Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Aug 1.
Published in final edited form as: J Invest Dermatol. 2022 Feb 7;142(8):2285–2288.e2. doi: 10.1016/j.jid.2022.01.009

IL-32 supports the survival of malignant T cells in cutaneous T cell lymphoma

Kenneth K Yu 1, Neal P Smith 2,3, Sernah V Essien 1, Jessica E Teague 1, Pablo Vieyra-Garcia 4, Ahmed Gehad 1, Qian Zhan 1, Jack D Crouch 1, Nega Gerard 1, Cecilia Larocca 1,5, Peter Wolf 4, Nicole R LeBoeuf 1,5, Marianne Tawa 1,5, Thomas S Kupper 1,5, Alexandra-Chloé Villani 2,3, Rachael A Clark 1,5,*
PMCID: PMC9329172  NIHMSID: NIHMS1778243  PMID: 35143819

To the Editor:

Cutaneous T-cell lymphoma is a heterogeneous collection of non-Hodgkin’s lymphomas derived from T cells tropic for the skin. Our understanding of this disease has been hampered by the fact that malignant T cells survive poorly in culture and do not survive well even in sophisticated xenograft models (Townsend et al., 2016). Isolation and culture of T cells from CTCL skin lesions leads to an overgrowth of benign infiltrating T cells and loss of the malignant T cell clone. It is an inescapable irony that malignant T cells capable of killing patients die in even the most supportive in vitro and in vivo environments. These cells appear to require a survival factor found in skin but not replicated by in vitro and in vivo murine systems.

We found that interleukin-32 (IL-32), a cytokine unique to humans, was the single most highly expressed interleukin in the skin lesions of mycosis fungoides (MF, Fig. 1a), in agreement with prior reports that IL-32 is increased in MF and correlates with disease progression (Ohmatsu et al., 2014, Suga et al., 2014, van Kester et al., 2012). Single-cell RNA sequencing (scRNAseq) of CD45+ hematopoetic cells from MF skin lesions demonstrated that IL-32 was expressed predominately by T cells in skin and was produced at highest levels by the malignant T cell clone and regulatory T cells (Fig. 1b,c). We confirmed expression of IL-32 at the protein level by multiplex immunostaining in both MF and leukemic CTCL/Sezary syndrome patients, utilizing malignant TCR Vβ immunostaining to discriminate malignant from benign infiltrating T cells (Fig. 1d,e). Malignant T cells expressed significantly more IL-32 than benign infiltrating T cells (Fig. 1f). Prior gene expression studies demonstrated that mRNA for the IL-32 β isoform was produced by circulating T cells in Sezary syndrome but production was not localized definitively to malignant T cells (Wang et al., 2015). Our immunostaining also demonstrated that IL-32 was produced by a subset of keratinocytes, as previously reported (Fig. 1g) (Suga et al., 2014).

Figure 1. IL-32 is the most highly expressed interleukin in CTCL skin lesions and malignant T cells are a major source of IL-32 production.

Figure 1.

Open in a new tab

(a) Mean gene expression counts in skin lesions from seven patients with Stage I-IIA CTCL. (b) ScRNAseq of CD45+ cells from skin lesions of 2 MF patients (c) demonstrated that the majority of IL-32 was produced by T cells. (d-f) Immunostaining of MF (d, f) and Sezary syndrome (e,f) lesions confirmed malignant T cells are a major source of IL-32. (f) Cells were quantified in a minimum of three 400X high-powered fields; mean and SEM are shown. (g) IL-32 was also produced by a subset of keratinocytes (white arrow). Scale bars: 10 μm (d, third panel), 50 μm (all other panels).

We next tested the ability of IL-32 to support malignant T cell survival in vitro (Fig. 2). Immunostaining for the TCR Vβ subunit of the malignant clone and analysis by flow cytometry was utilized to discriminate benign from malignant T cells. When purified T cells from the blood of patients with Sezary syndrome were cultured with IL-32 (25 ng/ml each of α,β,γ isoforms) and sub-mitogenic levels of IL-2 (25 IU/ml), there was no improvement in malignant T cell survival. However, when myeloid cells were included in the culture system, there was enhanced survival of both malignant and benign T cells (Fig. 2a). Culture of Sezary PBMC in IL-32β alone led to prolonged survival of the malignant T cell clone, along with enhanced survival of benign T cells and a population of CD14+ APC (Fig. 2b). High throughput TCR sequencing (HTS) confirmed that surviving TCR Vβ+ T cells represented the malignant T cell clone and not benign infiltrating T cells expressing the same TCR Vβ (Fig. 2c,d). Studies of eight CTCL donors, two studied by HTS (Fig. 2c,d) and six studied by flow cytometry (Fig. 2e), confirmed that IL-32β supported the survival of malignant T cells in vitro. We observed a population of large monocyte-derived CD14+ cells in IL-32 cultures (Fig. 2f,g), consistent with prior reports that IL-32 can induce differentiation of monocytes into dendritic cells and macrophages (Ohmatsu et al., 2017). The presence of these IL-32-induced CD14+ cells was required for the survival of malignant T cells. The number of surviving malignant T cells directly correlated with the number of CD14+ large feeder APC in cultures (Fig. 2h). Cell counts from two separate donors with and without IL-32β are included (Fig. 2h); regardless of the presence or absence of IL-32β, the number of large CD14+ cells was strongly correlated with the number of surviving malignant T cells. CTCL blood donors varied somewhat in the kinetics of enhanced malignant T cell survival; in all cases, optimal T cell survival was observed when a population of large CD14+ APC had developed in the cultures. Lastly, we demonstrated that improved malignant T cell survival resulted from decreased apoptosis in the presence of IL-32 induced APC (Fig. 2i). This work builds on prior studies demonstrating that monocytes can support malignant T cell survival in T cell lymphoproliferative disorders (Wilcox et al., 2009) and that immature DC have the potential to support malignant T cell survival in CTCL (Berger et al., 2002).

Figure 2. IL-32 induces CD14+ APC that support malignant T cell survival.

Figure 2.

Open in a new tab

(a) Survival of purified T cells vs. PBMC from a Stage IV CTCL patient after culture with IL-32 and sub-mitogenic IL-2. (b) IL-32β supported survival of malignant and benign T cells and CD14+ myeloid cells. Five additional donors showed similar results. (c,d) HTS confirmed persistence of malignant T cell TCRs in culture. (e) IL-32β enhanced survival of benign and malignant T cells. (f,g) IL-32β induced formation of large, activated CD14+ T cells. Similar results observed in five additional donors. (h) The number of surviving malignant T cells was directly proportional to the number of IL-32β induced large CD14+ cells. (i) IL-32β enhanced malignant T cell survival by reducing apoptosis (detected by SYTOX Green).

Our studies demonstrate that IL-32 can support the survival of malignant T cells in vitro by generating a population of CD14+ APC that support malignant T cell survival. IL-32 has been proposed as an autocrine growth factor in CTCL (Suga et al., 2014); we show that IL-32 supports the survival of malignant T cells by inducing a population of feeder APC that support the survival of both malignant and benign T cells.

These findings also have practical applications. The ability to maintain malignant T cells in culture using IL-32 is a technical advance that will facilitate the study of malignant T cells in this disease. Our work also suggests that mice genetically engineered to produce human IL-32 may be a useful model for the expansion and propagation of malignant T cells in CTCL.

In summary, our studies demonstrate that IL-32 is the most highly expressed of all interleukins in MF, with expression levels 13-fold higher than the next most prevalent interleukin. We utilized single cell RNA sequencing and multiplex malignant TCR Vβ immunostaining to definitively demonstrate that malignant T cells are a major source of IL-32. Our studies demonstrate that IL-32 induces a multi-cell collaboration between T cells and CD14+ APC that may be the in vitro reflection of the biology in CTCL skin lesions. We recently observed that malignant T cells exist in skin in close proximity to c-Kit+/OX40L+/CD40L+ dendritic cells, creating an inflammatory synapse that drives inflammation and likely provides pro-survival signals to malignant T cells (Vieyra-Garcia et al., 2019). Our ongoing studies are focused on characterizing in vitro generated, IL-32-induced APC and exploring the hypothesis that IL-32 produced by malignant T cells may act on APC in the skin to create a protective pro-survival niche. These findings could have implications for the treatment of CTCL as well as providing insights into how resident T cells survive long-term in skin in the absence of their cognate antigen.

Materials and Methods:

Skin and blood samples

This is an experimental laboratory study performed on human tissue samples. All studies were performed in accordance with the Declaration of Helsinki. Written consent was obtained from all patients before study entry and sample collection. Patients who were studied met the World Health Organization-European Organization for Research and Treatment of Cancer (WHO-EORTC) criteria for MF (Willemze et al. 2005). Biopsy specimens from Stage IA-IIB mycosis fungoides patients were obtained from patients treated at the Medical University of Graz or the Dana-Farber/Brigham and Women’s Cancer Center Cutaneous Lymphoma Program. Blood samples were obtained from patients with leukemic CTCL seen at the Dana-Farber/Brigham and Women’s Cancer Center Cutaneous Lymphoma Program. All tissues were collected with previous approval from relevant review boards: Medical University of Graz Ethical Committee and the Dana-Farber Cancer Institute Institutional Review Board. Translational studies were approved by the Institutional Review Board of the Partners Human Research Committee.

IL-32 mRNA measurements

RNA was isolated from CTCL skin samples using the RNeasy FFPE or RNeasy Mini Kit (Qiagen) as per manufacturer’s instructions. The expression of 770 inflammation-related genes was measured using the NanoString Human PanCancer Immune Profiling Panel. Normalization and analysis of NanoString data were carried out using nSolver software. Normalization factors were calculated based on 40 reference genes.

Single cell RNA sequencing

Skin biopsies were cut into 2×2×2 mm pieces and frozen in cryopreservation medium (10% DMSO, 40% FBS, 50% culture medium) until use. Skin fragments were digested overnight in RPMI-1640 medium supplemented with 10% fetal bovine serum and 1.6 mg/ml Type IV collagenase (catalog # LS004209, Worthington Biochemical Corp., Lakewood, NJ). Isolated cells were stained with Zombie NIR viability dye (1:250 in phosphate buffered saline; catalog # 423105, Biolegend, San Diego, CA) followed by staining with PE-conjugated pan-CD45 antibody (1:20; clone 2D1; R&D Systems, Minneapolis, MN) and TotalSeq-C anti-human Hashtag (barcoding) antibodies (1:100; catalog # 394661, 394663, 394665, or 394667; Biolegend). Viable lymphocytes were sorted by flow cytometry at the Brigham and Women’s Hospital Human Immunology Center Flow Cytometry Core Facility, counted manually, and cells from four individual biopsies were pooled in approximately equal numbers into a single sample. Single cell libraries were prepared by the Brigham and Women’s Hospital Single Cell Genomics Core on the droplet-based 10X Genomics Chromium Controller (10X Genomics, Pleasanton, CA) using Chromium Single Cell 5’ Library & Gel Bead Kit (PN-1000006), Chromium Single Cell 5’ Feature Barcode Library Kit (catalog # 1000080) and Chromium Single Cell V(D)J Enrichment Kit, Human T Cell (catalog # 1000005) as per manufacturer’s protocols. Pooled libraries were sequenced on Illumina NovaSeq 6000 Sequencing instrument, with a running program of PE150.

Single cell RNA-sequencing data analysis

Raw sequencing data was aligned to the GRCh38 reference genome with CellRanger (version 3.1.0) (Zheng et al. 2017). All analyses were performed using the python software package Pegasus (version 0.17.2) (Li et al. 2020). Low quality cells were excluded from analyses based on having < 500 genes detected and/or having > 10% of their UMIs correspond to mitochondrial genes. Count data was log-normalized (loge(CPM + 1)) and the top 2000 most highly-variable genes were scaled and used for principal component analysis. To ensure proper integration of data from multiple batches, the principal components were adjusted with Harmony via the Harmonypy python package (version 0.0.4)(Korsunsky et al. 2019). The top 50 principal components were used as input for UMAP dimensionality reduction to visualize the data in 2D space. Cell clusters were defined using the Leiden clustering algorithm with a resolution of 0.9. The top markers for each cluster were defined using the de_analysis function in Pegasus and clusters were annotated based on legacy knowledge of known markers.

IL-32 immunostaining and cell quantification

5 μm cryosections were fixed in acetone for 10 minutes, then rehydrated and permeabilized for 10 minutes in TBS with 0.5% Tween (TBST) detergent. Non-specific binding of endogenous IgG was minimized by 20-minute blocking with1.6% human IgG. Slides were incubated with primary antibodies for 2 hours at room temperature. After two washes in buffer, they were incubated with isotype-specific secondary antibodies for 35 minutes at room temperature. Slides were washed twice more after antibody treatment and mounted in Vectashield Hard Set Mounting Medium with DAPI (Vector Labs). Tissue was imaged immediately after mounting on a Mantra Quantitative Pathology Workstation (Akoya Biosciences) using Mantra Snap 1.0 imaging software and analyzed with inForm image analysis software (Akoya Biosciences).

All secondary antibodies were purchased from Invitrogen and used at a 1:2000 dilution. The antibodies, clones, and dilutions used were: CD3 (UCHT1, BioLegend, 1:40); IL-32 (034, Lifespan Biosciences, 1:100), TCR Vβ5.1 (IMMU157, Beckman-Coulter, 1:10), and TCR Vb12 (Ver 2.32.1, BeckmanCoulter, 1:10)

DNA isolation and HTS

DNA was isolated from cultured cells with the QIAamp DNA Mini kit (Qiagen) and studied by ImmunoSEQ™ (Adaptive Biotechnologies, Seattle, WA) from 100–400 ng of DNA template as previously described (Kirsch et al. 2015).

In vitro cultures

PBMC were obtained by Ficoll centrifugation from blood samples obtained from patients with leukemic CTCL. For cultures of T cells alone, T cells were further purified using negative magnetic bead selection on the AutoMACS Pro (Miltenyi Biotec) with the Pan T cell Isolation Kit (Miltenyi #130–096-535). 300,000–1,000,000 cells in 200 μL per well were cultured in round-bottom 96-well plates in the presence or absence of IL-2 (25 IU/mL, NCI), IL-32α (25 ng/mL, R&D Systems, #3040-IL-050), IL-32β (25 ng/mL, R&D Systems, #6769-IL-025), or IL-32γ (25 ng/mL, R&D Systems, #4690-IL-025/CF). Plates were incubated in a 5% CO2, 37° C incubator for the indicated time. Cultures were fed three times a week replacing half of the medium with fresh medium with or without cytokine. Nonadherent cells were harvested by aspiration and adherent cells were isolated after a 20-minute incubation at 37° C in Accutase Cell Detachment Solution (BioLegend, #423201). Adherent and non-adherent cells were pooled. Each sample was then blocked with 5 μL of Human TruStain FcX (Fc Receptor Blocking Solution) (BioLegend, #422302) and 5 μL of TrueStain Monocyte Blocker (BioLegend, #426103), then immunostained with directly conjugated fluorescent antibodies and the viability stain SYTOX Green (ThermoFisher, S7020). 25 μL of Count Bright Absolute Counting Beads (ThermoFisher, C36950) were added and the cells were analyzed on a Becton Dickinson FACS Canto flow cytometer and analyzed on BD FACS Diva software (version 8.0.1). Antibodies used: PE-TCR Vβ (clone varied, Beckman Coulter), CD3 (UCHT1, BioLegend), CD8a (SK1, BioLegend), CD14 (HCD14, BioLegend). Apoptotic cells were identified by gating on SYTOX Green+ cells located in the viable lymphocytes forward/side scatter gate.

Acknowledgments:

We are grateful to the patients who entrusted us with their clinical care and provided tissue specimens for this research. We also thank the Lubin Family Foundation, the David Lamb Fund and the NIH (NIH/NCI R01CA203721) for supporting this work.

Footnotes

Conflict of Interest Statement: The authors declare they have no conflicts of interest.

Patient Materials: All donors provided written, informed consent. Approval for all human studies was obtained from the Institutional review boards of Mass General Brigham and the Dana-Farber Cancer Institute.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Data availability Statement:

The scRNAseq data used for this project have been deposited at the National Center for Biotechnology Information database of Genotypes and Phenotypes (dbGaP) and Sequence Read Archive under accession phs002717.v1.p1.

High throughput T cell receptor sequencing data has been deposited at DOI 10.21417/KKY2021JID, URL is https://clients.adaptivebiotech.com/pub/yu-2021-jid

References:

  1. Berger CL, Hanlon D, Kanada D, Dhodapkar M, Lombillo V, Wang N, et al. The growth of cutaneous T-cell lymphoma is stimulated by immature dendritic cells. Blood 2002;99(8):2929–39. [PubMed] [Google Scholar]
  2. Ohmatsu H, Humme D, Gonzalez J, Gulati N, Mobs M, Sterry W, et al. IL-32 induces indoleamine 2,3-dioxygenase(+)CD1c(+) dendritic cells and indoleamine 2,3-dioxygenase(+)CD163(+) macrophages: Relevance to mycosis fungoides progression. Oncoimmunology 2017;6(2):e1181237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ohmatsu H, Humme D, Gulati N, Gonzalez J, Mobs M, Suarez-Farinas M, et al. IL32 is progressively expressed in mycosis fungoides independent of helper T-cell 2 and helper T-cell 9 polarization. Cancer immunology research 2014;2(9):890–900. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Suga H, Sugaya M, Miyagaki T, Kawaguchi M, Fujita H, Asano Y, et al. The role of IL-32 in cutaneous T-cell lymphoma. The Journal of investigative dermatology 2014;134(5):1428–35. [DOI] [PubMed] [Google Scholar]
  5. Townsend EC, Murakami MA, Christodoulou A, Christie AL, Koster J, DeSouza TA, et al. The Public Repository of Xenografts Enables Discovery and Randomized Phase II-like Trials in Mice. Cancer cell 2016;29(4):574–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. van Kester MS, Borg MK, Zoutman WH, Out-Luiting JJ, Jansen PM, Dreef EJ, et al. A meta-analysis of gene expression data identifies a molecular signature characteristic for tumor-stage mycosis fungoides. The Journal of investigative dermatology 2012;132(8):2050–9. [DOI] [PubMed] [Google Scholar]
  7. Vieyra-Garcia P, Crouch JD, O’Malley JT, Seger EW, Yang CH, Teague JE, et al. Benign T cells drive clinical skin inflammation in cutaneous T cell lymphoma. JCI insight 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Wang L, Ni X, Covington KR, Yang BY, Shiu J, Zhang X, et al. Genomic profiling of Sezary syndrome identifies alterations of key T cell signaling and differentiation genes. Nature genetics 2015;47(12):1426–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Wilcox RA, Wada DA, Ziesmer SC, Elsawa SF, Comfere NI, Dietz AB, et al. Monocytes promote tumor cell survival in T-cell lymphoproliferative disorders and are impaired in their ability to differentiate into mature dendritic cells. Blood 2009;114(14):2936–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Kirsch IR, Watanabe R, O’Malley JT, Williamson DW, Scott LL, Elco CP, Teague JE, Gehad A, Lowry EL, LeBoeuf NR, Krueger JG, Robins HS, Kupper TS, and Clark RA. 2015. ‘TCR sequencing facilitates diagnosis and identifies mature T cells as the cell of origin in CTCL’, Sci Transl Med, 7: 308ra158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Korsunsky I, Millard N, Fan J, Slowikowski K, Zhang F, Wei K, Baglaenko Y, Brenner M, Loh PR, and Raychaudhuri S. 2019. ‘Fast, sensitive and accurate integration of single-cell data with Harmony’, Nat Methods, 16: 1289–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Li B, Gould J, Yang Y, Sarkizova S, Tabaka M, Ashenberg O, Rosen Y, Slyper M, Kowalczyk MS, Villani AC, Tickle T, Hacohen N, Rozenblatt-Rosen O, and Regev A. 2020. ‘Cumulus provides cloud-based data analysis for large-scale single-cell and single-nucleus RNA-seq’, Nat Methods, 17: 793–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Willemze Rein, Jaffe Elaine S., Burg Gunter, Cerroni Lorenzo, Berti Emilio, Swerdlow Steven H., Ralfkiaer Elisabeth, Chimenti Sergio, Diaz-Perez Jose L., Duncan Lyn M., Grange Florent, Nancy Lee Harris Werner Kempf, Kerl Helmut, Kurrer Michael, Knobler Robert, Pimpinelli Nicola, Sander Christian, Santucci Marco, Sterry Wolfram, Vermeer Maarten H., Wechsler Janine, Whittaker Sean, and Meijer Chris J. L. M.. 2005. ‘WHO-EORTC classification for cutaneous lymphomas 10.1182/blood-2004-09-3502’, Blood, 105: 3768–85. [DOI] [PubMed] [Google Scholar]
  14. Zheng GX, Terry JM, Belgrader P, Ryvkin P, Bent ZW, Wilson R, Ziraldo SB, Wheeler TD, McDermott GP, Zhu J, Gregory MT, Shuga J, Montesclaros L, Underwood JG, Masquelier DA, Nishimura SY, Schnall-Levin M, Wyatt PW, Hindson CM, Bharadwaj R, Wong A, Ness KD, Beppu LW, Deeg HJ, McFarland C, Loeb KR, Valente WJ, Ericson NG, Stevens EA, Radich JP, Mikkelsen TS, Hindson BJ, and Bielas JH. 2017. ‘Massively parallel digital transcriptional profiling of single cells’, Nat Commun, 8: 14049. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The scRNAseq data used for this project have been deposited at the National Center for Biotechnology Information database of Genotypes and Phenotypes (dbGaP) and Sequence Read Archive under accession phs002717.v1.p1.

High throughput T cell receptor sequencing data has been deposited at DOI 10.21417/KKY2021JID, URL is https://clients.adaptivebiotech.com/pub/yu-2021-jid

RESOURCES