Single-cell sequencing shows cellular heterogeneity of cutaneous lesions in lupus erythematosus

Zheng, Meiling; Hu, Zhi; Mei, Xiaole; Ouyang, Lianlian; Song, Yang; Zhou, Wenhui; Kong, Yi; Wu, Ruifang; Rao, Shijia; Long, Hai; Shi, Wei; Jing, Hui; Lu, Shuang; Wu, Haijing; Jia, Sujie; Lu, Qianjin; Zhao, Ming

doi:10.1038/s41467-022-35209-1

Download PDF

Article
Open access
Published: 05 December 2022

Single-cell sequencing shows cellular heterogeneity of cutaneous lesions in lupus erythematosus

Meiling Zheng^1,2^na1,
Zhi Hu^1,2^na1,
Xiaole Mei³^na1,
Lianlian Ouyang^1,2^na1,
Yang Song^1,2^na1,
Wenhui Zhou^1,2,
Yi Kong^1,2,
Ruifang Wu^1,2,
Shijia Rao^1,2,
Hai Long^1,2,
Wei Shi⁴,
Hui Jing^1,2,
Shuang Lu^1,2,
Haijing Wu^1,2,
Sujie Jia⁵,
Qianjin Lu ORCID: orcid.org/0000-0002-1504-4896^1,2,3 &
…
Ming Zhao ORCID: orcid.org/0000-0002-1320-1093^1,2

Nature Communications volume 13, Article number: 7489 (2022) Cite this article

10k Accesses
11 Citations
18 Altmetric
Metrics details

Subjects

Abstract

Discoid lupus erythematosus (DLE) and systemic lupus erythematosus (SLE) are both types of lupus, yet the characteristics, and differences between them are not fully understood. Here we show single-cell RNA sequencing data of cutaneous lesions from DLE and SLE patients and skin tissues from healthy controls (HCs). We find significantly higher proportions of T cells, B cells and NK cells in DLE than in SLE. Expanded CCL20⁺ keratinocyte, CXCL1⁺ fibroblast, ISG^hiCD4/CD8 T cell, ISG^hi plasma cell, pDC, and NK subclusters are identified in DLE and SLE compared to HC. In addition, we observe higher cell communication scores between cell types such as fibroblasts and macrophage/dendritic cells in cutaneous lesions of DLE and SLE compared to HC. In summary, we clarify the heterogeneous characteristics in cutaneous lesions between DLE and SLE, and discover some specific cell subtypes and ligand-receptor pairs that indicate possible therapeutic targets of lupus erythematosus.

Mapping systemic lupus erythematosus heterogeneity at the single-cell level

Article 03 August 2020

Tubular cell and keratinocyte single-cell transcriptomics applied to lupus nephritis reveal type I IFN and fibrosis relevant pathways

Article 20 May 2019

Chromatin accessibility landscapes of skin cells in systemic sclerosis nominate dendritic cells in disease pathogenesis

Article Open access 17 November 2020

Introduction

Lupus erythematosus (LE) is a severe autoimmune disease characterized by the presence of many abnormal immune cells and a large number of autoantibodies and immune complexes, all of which lead to damage to multiple organs, such as the skin, kidney, and brain¹. Clinically, LE is mainly divided into two types. One type is known as cutaneous lupus erythematosus (CLE), which mainly presents as dermatological injuries and does not involve systemic damage². The other type of LE involves systemic manifestations, including cutaneous, respiratory, renal, cardiovascular and other symptoms, and is called systemic lupus erythematosus (SLE)³. Discoid lupus erythematosus (DLE) accounts for >80% of cases of CLE and is the most common type of CLE⁴. On the one hand, approximately 5% of CLE patients will convert to SLE⁵; on the other hand, ~20% of SLE patients exhibit CLE lesions at the time of diagnosis or in the years after diagnosis⁶. Although the manifestations of cutaneous lesions are distinct between DLE and SLE, the cell composition and the underlying molecular events in cutaneous lesions of DLE and SLE remain unclear.

Recently, single-cell RNA sequencing (scRNA-seq) studies were performed in peripheral blood mononuclear cells (PBMCs), kidney biopsies and lesions of lupus nephritis (LN) patients using a microfluidic device and CEL-Seq2. These results indicated single-cell transcriptional maps of peripheral blood and damaged organs in SLE and identified the important function of type I interferon genes in the pathogenesis of SLE^7,8,9. However, due to the throughput limitation of scRNA-seq on microfluidic chips, little information has been obtained about the cutaneous lesions of lupus patients. Besides, previous studies reported that the infiltration of plasmacytoid dendritic cells (pDCs) contributed to the overproduction of type I interferons, which have an important function in the cutaneous lesions of DLE and SLE^10,11. Several bulk RNA-seq analyses also showed enhanced activation of the interferon (IFN) signaling pathway, innate immune response pathways (including TLR, RIG-I, cytosolic DNA sensing, JAK/STAT) and high expression of some genes, such as GBP2, HLA-F, IFIT2, RSAD2, and ISG15, in cutaneous lesions of DLE patients^12,13. In addition, ultraviolet light (UV), immune cells, cytokines and the deposition of immunoglobulins have been reported to have a function in the development of skin inflammation and damage in SLE¹⁴. Macrophages and TNF and IFN/IFNR signaling participate in skin injury of SLE¹⁵. Although previous studies have provided general knowledge about the cutaneous lesions of lupus patients, an understanding of the precise pathological cell types in skin lesions, the functions and activities of each cell subset, and skewed communication networks have not been completely elucidated.

In this study, to further explore the differences between cutaneous lesions of DLE and SLE at a high-throughput single-cell transcriptional level, we collected 23 skin tissues from DLE and SLE patients and age- and sex-matched healthy controls (HCs), separated their epidermis and dermis and performed scRNA-seq on the single-cell suspensions using the Chromium System of 10× Genomics. We show important and detailed information on the differences between healthy skin tissues and skin biopsies of lupus patients and demonstrate the similarities and differences in the cutaneous lesions of DLE and SLE.

Results

Sample collection and scRNA-seq workflow

As the cell diameters of epidermal and dermal cells are different, to avoid the capturing preference of the 10× Genomics system, we separated the epidermis and dermis of skin tissues to obtain more transcriptional messages from the cells. In total, 23 skin biopsy samples from 8 DLE patients with an average age of 40.4 ± 11.0 years, 10 SLE patients with an average age of 41.4 ± 12.3 years and 5 HCs with an average age of 32.4 ± 6.0 years were collected. We successfully performed scRNA-seq on 14 epidermal single-cell suspensions (4 HCs, 5 DLE, and 5 SLE samples) and 16 dermal single-cell suspensions (4 HCs, 5 DLE, and 7 SLE) by the 10× Genomics Chromium system (Fig. 1a). The age, sex and other clinical information of all samples are summarized in Supplementary Data 1. The initial number of captured cells per skin biopsy sample, the average reads per cell and other sequencing information are shown in Supplementary Data 2.

**Fig. 1: scRNA-seq shows the main cell types in epidermis and dermis.**

Distinct cell compositions of DLE and SLE illustrated by scRNA-seq

After removing cell doublets, correcting for batch effects, and filtering low-quality cells (see Materials and methods), we acquired 107,428 epidermal cells and 191,701 dermal cells for the downstream analysis (Supplementary Fig. 1a, b, e, f). Through t-stochastic neighborhood embedding (t-SNE) clustering and analysis of differentially expressed genes (DEGs), all epidermal clusters were identified as keratinocytes by high expression of KRT14 and KRT1¹⁶, T cells by CD3D and CD3G, macrophages/dendritic cells (Macro/DCs) by LYZ and AIF1¹⁷, melanocytes by PMEL and MLANA¹⁷, NK cells by XCL2 and NKG7¹⁸ and B cells by CD79A and MS4A1¹⁹ (Fig. 1b, d, Supplementary Fig. 1d, and Supplementary Data 3). In addition to T cells, Macro/DCs, B cells and NK cells, other dermal clusters were labeled as fibroblasts by COL1A1 and COL3A1²⁰, endothelial cells by CDH5 and VWF¹⁷, mast cells by TPSAB1 and TPSB2²¹ and Schwann cells by CDH19 and MPZ²² (Fig. 1c, e, Supplementary Fig. 1h, and Supplementary Data 4). Overall, except for immune cells, other cell types identified in our scRNA-seq were consistent with those cell types found in full-thickness skin²³.

The proportions of T cells, B cells, Macro/DC, and NK cells were 29.5%, 1.7%, 10.2%, and 3.5%, respectively, in the epidermis of DLE patients, and 31.2%, 0.4%, 8.6%, and 1.9%, respectively, in the epidermis of SLE patients, which were higher than those in the epidermis of HC (T: 14.0%, B: 0.1%, Macro/DC: 1.8%, NK: 0.5%) (Fig. 2a). Moreover, the proportions of T cells, B cells and NK cells in the dermis of DLE patients were 43.7%, 16.7%, and 5.4%, respectively, which were higher than those in the dermis of SLE patients (T: 17.2%, B: 2.5%, and NK: 1.3%). The higher proportion of B cells in the lesional skin of DLE patients compared to SLE patients was consistent with a previous report²⁴. The proportions of dermal immune cells from both DLE and SLE patients were higher than those from HC (T: 11.5%, B: 0.8%, NK: 0.8%) (Fig. 2b). Regarding dermal Macro/DCs, the proportion in the dermis of SLE patients (8.3%) was higher than that in the dermis of DLE patients (6.7%), and HC (6.9%) (Fig. 2b). In addition, the compositions of cell types in individual samples further supported the findings of different cell proportions in HC, DLE patients and SLE patients (Fig. 2c and Supplementary Fig. 1c, g). Immunofluorescence staining of T cells (CD3), B cells (CD19) and NK cells (CD56) verified the differences in cell composition among HCs, DLE patients and SLE patients (Fig. 2d). Taken together, these results show that immune cells, such as T, B, and NK cells, account for higher cell proportions in DLE patients than in SLE patients or HCs.

**Fig. 2: Distinct cell compositions of HC, DLE, and SLE.**

CCL20⁺ keratinocyte and other amplified keratinocyte subtypes in epidermal samples of DLE and SLE

Keratinocytes, the most prominent cell type in the epidermis, have been reported to be involved in the pathogenesis of various autoimmune skin diseases, including lupus²⁵. Here, after removing the batch effect (Supplementary Fig. 2a, b), we identified and labeled 12 keratinocyte subtypes: 6 differentiating keratinocyte subtypes (Diff kera-1–6) with high expression of KRT1, KRTDAP, and SBSN²⁶, 2 basal keratinocyte subtypes (Basal kera-1,2) with high expression of COL17A1, DST, and KRT15²⁶, 1 channel-related keratinocyte subtype (Channel-related kera) with high expression of components of channel ATPases (ATP1B1) and GJB2¹⁶, 1 cycling keratinocyte subtype (Cycling kera) with high expression of mitotic markers (UBE2C and TOP2A²⁶) and proliferative markers (MKI67 and PCNA), 1 CCL20⁺ keratinocyte subtype (CCL20⁺ kera) with high expression of CCL20 and 1 terminally differentiating keratinocyte subtype (Terminal diff kera) with high expression of terminal keratinocyte markers (LCE1A and LCE1B) (Fig. 3c, d). Among all subtypes of keratinocytes, we found that keratinocytes from DLE and SLE groups accounted for more than 75% of Diff kera-3, CCL20⁺ kera, Diff kera-4, Diff kera-5 and Diff kera-6, which suggested their amplifications in the epidermis of DLE and SLE patients (Fig. 3b and Supplementary Fig. 2c).

**Fig. 3: Sub-clustering analysis of keratinocytes identifies major keratinocyte subtypes in epidermis.**

To identify the transcriptome differences between those DLE and SLE-expanded keratinocyte subtypes, we adopted Gene Ontology (GO) analysis based on the DEGs. Notably, Diff kera-3, characterized by high expression of S100A6 and ACTN1, showed enrichment in epithelial cell proliferation, regulation of actin cytoskeleton organization and muscle organ development (Fig. 3d, f). CCL20⁺ kera highly expressed CCL20 gene (Fig. 3d), which has been reported to be a ligand for the chemokine receptor CCR6 to recruit immune cells²⁷. The immunofluorescence staining of CCL20 and KRT10 further confirmed the expansion of CCL20⁺ kera in the epidermis of DLE and SLE patients (Fig. 3e). Furthermore, Diff kera-4 was characterized by high expression of the interferon-stimulated genes (ISGs) IFIH1, IFIT2, and IFITM3, the chemokines CXCL10 and CXCL11, and antigen processing and presentation related molecules HLA-DRA and HLA-DRB1 (Fig. 3d, f and Supplementary Data 5–6). Diff kera-5 with high expression of CYP1B1 and GPX4 was involved in the regulation of lipid metabolic processes (Fig. 3f and Supplementary Data 5–6). Diff kera-6 expressed the chemokine ligands CCL4 and CCL5, and the chemokine receptor CXCR4, which enriched leukocyte cell adhesion and migration (Fig. 3d, f and Supplementary Data 5–6). Interestingly, the pathways of neutrophil activation and neutrophil degranulation, which have been reported to have an important function in the development of lupus²⁸, were commonly enriched in CCL20⁺ kera, Diff kera-4, Diff kera-5 and Diff kera-6 (Fig. 3f and Supplementary Data 5–6).

Due to the existence of similar and distinct GO enrichments in the lupus-expanded keratinocyte subtypes, we wondered whether there was a differentiating association between those subtypes. Thus, we performed pseudotime trajectory analysis by Monocle2. Intriguingly, the results showed a potential differentiated trajectory from Diff kera-2 to Diff kera-1, Diff kera-6 and finally to Terminal diff kera or CCL20⁺ kera and Diff kera-4 (Fig. 3g). Diff kera-2 was located in the initial position of the trajectory with high expression of genes such as GRHL3 and SERPINB2. Both Diff kera-1 with high expression of ID1, FGFR3, and CHP2 and Diff kera-6 with high expression of RUNX1, RGS1, and CXCR4 was located in the middle of the trajectory. Terminal diff kera was in one end stage of the trajectory with high expression of LCE1C, LCE2B, and LCE1B. CCL20⁺ kera and Diff kera-4 were located in the other end stage with high expression of IFI3, IFITM1, and CXCL10 (Fig. 3h). The above findings suggested a potential differentiation direction of keratinocytes in lesional skin of lupus.

CXCL1⁺ fibroblasts, HLA⁺ fibroblasts and other expanded fibroblast subtypes in the dermis of SLE patients

Recent reports have shown that fibroblasts participate in skin wound healing and the progression of scleroderma^29,30. However, the function of fibroblasts in lupus is poorly understood. In our study, after integrating 92,162 fibroblasts from DLE patients, SLE patients and HCs and removing batch effects, we generated and labeled 9 fibroblast subtypes (Fig. 4a and Supplementary Fig. 3a–c). The cells from SLE occupied for more than 75% of CXCL1⁺ Fib, HLA⁺ Fib1, Pericyte2, and Fib4, which indicated their amplification in the dermis of SLE (Fig. 4b and Supplementary Fig. 3d). Although we were unable to identify fibroblasts by the known marker genes of the classic fibroblast subtype^31,32 (Supplementary Fig. 3e), two fibroblast genes, COL1A1 and DCN²⁰, were highly expressed in 7 fibroblast subtypes (Fib1, Fib2, CXCL1⁺ Fib, HLA⁺ Fib1, Fib3, Fib4, and HLA⁺ Fib2). Pericyte1 and Pericyte2 were characterized by high expression of ACTA2 and TPM2, which indicated a pericyte subtype (a special subtype of fibroblasts³²) (Fig. 4c, d). CXCL1⁺ Fib was found to have high expression of CXCL1, which has been reported to promote the infiltration of immune cells, especially neutrophils³³ (Fig. 4c, d). In addition, we identified a marker gene CXCL12 of activated fibroblasts²⁹ to be highly expressed in CXCL1⁺ Fib and HLA⁺ Fib2. Interestingly, major histocompatibility complex-II family member genes (HLA-DRB1 and HLA-DRA1), which are usually expressed in antigen-presenting cells, were identified to be highly expressed in HLA⁺ Fib1 and HLA⁺ Fib2, suggesting that these fibroblasts may function as nonclassical antigen-presenting cells (APCs) (Fig. 4c, d and Supplementary Data 7). Although the CXCL1⁺ Fib and HLA⁺ Fib1 subtypes were not expanded in DLE, we observed that CXCL1 and HLA-DRB1 were highly expressed in DLE fibroblasts (Supplementary Fig. 3f). The immunofluorescence staining of CXCL1 and HLA-DRB1 with Vimentin also verified their overexpression in SLE and DLE (Fig. 4e, f).

**Fig. 4: Sub-clustering analysis of fibroblasts identifies major fibroblast subtypes in dermis.**

Furthermore, GO function analysis based on DEGs in expanded fibroblast subtypes showed that the biological processes (BPs) of response to oxygen levels, response to hypoxia and neutrophil degranulation were enriched in HLA⁺ Fib1, Fib4, and CXCL1⁺ Fib cells, while the regulation of viral process and the immune response-activating signal transduction were enriched in HLA⁺ Fib1 and Fib4 cells. In addition, HLA⁺ Fib1 and CXCL1⁺ Fib showed enrichment of leukocyte cell-cell adhesion, positive regulation of leukocyte activation and positive regulation of cytokine production. Strikingly, type I interferon signaling pathway and response to virus were separately enriched in Pericyte2 and CXCL1⁺ Fib (Fig. 4g and Supplementary Data 8).

We also performed pseudotime trajectory analysis on fibroblast subtypes to explore whether there was a differential trajectory in these expanded fibroblast subtypes. The pseudotime trajectory analysis presented a potentially differentiated trajectory of fibroblasts, in which Fib2 was located in the initial part and HLA⁺ Fib1 and HLA⁺ Fib2 expressing HLA-DRB1, CD74, and HLA-DRA were located in the first middle stage. Part of the Fib1 expressing PDGFRA and FBLN was in one end stage of the trajectory, and CXCL1⁺ Fib with CYP1B1 and CXCL1 expression was located in the other end stage (Fig. 4h, i).

ISGs^hi T cells, GPR183⁺ T cells, and other expanded T cell subclusters in cutaneous lesions of DLE and SLE patients

Abnormal T cells in the peripheral blood of patients with lupus have been widely reported^34,35. However, the T cell subset changes in cutaneous lesions of DLE and SLE are not fully understood. To evaluate T cell subset transcriptome changes, we integrated a total of 23,982 epidermal T cells (HC: 4,286; DLE: 11,429; SLE: 8,267) and yielded 7 T cell subclusters (SCs) (T_SC0 to T_SC6) (Fig. 5a and Supplementary Fig. 4a–c). The cells from DLE and SLE accounted for more than 75% of the 5 epidermal T cell SCs (T_SC1-T_SC5) (Fig. 5b and Supplementary Fig. 4d). In addition, a total of 40,047 dermal T cells (HC: 5,330; DLE: 20,327; SLE: 14,390) yielded 7 dermal T cell SCs (T_SC0-T_SC6) (Fig. 5f and Supplementary Fig. 4f–h). DLE and SLE accounted for 75% of all dermal T cell SCs except T_SC2 (Fig. 5g and Supplementary Fig. 4i).

**Fig. 5: Sub-clustering analysis of T cells identifies major T cell subclusters in skin tissues.**

According to the classical marker genes and DEGs profile, there were several T cell SCs with similar characteristics in both epidermis and dermis. First, ISGs (IFI6, IFI44, IFIH1, and DDX58) were highly expressed in epidermal T_SC1 and dermal T_SC5, which are identical to the previously reported T-SC4 in PBMCs of lupus patients⁹ (Fig. 5c, d, h, i, Supplementary Fig. 4e, j, and Supplementary Data 9). Moreover, a gene panel of ISGs (IFI6, IFI44, IFITM3, ISG20, IFI27, ISG15, and IFI44L) showed significantly higher expression in dermal T cells of DLE and SLE patients than in HCs (Supplementary Fig. 4k). Second, epidermal T_SC3 and dermal T_SC3 were both identified by high expression of Treg marker genes (IKZF2, IL2RA and FOXP3³⁶; Fig. 5c, d, h, i and Supplementary Data 9). Epidermal T_SC2 and dermal T_SC5 highly expressed marker genes of Th1 (IFNG and TBX21³⁷) and CTLs (cytotoxic T cells: XCL2, XCL1, and GZMA⁹) (Fig. 5d, i), which suggests the potential effector functions of these two T cell subsets. Notably, GPR183, which has been reported to be involved in the migration, expansion, and infiltration of B cells and T follicular helper cells (Tfh)^38,39, showed high expression in dermal T_SC0 (Fig. 5h), suggesting that this T cell subset may be related to B cell accumulation and ectopic germinal center formation in cutaneous lesions of DLE and SLE.

To determine the distinct transcriptome functions of T cells in skin tissues among HCs, DLE patients and SLE patients, we conducted GO analysis on DEGs. Intriguingly, T cells in the epidermis of DLE and SLE patients showed enrichment of some BPs, including cellular response to type I interferon, positive regulation of defense response, positive regulation of leukocyte cell-cell adhesion and type I interferon signaling pathway, which were distinct from T cells in the epidermis of HC with enrichment of RNA catabolic process and translational initiation (Fig. 5e). In contrast to epidermal T cells, both cell adhesion and regulation of T cell activation were enriched in dermal T cells of DLE patients, SLE patients and HCs. However, response to virus and cellular response to heat were specifically enriched in dermal T cells from DLE, and type I interferon signaling pathway, response to interferon gamma and regulation of inflammatory response were enriched in dermal T cells from SLE (Fig. 5j). These GO BP results suggest the common and distinct functions of epidermal and dermal T cells among HCs, DLE patients and SLE patients.

To further understand the relationship among dermal T cells of HCs, DLE patients and SLE patients, we performed pseudotime trajectory analysis to assess the potential differentiation of dermal T cells. The trajectory results showed that dermal T cells of HC were located in the initial stage with high expression of GAS5, TLE5 and SNHG29, dermal T cells highly expressing ISGs (IFITM1, IFI44L, ISG15, and IFITM3) from DLE and SLE were located in the middle stage, and dermal T cells mainly from SLE with high expression of H1FX, HLA-A, and TGFBR2 were located in the end stage (Supplementary Fig. 4l, m).

ISG^hi plasma, HSP^hi B and other expanded B cell SCs in cutaneous lesions of DLE and SLE patients

B cell selection defects and autoreactive B cell activation lead to the overproduction of autoreactive antibodies in lupus⁴⁰. In our study, 864 epidermal B cells (HC: 41; DLE: 678; SLE: 105) yielded 5 B cell SCs (B_SC0-B_SC4), almost all of which were composed of cells from DLE and SLE (Fig. 6a, b and Supplementary Fig. 5a–d). In addition, 9,494 dermal B cells (HC:338; DLE: 7,190; SLE: 1966) generated 7 B cell SCs (B_SC0-B_SC6) (Fig. 6f and Supplementary Fig. 5g–i). Dermal B cells from DLE and SLE accounted for more than 75% in each dermal B cell SC (Fig. 6g and Supplementary Fig. 5j), which is consistent with a previous report of B cell expansion in CLE lesions^24,41.

**Fig. 6: Sub-clustering analysis of B cells implicates major B cell subclusters.**

To clarify the characteristics of epidermal and dermal B cell subclusters, we represented the classic B cell subset markers and DEGs expression in B cell SCs. Notably, B cell SCs from the epidermis and dermis presented similar features. We found high expression of ISGs in epidermal B_SC3 and dermal B_SC4 and B_SC5, which were identified as plasma cells by the gene expression of CD38, IGHG4, and IGHG1⁴² (Fig. 6c, d, h, i and Supplementary Data 10). These three SCs were similar to the P-SC0 in the PBMCs of lupus patients identified in a previous scRNA-seq study⁹ (Supplementary Fig. 5e, k). Hsp70-coding genes, which are involved in antigen presentation^43,44, were highly expressed in epidermal B_SC0, dermal B_SC3 and B_SC6 (Fig. 6d, h), suggesting that these B_SCs may participate in antigen presentation. Age-associated B cells (ABC) have been reported to be abnormally expanded in SLE^7,45. Therefore, we analyzed the expression of ABC marker genes (ITGAX, TBX21, FCRL2⁴⁶) in B cell subsets and found that these marker genes were expressed in multiple B cell subsets but not in a specific subset (Supplementary Fig. 5f, l).

Next, we performed GO analyses on epidermal and dermal B cells of HCs, DLE patients and SLE patients. The results showed that epidermal B cells in HC were involved in RNA catabolic processes and T cell activation, while epidermal B cells in DLE and SLE were involved in more BPs, including the type I interferon signaling pathway, response to virus, response to interferon gamma, regulation of innate immune response, neutrophil activation involved in immune response and leukocyte cell-cell adhesion (Fig. 6e). In addition, epidermal B cells of DLE were characterized by macroautophagy, which has been reported previously to be activated in B cells of SLE lesions and required for plasmablast development⁴⁷ (Fig. 6e). Different from epidermal B cells, the response to interferon gamma and viral life cycle were enriched in dermal B cells in each sample group. Moreover, dermal B cells from HCs and DLE patients were enriched in lymphocyte differentiation, leukocyte cell-cell adhesion and T cell activation, and only dermal B cells from SLE patients were enriched in the type I interferon signaling pathway (Fig. 6j). Taken together, these findings indicate that epidermal and dermal B cells, including ISG^hi plasma cells and HSP^hi B cells, were expanded in the epidermis and dermis of DLE and SLE, especially in the dermis of DLE.

pDC, ISGs^hi Macro/DC and other expanded Macro/DC SCs in cutaneous lesions of DLE and SLE patients

Macrophages and DCs have been reported to be defective in clearing apoptotic cells in lupus⁴⁸. In our study, 6,595 epidermal Macro/DCs (HC: 546; DLE: 3832; SLE: 2217) and 11,864 dermal Macro/DCs (HC: 3059; DLE: 2874; SLE: 5931) were used for sub-clustering analysis (Supplementary Fig. 6a, e). More than 75% of the 5 epidermal Macro/DC SCs, except M_SC5 and dermal M_SC3, M_SC5 and M_SC6, were contributed by the DLE and SLE samples (Fig. 7b, g, Supplementary Fig. 6b–d, f–h).

To discover transcriptional differences among Macro/DC SCs of HCs, DLE patients and SLE patients, we performed DEGs analysis. We found that ISGs were highly expressed in epidermal M_SC1 cells, which were identified as a mix of macrophages and DCs because of the co-expression marker genes of CD16⁺ DCs (FCGR3A, CXCL10, CXCL11⁴¹) and macrophages (MRC1, CD163, CD68⁴⁹) (Fig. 7d). Dermal M_SC6 highly expressed ISGs and marker genes of CD16⁺ DCs (Fig. 7i). Epidermal M_SC2 and dermal M_SC3 showed high expression of marker genes of pDCs (JCHAIN and MZB1⁴⁸) (Fig. 7c, d, h, i, Supplementary Data 11), which have been reported to accumulate and help produce spontaneous germinal centers in lupus⁵⁰. As classic APCs, multiple Macro/DC SCs expressed HSP70-coding genes. In addition, epidermal-specific Langerhans cells were identified by CD207 expression in epidermal M_SC5 (Fig. 7d).

**Fig. 7: Sub-clustering analysis of macrophage/dendritic cells (Macro/DCs) identifies major Macro/DCs subclusters.**

Further GO analysis showed that antigen processing and presentation, neutrophil activation involved in immune response and T cell activation were common enriched in epidermal Macro/DCs in HCs, DLE patients and SLE patients. The differences were that epidermal Macro/DCs of DLE contained some BPs, including cell killing, positive regulation of leukocyte cell-cell adhesion and response to oxidative stress, and epidermal Macro/DCs of SLE were enriched in the type I interferon signaling pathway (Fig. 7e). Dermal Macro/DCs from HCs, DLE patients and SLE patients were all involved in antigen processing and presentation and T cell activation, which was similar to epidermal Macro/DCs. Although type I interferon signaling pathway, type I interferon production and response to virus were enriched in dermal Macro/DCs of both DLE and SLE, macroautophagy, which was reported to be impaired in SLE macrophages⁵¹, and regulation of inflammatory response were enriched in dermal Macro/DCs of SLE (Fig. 7j). These differences in Macro/DCs further suggest a possible difference in the pathogenesis of cutaneous lesions between DLE and SLE.

ISGs and HSP-coding gene overexpression in NK cells of DLE and SLE patients

NK cells have rarely been reported in cutaneous lesions of lupus. In our study, 2152 epidermal NK cells (HC: 214; DLE: 1398; SLE: 540) and 3489 dermal NK cells (HC: 282; DLE: 2165; SLE: 1042) were ultimately involved in sub-clustering analysis (Supplementary Figure 7a, e). Except for epidermal and dermal N_SC3, cells from DLE and SLE composed more than 75% of other epidermal and dermal NK SCs (Fig. 8b, g and Supplementary Fig. 7b–d, f–h).

According to DEGs analysis, we found that ISGs were more highly expressed in NK cells from DLE and SLE patients than in those from HCs, and HSP70-coding genes were highly expressed in epidermal N_SC0 and dermal N_SC1 cells (Fig. 8c, d, h, i and Supplementary Data12). Further GO functional analysis showed the enrichment of the type I interferon signaling pathway in epidermal NK cells from DLE and SLE patients (Fig. 8e). In contrast to dermal NK cells of SLE patients enriching the type I interferon signaling pathway and the regulation of inflammatory pathway, only the defense response to virus and the positive regulation of cytokine production were enriched in dermal NK cells of DLE patients (Fig. 8j).

**Fig. 8: Sub-clustering analysis of NK cells identifies major NK subclusters.**

Active cell communication in cutaneous lesions of DLE and SLE patients

The interaction between cells based on receptors and ligands has an important function in physiological and pathological processes²⁶. To investigate the interaction among different cell types identified by scRNA-seq in lupus cutaneous lesions, we conducted cell communication analysis by CellPhoneDB⁵². Based on the number of ligand‒receptor interaction pairs, we calculated the interaction scores. The cell communication scores of keratinocytes and macro/DCs were significantly higher in the epidermis of DLE and SLE patients than in HCs (Fig. 9a). Notably, the scores of keratinocytes, Macro/DCs and melanocytes in DLE were higher than those in SLE (Fig. 9a). Moreover, the cell communication scores of endothelial cells, fibroblasts, and Macro/DCs in the dermis of DLE were higher than those in SLE and HC (Fig. 9b). These findings suggested a potential connection of keratinocytes, endothelial cells, fibroblasts, and Macro/DCs in the immune microenvironment in cutaneous lesions of DLE and SLE.

**Fig. 9: Cell-cell interaction in epidermis and dermis.**

Furthermore, the more highly expressed chemokine ligands and receptors of cell types in DLE and SLE are shown in Supplementary Data13. Immune cells, such as T cells, NK cells, Macro/DCs and B cells in the epidermis of DLE and SLE, highly expressed CXCR3 (Fig. 9c), which is a receptor for the chemokine ligand CCL20 with higher expression levels in keratinocytes of DLE and SLE patients than in those of HCs. In addition, other ligand-receptor pairs in keratinocytes and immune cells, such as TNFSF9_TNFRSF9 and SPN_ICAM1, were also highly expressed in the epidermis of DLE and SLE (Fig. 9c). Macro/DCs, as antigen-presenting cells, interact most strongly with other immune cells, in which FAM3C_CLEC2D indicates the possible interaction between Macro/DCs and T or B cells in DLE and SLE (Fig. 9c). In addition, FAM_3C_ HLA_C was expressed at higher levels in immune cells (B, NK, T, Macro/DCs) and keratinocytes of DLE than SLE (Fig. 9c).

Regarding the dermis, the receptors of CCL19 (CCR7, CCR2, and CXCR3), CXCL12 (CXCR4), FAM3C (CLEX2D), and TNFSF13B (CD40) in fibroblasts showed the highest expression in B, Macro/DCs, NK, and T cells in the DLE group among the three sample groups (Fig. 9d). The interaction between Macro/DCs and other cells was represented by the expression of TNARSF13B_CD40, which also showed the highest expression in DLE (Fig. 9d). Immunochemical staining verified the increased CCL19 expression in fibroblasts (vimentin) and the increased CCR7 expression in macrophages (CD68) in cutaneous lesions of DLE and SLE patients (Fig. 9e). Together, the overexpression of these ligands and receptors may be involved in the activation, migration, and residence of immune cells in cutaneous lesions of DLE and SLE patients.

Discussion

Lupus is a spectrum of autoimmune diseases, including DLE on one end and SLE on the other⁵³. Although we know that there are distinct skin phenotypes between DLE and SLE and understand the basic immune changes that occur in cutaneous lesions, the precise cell composition and their functions in cutaneous lesions in DLE and SLE remain unclear. Here, we identified detailed cell types and their constitutions in the epidermis and dermis of cutaneous lesions in DLE and SLE by scRNA-seq. Our dataset comprised transcriptional database of 107,428 epidermal cells and 191,701 dermal cells from 23 skin tissues of HCs, DLE patients and SLE patients. We compared the difference in cell type composition, further described the cell subclusters and their functions, and finally investigated cell–cell interactions.

Based on scRNA-seq data analysis and immunofluorescent staining, we identified that there are more immune cells, such as T, B and NK cells, in the cutaneous lesions of DLE than SLE. The B cell expansion in DLE was consistent with a previous study showing that B cells account for a greater proportion in DLE than in SLE²⁴. These results suggested the distinct pathogenesis of DLE and SLE and provided a potential way to distinguish DLE and SLE. Furthermore, we identified the subclusters of some cell types and showed that the ISGs signature, autophagy signaling and neutrophil activation were enriched in DLE or SLE cells. Notably, we defined some epidermal and dermal cell subtypes with increased proportions/expressions in DLE and SLE, including CCL20⁺ keratinocytes, CXCL1⁺ fibroblasts, ISG^hi CD4/CD8 T, ISG^hi plasma, pDCs, and NK SCs. We also identified some important genes related to immune microenvironments in cutaneous lesions of lupus patients. For example, the G protein-coupled receptor GPR183, which has been reported to be involved in the migration and localization of multiple immune cells⁵⁴, was shown to be highly expressed in dermal T_SC0 in our dataset.

Combining DEGs and GO enrichment analysis, we found that some keratinocytes and fibroblasts in cutaneous lesions of lupus might function as immune cells. Diff era-4, Diff kera-5 and Diff kera-6 participated in some BPs, such as the regulation of innate immune response and antigen processing and presentation, and HLA⁺Fib1 and CXCL1⁺Fib were involved in the processes of leukocyte adhesion and activation⁵⁵. HSP70-coding genes have been reported to assist in the delivery of antigens in both epidermis and dermis and to have an important function in immunoregulation of autoimmune diseases⁵⁶. In this study, we found that HSP70-coding genes were highly expressed in epidermal T_SC2, B_SC0, M_SC2, and N_SC0 and dermal T_SC5, B_SC3, M_SC3, and N_SC1, which suggests that HSP70-coding genes may have an important function in cutaneous lesions of DLE and SLE.

Extensive evidence has shown the involvement of the interferon signaling pathway in the pathogenesis of lupus^28,57,58. Our study identified some cell subtypes with high expression of ISGs, such as Diff kera-4, Pericyte2, epidermal T_SC1, dermal T_SC5, epidermal B_SC3, dermal B_SC4 and B_SC5, epidermal M_SC1, dermal M_SC6, epidermal N_SC2 and dermal N_SC1, which is consistent with a previous report that a strong interferon signature was identified in multiple cell subtypes of CLE⁴¹. Thus, studying the functions and mechanism of the activation of the interferon signaling pathway in lupus may help identify more intervention targets for lupus treatment.

Previous studies have shown that neutrophil dysfunction is related to the pathogenesis of SLE⁵⁹. In this study, although we did not detect any neutrophil infiltration due to the preference error of the droplet-based 10× Genomic system, the neutrophil activation pathway was found in multiple cell types according to GO BP analysis. Furthermore, we identified neutrophils in cutaneous lesions of DLE and SLE by immunofluorescence staining for ELANE (a neutrophil elastase; Supplementary Fig. 8), suggesting that neutrophils are involved in the formation of cutaneous lesions of lupus. In addition, we found that the autophagy pathway, which has been reported to facilitate skin differentiation and epidermal proliferation and accelerate inflammation^60,61, was enriched in epidermal B cells of DLE and dermal Macro/DCs of SLE.

Cell communication analysis among different cell types identified the enhanced interactions of keratinocytes and Macro/DCs with other immune cells in the epidermis of DLE and SLE, which may contribute to an increased number and activation of immune cells in cutaneous lesions of DLE and SLE. Furthermore, ligand-receptor pairs, such as CCL20_CXCR3, FAM3C_CLEC2D, FAM3C_HLA_C, SPN_ICAM1, and TNFSF9_TNFRSF9, had higher mean expression levels in epidermal cells of DLE patients and SLE patients than in those of HCs. Similarly, higher interaction scores were found in endothelial cells, fibroblasts, and Macro/DCs in the dermis of DLE patients than in SLE patients and HCs, in which many receptors and ligands, such as CCR7, CCR2, CXCR3, and CCL19, were found to be highly expressed in DLE. Therefore, the potential cell-cell interactions found in our study will help understand the changes in the immune microenvironment and cutaneous lesion formation in DLE and SLE.

While this study discovered critical transcriptional features of cells in cutaneous lesions of DLE and SLE, there were several limitations. We admit that because of mechanical separation and sequencing data calculation deviation, we could not completely distinguish the cell types in the epidermis and dermis and had to remove the overlapping cells in the clustering analysis of the epidermis and dermis and in the sub-clustering analysis of cell types. Transcriptome analysis has a limited ability to identify immune cell subclusters, and the use of some protein markers will help to identify cell types more accurately.

In summary, we have shown that the use of scRNA-seq by 10× Genomics is feasible and informative in the study of lupus, especially in distinguishing different cutaneous lesions of lupus patients. Our findings provide much information about DLE- and SLE-specific cellular and molecular signatures in cutaneous lesions compared to that of HCs, and we described in detail the similarities and differences in cell compositions between DLE and SLE, which will help understand the pathogenesis of DLE and SLE and develop novel and precise therapeutic targets for lupus erythematosus.

Methods

Clinical sample collection

All skin biopsies were collected at the dermatology biopsy center in the Second Xiangya Hospital of Central South University, Xiangya Hospital of Central South and University and Institute of Dermatology of Chinese Academy of Medical Sciences and Peking Union Medical College. Our study was approved by the Ethics Committee of the Second Xiangya Hospital, Xiangya Hospital and University and Institute of Dermatology of Chinese Academy of Medical Sciences and Peking Union Medical College. All patients and healthy controls involved in this study signed informed consent forms. Skin biopsy tissue samples (4 × 4 mm) were obtained from patients with clinically indicated LE by the classification criteria for SLE (1997/2012)⁶² and CLE⁶³. The average tissue mass of cutaneous lesions was 7 mg (3–10 mg).

Preparation of single-cell suspensions

Once the pathological cutaneous lesion samples were removed from the patients, we immediately immersed them in tissue separation solution consisting of 10 mg dispase enzyme (Sigma, D4693-1G), 1 mL sterile PBS, and 2 mL sterile 1640 medium or physiological saline. Then, the biopsy tissue samples were transported to the laboratory within 30 minutes. The cutaneous lesion sample was placed in a petri dish with the epidermis facing up and the dermis facing down at 4 °C overnight. Skin biopsy tissue sample isolation and single-cell suspension preparation were performed the next day. We used an enzyme isolation method to make single-cell suspensions as previously described⁶⁴. In short, enzyme reagents in a whole-skin dissociation kit (Miltenyi Biotec, 130-101-540) were mixed into EP tubes at a certain ratio according to the operation manual. The dermis and epidermis were separated by sterile tweezers and scissors and transferred into mixed enzyme solutions. Both the dermis and epidermis were subjected to a water bath at 37 °C for 3–4 hours. After the enzymatic reaction, the tissue residue solution was filtered by a strainer, and the filtrate was centrifuged at 300 × g for 10 min at 4 °C. Furthermore, we separately added 100 μL magnetic beads to cells of the epidermis and dermis, incubated them at room temperature for 15 min, and then filtered them on a magnetic stand. The filtrate was collected, centrifuged at 300 × g for 10 min at 4 °C, and then resuspended in 10% FBS. Ten microliters of the cell suspension were mixed with the same volume of Trypan blue for counting. The loading volume was verified by the cell concentration.

scRNA-seq

scRNA-seq mainly includes GEM (gel bead-in-emulsion) generation, barcoding, cDNA amplification, library construction and sequencing. These steps were completed according to the user’s instructions of Chromium Single Cell 3ʹ Reagent Kits v3.1 (10× Genomics, product code: 1000268, 1000215, 1000120) (https://www.10xgenomics.com/support/single-cell-gene-expression). Libraries were sequenced by an Illumina NovaSeq6000 System. Approximately 10,000 cells (targeting 5000–12,000) per sample were used for single-cell RNA sequencing.

scRNA-seq data processing

Raw sequencing data were first filtered and then mapped to GRCh38 by using Cell Ranger Version 3.0.0 (https://www.10xgenomics.com). After the Cell Ranger indicator evaluation, cell doublets were removed by Scrublet⁶⁵. Then, the Seurat v4 package⁶⁶ was used as the main tool for single-cell RNA sequencing analysis. Quality control was performed to remove low-quality cells with nFeature_RNA values less than 200 or greater than 5000, with percent.mt values over 20% and with percent.redcell values over 10%. The filtered gene expression matrix was normalized and scaled by a scaling factor of 10,000. Then, the variable genes were generated by the function FindVariableFeatures. Furthermore, the first 2,000 highly variable genes were selected for principal component analysis (PCA). The batch effect among different patients was removed by the Harmony package⁶⁷ (https://github.com/immunogenomics/harmony). Then, the ElbowPlot function was performed to identify meaningful PCs. The first 20–40 PCs were used for t-SNE (t-distributed stochastic neighbor embedding) clustering analysis. The DEGs of each cluster were determined by the FindAllMarkers function, which compared one cluster with all other cells. The function AverageExpression was applied to calculate the average gene expression of each cluster or cell type. The distribution of single cells in clusters, sample groups or each sample was visualized by t-SNE plots. Some specific gene expression levels were plotted by feature plots and stacked violin plots. Other R packages, such as pheatmap, ggplot2, dplyr and RColorBrewer, were used for stacked bar plots and heatmaps. We followed the same steps as above for the sub-clustering analysis. Due to sequencing bias, there were some other cells involved in the second clustering analysis of a cell type. To eliminate their interference, we removed these cells before sub-clustering analysis.

Pseudotime trajectory analysis

Monocle2⁶⁸ was adopted to infer potential differentiation associations between some expanded subclusters of cell types. The genes used for pseudotime trajectory analysis came from the FindAllMarkers function in the Seurat package based on the Wilcoxon rank sum test and were filtered by pct.1 > 0.5 and pct.2 < 0.5. The newCellDataSet function in Monocle2 was applied to generate a data structure for trajectory analysis. Then, after normalization, reduceDimension and orderCells were used to reduce the dimensionality and order cells in the pseudotime trajectory.

GO biological process analysis

The clusterProfiler package was used for the GO functional analysis of biological processes as previously described⁶⁹. Briefly, the marker gene list of subclusters or samples was provided for ID conversion from gene symbols to Entrez ID, and the enriched GO functions were used for GO biological process analysis with a p value <0.01 and a q-value <0.05. The bubble chart in this study visualizes the partially enriched biological processes of subclusters or samples.

Similarity scores

The similarity scores were calculated based on the differentially expressed genes of each subcluster in our study and subclusters (SCs) reported in PBMCs of lupus⁹. We adopted the intersecting function of R to find the common number of differentially expressed genes between the subclusters in our study and SCs in PBMCs, and then we divided by the sum of two differentially expressed genes. The value obtained was the similarity score. The pheatmap package was used for visualization.

Immunofluorescence

T cells, B cells and NK cells were separately stained for CD3 (MAB-0740, MXB Biotech), CD19 (1:1000, ab134114, Abcam) and CD56 (1:400, ab75813, Abcam) expression. CCL20⁺ keratinocytes were stained with CCL20 (1:50, ab224188, Abcam) and KRT10 (1:500, ab76318, Abcam). CXCL1⁺ fibroblasts and HLA⁺ Fib1 were identified by staining CXCL1 (1:100, ab89318, Abcam) and HLA-DRB1 (1:1000, ab133578, Abcam) with Vimentin (1:20000, ab92547, Abcam). ELANE (1:2000, ELA-2, ab131260, Abcam) was used to stain neutrophils. CCL19 (1:500, 13397-1-AP, ProteinTech) staining was performed in fibroblasts, and CCR7 (1:500, 25898-1-AP, ProteinTech) staining was performed in macrophages (CD68, 1:400, ab955, Abcam). The second antibodies used in this study came from Abcam (Donkey Anti-Rabbit IgG, 1:1000, ab205722, Abcam). All immunofluorescence experiments were conducted as follows⁷⁰. Briefly, skin biopsy samples were embedded in paraffin and sectioned into 4 µm thick sections. After the paraffin tissue sections were baked, dewaxed, and rehydrated, they were used for antigen retrieval under acidic conditions in a pressure cooker. After peroxidase enzymes were removed, the cutaneous lesions were sealed with 5% serum (Well-Biology, CN). Then, the samples were incubated with primary antibodies at room temperature for 1 h or at 4 °C overnight. After incubation with secondary antibodies and staining with reagents from an Opal 7-color IHC detection kit (NEL811001KT, PerkinElmer, Hopkinton, MA), histologic sections were observed with an LSM 510 confocal microscopy system (Leica) or PerkinElmer Vectra (Hopkinton, MA). All images were analyzed using inForm software compatible with the PerkinElmer Vectra.

Cell-cell communication analysis and interaction scores

Cell-to-cell communication analysis of potential receptor-ligand pairings was conducted using CellPhoneDB⁵². For calculation of the average expression and product value of ligands and receptors, each cell type was labeled and randomly arranged 1000 times. To identify the meaningful interaction pairs, we calculated the p value and mean expression, and then we eliminated the pairs with a cutoff p value lower than 0.05. The visualizations were performed in pheatmap and ggplot2. Furthermore, we counted the numbers of ligand-receptor pairs enriched between cell types, and the interaction scores were generated by the pheatmap function with the parameter scale = ‘row’ and clustering_distance_rows = “correlation” (Pearson correlation).

Statistical analysis

The data are shown as the mean ± SD, and the statistical significance of the proportional differences in cell types among HC, DLE, and SLE was determined by one-way ANOVA test with Bonferroni’s multiple comparisons test. Other biologically significant differences were determined by the default algorithm of R packages.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The processed data of this study have been deposited in the GEO database under accession code GSE179633. These data generated in this study are available within the Article, Supplementary Information or from the corresponding author upon reasonable request. Source data are provided with this paper.

Code availability

All the codes used for processing and analyzing the data in this study have been deposited in an available GitHub repository⁷¹ (https://doi.org/10.5281/zenodo.7193545).

References

Durcan, L., O’Dwyer, T. & Petri, M. Management strategies and future directions for systemic lupus erythematosus in adults. Lancet 393, 2332–2343 (2019).
Article Google Scholar
Mande, P. et al. Fas ligand promotes an inducible TLR-dependent model of cutaneous lupus-like inflammation. J. Clin. Invest. 128, 2966–2978 (2018).
Article Google Scholar
Kiriakidou, M., Cotton, D., Taichman, D. & Williams, S. Systemic lupus erythematosus. Ann. Intern. Med. 159, Itc4–1 (2013).
Article Google Scholar
Zhang, Y. P., Wu, J., Han, Y. F., Shi, Z. R. & Wang, L. Pathogenesis of cutaneous lupus erythema associated with and without systemic lupus erythema. Autoimmun. Rev. 16, 735–742 (2017).
Article CAS Google Scholar
Grönhagen, C. M., Fored, C. M., Granath, F. & Nyberg, F. Cutaneous lupus erythematosus and the association with systemic lupus erythematosus: a population-based cohort of 1088 patients in Sweden. Br. J. Dermatol. 164, 1335–1341 (2011).
Article Google Scholar
Li, Q., Wu, H., Zhou, S., Zhao, M. & Lu, Q. An update on the pathogenesis of skin damage in lupus. Curr. Rheumatol. Rep. https://doi.org/10.1007/s11926-020-00893-9 (2020).
Arazi, A. et al. The immune cell landscape in kidneys of patients with lupus nephritis. Nat. Immunol. 20, 902–914 (2019).
Article CAS Google Scholar
Der, E. et al. Tubular cell and keratinocyte single-cell transcriptomics applied to lupus nephritis reveal type I IFN and fibrosis relevant pathways. Nat. Immunol. 20, 915–927 (2019).
Article CAS Google Scholar
Nehar-Belaid, D. et al. Mapping systemic lupus erythematosus heterogeneity at the single-cell level. Nat. Immunol. https://doi.org/10.1038/s41590-020-0743-0 (2020).
Farkas, L., Beiske, K., Lund-Johansen, F., Brandtzaeg, P. & Jahnsen, F. L. Plasmacytoid dendritic cells (natural interferon- alpha/beta-producing cells) accumulate in cutaneous lupus erythematosus lesions. Am. J. Pathol. 159, 237–243 (2001).
Article CAS Google Scholar
Furie, R. et al. Monoclonal antibody targeting BDCA2 ameliorates skin lesions in systemic lupus erythematosus. J. Clin. Invest. 129, 1359–1371 (2019).
Article Google Scholar
Vorwerk, G., Zahn, S., Bieber, T. & Wenzel, J. NKG2D and its ligands as cytotoxic factors in cutaneous lupus erythematosus. Exp. Dermatol. https://doi.org/10.1111/exd.14311 (2021).
Xiang, M. et al. Bioinformatic analysis of key biomarkers and immune filtration of skin biopsy in discoid lupus erythematosus. Lupus 30, 807–817 (2021).
Article CAS Google Scholar
Deng, G. M. & Tsokos, G. C. Pathogenesis and targeted treatment of skin injury in SLE. Nat. Rev. Rheumatol. 11, 663–669 (2015).
Article CAS Google Scholar
Deng, G. M. Pathogenesis of skin injury of systemic lupus erythematosus. Curr. Rheumatol. Rep. 20, 5 (2018).
Article Google Scholar
He, H. et al. Single-cell transcriptome analysis of human skin identifies novel fibroblast subpopulation and enrichment of immune subsets in atopic dermatitis. J. Allergy Clin. Immunol. https://doi.org/10.1016/j.jaci.2020.01.042 (2020).
Tabib, T., Morse, C., Wang, T., Chen, W. & Lafyatis, R. SFRP2/DPP4 and FMO1/LSP1 define major fibroblast populations in human skin. J. Invest. Dermatol. 138, 802–810 (2018).
Article CAS Google Scholar
Li, J. et al. Single-cell characterization of the cellular landscape of acral melanoma identifies novel targets for immunotherapy. Clin. Cancer Res. 28, 2131–2146 (2022).
Article CAS Google Scholar
Ma, F. et al. The cellular architecture of the antimicrobial response network in human leprosy granulomas. Nat. Immunol. 22, 839–850 (2021).
Article CAS Google Scholar
Deng, C. C. et al. Single-cell RNA-seq reveals fibroblast heterogeneity and increased mesenchymal fibroblasts in human fibrotic skin diseases. Nat. Commun. 12, 3709 (2021).
Article CAS ADS Google Scholar
Xing, X. et al. Decoding the multicellular ecosystem of lung adenocarcinoma manifested as pulmonary subsolid nodules by single-cell RNA sequencing. Sci. Adv. https://doi.org/10.1126/sciadv.abd9738 (2021).
Direder, M. et al. Schwann cells contribute to keloid formation. Matrix Biol. 108, 55–76 (2022).
Article CAS Google Scholar
Sun, Z. et al. A Bayesian mixture model for clustering droplet-based single-cell transcriptomic data from population studies. Nat. Commun. 10, 1649 (2019).
Article ADS Google Scholar
Abernathy-Close, L. et al. B cell signatures distinguish cutaneous lupus erythematosus subtypes and the presence of systemic disease activity. Front. Immunol. 12, 775353 (2021).
Article CAS Google Scholar
Lou, F. et al. Excessive polyamine generation in keratinocytes promotes self-RNA sensing by dendritic cells in psoriasis. Immunity 53, 204–216.e10 (2020).
Article CAS Google Scholar
Ji, A. L. et al. Multimodal analysis of composition and spatial architecture in human squamous cell carcinoma. Cell https://doi.org/10.1016/j.cell.2020.05.039 (2020).
Nagao, K. et al. Stress-induced production of chemokines by hair follicles regulates the trafficking of dendritic cells in skin. Nat. Immunol. 13, 744–752 (2012).
Article CAS Google Scholar
Goel, R. R. et al. Interferon lambda promotes immune dysregulation and tissue inflammation in TLR7-induced lupus. Proc. Natl Acad. Sci. USA 117, 5409–5419 (2020).
Article CAS ADS Google Scholar
Mahmoudi, S. et al. Heterogeneity in old fibroblasts is linked to variability in reprogramming and wound healing. Nature 574, 553–558 (2019).
Article CAS ADS Google Scholar
Showalter, K. et al. Machine learning integration of scleroderma histology and gene expression identifies fibroblast polarisation as a hallmark of clinical severity and improvement. Ann. Rheum. Dis. 80, 228–237 (2021).
Article CAS Google Scholar
Salzer, M. C. et al. Identity noise and adipogenic traits characterize dermal fibroblast aging. Cell 175, 1575–1590.e1522 (2018).
Article CAS Google Scholar
Lynch, M. D. & Watt, F. M. Fibroblast heterogeneity: implications for human disease. J. Clin. Invest. 128, 26–35 (2018).
Article Google Scholar
Adrover, J. M. et al. A neutrophil timer coordinates immune defense and vascular protection. Immunity 50, 390–402.e310 (2019).
Article CAS Google Scholar
Davidson, A. & Diamond, B. Autoimmune diseases. N. Engl. J. Med. 345, 340–350 (2001).
Article CAS Google Scholar
Kaul, A. et al. Systemic lupus erythematosus. Nat. Rev. Dis. Prim. 2, 16039 (2016).
Article Google Scholar
Zhang, J. Y. et al. Single-cell landscape of immunological responses in patients with COVID-19. Nat. Immunol. 21, 1107–1118 (2020).
Article CAS Google Scholar
Cano-Gamez, E. et al. Single-cell transcriptomics identifies an effectorness gradient shaping the response of CD4(+) T cells to cytokines. Nat. Commun. 11, 1801 (2020).
Article CAS ADS Google Scholar
Hannedouche, S. et al. Oxysterols direct immune cell migration via EBI2. Nature 475, 524–527 (2011).
Article CAS Google Scholar
Pereira, J. P., Kelly, L. M., Xu, Y. & Cyster, J. G. EBI2 mediates B cell segregation between the outer and centre follicle. Nature 460, 1122–1126 (2009).
Article CAS ADS Google Scholar
Tipton, C. M., Hom, J. R., Fucile, C. F., Rosenberg, A. F. & Sanz, I. Understanding B-cell activation and autoantibody repertoire selection in systemic lupus erythematosus: a B-cell immunomics approach. Immunol. Rev. 284, 120–131 (2018).
Article CAS Google Scholar
Billi, A. C. et al. Nonlesional lupus skin contributes to inflammatory education of myeloid cells and primes for cutaneous inflammation. Sci. Transl. Med. 14, eabn2263 (2022).
Article CAS Google Scholar
Peng, L. et al. Integrating single-cell RNA sequencing with spatial transcriptomics reveals immune landscape for interstitial cystitis. Signal Transduct. Target Ther. 7, 161 (2022).
Article CAS Google Scholar
Chen, D. & Androlewicz, M. J. Heat shock protein 70 moderately enhances peptide binding and transport by the transporter associated with antigen processing. Immunol. Lett. 75, 143–148 (2001).
Article CAS Google Scholar
Suzue, K., Zhou, X., Eisen, H. N. & Young, R. A. Heat shock fusion proteins as vehicles for antigen delivery into the major histocompatibility complex class I presentation pathway. Proc. Natl Acad. Sci. USA 94, 13146–13151 (1997).
Article CAS ADS Google Scholar
Chen, Z. et al. IL-13Rα1-mediated signaling regulates age-associated/autoimmune B-cell expansion and lupus pathogenesis. Arthritis Rheumatol. https://doi.org/10.1002/art.42146 (2022).
Dominguez Conde, C. et al. Cross-tissue immune cell analysis reveals tissue-specific features in humans. Science 376, eabl5197 (2022).
Article CAS Google Scholar
Clarke, A. J. et al. Autophagy is activated in systemic lupus erythematosus and required for plasmablast development. Ann. Rheum. Dis. 74, 912–920 (2015).
Article Google Scholar
Sisirak, V. et al. Digestion of chromatin in apoptotic cell microparticles prevents autoimmunity. Cell 166, 88–101 (2016).
Article CAS Google Scholar
Zhang, L. et al. Single-cell analyses inform mechanisms of myeloid-targeted therapies in colon cancer. Cell 181, 442–459.e429 (2020).
Article CAS Google Scholar
Sisirak, V. et al. Genetic evidence for the role of plasmacytoid dendritic cells in systemic lupus erythematosus. J. Exp. Med. 211, 1969–1976 (2014).
Article CAS Google Scholar
Hsu, H. C., Chen, Y. H., Lin, T. S., Shen, C. Y. & Hsieh, S. C. Systemic lupus erythematosus is associated with impaired autophagic degradation via interleukin-6 in macrophages. Biochim. Biophys. Acta Mol. Basis Dis. 1867, 166027 (2021).
Article CAS Google Scholar
Efremova, M., Vento-Tormo, M., Teichmann, S. A. & Vento-Tormo, R. CellPhoneDB: inferring cell–cell communication from combined expression of multi-subunit ligand–receptor complexes. Nat. Protoc. 15, 1484–1506 (2020).
Article CAS Google Scholar
Wieczorek, I. T., Propert, K. J., Okawa, J. & Werth, V. P. Systemic symptoms in the progression of cutaneous to systemic lupus erythematosus. JAMA Dermatol. 150, 291–296 (2014).
Article Google Scholar
Li, J., Lu, E., Yi, T. & Cyster, J. G. EBI2 augments Tfh cell fate by promoting interaction with IL-2-quenching dendritic cells. Nature 533, 110–114 (2016).
Article CAS ADS Google Scholar
Wei, K. et al. Notch signalling drives synovial fibroblast identity and arthritis pathology. Nature 582, 259–264 (2020).
Article CAS ADS Google Scholar
Fürnrohr, B. G. et al. Polymorphisms in the Hsp70 gene locus are genetically associated with systemic lupus erythematosus. Ann. Rheum. Dis. 69, 1983–1989 (2010).
Article Google Scholar
Sarkar, M. K. et al. Photosensitivity and type I IFN responses in cutaneous lupus are driven by epidermal-derived interferon kappa. Ann. Rheum. Dis. 77, 1653–1664 (2018).
Article CAS Google Scholar
Banchereau, J. & Pascual, V. Type I interferon in systemic lupus erythematosus and other autoimmune diseases. Immunity 25, 383–392 (2006).
Article CAS Google Scholar
Mistry, P. et al. Transcriptomic, epigenetic, and functional analyses implicate neutrophil diversity in the pathogenesis of systemic lupus erythematosus. Proc. Natl Acad. Sci. USA 116, 25222–25228 (2019).
Article CAS ADS Google Scholar
Frangou, E. et al. REDD1/autophagy pathway promotes thromboinflammation and fibrosis in human systemic lupus erythematosus (SLE) through NETs decorated with tissue factor (TF) and interleukin-17A (IL-17A). Ann. Rheum. Dis. 78, 238–248 (2019).
Article CAS Google Scholar
Koenig, U. et al. Cell death induced autophagy contributes to terminal differentiation of skin and skin appendages. Autophagy 16, 932–945 (2020).
Article CAS Google Scholar
Hartman, E. A. R., van Royen-Kerkhof, A., Jacobs, J. W. G., Welsing, P. M. J. & Fritsch-Stork, R. D. E. Performance of the 2012 Systemic Lupus International Collaborating Clinics classification criteria versus the 1997 American College of Rheumatology classification criteria in adult and juvenile systemic lupus erythematosus. A systematic review and meta-analysis. Autoimmun. Rev. 17, 316–322 (2018).
Article Google Scholar
Kuhn, A., Sticherling, M. & Bonsmann, G. Clinical manifestations of cutaneous lupus erythematosus. J. Dtsch Dermatol. Ges. 5, 1124–1137 (2007).
Article Google Scholar
Der, E. et al. Single cell RNA sequencing to dissect the molecular heterogeneity in lupus nephritis. JCI Insight https://doi.org/10.1172/jci.insight.93009 (2017).
Wolock, S. L., Lopez, R. & Klein, A. M. Scrublet: computational identification of cell doublets in single-cell transcriptomic data. Cell Syst. 8, 281–291.e289 (2019).
Article CAS Google Scholar
Stuart, T. et al. Comprehensive integration of single-cell data. Cell 177, 1888–1902.e1821 (2019).
Article CAS Google Scholar
Korsunsky, I. et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nat. Methods 16, 1289–1296 (2019).
Article CAS Google Scholar
Qiu, X. et al. Reversed graph embedding resolves complex single-cell trajectories. Nat. Methods 14, 979–982 (2017).
Article CAS Google Scholar
Yu, G., Wang, L. G., Han, Y. & He, Q. Y. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 16, 284–287 (2012).
Article CAS Google Scholar
Wang, Z. et al. E4BP4-mediated inhibition of T follicular helper cell differentiation is compromised in autoimmune diseases. J. Clin. Invest. 130, 3717–3733 (2020).
Article CAS Google Scholar
Zheng, M. et al. Single-cell sequencing shows cellular heterogeneity of cutaneous lesions in lupus erythematosus. Github. https://doi.org/10.5281/zenodo.7193545 (2022).

Download references

Acknowledgements

We thank the dermatology biopsy centers in the second Xiangya Hospital of Central South University, Xiangya Hospital of Central South University and Institute of Dermatology of Chinese Academy of Medical Sciences and Peking Union Medical College for providing the skin biopsies for this study. This work was supported by the National Natural Science Foundation of China (No. 82030097 and No. 81874243, Ming Zhao), CAMS Innovation Fund for Medical Sciences (CIFMS) (2019-I2M-5-033, M.Z.), the Key project for international and regional cooperation in science and technology innovation of Hunan province (2019WK2081, M.Z.), and the Project for leading talents in science and technology in Hunan province (2019RS3003, Ming Zhao).

Author information

These authors contributed equally: Meiling Zheng, Zhi Hu, Xiaole Mei, Lianlian Ouyang, Yang Song.

Authors and Affiliations

Department of Dermatology, Hunan Key Laboratory of Medical Epigenomics, Second Xiangya Hospital, Central South University, 410011, Changsha, China
Meiling Zheng, Zhi Hu, Lianlian Ouyang, Yang Song, Wenhui Zhou, Yi Kong, Ruifang Wu, Shijia Rao, Hai Long, Hui Jing, Shuang Lu, Haijing Wu, Qianjin Lu & Ming Zhao
Research Unit of Key Technologies of Diagnosis and Treatment for Immune-related Skin Diseases, Chinese Academy of Medical Sciences, 410011, Changsha, China
Meiling Zheng, Zhi Hu, Lianlian Ouyang, Yang Song, Wenhui Zhou, Yi Kong, Ruifang Wu, Shijia Rao, Hai Long, Hui Jing, Shuang Lu, Haijing Wu, Qianjin Lu & Ming Zhao
Institute of Dermatology, Chinese Academy of Medical Sciences and Peking Union Medical College, 210042, Nanjing, China
Xiaole Mei & Qianjin Lu
Department of Dermatology, Xiangya Hospital, Central South University, 410008, Changsha, China
Wei Shi
Department of Pharmacy, The Third Xiangya Hospital, Central South University, 410011, Changsha, China
Sujie Jia

Authors

Meiling Zheng
View author publications
You can also search for this author in PubMed Google Scholar
Zhi Hu
View author publications
You can also search for this author in PubMed Google Scholar
Xiaole Mei
View author publications
You can also search for this author in PubMed Google Scholar
Lianlian Ouyang
View author publications
You can also search for this author in PubMed Google Scholar
Yang Song
View author publications
You can also search for this author in PubMed Google Scholar
Wenhui Zhou
View author publications
You can also search for this author in PubMed Google Scholar
Yi Kong
View author publications
You can also search for this author in PubMed Google Scholar
Ruifang Wu
View author publications
You can also search for this author in PubMed Google Scholar
Shijia Rao
View author publications
You can also search for this author in PubMed Google Scholar
Hai Long
View author publications
You can also search for this author in PubMed Google Scholar
Wei Shi
View author publications
You can also search for this author in PubMed Google Scholar
Hui Jing
View author publications
You can also search for this author in PubMed Google Scholar
Shuang Lu
View author publications
You can also search for this author in PubMed Google Scholar
Haijing Wu
View author publications
You can also search for this author in PubMed Google Scholar
Sujie Jia
View author publications
You can also search for this author in PubMed Google Scholar
Qianjin Lu
View author publications
You can also search for this author in PubMed Google Scholar
Ming Zhao
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

M. Zheng processed the data, performed the analysis, and wrote the manuscript. X.M., L.O., and Y.S. conducted the immunofluorescence staining of skin tissues. M.Zhao and Z.H. helped M. Zheng to analyze the subclusters. W.Z. and H.J. helped M. Zheng to collect and handle the clinical information of patients. Z.H., Y.K., and R.W. performed the treatment of single suspensions. Q.L., H.L., S.R., and W.S. performed the diagnosis of DLE and SLE and collected the skin tissues. S.L., H.W. and S.J. help collect samples of DLE and SLE and immunofluorescence staining of skin tissues. M. Zhao conceptualized and supervised this study and revised the manuscript. Q.L. conceptualized and supervised this study and revised the manuscript. All authors reviewed and approved the manuscript.

Corresponding authors

Correspondence to Qianjin Lu or Ming Zhao.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Marta Alarcon-Riquelme, Sarah Pringle and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files

Supplementary Data 1

Supplementary Data 2

Supplementary Data 3

Supplementary Data 4

Supplementary Data 5

Supplementary Data 6

Supplementary Data 7

Supplementary Data 8

Supplementary Data 9

Supplementary Data 10

Supplementary Data 11

Supplementary Data 12

Supplementary Data 13

Supplementary Data 14

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Zheng, M., Hu, Z., Mei, X. et al. Single-cell sequencing shows cellular heterogeneity of cutaneous lesions in lupus erythematosus. Nat Commun 13, 7489 (2022). https://doi.org/10.1038/s41467-022-35209-1

Download citation

Received: 15 July 2021
Accepted: 23 November 2022
Published: 05 December 2022
DOI: https://doi.org/10.1038/s41467-022-35209-1

This article is cited by

COX5A as a potential biomarker for disease activity and organ damage in lupus
- Minglong Cai
- Yi Qin
- Zhu Chen
Clinical and Experimental Medicine (2023)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.