Abstract

Background

   Ulcerative colitis (UC) is a chronic autoimmune-related disease that
   causes inflammation of the intestine. Ankylosing spondylitis (AS) is a
   common extraintestinal complication of UC involving the sacroiliac
   joint. However, the pathogenesis of AS secondary to UC has not been
   studied. This study aimed to investigate the shared pathways and
   potential common biomarkers of UC and AS.

Methods

   Microarray data downloaded from the Gene Expression Omnibus (GEO)
   database were used to screen differentially expressed genes (DEGs) in
   the UC and AS datasets. Weighted gene co-expression network analysis
   (WGCNA) was performed to identify co-expression modules related to UC
   and AS. Shared genes were then further analyzed for functional pathway
   enrichment. Next, the optimal common biomarker was selected using
   SVM-RFF and further validated using two independent GEO datasets.
   Finally, immune infiltration analysis was used to investigate the
   correlation of immune cell infiltration with common biomarkers in UC
   and AS.

Results

   A total of 4428 and 2438 DEGs in UC and AS, respectively, were
   screened. Four modules were identified as significant for UC and AS
   using WGCNA. A total of 25 genes overlapped with the strongest positive
   and negative modules of UC and AS. KEGG analysis showed these genes may
   be involved in the mitogen-activated protein kinase (MAPK) signaling
   pathway. GO analysis indicated that these genes were significantly
   enriched for RNA localization. PAN3 was selected as the optimal common
   biomarker for UC and AS. Immune infiltration analysis showed that the
   expression of PAN3 was correlated with changes in immune cells.

Conclusion

   This study first explored the common pathways and genetic diagnostic
   markers involved in UC and AS using bioinformatic analysis. Results
   suggest that the MAPK signaling pathway may be associated with both
   pathogeneses and that PAN3 may be a potential diagnostic marker for
   patients with UC complicated by AS.

   Keywords: ulcerative colitis, ankylosing spondylitis, weighted gene
   co-expression network analysis, mitogen-activated protein kinase, PAN3

1. Introduction

   Ulcerative colitis (UC) is a chronic autoimmune-related inflammatory
   disease with symptoms occurring both inside and outside the
   gastrointestinal tract ([35]1). The incidence of UC is increasing
   annually with changes in economic growth, diet, and lifestyle habits
   ([36]2). Clinically, UC is characterized by recurrent diarrhea, bloody
   purulent stools, and abdominal pain ([37]3). Some patients also have
   extraintestinal manifestations (EIM), with primary involvement of the
   musculoskeletal system, skin, hepatobiliary tract, and eyes, which
   seriously affects the patient’s quality of life. However, the
   pathogenesis of EIM is poorly understood. It has been suggested that
   EIM may be linked to intestinal immune response dysfunction ([38]4).
   Additionally, it is thought EIM may be an independent inflammatory
   event triggered by the persistence of inflammatory bowel disease (IBD)
   ([39]5). Moreover, environmental, and genetic factors may also
   contribute to the progression ([40]5). Ankylosing spondylitis (AS) is a
   common autoimmune disease characterized by inflammation of the
   sacroiliac joints and spinal attachments ([41]6). A German study
   suggested that 5–10% of AS is associated with IBD, and researchers
   found a greater proportion of AS patients had subclinical bowel
   inflammation ([42]7). Consistently, inflammatory arthropathy is the
   most common extraintestinal complication in IBD patients, with a
   prevalence of 20%-50% for axial inflammation ([43]8), and 5%-20% for
   peripheral arthritis ([44]4). The prevalence of AS in patients with IBD
   is estimated to be 3% ([45]9). Human leukocyte antigen B27 (HLA-B27)
   positivity is associated with idiopathic AS in over 90% of cases;
   however, HLA-B27 is less specific and sensitive in spondylitis
   complicated by IBD ([46]10). Moreover, the progression of AS, from
   inflammatory back pain to radiographic sacroiliitis often takes years
   to develop ([47]11). Therefore, finding more specific diagnostic
   markers for IBD patients suffering from back pain, and carrying out
   earlier treatment interventions for IBD patients with AS is of great
   clinical significance. In this study, published gene expression data in
   the GEO database was reviewed. Shared gene pathways and diagnostic
   markers between UC and AS were then explored using a systems biology
   approach. Using this data, we aimed to determine new potential
   diagnostic and treatment strategies for patients with AS secondary to
   UC.

2. Methods

2.1. Datasets and data pre-processing

   We obtained raw gene expression profile data and clinical information
   from the GEO database. We searched the GEO database for RNA-seq
   profiles using the keywords “ulcerative colitis” and “ankylosing
   spondylitis”. The following filtering criteria were used: The tissue
   used for sequencing should be peripheral blood mononuclear cells
   (PBMCS), and the number of samples per group should not be less than 10
   to ensure the accuracy of WGCNA.

   Finally, The GEO datasets numbered [48]GSE73754, [49]GSE126124,
   [50]GSE25101, and [51]GSE3365 were obtained. The [52]GSE73754 dataset
   was used on the [53]GPL10558 platform. The dataset contained 72
   samples, including 52 peripheral blood samples from AS patients and 20
   from healthy controls. The [54]GSE126124 dataset was used on the
   [55]GPL6244 platform. This dataset contained 50 samples, including 18
   peripheral blood samples from UC patients and 32 peripheral blood
   samples from healthy controls. Additionally, the [56]GSE25101 dataset
   based on [57]GPL6947 was regarded as an external validation set from
   the GEO database with 16 cases of AS patients as the experimental group
   and 16 cases of normal samples as the control group. Similarly, the
   [58]GSE3365 dataset was also regarded AS the external validation set of
   the GEO database based on the [59]GPL96 platform, with 26 UC patients
   as the experimental group and 42 normal samples as the control group
   ([60]Figure S1).

2.2. Differential genetic screening

   Differentially expressed genes (DEGs) were screened in [61]GSE73754 and
   [62]GSE126124 data sets. Limma R package was used to screen DEGs, and
   p. adj. value < 0.05 was selected as the cut-off standard.

2.3. Construction and module analysis of weighted gene co-expression network
(WGCNA)

   The WGCNA (Weighted Gene Co-expression Network Analysis) software
   package R.4.0.3 is used to perform WGCNA analysis. WGCNA R package was
   used to construct co-expression networks with corresponding clinical
   characteristics for DEGs of UC and AS. Prior to analysis, hierarchical
   cluster analysis was performed using the Hclust function in R language
   to exclude outlier samples. Then, according to the standard of
   scale-free network, the “pickSoftThreshold” function in the WGCNA
   software package is used to select the appropriate soft power β
   (ranging from 1 to 20) for automatic network construction. The results
   are clustered by topological overlap matrix analysis, which contains
   module assignments labeled by color and module feature (ME). In
   addition, Pearson’s correlation test was used to calculate the
   correlation between ME and clinical features. The modules with a
   |ME|> 0.3 and a p-value < 0.05 were considered in the interaction with
   clinical features is of great significance.

2.4. Enrichment analysis

   Gene Ontology (GO) enrichment analysis is a commonly used
   bioinformatics method to search for comprehensive information on
   large-scale gene data. In addition, KEGG pathway enrichment analysis
   has been widely used to understand biological mechanisms and functions.
   The GOplot program package was used to visualize the GO and KEGG
   pathways. Finally, the clusterprofile package and GSVA package are used
   to further explore important signaling pathways. In addition, the gene
   sets and gene expression matrices of UC and AS were analyzed by GSEA to
   explore the possible regulatory pathways involved.

2.5. Identification and validation of potential shared diagnostic markers

   Support vector machine recursive feature elimination is a machine
   learning method based on support vector machine, which finds the
   optimal core gene by deleting the feature vectors generated by support
   vector machine. In the intersection genes of DEGs and WCGNA mentioned
   above, SVM method was used to screen the core markers. Finally, the
   area under the ROC curve (AUC) was used to evaluate the diagnostic
   efficacy of core markers in the modeling set and validation set of UC
   and AS.

2.6. Extraction of peripheral blood mononuclear cell(PBMC)

   We used purple anticoagulation tubes to collect 6 ml of blood samples
   from 3 UC patients, 3 AS patients, and 3 normal controls. I the
   ultra-clean table, human peripheral blood, and phosphate buffer saline
   (PBS) evenly were mixed in a ratio of 1:1 (6 ml: 6 ml). 3 ml of
   lymphocyte separation medium was added (BL590 Biosharp) into each 15 ml
   centrifuge tube and 6 ml of diluted blood was gently added on top of
   lymphocyte isolation medium. Then we placed the centrifuge tubes in the
   centrifuge for centrifugation for 20-25 min (400 g, 25 °C, ACCECL 2,
   DECELE 2).The cloud layer at the middle and upper interface was
   carefully collected and transferred to a new 15 mL tube. Finally, we
   washed PBMC with PBS, and discarded supernatant after centrifugation.
   Each patient signed the informed consent form in advance and approved
   by the Ethics Committee of the First People’s Hospital of Huai’an.

2.7. Quantitative real-time PCR (qRT-PCR)

   Total RNA was extracted from Peripheral blood mononuclear cell (PBMC)
   by RNA simple total RNA kit (TIANGEN, DP419). 2 μL of total RNA was
   pipetted for measurement by Nanodrop. RNA was then transcribed into
   cDNA using a PrimeScript™ RT Master Mix kit (TAKARA, Ref. RR036A).
   qRT-PCR was performed using a LightCycler Instrument. PCR system using
   SYBR○Rpremix Ex Taq™ (TAKARA, RR420A) following manufacturer’s
   instructions. The relative mRNA levels were normalized to those of
   GAPDH using the 2-ΔΔCt method. (PAN-Fwd: 5’-GAAACTGTTGGTGGGACGACTTA-3’;
   PAN-Rev: 5’-TGAGGTGCAGTTGGAGGATAAAT-3’; GAPDH-Fwd:
   5’-ATGGGGAAGGTGAAGGTCG-3’; GAPDH-Rev: 5’-CTCCACGACGTACTCAGCG-3’).

2.8. Immune analysis algorithm

   The CIBERSORT algorithm calculates the proportion of different immune
   cell types based on the expression levels of immune cell-related genes.
   The output results of 22 infiltrating immune cells were integrated, and
   the component matrix of immune cells was generated for analysis. Using
   the results of immune infiltration analysis of the above shared
   markers, the correlation between core markers and the expression of
   immune infiltrating cells was analyzed by non-parametric correlations
   (spearman) method.

3. Results

3.1. Differential genetic screening

   A total of 4428 differentially expressed genes (DEGs) were screened
   between UC patients and normal controls, with 1912 upregulated and 2516
   downregulated DEGs shown in volcano plots ([63]Figure 1A). Of these
   DEGs, the top 20 differentially expressed genes are displayed in
   heatmaps ([64]Figure 1B). Additionally, 2438 DEGs were identified
   between AS and normal controls, including 1037 upregulated genes and
   1401 downregulated genes ([65]Figure 1C). The top 20 genes are listed
   in [66]Figure 1D. KEGG pathway enrichment analysis was used to analyze
   the DEGs. Interestingly, KEGG analysis showed that both AS and UC were
   enriched in the mitogen-activated protein kinase (MAPK) pathway
   ([67]Figures 1E, F).

Figure 1.

   [68]Figure 1
   [69]Open in a new tab

   (A) The volcano plots of DEGs in [70]GSE126124(n=50, p < 0.05). (B) The
   heatmap plots of DEGs in [71]GSE126124. (C) The volcano plots of DEGs
   in [72]GSE73754 (n=68, p < 0.05). (D) The heatmap plots of DEGs in
   [73]GSE73754. (E) The GO term analysis of DEGs in [74]GSE126124. (F)
   The GO term analysis of DEGs in [75]GSE73754.

3.2. Construction and module analysis of weighted gene co-expression network
(WGCNA)

   WGCNA was used to identify clusters of co-expressed genes whose
   expression differed between UC and AS, and the correlation of combined
   modules with disease characterization was calculated. It was determined
   that a soft threshold power of β=35 was set in the UC model, and the
   soft threshold power in the AS modeling set was 15 ([76]Figures 2A, C).
   After merging similar gene modules, seven modules were identified in
   the UC model set and 14 modules in the AS model set. As shown in
   [77]Figure 2B, the brown module had the strongest positive correlation
   with the occurrence of UC (r=0.47) and the green module had the
   strongest negative correlation with the occurrence of UC (r=-0.54).
   Further, in the AS modeling set, the yellow module in the AS modeling
   set had the strongest positive correlation with AS occurrence (r=0.46),
   and the brown module had the strongest negative correlation with AS
   occurrence (r=-0.55) ([78]Figure 2D).

Figure 2.

   [79]Figure 2
   [80]Open in a new tab

   (A) The cluster dendrogram of co-expression in UC. (B) Correlation
   between modules and clinical traits in UC. (C) The cluster dendrogram
   of co-expression in AS. (D) Correlation between modules and clinical
   traits in AS.

3.3. Identification of shared genes and shared pathways

   A total of 25 genes overlapped in the strongest positive and negative
   modules of UC and AS ([81]Figure 3A). These genes may be related to the
   pathogenesis of UC and AS. Enrichment analysis was performed again for
   these 25 genes. KEGG analysis showed that these genes may be involved
   in the MAPK signaling pathway and other pathways ([82]Figure 3B). In
   addition, GO analysis showed that these genes were significantly
   enriched in RNA localization ([83]Figure 3C). Subsequently, we
   performed GSEA on AS and UC samples and found that inflammatory
   responses were involved in a common pathogenic process ([84]Figures 3D,
   E). Therefore, we propose a bold conjecture that the occurrence of AS
   and UC may be jointly driven by the MAPK pathway and immune response.

Figure 3.

   [85]Figure 3
   [86]Open in a new tab

   (A) The Venn diagram shows an overlap of 25 genes in modules between UC
   and AS. (B) The GO term analysis of 25 shared genes between UC and AS
   screened by WGCNA. (C) The KEGG analysis of 25 shared genes between UC
   and AS screened by WGCNA. (D) Single-gene GSEA analyses result for DEGs
   of UC. (E) Single-gene GSEA analyses result for DEGs of AS.

3.4. Identification and validation of potential shared diagnostic markers

   SVM-REF is a machine learning method based on a support vector machine.
   The optimal core gene is found by deleting the feature vector generated
   by the support vector machine. Based on the 25 shared genes, four
   possible diagnostic markers were identified in [87]GSE126124
   ([88]Figure 4A), and eight possible diagnostic markers were identified
   in [89]GSE73754 ([90]Figure 4B). Ultimately, we determined that PAN3
   may be the best diagnostic biomarker for the development of UC and AS
   ([91]Figure 4C). We further identified the diagnostic power of the
   shared markers based on ROC analysis. In [92]GSE126124, we obtained an
   AUC of 0.903 ([93]Figure 4D). Additionally, the same ROC analysis was
   performed again for [94]GSE73754, and the result showed that the AUC
   was 0.816 ([95]Figure 4E). Interestingly, differential expression
   analysis showed that PAN3 was significantly overexpressed in AS, but
   expression significantly decreased in UC ([96]Figure 4F).

Figure 4.

   Figure 4
   [97]Open in a new tab

   (A) SVM-RFE to screen diagnostic markers in [98]GSE126124. (B) SVM-RFE
   to screen diagnostic markers in [99]GSE73754. (C) The Venn diagram
   shows the diagnostic biomarkers between UC and AS. (D) ROC curve
   evaluating the PAN3 diagnostic efficacy in [100]GSE126124. (E) ROC
   curve evaluating the PAN3 diagnostic efficacy in [101]GSE73754. (F)
   PAN3 showed significant differences in [102]GSE126124 and [103]GSE73754
   (P<0.05).

3.5. The confirmation of the diagnostic marker gene

   We verified the diagnostic efficacy of PAN3 in an independent UC
   dataset [104]GSE3365. The differential expression analysis showed a
   significant difference in PAN3, similar to the results reported above
   ([105]Figure 5A), AUC= 0.701 ([106]Figure 5B). In another independent
   AS validation set [107]GSE25101, the expression levels of PAN3 in
   different samples were significantly different ([108]Figure 5C), and
   the AUC was 0.625 ([109]Figure 5D). Additionally, we collected fresh
   full blood samples from nine patients (3 UC, 3 AS, 3 Controls),
   extracted PBMC and performed quantitative qPCR analysis to further
   verify the differential expression of PAN3 in the patient samples. The
   results showed that PAN3 expression was decreased in UC patients and
   increased in AS patients compared to normal subjects ([110]Figures 5E,
   F), suggesting that PAN3 has the potential to be a genetic diagnostic
   marker for both diseases.

Figure 5.

   [111]Figure 5
   [112]Open in a new tab

   (A) PAN3 showed significant differences in [113]GSE3365. (B) ROC curve
   evaluating the PAN3 diagnostic efficacy in [114]GSE3365. (C) PAN3
   showed significant differences in [115]GSE25101. (D) ROC curve
   evaluating the PAN3 diagnostic efficacy in [116]GSE25101. (E–F)
   Real-time PCR analysis for the mRNA expression levels of PAN3 in PBMC
   from patients and healthy controls. **P < 0.01; ***P < 0.001.

3.6. Immune infiltration analysis of shared markers

   Considering the important roles of immune and inflammatory responses in
   the development of UC and joint involvement of AS, we used the
   CIBERSORT algorithm in enrichment analysis. This enabled analysis of
   the content of immune cells in different samples. The histograms showed
   that the proportion of B cell populations in the AS samples was
   significantly different from that in the UC samples ([117]Figures 6A,
   C). In addition, compared with normal tissues, AS samples had higher
   levels of CD8 T cells, CD4 naive T cells, CD4 memory activated T cells,
   regulatory T cells (Tregs), gamma delta T cells, resting NK cells, and
   neutrophils ([118]Figure 6B). However, comparison of UC and normal
   tissues showed differences in the contents of plasma cells including
   CD4 naïve T cells, CD4 memory T cells, resting monocytes, and
   macrophages (M0) ([119]Figure 6D). Interestingly, there were
   similarities in the differences in T cell subsets between the two
   diseases.

Figure 6.

   [120]Figure 6
   [121]Open in a new tab

   (A) The Bar plot of immune cell infiltration between AS and normal. (B)
   The Violin plot of the 22 types of immune cells proportions between AS
   and normal. (A, C) The Bar plot of immune cell infiltration between UC
   and normal. (D) The Violin plot of the 22 types of immune cells
   proportions between UC and normal.

3.7. Correlation between shared markers and immune cell content

   The correlation between shared markers and immune cell content was also
   explored. In AS samples, PAN3 expression was significantly correlated
   with naive B cells, CD8 T cells, CD4 memory T cells, gamma delta T
   cells, resting NK cells, M0s, polarizing macrophages (M2), and
   neutrophils ([122]Figure 7A). In UC samples, PAN3 was significantly
   associated with plasma cells, CD4 memory resting T cells, resting NK
   cells, monocytes, macrophages, eosinophils, and neutrophils
   ([123]Figure 7B).

Figure 7.

   [124]Figure 7
   [125]Open in a new tab

   (A) Correlation analysis between PAN3 and infiltrating immune cells in
   AS samples where red represented the positive correlation with a
   significant difference. (B) Correlation analysis between PAN3 and
   infiltrating immune cells in UC samples where red represented the
   positive correlation with a significant difference.

4. Discussion

   Existing studies have shown a potential link between gut and joint
   inflammation ([126]12–[127]14). The similar pathogenesis of
   inflammatory bowel disease and spondylitis (SpA) has led researchers to
   propose the theory of the gut-joint axis ([128]15). Currently, a causal
   relationship between SpA and IBD cannot be ascertained. Therefore,
   identifying common features of these two diseases allows exploration of
   the causal relationship between them. In 1973, investigators found that
   HLA-B27 was highly associated with AS contributing to approximately
   20.1% of AS heritability ([129]16). However, HLA-B27 is less specific
   in AS associated with IBD than in idiopathic AS ([130]7). In this
   study, we explored the common disease pathways and diagnostic markers
   involved in UC and AS using bioinformatic analysis. Based on the
   functional analysis of DEGs screened from different datasets, we found
   that the occurrence of AS and UC is closely related to the MAPK
   pathway. Gene clustering analysis was used to screen disease-related
   modules and shared genes, and functional analysis confirmed the
   important role of the MAPK signaling pathway. We speculated that the
   MAPK signaling pathway may be involved in the co-pathogenesis of AS and
   UC. In addition, we screened the optimal core genes based on the
   SVM-RFE algorithm, and identified PAN3 as a possible diagnostic marker
   of UC and AS.

   The MAPK signaling pathway is a classic pathway involved in cell
   growth, differentiation, stress, and inflammatory responses. The MAPK
   family members can be divided into two groups: classical MAPKs,
   including ERK1/2, P38, JNK, and ERK5 ([131]17), and atypical MAPKs,
   consisting of ERK3, ERK4, ERK7, and NLK ([132]18), with independent or
   overlapping effects. Many preclinical studies have confirmed that MAPK
   inhibitors have significant efficacy in experimental colitis animal
   models ([133]19–[134]21). Moreover, these results suggest the MAPK
   signaling pathway may play an important role in IBD. In addition, it
   has been shown that levels of all three phosphorylated MAPKs are
   significantly elevated in the inflamed intestinal mucosa of IBD
   patients compared to normal controls, and non-inflamed sites in IBD
   patients ([135]22). Exploration of potential therapeutic agents for UC,
   showed that many drugs exert their effects by inhibiting MAPK-related
   pathways. In a study by Zhou et al., mTOR-dependent autophagy was found
   to be involved in the regulation of active UC through upstream
   TLR4-MyD88-MAPK signaling and the downstream NF-kB pathway ([136]23).
   This demonstrates a close relationship between the MAPK signaling
   pathway and the mTOR-autophagy axis, providing mechanistic evidence for
   mTOR inhibitors as candidate therapies for IBD. Wang et al. (2019) used
   methane-rich saline to protect against acetic acid-induced UC by
   inhibiting the TLR4/NF-KB/MAPK signaling pathway and promoting the
   IL-10/JAK1/STAT3-mediated anti-inflammatory response ([137]24).

   The MAPK signaling pathway is also strongly associated with the
   development of AS. A study has shown that simulated inflammatory
   environments can activate MAPK and TLR signaling pathways in
   mesenchymal stem cells (MSCs) derived from AS patients, causing
   upregulation of inflammatory gene expression ([138]25). Moreover, in a
   study by Yong et al. ([139]26) MAPK7 and NDUFS4 were shown to play an
   important role in AS through the cAMP-mediated signaling pathways; thus
   these may be potential targets for indomethacin treatment for AS
   ([140]26). This study also identified MAPK as a potential common
   pathway involved in the pathogenesis of both AS and UC. Thus, further
   investigation of the pathogenesis of extra-intestinal joint
   inflammatory complications in patients with IBD should be performed.

   In this study, PAN3 was found to be a common diagnostic marker involved
   in the development of UC and AS, with PAN3 expression being
   significantly high in patients with AS; though, significantly reduced
   in patients with UC. Pan3 and Pan2 form a protein complex involved in
   the deadenylation of the 3’PolyA tail of eukaryotic mRNA to regulate
   mRNA stability. Two Pan3 cells comprise an asymmetric dimer, providing
   a scaffold for Pan2. The Pan2 subunit has nucleic acid exonuclease
   activity, which is necessary for deadenylation; however, effective
   deadenylation requires the participation of Pan3 ([141]27). However,
   the mechanism of Pan3 remains unknown. A study published in 2019 showed
   high expression of immune cell-associated circular RNA circPan3 in mice
   and human Lgr5+ intestinal stem cells (ISC). Circular RNA (circRNA)
   circPan3 is a transcript derived from the PAN3 gene. CircPan3 in
   Lgr5-GFR^+ ISCs promotes the self-renewal ability of ISCs through
   cryptic ILC2-mediated IL-13-IL-13R signaling. The circRNA circPan3
   plays an important role in the self-renewal of ISC ([142]28). Though,
   whether PAN3 plays a role in patients with UC complicated by AS through
   mediating effects on intestinal stem cells needs to be further
   explored.

   Studies have shown an immune response and inflammatory activity in both
   UC and AS diseases with multiple immune cell types ([143]29, [144]30).
   In our study, different proportions of B cell populations were involved
   in the AS and UC samples calculated by CIBERSORT. B cells play an
   important role in intestinal homeostasis in UC. The IgA-dominated
   B-cell response is an essential part of the intestinal mucosa that
   maintains homeostasis and defends against pathogens. It has been
   demonstrated that in patients with UC, the number of B cells is reduced
   and the functional activity of B (Breg) cells is decreased ([145]31).
   Targeted therapies targeting the B-cell compartment have been suggested
   to be potentially promising in the treatment of IBD ([146]32). The role
   of B cells in the pathogenesis of AS has rarely been studied, but there
   is growing evidence for the involvement of B cells in the pathogenesis
   of AS ([147]33). Significantly elevated B cell counts were found in
   tissue biopsies from patients with AS with persistent inflammatory
   activity ([148]34). In addition, large ectopic infiltration of CD20^+ B
   cells was observed in the bone marrow and fibrous tissue of patients
   with AS. This evidence suggests the involvement of active B cells in AS
   pathogenesis. A clinical trial showed that the removal of B cells
   through rituximab therapy (anti-CD20) could benefit some AS patients
   who had not received TNF-α therapy ([149]35). Targeted B-cell therapy
   holds promise as a new research direction for the treatment of patients
   with UC and AS. In addition to altered B cell content, many
   T-lymphocyte immune functions were altered in patients with UC and AS
   in this study. An imbalance of CD4^+ T cell subsets is strongly
   associated with the development of UC ([150]36). Naive CD4^+ T cells
   interacting with dendritic cells carrying homologous antigens can
   differentiate into different effector cell subtypes, including Treg,
   Th1, Th2, Th9, Th17, and Th22, each of which produces characteristic
   cytokines to perform their corresponding biological functions
   ([151]37). A meta-analysis of peripheral blood lymphocytes from
   patients with AS by Liu et al., showed no significant difference in the
   percentage of T cells, NK cells, and NKT cells between AS patients and
   healthy controls. However, there was a significant increase in B cells
   ([152]38), which was consistent with the results of our partial
   findings.

   Interestingly, we found that monocytes and neutrophils showed similar
   trends in UC and AS. In a study by Furukawa et al. (2021), peripheral
   blood monocyte counts were shown to be independently and negatively
   associated with clinical remission and mucosal healing (MH) in Japanese
   UC patients. Peripheral blood mononuclear cell counts could also be
   used as complementary blood markers for MH in UC patients with low
   C-reactive protein (CRP) levels ([153]39). Monocytes have also been
   shown to play a key role in the pathogenesis of SpA. Moreover, monocyte
   subpopulations were found to be positively correlated with AS disease
   activity parameters and C-reactive protein levels, and monocytes in AS
   patients significantly increased phagocytic activity than controls
   ([154]40). Recent evidence has shown that NETs are involved in the
   pathogenesis of IBD. NETs are the first immune cells to infiltrate the
   intestinal barrier, and extensive infiltration of NETs is positively
   correlated with active UC ([155]41). Jiang et al. reported that the
   upregulation of neutrophils may be a key factor in the progression of
   AS ([156]42). Interestingly, most of the evidence accumulated thus far
   points to adaptive immunity being related to AS pathogenesis. However,
   we have provided evidence to further explore the involvement of innate
   immunity in the pathogenesis of AS, through exploring the common
   pathways and diagnostic markers of UC and AS. The question of whether
   AS is predominantly autoinflammatory or autoimmune is still being
   explored by researchers ([157]43). However, it is indisputable that our
   study has some limitations, as only lymphocytes in peripheral blood
   were studied. As such, we cannot provide a comprehensive response to
   the inflammatory status of UC and AS.

5. Conclusion

   This is the first study to use bioinformatic analysis to explore the
   common pathways and genetic diagnostic markers involved in UC and AS
   ([158]Figure 8). Results suggest that the MAPK signaling pathway may be
   associated with the pathogenesis of both AS and UC. Moreover, PAN3 may
   be a potential diagnostic marker for patients with UC complicated by
   AS. Additionally, immune infiltration correlation analysis suggests the
   pathogenesis of UC and AS may be closely related to innate immune
   imbalances. This study provides a new perspective for investigating the
   possible mechanism of AS secondary to UC. In future studies, we will
   further investigate the mechanism of the MPAK pathway and PAN3
   expression changes in conjunction with relevant in vitro and Vivo
   experiments.

Figure 8.

   [159]Figure 8
   [160]Open in a new tab

   The model diagram of the shared pathway and common biomarker analysis
   in UC and AS.

Data availability statement

   The datasets presented in this study can be found in online
   repositories. The names of the repository/repositories and accession
   number(s) can be found in the article/[161]Supplementary Material.

Ethics statement

   The studies involving human participants were reviewed and approved by
   The Ethics Committee of the Affiliated Huai’an No.1 People’s Hospital
   with Nanjing Medical University. The patients/participants provided
   their written informed consent to participate in this study. Written
   informed consent was obtained from the individual(s) for the
   publication of any potentially identifiable images or data included in
   this article.

Author contributions

   MZ, JZ, and HW designed the experiment. LH and JW analyzed the data. MZ
   and JZ wrote and prepared the manuscript. XY and XZ reviewed the
   manuscript. All authors contributed to the article and approved the
   submitted version.

Funding Statement

   This study was supported by Huai’an Key Laboratory of Aging and
   Geriatric Syndrome Research (HAP202105).

Conflict of interest

   The authors declare that the research was conducted in the absence of
   any commercial or financial relationships that could be construed as a
   potential conflict of interest.

Publisher’s note

   All claims expressed in this article are solely those of the authors
   and do not necessarily represent those of their affiliated
   organizations, or those of the publisher, the editors and the
   reviewers. Any product that may be evaluated in this article, or claim
   that may be made by its manufacturer, is not guaranteed or endorsed by
   the publisher.

Supplementary material

   The Supplementary Material for this article can be found online at:
   [162]https://www.frontiersin.org/articles/10.3389/fimmu.2023.1089622/fu
   ll#supplementary-material
   Supplementary Figure 1

   The Venn diagram shows 7494 common genes in four different databases.
   [163]Click here for additional data file.^ (2MB, jpg)

References