA Expression and RFI
      Cell Type% CLA+ CellsRFI
      My-LA10093
      CD3+CD4+ healthy donor cells3334
      2733
      3335
      Abbreviations: CLA, cutaneous lymphocyte antigen; RFI, relative fluoresence intensity.

      References

        • Andrique L.
        • Poglio S.
        • Prochazkova-Carlotti M.
        • Kadin M.E.
        • Giese A.
        • Idrissi Y.
        • et al.
        Intrahepatic xenograft of cutaneous T-cell lymphoma cell lines: A useful model for rapid biological and therapeutic evaluation.
        Am J Pathol. 2016; 186: 1775-1785
        • Armstrong F.
        • Brunet de la Grange P.
        • Gerby B.
        • Rouyez M.C.
        • Calvo J.
        • Fontenay M.
        • et al.
        NOTCH is a key regulator of human T-cell acute leukemia initiating cell activity.
        Blood. 2009; 113: 1730-1740
        • Berg E.L.
        • Yoshino T.
        • Rott L.S.
        • Robinson M.K.
        • Warnock R.A.
        • Kishimoto T.K.
        • et al.
        The cutaneous lymphocyte antigen is a skin lymphocyte homing receptor for the vascular lectin endothelial cell-leukocyte adhesion molecule 1.
        J Exp Med. 1991; 174: 1461-1466
        • Brouwer I.J.
        • Out-Luiting J.J.
        • Vermeer M.H.
        • Tensen C.P.
        Cucurbitacin E and I target the JAK/ STAT pathway and induce apoptosis in Sézary cells.
        Biochem Biophys Rep. 2020; 24100832
        • Campbell J.J.
        • Clark R.A.
        • Watanabe R.
        • Kupper T.S.
        Sezary syndrome and mycosis fungoides arise from distinct T-cell subsets: a biologic rationale for their distinct clinical behaviors.
        Blood. 2010; 116: 767-771
        • Chang D.K.
        • Sui J.
        • Geng S.
        • Muvaffak A.
        • Bai M.
        • Fuhlbrigge R.C.
        • et al.
        Humanization of an anti-CCR4 antibody that kills cutaneous T-cell lymphoma cells and abrogates suppression by T-regulatory cells.
        Mol Cancer Ther. 2012; 11: 2451-2461
        • Clark R.A.
        • Chong B.
        • Mirchandani N.
        • Brinster N.K.
        • Yamanaka K.
        • Dowgiert R.K.
        • et al.
        The vast majority of CLA+ T cells are resident in normal skin.
        J Immunol. 2006; 176: 4431-4439
        • Cristofoletti C.
        • Bresin A.
        • Picozza M.
        • Picchio M.C.
        • Monzo F.
        • Helmer Citterich M.
        • et al.
        Blood and skin-derived Sezary cells: differences in proliferation-index, activation of PI3K/AKT/mTORC1 pathway and its prognostic relevance.
        Leukemia. 2019; 33: 1231-1242
        • De Boer O.J.
        • Horst E.
        • Pals S.T.
        • Bos J.D.
        • Das P.K.
        Functional evidence that the HECA-452 antigen is involved in the adhesion of human neutrophils and lymphocytes to tumour necrosis factor-alpha-stimulated endothelial cells.
        Immunology. 1994; 81: 359-365
        • de Masson A.
        • Beylot-Barry M.
        • Bouaziz J.D.
        • Peffault de Latour R.P.
        • Aubin F.
        • Garciaz S.
        • et al.
        Allogeneic stem cell transplantation for advanced cutaneous T-cell lymphomas: a study from the French Society of Bone Marrow Transplantation and French Study Group on Cutaneous Lymphomas.
        Haematologica. 2014; 99: 527-534
        • Fargnoli M.C.
        • Edelson R.L.
        • Berger C.L.
        • Chimenti S.
        • Couture C.
        • Mustelin T.
        • et al.
        Diminished TCR signaling in cutaneous T cell lymphoma is associated with decreased activities of Zap70, Syk and membrane-associated CSK.
        Leukemia. 1997; 11: 1338-1346
        • Fink-Puches R.
        • Zenahlik P.
        • Bäck B.
        • Smolle J.
        • Kerl H.
        • Cerroni L.
        Primary cutaneous lymphomas: applicability of current classification schemes (European Organization for Research and Treatment of Cancer, World Health Organization) based on clinicopathologic features observed in a large group of patients.
        Blood. 2002; 99: 800-805
        • Fuhlbrigge R.C.
        • Kieffer J.D.
        • Armerding D.
        • Kupper T.S.
        Cutaneous lymphocyte antigen is a specialized form of PSGL-1 expressed on skin-homing T cells.
        Nature. 1997; 389: 978-981
        • Guerrero P.E.
        • Miró L.
        • Wong B.S.
        • Massaguer A.
        • Martínez-Bosch N.
        • Llorens R.
        • et al.
        Knockdown of α2,3-sialyltransferases impairs pancreatic cancer cell migration, invasion and E-selectin-dependent adhesion.
        Int J Mol Sci. 2020; 21: 6239
        • Herrera A.
        • Cheng A.
        • Mimitou E.P.
        • Seffens A.
        • George D.D.
        • Bar-Natan M.
        • et al.
        Multimodal single-cell analysis of cutaneous T cell lymphoma reveals distinct sub-clonal tissue-dependent signatures.
        Blood. 2021; 138: 1456-1464
        • Kamarashev J.
        • Burg G.
        • Kempf W.
        • Hess Schmid M.H.
        • Dummer R.
        Comparative analysis of histological and immunohistological features in mycosis fungoides and Sezary syndrome.
        J Cutan Pathol. 1998; 25: 407-412
        • Kasamon Y.L.
        • Chen H.
        • de Claro R.A.
        • Nie L.
        • Ye J.
        • Blumenthal G.M.
        • et al.
        FDA approval summary: Mogamulizumab-kpkc for mycosis fungoides and Sézary syndrome.
        Clin Cancer Res. 2019; 25: 7275-7280
        • Kim Y.H.
        • Bagot M.
        • Pinter-Brown L.
        • Rook A.H.
        • Porcu P.
        • Horwitz S.M.
        • et al.
        Mogamulizumab versus vorinostat in previously treated cutaneous T-cell lymphoma (MAVORIC): an international, open-label, randomised, controlled phase 3 trial.
        Lancet Oncol. 2018; 19 ([published correction appears in Lancet Oncol 2018;19:e581]): 1192-1204
        • Klemke C.D.
        • Brenner D.
        • Weiss E.M.
        • Schmidt M.
        • Leverkus M.
        • Gülow K.
        • et al.
        Lack of T-cell receptor–induced signaling is crucial for CD95 ligand up-regulation and protects cutaneous T-cell lymphoma cells from activation-induced cell death.
        Cancer Res. 2009; 69: 4175-4183
        • Kummitha C.M.
        • Shirure V.S.
        • Delgadillo L.F.
        • Deosarkar S.P.
        • Tees D.F.J.
        • Burdick M.M.
        • et al.
        HECA-452 is a non-function blocking antibody for isolated sialyl Lewis x adhesion to endothelial expressed E-selectin under flow conditions.
        J Immunol Methods. 2012; 384: 43-50
        • Maverakis E.
        • Kim K.
        • Shimoda M.
        • Gershwin M.E.
        • Patel F.
        • Wilken R.
        • et al.
        Glycans in the immune system and the Altered Glycan Theory of Autoimmunity: a critical review.
        J Autoimmun. 2015; 57: 1-13
        • Mondal N.
        • Buffone A.
        • Stolfa G.
        • Antonopoulos A.
        • Lau J.T.Y.
        • Haslam S.M.
        • et al.
        ST3Gal-4 is the primary sialyltransferase regulating the synthesis of E-, P-, and L-selectin ligands on human myeloid leukocytes.
        Blood. 2015; 125: 687-696
        • Natoni A.
        • Farrell M.L.
        • Harris S.
        • Falank C.
        • Kirkham-McCarthy L.
        • Macauley M.S.
        • et al.
        Sialyltransferase inhibition leads to inhibition of tumor cell interactions with E-selectin, VCAM1, and MADCAM1, and improves survival in a human multiple myeloma mouse model.
        Haematologica. 2020; 105: 457-467
        • Olsen E.
        • Vonderheid E.
        • Pimpinelli N.
        • Willemze R.
        • Kim Y.
        • Knobler R.
        • et al.
        Revisions to the staging and classification of mycosis fungoides and Sezary syndrome: a proposal of the International Society for Cutaneous Lymphomas (ISCL) and the cutaneous lymphoma task force of the European Organization of Research and Treatment of Cancer (EORTC).
        Blood. 2007; 110 ([published correction appears in Blood 2008;111:4830]): 1713-1722
        • Picker L.J.
        • Michie S.A.
        • Rott L.S.
        • Butcher E.C.
        A unique phenotype of skin-associated lymphocytes in humans. Preferential expression of the HECA-452 epitope by benign and malignant T cells at cutaneous sites.
        Am J Pathol. 1990; 136: 1053-1068
        • Poglio S.
        • Prochazkova-Carlotti M.
        • Cherrier F.
        • Gros A.
        • Laharanne E.
        • Pham-Ledard A.
        • et al.
        Xenograft and cell culture models of Sézary syndrome reveal cell of origin diversity and subclonal heterogeneity.
        Leukemia. 2021; 35: 1696-1709
        • Ramelyte E.
        • Dummer R.
        • Guenova E.
        Investigative drugs for the treatment of cutaneous T-cell lymphomas (CTCL): an update.
        Expert Opin Investig Drugs. 2019; 28: 799-809
        • Roelens M.
        • Delord M.
        • Ram-Wolff C.
        • Marie-Cardine A.
        • Alberdi A.
        • Maki G.
        • et al.
        Circulating and skin-derived Sézary cells: clonal but with phenotypic plasticity.
        Blood. 2017; 130: 1468-1471
        • Seto A.G.
        • Beatty X.
        • Lynch J.M.
        • Hermreck M.
        • Tetzlaff M.
        • Duvic M.
        • et al.
        Cobomarsen, an oligonucleotide inhibitor of miR-155, co-ordinately regulates multiple survival pathways to reduce cellular proliferation and survival in cutaneous T-cell lymphoma.
        Br J Haematol. 2018; 183: 428-444
        • Silic-Benussi M.
        • Saponeri A.
        • Michelotto A.
        • Russo I.
        • Colombo A.
        • Pelizzo M.G.
        • et al.
        Near infrared photoimmunotherapy targeting the cutaneous lymphocyte antigen for mycosis fungoides.
        Expert Opin Biol Ther. 2021; 21: 977-981
        • Sokolowska-Wojdylo M.
        • Wenzel J.
        • Gaffal E.
        • Lenz J.
        • Speuser P.
        • Erdmann S.
        • et al.
        Circulating clonal CLA(+) and CD4(+) T cells in Sezary syndrome express the skin-homing chemokine receptors CCR4 and CCR10 as well as the lymph node-homing chemokine receptor CCR7.
        Br J Dermatol. 2005; 152: 258-264
        • Templeton K.
        • Ramos M.
        • Rose J.
        • Le B.
        • Zhou Q.
        • Cressman A.
        • et al.
        Mesenchymal stromal cells regulate sialylations of N-glycans, affecting cell migration and survival.
        Int J Mol Sci. 2021; 22: 6868
        • Tinoco R.
        • Otero D.C.
        • Takahashi A.A.
        • Bradley L.M.
        PSGL-1: a new player in the immune checkpoint landscape.
        Trends Immunol. 2017; 38: 323-335
        • Wilcox R.A.
        Cutaneous T-cell lymphoma: 2016 update on diagnosis, risk-stratification, and management.
        Am J Hematol. 2016; 91: 151-165