Esophageal achalasia (EA) is a rare primary esophageal motility disorder, which is characterized by aberrant peristalsis in distal esophageal and insufficient relaxation of lower esophageal sphincter (LES).^1,2 EA is caused by destruction of ganglion cells in myenteric plexus of esophageal. The majority of affected patients present with dysphagia to solids and liquids, regurgitation, and chest pain or weight loss. However, the specific pathogenesis of EA remains largely unclear, all current treatment options in EA are palliative in nature.³

EA is a neurodegenerative disease of enteric nervous system with a hypothesis that immune-mediated ganglionitis triggered by potential virus infections has been proposed to underlie the loss of myenteric neurons in EA, particularly in genetically susceptible individuals.^4,5 Genetic association studies have demonstrated a strong MHC association by imputing human leukocyte antigen haplotypes variant^6,7 and other immune common loci.^8-12 Compared to controls, patients with EA are more likely to suffer from an autoimmune disease.^13,14 What is more, there is a prominent pattern of inflammatory cells infiltrating, such as eosinophils, mast cells, T cells, and elevated of cytokines in both in circulation and the LES.^15-17 These studies suggest that inflammatory response plays an essential role in development of EA. However, there are still no studies mapping the peripheral leukocytes for patients with EA.

In the present study, in comparison to controls, we addressed this question by analyzing in detail the phenotype of peripheral leukocytes of EA by flow cytometry and then verified the expression of significantly different clusters histologically. Furthermore, we performed RNA sequencing analysis (RNA-seq) of peripheral blood mononuclear cells (PBMCs) to detect transcriptomes changes in EA. Our results provided important clues for cell clusters heterogeneity in both the number and expression levels.

Participants Cohort

Blood and tissue samples used in this study were collected from Zhongshan Hospital, Fudan University following approval (IRB No. B2021-438R) from the ethics committee. In this study, 76 EA patients (mean age ± SD: 44.67 ± 15.92 years) and 50 age-matched healthy controls (41.27 ± 14.25 years) were included in flow cytometry analysis and RNA extraction. All patients underwent a professional clinical assessment by experienced endoscopists. Patients were diagnosed with EA based on high-resolution manometry, upper-gastrointestinal endoscopy, and barium esophagogram. All patients did not have a history of prior treatments like dilation, botulinum toxin injection, stent, Heller myotomy, or peroral endoscopic myotomy (POEM). No participant had a history of cancer, cerebrovascular disease, genetic disorders, infectious disease, and other acute inflammatory-related diseases, including the active stage of allergy, celiac disease, glomerulonephritis, inflammatory bowel disease. Detailed patient demographics and clinical data are summarized in Supplementary Table 1. All participants were informed about the study and agreed to participate by signing an informed consent form.

Human Blood Processing

Five milliliters of blood samples were collected in a lithium heparin tube (BD Biosciences, San Jose, CA, USA), placed on ice, and processed within 3 hours. 100 μL/sample were processed for flow cytometry in each panel. Leukocyte were obtained after 10-15 minutes red blood cell lysis. After washing, cells were resuspended in 300 μL of staining buffer for staining.

PBMCs were isolated from the rest whole blood samples by Ficoll-Paque (GE Healthcare, Pittsburgh, PA, USA) gradient centrifugation according to the manufacturer’s instructions, and then were counted. After that, about 1-2 × 10⁶ PBMCs were taken from the samples and lysed in 1 mL TRIzol for gene expression analysis. The samples were stored at –80℃.

Leukocyte Staining and Data Acquisition

For staining, assays were carried out with staining buffers and antibodies from BD Biosciences. Cells were surface stained with panel 1 (anti-CD3-BUV395, anti-HLA-DR-FITC, anti-CD11c-BV421, anti-CD123-APC, anti-CD56-PE, anti-CD19-PE-Cy7, anti-CD14-BV510, and anti-CD15-APC-Cy7) and panel 2 (anti-CD3-BV421, anti-CD4-APC-Cy7, anti-CD8-PerCP-Cy5.5, anti-CD25-PE, anti-γδTCR-APC, and anti-CD127-FITC).

As for intracellular cytokine staining, assays were carried out with staining buffer and antibodies from BD Biosciences. Briefly, cells were seeded 1 × 10⁶ cell per well in a 96-well plate and stimulated with leukocyte activation cocktail (BD GolgiPlug; BD Biosciences) for 4 hours. After stimulation, cells were fixed with BD Cytofix/Cytoperm fixation and washed with BD Perm/Wash buffer. Cells were stained by a separate panel 3 including anti-CD3-BV421, anti-CD8-PerCP-Cy5.5, anti-IFNγ-FITC, anti-IL4-APC, and anti-IL17-PE.

All stained cells were analyzed with a BD LSR II cytometer. Isotype controls or fluorescence minus 1 was used for markers or gate settings.

High-dimensional Flow Cytometry Data Analysis

Flow Cytometry Standard files were imported into FlowJo software version 10. First, we cleaned up the data with FlowAI, and standard gated to remove aggregates and dead cells. After compensation, data were reduced by the dimensionality of the data-set for visualization by a Uniform Manifold Approximation and Projection (UMAP) algorithm. Cell population identification was conducted by combining supervised and FlowSOM Unsupervised Clustering approaches¹⁸ for high-dimensional cytometry data. All visualizations of clustering solutions were through the use of UMAP heatmap.

Tissue Sample Collection

Esophageal achalasia is characterized by aberrant peristalsis in the distal esophageal and insufficient relaxation of LES, and the tissue biopsy was collected at LES high-pressure zone muscle tissue, which can be considered a representative of the typical reflection of achalasia. Tissue biopsies were collected in the operating room from patients undergoing POEM and controls undergoing submucosal tunneling endoscopic resection for esophageal benign submucosal tumor. The samples of LES were immediately fixed in 10% formalin and then embedded in paraffin. All participants received detailed explanations about biopsy for medical research and signed an informed consent before the operation.

Hematoxylin and Eosin and Immunohistochemical Staining

To observe morphological and immune cells infiltrate in the LES, the formalin-fixed paraffin-embedded (FFPE) specimens were sectioned at 6 μm thickness for H&E, immunohistochemical and immunofluorescence. For immunohistochemical, the FFPE slides were incubated with primary antibodies myeloperoxidase (ab45977; Abcam, Cambridge, UK) or major basic protein (MBP) (BA3806; Boster, Wuhan, China) at 4℃ overnight and biotinylated secondary antibody for 1 hour. Slides were analyzed using a bright field microscope. For immunofluorescence, the FFPE slides were incubated with primary antibodies CD8 (ab4055; Abcam), and CD4 (ab25804; Abcam) antibodies, overnight at 4℃ in the dark, and Alexa-488-labeled and Cy3-labeled secondary antibody for 1 hour. Slides were analyzed using a fluorescent microscope. All stainings were counted according to the number of positive staining cells around the ganglion.

RNA Extraction, Quantification, and Sequencing

Total RNAs were extracted from PBMC cells using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer instructions. Total RNA was qualified and quantified using a Nano Drop and Agilent 2100 bioanalyzer (Thermo Fisher Scientific, Waltham, MA, USA). Samples with RNA concentration > 10 ng/μL and RNA integrity/quality numbers > 7 were selected for the following RNA-seq library preparations. RNA sequencing was performed by using a BGIseq500 sequencer (BGI, Shenzhen, Guangdong, China), with an average yield of 1.17 G data per sample. After quality control, quantitative gene analysis was performed. The average alignment ratio of the sample comparison genome was 95.16%, which was comparative. The average alignment of the gene set was 69.14%. The filtered clean reads are aligned to the reference sequence. Several analyses based on gene expression levels (principal components, correlation, and differential gene screening) were performed. A total of 18 852 mRNA were detected.

Proportions of Immune Cell Types According to Bulk RNA Sequencing Data

RNA-seq reads were quantified to transcripts per million, which were suitable for use with Cell-type Identification by Estimating Relative Subsets of RNA Transcripts (CIBERSORT).^19,20 While CIBERSORT estimated the proportions of 20 immune cell types, we recategorized these 20 cell types into 10 major cell types by summing the proportions as appropriate.

Differentially Expressed Genes Identification, Biological and Functional Analysis, Network-based Meta-analysis, and Selection of Hub Genes

Differentially expressed genes (DEGs) were defined as genes with a Q-value < 0.05, log2 fold change > 1 or < –1. GO functional significance enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment were performed by programming language R. The correlation between differentially expressed transcription factors was assembled using protein-protein interactions by Search Tool for the Retrieval of Interacting genes (STRING) (http://string-db.org) and visualized by Cytoscape. Hub genes were selected based on by the CytoHubba application in Cytoscape and the top 10 hub genes evaluated by the 5 calculation methods, including maximal clique centrality, maximum neighborhood component, edge percolated component, node connect degree, and Eccentricity.

Statistical Methods

All the analyses and figures were performed with programming language R 4.0.2, and Prism (version 8.2; Graphpad, San Diego, CA, USA). EA samples were clustering into subgroups according to their median. Differences in frequency of subsets were analyzed by the 2-tailed Student’s t test, and P < 0.05 was considered significant.

Access to Data

All authors had access to the all data and have reviewed and approved the final manuscript.

Transcript Profiling

The United States National Library of Medicine (NLM) BioProject (PRJNA761345).

High-dimensional Flow Cytometry Analysis of Peripheral Leukocytes Identifies Esophageal Achalasia Immunophenotypes

An 8-color flow cytometry panel was developed to identify leukocyte populations from peripheral blood of 50 healthy age matched controls and 76 EA patients (Fig. 1A and Supplementary Tables 1 and 2). The cases consisted of 40 males and 36 females (44.67 ± 15.92 years). The mean preoperative Eckardt score was 6 (range 4-8), and median disease duration was 4 years (range 2-10 years). We concluded 51 patients for IgG test of viral infection, found 6 patients with IgG positive of Epstein–Barr virus, 48 patients with Cytomegalovirus, 27 patients with herpes simplex virus, and 43 patients with Rubella virus.

Figure 1. High-dimensional flow cytometry analysis of peripheral leukocytes in a learning cohort of achalasia. (A) The experimental workflow of peripheral blood sample collection and analysis. (B) Non-supervised and color-coded Uniform Manifold Approximation and Projection (UMAP) map depicting 10 clusters from 126 donors (achalasia, 76; control, 50). (C) Cell surface marker expression in the UMAP analysis. (D) UMAP map of all patients with controls (left) and achalasia (right). (E) Pie chart depicting the cellular composition of all leukocytes. (F) Stacked column graph of the frequencies of the 10 clusters in healthy controls and achalasia samples. RBC, red blood cell; NK, natural killer; mDC, classical dendritic cell; pDC, plasmacytoid dendritic cell.

Ten distinct subpopulations were identified from peripheral leukocytes, visualized using UMAP, which corresponded to neutrophils, monocytes, T cells, natural killer (NK) cells, B cells, classical dendritic cells, plasmacytoid dendritic cells, eosinophils, and basophils (Fig. 1B-E). Clusters were based on the surface staining compared to all other cells (Supplementary Table 3). Bar graph of the frequencies of the 10 clusters indicated that they are substantially different at the high-dimensional single cell level (Fig. 1F).

Many clusters were differentially represented in controls and patients (Fig. 2A and Table). Subpopulation overrepresented in patients were neutrophils, and underrepresented were T cells and eosinophils (Fig. 2B and 2C). The remaining (7/10) subpopulations were present in similar proportions across the 2 groups, and a reduction of eosinophils with high Eckardt (Eckardt score ≥ 6) (P-value = 0.001) (Supplementary Table 4). Besides, patients with EA also exhibited an expansion in the absolute numbers of neutrophils and eosinophil in LES infiltration (Fig. 2D-F). Thus, our results revealed an alteration of immune cell clusters in EA peripheral leukocyte and innate cells infiltration in LES.

Figure 2. Altered immune clusters in peripheral and lower esophageal sphincter (LES) in patients with achalasia. (A) Violin plot comparing the proportion of clusters (n = 9) across the individuals. Controls are shown in blue, and achalasia in red. P-values were calculated using t tset or Wilcoxon test. (B, C) Zebra plot show flow cytometry analysis of peripheral neutrophils and eosinophils. (D) Neutrophilic and eosinophilic infiltration of the LES was assessed by immunostaining for myeloperoxidase (MPO) and major basic protein (MBP) in patients and controls (ganglion cells, blue arrows; MPO or MBP positive cells, red arrows) (controls, n = 15; achalasia, n = 20). (E, F) Number of MPO and MBP-labeled neutrophils and eosinophils per high-powered field from LES (controls, n = 10; achalasia, n = 10). NK, natural killer; mDC, classical dendritic cell; pDC, plasmacytoid dendritic cell. *P < 0.05; **P < 0.01.

Table. Frequency of Each Subsets

Cell type	All samples (N = 126)	Control (n = 50)	Achalasia (n = 76)	P-value
Panel 1 (WBC%)
Neutrophils	44.79 ± 12.20	40.89 ± 12.81	47.36 ± 11.13	0.003
B cell	3.88 (2.68-5.33)	3.89 (2.50-5.32)	3.84 (2.73-5.47)	0.942
NK cell	7.12 (4.63-9.46)	7.22 (5.21-10.55)	6.93 (4.34-9.37)	0.322
Monocyte	2.18 (1.50-3.08)	2.28 (1.50-3.08)	2.10 (1.40-3.06)	0.423
T cell	35.28 ± 10.21	38.17 ± 11.58	33.38 ± 8.77	0.009
Eosinophil	0.11 (0.059-0.244)	0.190 (0.070-0.310)	0.091 (0.058-0.180)	0.004
Basophils	0.130 (0.069-0.200)	0.130 (0.069-0.200)	0.130 (0.063-0.200)	0.970
mDC	0.037 (0.021-0.055)	0.037 (0.022-0.056)	0.037 (0.020-0.055)	0.673
pDC	0.026 (0.015-0.039)	0.026 (0.015-0.038)	0.025 (0.013-0.040)	0.891
Panel 2 (WBC%)
γδT	0.17 (0.10-0.41)	0.19 (0.11-0.35)	0.16 (0.10-0.43)	0.729
CD8+	7.80 (5.83-10.85)	7.02 (5.12-9.29)	8.34 (6.05-10.93)	0.041
CD4+	11.90 (9.14-15.33)	11.31 (8.56-15.48)	12.39 (9.38-15.30)	0.487
Treg (CD4%)	6.48 (5.30-7.93)	6.00 (4.90-7.45)	6.84 (5.50-8.30)	0.023
Panel 3 (CD4%)
Th1	23.00 (17.30-31.00)	21.00 (16.90-26.65)	25.30 (17.80-33.25)	0.057
Th2	1.50 (1.01-2.22)	1.26 (1.01-1.95)	1.68 (1.02-2.37)	0.102
Th17	1.83 (1.30-2.55)	1.58 (1.27-2.59)	1.95 (1.30-2.50)	0.584
Pathological Th17	0.25 (0.14-0.42)	0.28 (0.15-0.48)	0.24 (0.10-0.40)	0.398

WBC, white blood cell; NK, natural killer; mDC, classical dendritic cell; pDC, plasmacytoid dendritic cell; Treg, T regulatory cell; Th, T helper cell.

Data are presented as mean ± SD or median (range).

T cells Heterogeneity in Esophageal Achalasia

To better characterize the diversity occurring at the level of the T cell compartment specifically within EA, we developed 2 polychromatic panels capable of simultaneously investigating 10 parameters (Supplementary Table 2). The analysis of T cell subpopulations revealed that the percentage of CD8+ T cells in peripheral leukocyte was higher in patients, though percentage of total T cells was decreased (P < 0.05) (Fig. 3A and 3B). In addition, the percentage of CD4+ T cells (Fig. 3A and 3C) and γδ T cells (data not shown) had no difference. There was also a tendency toward increased numbers of CD8+ T cells but not CD4+ T cells and γδT cells (Fig. 3D and 3E, Table).

Figure 3. The expression of CD8+ T cell and CD4+ T cells in peripheral and lower esophageal sphincter (LES) infiltration. (A) Zebra plot showed frequency of peripheral CD8+ T cells and CD4+ T cells measured by flow cytometry analysis. (B) Scatter plot showed analysis comparing the proportion of CD8+ T cells across the individuals. (C) Scatter plot showed analysis comparing the proportion of CD4+ T cells across the individuals. (D) Immunofluorescence staining of CD8+ T cells and CD4+ T cells in LES. (E) Number of CD8+ T cells per high-powered field (HPF) from LES. (F) Number of CD4+ T cells per HPF from LES. P-values were calculated using t test or Wilcoxon test. WBC, white blood cell. *P < 0.05.

As for CD4 subsets may be closely related with EA,^14,21 we further used chemokines, IFN-γ, IL-4, and IL-17, to distinguish between T helper subtypes in panel 3 (Fig. 4A-C and 4E-H), and defined T regulatory cells (Tregs) as CD3⁺CD8^–CD25^high+CD127^low cells in panel 2 (Fig. 4D and 4I). We did not consider T helper type 9 cell (Th9) and T helper type 22 cell (Th22) as they are generally present in very low percentages in the blood and it was impractical for us. Here, the percentage of circulating Treg from EA patients were conspicuously higher when compared with healthy individuals (Fig. 4B-E and Table, P < 0.05). The percentage of circulating Th1 cells from EA patients were expanded in some EA patients without reaching statistical significance (P = 0.057). Other subsets of CD4+ T cells had similar proportions across the 2 groups. Treg have no significant difference among subgroups (Supplementary Table 4).

Figure 4. The expression of Peripheral CD4+ T cell subsets in control and achalasia group. (A) A CD3⁺CD8^– gate used to identify CD4+ T cells. (B, C) Representative flow cytometric analysis of CD4 subsets, including T helper type 1 (Th1) cells, Th17 cells, pathological Th17 cells, Th2 cells, and T regulatory cells (Treg). (E-I) Scatter plot analysis comparing the proportion of CD4+ T cells across the individuals. Controls are shown in blue, and achalasia in red. P-values were calculated using t test or Wilcoxon test. *P < 0.05.

Cell-type Distribution and Differential Transcriptome Analysis of Peripheral Blood Mononuclear Cells From Patients

PBMCs of 21 patients and 14 controls were selected into RNA-seq analysis. High consistency among individual samples of each group was confirmed by multidimensional scaling analysis (Supplementary Figure). We first compared cell-type distribution among patients and controls by CIBERSORT. Twenty major types of immune cells were estimated in each sample, and recategorized into 11 subsets, including naive B cells, memory B cells, CD8+ T cells, naive CD4+ T cells memory resting CD4+ T cells, memory activated CD4+ T cells, Tregs, resting NK cells, activated NK cells, monocytes, and undefined subset (Fig. 5A and Supplementary Table 5). Accordingly, the proportion of CD8+ T cells was significantly increased in patients (Fig. 5B). Other immune cells were comparable between the 2 groups.

Figure 5. Proportions of the 10 major immune cell types among samples from controls and patients. (A) Fraction of leukocyte type content as inferred by Cell-type Identification by Estimating Relative Subsets of RNA Transcripts (CIBERSORT) between 21 patients and 14 controls. (B) Comparison of leukocyte cell subpopulations among controls and patients. NK cells, natural killer cells.

A total of 170 DEGs were detected, of which 45 genes were upregulated (higher expression in EA) and 125 genes were downregulated (Fig. 6A and 6B). There was dysregulated gene among different types of EA. Top 10 upregulated and downregulated DEGs were illustrated (Supplementary Table 6). To gain further insight into the specific cellular biological processes, DEGs were assessed by functional enrichment analysis both in upregulated and downregulated genes. Gene Ontology (GO) analysis revealed enrichment of 305 Biological Process, 15 Cellular Component, and 22 Molecular Function terms with Q value < 0.05 (Supplementary Table 7). Most significantly enriched components mainly focus on Biological Process terms, and top 5 GO term included myeloid leukocyte migration, synapse pruning, cell junction disassembly, leukocyte migration, and cell chemotaxis (Fig. 6C and 6D). KEGG analysis revealed enrichment of 18 pathway with Q value < 0.05 (Supplementary Table 8). Top 5 KEGG pathways enriched in pertussis, malaria, Chagas disease, complement, and coagulation cascades, as well as cytokine-cytokine receptor interaction. These significantly enriched GO terms and pathways could help us to understand key molecules in the development of EA (Fig. 6E and 6F).

Figure 6. Distribution of differentially expressed genes (DEGs) and functional analysis. (A) The volcano plot of DEGs. Each point represents a gene. Green and red dots represent downregulated and upregulated DEGs, respectively. (B) Heatmap of the DEGs. The horizontal and vertical axes represent the samples (achalasia, red; control, blue) and DEGs (red, upregulated; blue, downregulated), respectively. (C) Histogram of Gene Ontology (GO) classification of DEGs. (D) The chord diagram of GO analysis. (E) The Bubble Chart of DEGs Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment results. (F) Cnetplot in DEGs KEGG pathway enrichment results. BP, biological process; CC, cellular component; MF, molecular function; P. adjust, P. adjust value; TRP, transient receptor potential ion channels.

Totally 160 nodes and 250 interactions were obtained in protein-protein interaction network using the STRING database, with a minimum required interaction score > 0.4 (medium confidence) and the disconnected proteins have not been displayed (Fig. 7A). The top 10 genes evaluated by the 5 calculation methods (maximal clique centrality, maximum neighborhood component, edge percolated component, node connect degree, and Eccentricity) were listed by CytoHubba application in Cytoscape (Fig. 7B and Supplementary Table 9). IL-6, toll like receptor 2, IL-1β, TNF, complement C3, complement C1q B chain (C1QB), and complement C1q A chain (C1QA), as significant hub genes. And only IL-6 and IL-1β were upregulated in patients, while others were downregulated. All these hub genes were enriched in Pertusis, systemic lupus erythematosus, Chagas disease, and Legionellosis (Fig. 7C). However, there was no significant difference regarding hub genes among EA subgroups (Supplementary Table 10).

Figure 7. Identification of Hub Genes and Pathways through differentially expressed genes (DEGs) protein-protein interaction (PPI) Network Analysis. (A) The PPI network based on DEGs between the patients and controls. The green ellipses represented down-regulated DEGs and the red ellipses represented up-regulated DEGs. (B) Venn plot of the overlapping genes for the top 10 genes selected based on the 5 ranking methods, maximal clique centrality (MCC), maximum neighborhood component (MNC), edge percolated component (EPC), node connect degree (Degree), and Eccentricity. (C) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways displayed as an interaction network using Cytoscape plug-in ClueGO.

EA is an incurable disease, and the underlying etiology remains unknown. Based on previous studies, immunity is critical in the loss of neurons in the myenteric plexus of the distal esophagus and LES of EA.²² Though systemic and local immunity dysfunction have been reported to have effect on the development of EA, information on the phenotype and function of peripheral leucocyte is sparse. Here, we found a clear distinction in quantity and expression levels of adaptive and innate immune cells between patients and healthy individuals. We confirmed that patients with EA exhibited a leukocyte profile characterized by proliferating neutrophils, CD8+ cytotoxic T cells and Tregs. While EA has been associated with an increase infiltration of eosinophils in LES, we observed a diminishing of circulating eosinophils. Besides, comparative transcriptomics of PBMC gives insights into disruption of immune cells migration and immune dysfunction. The compromised cytokines and chemokines expression may affect host immunity homeostasis.

Neutrophils are the most abundant cells in innate immunity. The pathogenic roles of neutrophils are also believed to be involved in systemic autoimmune diseases and neurodegenerative disease, including systemic lupus erythematosus, rheumatoid arthritis, psoriasis and multiple sclerosis, with increased quantity, enhanced neutrophil activation, reduced apoptosis, as well as higher levels of neutrophil extracellular traps in serum.^23-28 Besides, neutrophils also release several pro-inflammatory cytokines and activate the complement system.²⁹ Here, for the first time, our study inferred the tight relationship with neutrophils in EA. Perhaps further analysis of neutrophil function and neutrophil extracellular trap formation in EA need to be investigated.

Our results are consistent with previous study that eosinophils are reported to increasingly infiltrate LES, but also demonstrated a decrease in periphery. Besides, a reduction in the number of peripheral eosinophils with high Eckardt score. In a healthy gastrointestinal tract, eosinophils can be found from the stomach to the colon, but no eosinophils are present in the normal esophagus.³⁰ Esophagus eosinophilia usually results from increased recruitment of activated eosinophils from the bloodstream. Activated eosinophils have the capability to cause tissue damage and dysfunction by releasing cytotoxic eosinophil proteins, cytokines, chemokines, and lipid mediators.³¹ We further validated that eosinophilic infiltration in LES in this study, according to previously published studies,^32-34 which infer eosinophilic infiltration in LES form periphery may result in the loss of myenteric ganglion cells in EA.

Apart from innate immune dysfunction, our current study further affirms the relationship between adaptive immunity and EA, for T cell dysfunction.^35,36 The decrease in peripheral T cell count was more likely to be a response to the increased neutrophil concentration in EA. Accumulation of CD8+ T cells in systemic inflammation and tissue indirectly implicates to kill surrounding cells and lead to hypersecretion of inflammatory cytokines. CD8+ T cells express granzyme B, which suggests that they have cytotoxic activity, and also produce the pro-inflammatory cytokines in the periphery and in the CNS.^37,38 PBMCs RNA-seq characterized a host immune response in EA patients. Almost DEGs fall into categories including humoral immune response, lymphocyte mediated immunity, and complement activation. Complement, one of the crucial components of the innate immune system, acting like a bridge of connection for adaptive response, established an imbalance may play a crucial role in promoting inflammation in EA by the clues from hub gene complement C3, C1QB, and C1QA. Cytokines and chemokines play important roles in regulating immune cells for their recruitment, migration, proliferation, and differentiation and executing tasks. We have shown that patients with EA exhibited a pro-inflammatory profile with increased cytokines and chemokines.³⁹ Here, elevated IL-6, and IL-1β levels may be cellular communicators in EA.

It is crucial to figure out the role of immune cells in the pathogenesis in EA. Regarding study limitations, this cross-sectional study design restricts causal conclusions. It is not known whether the changes in many of the immunological indicators observed in this study played an active role or just occurred passively, it was a snapshot of the current state rather than an investigation into the pathophysiology. What is more, though the tissue biopsy was collected at LES high-pressure zone muscle tissue, which can be considered a representative of the typical reflection of achalasia, biopsy specimens may not have accurately represented the full scope of the esophageal pathology for each patient. As the number of patients was limited, it is difficult to analyze the possible relationship between subtypes of EA and the degree and distribution of immune infiltration.

In conclusion, patients with EA had dysfunction both in innate and adaptive immunity. Our study also reveals that 170 DEGs dysregulated in EA. Complement and cytokines may play a critical role in development of EA. The results could also help us get a better understanding of peripheral immunity dysfunction and the key immunological factors in EA. Further studies of model organisms to identify the biological functions of these immune cells and cytokines are required to elucidate these complexities.

This study was supported by grants from the National Natural Science Foundation of China (82170555, 82000507, 81873552, and 81670483), Shanghai Rising-Star Program (19QA1401900), Major Project of Shanghai Municipal Science and Technology Committee (19441905200), Yangfan Program of Shanghai Municipal Science and Technology Committee (S2020-016), and Youth Foundation of Zhongshan Hospital, Fudan University (2020ZSQN16).

Note: To access the supplementary tables and figure mentioned in this article, visit the online version of Journal of Neurogastroenterology and Motility at http://www.jnmjournal.org/, and at https://doi.org/10.5056/jnm21246.

None.

Lu Yao, Zuqiang Liu, and Weifeng Chen contributed equally to this article; Quanlin Li and Pinghong Zhou designed and conducted the study; Lu Yao, Zuqiang Liu, and Weifeng Chen performed the experiments and analysis; Lu Yao, Zuqiang Liu, and Quanlin Li wrote and revised the manuscript; and Jiaqi Xu, Xiaoyue Xu, Jiaxin Xu, Liyun Ma, and Xiaoqing Li contributed to human sample and data collection. All authors made the final approval of manuscript.