Next Article in Journal
Baricitinib Liposomes as a New Approach for the Treatment of Sjögren’s Syndrome
Previous Article in Journal
Formulation and Evaluation of the In Vitro Performance of Topical Dermatological Products Containing Diclofenac Sodium
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Pharmaceutics
Volume 14
Issue 9
10.3390/pharmaceutics14091893
pharmaceutics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Identification of Novel Loci Involved in Adalimumab Response in Crohn’s Disease Patients Using Integration of Genome Profiling and Isoform-Level Immune-Cell Deconvoluted Transcriptome Profiling of Colon Tissue

by
Mario Gorenjak
1,*,
Gregor Jezernik
1,
Martina Krušič
1,
Pavel Skok
2 and
Uroš Potočnik
1,3,4,*
1
Center for Human Molecular Genetics and Pharmacogenomics, Faculty of Medicine, University of Maribor, Taborska ulica 8, 2000 Maribor, Slovenia
2
Department of Gastroenterology, University Clinical Centre Maribor, 2000 Maribor, Slovenia
3
Faculty of Chemistry and Chemical Engineering, University of Maribor, Smetanova ulica 17, 2000 Maribor, Slovenia
4
Department for Science and Research, University Clinical Centre Maribor, Ljubljanska ulica 5, 2000 Maribor, Slovenia
*
Authors to whom correspondence should be addressed.
Pharmaceutics 2022, 14(9), 1893; https://doi.org/10.3390/pharmaceutics14091893
Submission received: 10 August 2022 / Revised: 25 August 2022 / Accepted: 2 September 2022 / Published: 7 September 2022
(This article belongs to the Section Biologics and Biosimilars)
Browse Figures
Versions Notes

Abstract

:
Crohn’s disease is a consequence of dysregulated inflammatory response to the host’s microbiota. Although anti-TNF treatment improves the quality of the patient’s life, a large proportion of patients lose response to the treatment. The past decade of research has led to a continuum of studies showcasing the heterogeneity of anti-TNF response; thus, the aim of the present study was to dissect transcriptome-wide findings to transcript isoform specific levels and combine the analyses with refined information of immune cell landscapes in colon tissue, and subsequently select promising candidates using gene ontology and genomic integration. We enrolled Slovenian Crohn’s disease patients who were naïve with respect to adalimumab treatment. We performed colon tissue RNA sequencing and peripheral blood mononuclear cell DNA genotyping with a subsequent contemporary integrative approach to combine immune cell deconvoluted isoform transcript specific transcriptome analysis, gene ontology layering and genomic data. We identified nine genes (MACF1, CTSE, HDLBP, HSPA9, HLA-DMB, TAP2, LGMN, ANAPC11, ACP5) with 15 transcripts and 16 variants involved in the adalimumab response. Our study identified loci, some of which were previously shown to contribute to inflammatory bowel disease susceptibility, as novel loci involved in adalimumab response in Crohn’s disease patients.

1. Introduction

A dysregulated inflammatory response to the host’s microbiota is a key etiology of Crohn’s disease and ulcerative colitis, and leads to chronic inflammation [1,2,3]. Crohn’s disease incidence in Europe is as high as 12.7 per 100,000 persons per year, or 322 per 100,000 persons, making this disease far from rare [4]. Established explanatory concepts of Crohn’s disease have shown a greater role for genetic predisposition as opposed to environmental factors (cigarette smoking and antibiotic use) in comparison to ulcerative colitis [5,6]. The basis of the chronic inflammatory process is now assumed to be commensal microbiota and dysbiosis [7,8,9]. In addition to mucosal barrier disorder, various defects of bacterial recognition, autophagy, endoplasmatic reticulum stress and immune cells exert an effect on antimicrobial defense and alter the microbiome [9,10]. Subsequently invading bacteria induce the inflammatory response, which is thought to be dysregulated [11]. With tumor necrosis factor alpha (TNF-α) being a key cytokine of inflammation, anti-TNF therapies such as infliximab and adalimumab have started to dominate the treatment of Crohn’s disease [12,13]. Anti-TNF therapy significantly improves the quality of patients’ lives, but a large proportion of patients do not respond to anti-TNF therapy or lose the response after the initial benefit [12,14,15,16]. In recent years, novel biological agents such as vedolizumab, ustekinumab and risankizumab in the treatment of Crohn’s disease have emerged [17,18,19,20]. Even after the advent of novel biological drugs targeting other cytokines, adalimumab and its biosimilars remain widely used biological therapy in inflammatory bowel disease [21]. Recently it was also shown that treatment choices for Crohn’s disease must be individualized for each patient, starting with either infliximab with azathioprine or adalimumab as a first-line therapy for the induction of clinical remission [22].
Almost a decade of research has led to a continuum of studies showcasing the heterogeneity of the anti-TNF response [23,24,25,26,27,28,29,30,31,32]. However, most genomic markers for anti-TNF response in Crohn’s disease do not reach sufficient thresholds and, therefore, no reproducibility between genetic and expression data exists [33]. Recent approaches have identified new anti-TNF candidates using integrative transcriptomic-genomic analyses while accounting for deconvoluted immune cell composition in bulk RNA samples from peripheral blood monocytes [30]. Moreover, it was recently shown that deconvolution of whole-tissue gene expression data yields refined information concerning the immune cell landscape in inflammatory bowel disease [34]. Nevertheless, to date, genetic studies of anti-TNF research in Crohn’s disease focused merely on genes, with disregard to specific isoforms; therefore, the aim of the present study was to:
  • Perform isoform-level transcriptome profiling of colon tissue.
  • Combine transcriptome profiling with refined information concerning immune cell landscapes in colon tissue.
  • Stringently select promising gene candidates based on gene ontology analysis.
  • Integrate and functionally annotate the findings from genome profiling data.

2. Materials and Methods

2.1. Enrolled Subjects

We enrolled 84 Slovenian patients of Caucasian Central European ethnicity with Crohn’s disease (55 responders and 29 non-responders) who were naïve in respect to treatment with adalimumab and fulfilled the criteria for adalimumab (Humira, Abbott Laboratories, Chicago, IL, USA) therapy initiation. Crohn’s disease diagnosis was made based on a combination of endoscopic, histological, clinical, biochemical, stool and imaging investigations based on ECCO-ESGAR guidelines [35]. Baseline demographics of the enrolled patients is available in previous study [30]. Inclusion criteria consisted of adverse effects to corticosteroids, refractoriness to corticosteroids and previous loss of response to infliximab, with a mandatory minimum eight-week wash-out period [24]. Exclusion criteria were defined as the presence of stenosis, abscesses, total colectomy, history of murine proteins allergy, active tuberculosis or a serious infection in the previous three months, pregnancy, lactation or malignancy, history of Listeria infection, HIV, demyelinating disease, chronic viral hepatitis, uncontrolled diabetes mellitus, unstable ischemic heart disease or congestive heart failure, cerebrovascular accidents, history of drug or alcohol abuse, previous treatment with natalizumab or any other conditions which would put the patient at risk based on clinician’s opinion [36]. Initiation of adalimumab treatment consisted of a loading dose of 160 mg followed by 80 mg after two weeks and then a maintenance dose of 40 mg every other week. If the dosage of adalimumab was stable in the last three months, concomitant treatment with azathioprine, 5-amisalycylates, corticosteroids or antibiotics was allowed. Response to the treatment was measured using an IBD questionnaire (IBDQ) after 12 weeks of initiation of the therapy, and response was defined as an increase of the IBDQ score of >22 points after baseline score or a total score of >170 points [37,38]. Healthy (N = 11) and inflamed (N = 11) colon tissue biopsies were also collected for subsequent genetic analyses from 22 enrolled individuals who were followed up 12, 20 and 30 weeks after initiation of the adalimumab therapy and were consistent responders or non-responders.

2.2. Extraction of Nucleic Acids

DNA was extracted from peripheral blood mononuclear cells using TRI-reagent (Merck, Darmstadt, Germany) according to manufacturer’s instructions. Before RNA was extracted from colon tissue biopsies, the tissue was first homogenized using a Bullet Blender homogenizer (Next Advance Inc., Troy, NY, USA) followed by immediate extraction of RNA using TRI-reagent (Merck) according to manufacturer’s instructions. Concentration, purity and integrity of nucleic acids were estimated using a Synergy 2 spectrophotometer (Biotek, Winooski, VT, USA) and a 2100 Bioanalyzer Instrument (Agilent, Santa Clara, CA, USA) with an RNA 6000 Nanochip.

2.3. Transcript Specific Tissue RNA-Seq Analysis

RNA sequencing was performed on 22 samples with estimated an RIN number >9.0 and a 28S/18S ratio between 1.5 and 2.0. Paired-end RNA sequencing was performed at BGI China (BGI, Shenzhen, China) using a DNBSEQ-G400 apparatus and 2 × 100 bp MGIEasy rRNA depletion kit (MGITech, Shenzen, China) and MGIEasy RNA library prep set (MGITech).
Raw fastq files were first assessed for quality using FastQC software (version 0.11.9, Babraham Bioinformatics, Cambridge, UK) with subsequent trimming of technical sequences using the Trimmomatic tool (version 0.39, USADEL LAB, Aachen, Germany) [39,40]. Transcript specific RNA-seq analysis was performed using a Kallisto pseudoalignment program (Pachter Lab, Berkeley, CA, USA) with GRCh37 reference genome [41]. Estimated counts and tpms were further processed using the R 4.2.1 environment (R Core Team 2020, Vienna, Austria) and using a pipeline described elsewhere [30]. Additional conventional alignment of raw reads to the GRCh37 reference genome was performed to obtain estimated meta-feature raw counts and tpms for deconvolution of immune cells in the bulk RNA from tissue biopsies, as described previously [30,42]. CIBERSORTx (Alizadeh & Newman Lab, Stanford, CA, USA) and an LM22 signature matrix were used to correct for immune cell infiltration in the colon tissue [43].
Transcript specific RNA-seq data analysis was independently performed on healthy tissue samples (four non-responders relative to seven responders) and inflamed tissue samples (seven non-responders relative to four responders). A linear regression model and empirical bayes were fitted and covariate-corrected using four different approaches: (I) corrected for sex, age at diagnosis, corticosteroid use, azathioprine use and aminosalicylates use; (II) corrected for sex, age at diagnosis, corticosteroid use, azathioprine use and deconvolution abundance fraction of T cells; (III) corrected for sex, age at diagnosis, corticosteroid use, azathioprine use and deconvolution abundance fraction of monocytes/macrophages; (IV) corrected for sex, age at diagnosis, corticosteroid use, azathioprine use and deconvolution abundance fraction of dendritic cells.
Results from separate healthy and inflamed tissue analyses were subsequently combined using the MetaVolcanoR R package (version 1.2.0, Cesar Prada, São Paulo, Brasil) to obtain meta-statistically significant differentially expressed transcripts showing the same direction of expression and consistence of perturbation [44]. Additionally, statistically significant differentially expressed transcripts showing opposite direction of expression in healthy and inflamed were also retained for further analyses. Statistically significant differential expression was considered for transcripts with p value < 0.05 and Log2FC > 2 or < −2.

2.4. Gene Ontology Analysis

Gene ontology (GO) analysis was carried out using the clusterProfiler R package (version 4.4, Guangchuang Yu, Guangzhou, China) [45]. Enrichment of terms was performed for biological processes, cellular components and molecular functions. Thresholds were set as p value < 0.01 and q value < 0.05.

2.5. Genome-Wide Association Analysis

DNA obtained from 84 enrolled patients was genotyped using the genotyping microarray Infinium Global Screening Array (GSA_24v1) (Illumina, San Diego, CA, USA). Quality control, genotype imputation, subsequent association analysis of the obtained genotype data and subsequent integration with transcriptomic data were performed as described previously [30]. Binary logistic Wald regression with adalimumab response at week 12 set as the dependent variable was carried out corrected to age at diagnosis, sex, azathioprine use, use of aminosalycylates, use of corticosteroids and the first four principal components. A statistically significant signal was considered for variants with adjusted p value < 0.05. Where more than one variant was present at the same locus, variants were pruned using the SNPclip tool from LDlink software (version 5.4, NIH, Bethesda, MD, USA) with R2 set to 0.5 r to obtain independent genomic signals [46]. Regional manhattan plots for gene regions were constructed using LocusZoom [47].

2.6. Functional Annotation and eQTL Estimation

Functional annotation was carried out using the HaploReg v4.1 database (accessed on 25 July 2022) [48]. Tissue eQTLs were estimated using the GTExPortal and ENSEMBL database (accessed on 25 July 2022) [49,50]. Transcript-wise eQTLs in our data were calculated using the Kruskal-Wallis H test and were calculated independently for all tissues, healthy tissue and inflamed tissue. A p value < 0.05 was considered as a statistically significant eQTL observation.

2.7. Construction of Gene Interaction Network and Visualization

Interaction grids of genes of interest were generated and illustrated using the software package CytoScape (version 3.8.2, CytoScape Team) with the integrated application ClueGO (version 2.5.8, Laboratory of Integrative Cancer Immunology (Team 15), Paris, France) [51,52]. To obtain curated data of gene interaction, we expanded our list of genes of interest by adding genes interacting with at least two investigated genes (i.e., genes associated with response to adalimumab) which we obtained from the BIOGRID database (accessed on 28 July 2022) using the biogridR R package (github.com/npjc, Vancouver, BC, Canada) [53,54].

2.8. Machine Learning Validation

Validation of identified transcripts and variants was further assessed using machine learning random forest algorithm using the randomForest R package (version 4.6–14, Merck Research Laboratories, Kenilworth, NJ, USA) [55]. Machine learning validation was done using all 15 identified transcripts, transcripts of LGMN and ACP5 genes and three transcripts of the HLA-DMB gene. Additionally, validation was made for variants where eQTL association was observed. Results are presented as probabilities after receiver operating characteristics (ROC) analysis using pROC R package (version 1.18.0, Swiss Institute of Bioinformatics, Geneva, Switzerland) [56].

3. Results

3.1. Transcript Specific RNA-Seq Analysis

Transcript specific RNA-seq data analysis was independently performed on healthy tissue and inflamed tissue. A regression model was fitted and corrected using four different approaches as aforementioned. Deconvolution fractions used in the approaches are presented in Figure S1.
Using the first approach, we first combined the datasets of the two separate analyses (differentially expressed isoforms (DEIs) between non-responders and responders identified in normal tissue and DEIs between non-responders and responders identified in inflamed tissue) in order to obtain meta-statistically significant DEIs showing the same direction of expression and consistence of perturbation. The analysis identified 62 statistically significant DEIs (Table S1) as evident from Figure 1. Additionally, 26 statistically significant DEIs between non-responders and responders showing the opposite direction in healthy and inflamed tissue were also retained from the two separate analyses (Table S2 and Figure S2).
Using correction for the T cell deconvolution fraction, we observed 52 meta-statistically significant DEIs with conserved perturbation in healthy and inflamed tissues (Table S3) as shown in Figure 1. Thirty-six statistically significant DEIs showing the opposite direction in healthy and inflamed tissue were also observed using the model with correction for T cell fraction (Table S4 and Figure S3).
For determination of meta-statistically DEIs using the regression model with correction for the monocytes/macrophages deconvolution fraction, we used a more stringent threshold (p value < 0.005 and Log2FC > 2 or < −2), and identified 134 transcripts with a conserved perturbation in healthy and inflamed tissue as shown in Figure 1 (Table S5). Additionally, four transcripts showed statistically significant differential opposite expression in healthy and inflamed tissue (Table S6 and Figure S4).
Using the deconvoluted dendritic cell fraction for correction of the regression model, we observed 98 meta-statistically significant DEIs with conserved perturbation in healthy and inflamed tissue as evident from Figure 1 (Table S7). We additionally observed 157 isoform transcripts with statistically significant differential expression in the opposite directions in healthy and inflamed tissue (Table S8 and Figure S5).

3.2. Gene Ontology Enrichment Analysis

Using the results obtained from approaches of isoform specific RNA-seq data analysis, we performed GO analysis for each set of identified isoform transcripts, and their corresponding genes, separately. GO analysis was performed for biological processes, cellular components and molecular function. The most significant enriched terms for each approach are presented in Table 1, and all enriched terms with corresponding gene ratios and gene IDs are presented in Table S9. Enriched terms are mostly correlated to antigen preprocessing and peptide antigen presentation in relation to MHC class I complexes. Furthermore, cytoplasmic translation, lumenal side of endoplasmatic reticulum membrane, ribosome and endocytic vesicle, are also most enriched terms. Altogether, using all statistically significant enriched terms, there are 111 unique gene IDs presented in Table S10.

3.3. GWAs Integration, Gene Selection and Functional Annotation

Genomic regions of ± 100 kbp from 111 genes identified in transcriptomics and subsequent GO analyses were further used to integrate transcriptomic and genomic data for stringent gene selection. Integration revealed 54 statistically significant signals in 9 (MACF1, CTSE, HDLBP, HSPA9, HLA-DMB, TAP2, LGMN, ANAPC11, ACP5) out of 111 selected gene regions after application of empirical autocorrelation adjustment (Table S11, Figures S6–S14). All nine genes were represented with altogether 15 isoform transcripts.
Where more variants at the same loci were present, they were additionally pruned by linkage disequilibrium (R2 > 0.5) to obtain independent genomic signals presented in Table 2. Independent variants were further functionally annotated to identify possible function on gene expression through epigenetic modifications, chromatin remodeling or protein/motif binding alternations. It was shown that protein binding and motif changes are listed for variants rs10888633, rs35441329, rs28396806, rs28386947, rs10132165 and rs28454947. For variant rs7761882 identified in HLA-DMB and TAP2 gene regions and rs2229531 in the ACP5 gene region, enhancer histone marks in immune cells from peripheral blood and motif changes are listed. Moreover, variant rs2229531 is a missense variant and flagged as constrained by GERP and SiPhy-omega algorithms. In the variant rs144948822 DNase experiments, motif changes are listed. Variants in the gene region of LGMN were the most annotated variants and are listed at promoter and enhancer histone marks in peripheral blood immune cells, intestinal or colon tissue. In DNase experiments, binding of various proteins and motif changes are listed at these variants. Detailed annotation is available in Table S12.
Variants were further analyzed using the eQTL database (accessed on 25 July 2022). Using the genotype-tissue expression GTEx portal database (accessed on 25 July 2022), only rs10137934 (LGMN) and rs2229531 (ACP5) were listed as significant eQTLs for identified genes in whole blood, small intestines and colon tissue. All other identified variants were listed as significant eQTLs for other neighboring genes in identified gene regions (Table S13). Interestingly, none of these neighboring genes were statistically significant differentially expressed in transcript specific RNA-seq analyses. To assess the eQTLs transcript-wise in our data, we statistically calculated genotype-transcript expression association separately in all tissues, healthy tissue and inflamed tissue (Table 3). Despite the small sample size, we were able to observe statistically significant eQTL for LGMN gene (ENST00000555169.1) and rs58531216 in all tissues (p = 0.004) and tendency for association in healthy (p = 0.053) and inflamed tissue (p = 0.052). Tendency for eQTL association was also observed for the LGMN gene (ENST000005551669.1) and rs11621843 in inflamed tissue (p = 0.052).

3.4. Interactions and Machine Learning Validation of Selected Genes

The aforementioned nine genes were also assessed for interplay using BioGRID database and visualization of interactions. It was shown that all genes except HLA-DMB indirectly interact with each other through a network of auxiliary gene nodes. Interestingly the distance of indirect interactions between target genes is not more than one auxiliary node, as evident from Figure 2.
Association of gene expression of the selected transcripts with response to adalimumab was additionally assessed using a Random Forest machine learning algorithm. Assessment was made using the information of tissue state (healthy/inflamed) and expression of: (I) all 15 identified transcripts corresponding to the nine genes; (II) transcripts of LGMN and ACP5 genes with proven eQTLs, and (III) transcripts of HLA-DMB gene (ENST00000383231.2, ENST00000428420.2, ENST00000440078.2), which was left out of the BioGRID interactions. All three assessments yielded an AUC of 1 in ROC analyses, thus further confirming the involvement of selected genes in response to adalimumab treatment. Moreover, genotypes from selected variants for which eQTL association was proven (rs58531216, rs10137934, rs2229531) were also assessed for involvement in adalimumab response and yielding AUC of 0.827 (CI95: 0.740–0.913), which further supports the involvement of LGMN and ACP5 genes in the adalimumab response (Figure 3).

4. Discussion

The present study investigated colon tissue isoform transcript specific association with response to adalimumab treatment in Slovenian Crohn’s disease patients who were naïve with respect to adalimumab treatment. Using an integrative approach (Figure S15) to combine immune cell deconvoluted isoform transcript specific transcriptome analysis, gene ontology layering and genomic data, we identified nine genes (MACF1, CTSE, HDLBP, HSPA9, HLA-DMB, TAP2, LGMN, ANAPC11, ACP5) with 15 transcripts involved in adalimumab response. Additional genomic integration identified 16 variants residing in aforementioned gene regions associated with adalimumab response. It has been recently shown that deconvolution of whole-tissue gene expression data yields refined information of the immune cell landscape in inflammatory bowel disease [34]. The latter is also supported by previous findings where increased robustness of analyses was shown when deconvolution information was used in regression models [29,42].
Gene MACF1 was identified as oppositely expressed with one transcript ENST00000372925.2 corresponding to MACF1 transcript variant 3. The isoform was up-regulated in non-responders relative to responders in healthy tissue, and down-regulated in non-responders relative to responders in inflamed tissue. MACF1 is also known as ACF7 and belongs to the spectraplakin family of proteins, which are evolutionarily conserved and are involved in cytoskeletal organization, cell/cell junctions and integrin-mediated epidermal attachments [57]. It was also proven that MACF1 regulates the molecular mechanism of cytoskeletal coordination and subsequent cell adhesion regulation in intestinal wound repair, and contributes to the development of inflammatory bowel disease [58]. The opposite direction of the expression of MACF1 transcript between non-responders relative to responders in healthy and inflamed colon tissue additionally suggests an important role of this gene in anti-TNF response in Crohn’s disease.
Two transcripts of the CTSE gene corresponding to CTSE transcript variant 2 (ENST00000360218.2, ENST00000581049.1-alternate locus) were both up-regulated in non-responders relative to responders. CTSE is cathepsin E and belongs to the A1 family of peptidases. CTSE was previously identified among the genes, which are uniquely expressed in ulcerative colitis phenotype of the inflammatory bowel disease [59]. Moreover, CTSE was also identified in a cluster of the REG4 regulatory network [60]. CTSE was also associated with intestinal fibrosis in persistent Salmonella infections [61].
Transcript ENST00000310931.4 corresponding to HDLBP (high-density lipoprotein binding protein) gene transcript variant 3 was found to be up-regulated in non-responders relative to responders in healthy tissue, and down-regulated in non-responders relative to responders in inflamed tissue. Protein encoded by HDLBP binds high-density lipoproteins and regulate excess cholesterol levels in the cells. HDLBP is also known as Vigilin, and has a general function in endoplasmatic reticulum translation and in tumor progression [62]. Furthermore, HDLBP was also shown to interact with TCS2 and thus regulates the formation of stress granules [63].
The HSPA9 gene belongs to the heat shock protein family A. Transcript ENST00000501917.2 belonging to HSPA9 processed transcript variant 12 was observed to be down-regulated in non-responders relative to responders. It was shown that HSPA9 plays a role in major up-regulated metabolic pathway during endoplasmatic reticulum stress in colonic goblet cells in a Winnie murine model [64]. Interestingly, HSPA9 was found to be differentially expressed between steroid responders and non-responders in ulcerative colitis, but did not reach sufficient predictive power to be highlighted [65].
For the HLA-DMB gene, three MHC haplotype-specific transcripts corresponding to transcript variant 1 (ENST00000383231.2 MHC haplotype QBL, ENST00000428420.2 MHC haplotype MANN, and ENST00000440078.2 MHC haplotype DBB) were observed to be up-regulated in non-responders relative to responders. HLA-DMB is major histocompatibility complex class II beta chain paralogue. This gene is known to be involved in pathways for presentation of viral and self-antigens to T cells and these pathways were found to be unique to the Crohn’s disease phenotype when compared to ulcerative colitis by colon transcriptome [66]. It is also one of the known inflammatory bowel disease genes found to be associated with autoimmune diseases [59]. A previous study also observed that variations within HLA genes had a predominant effect in disease susceptibility as opposed to other genes in MHC gene region [67].
The TAP2 gene, the neighbor of HLA-DMB, was present with four MHC haplotype specific transcripts corresponding to transcript variant 1. Transcript ENST00000443713.2 MCH haplotype COX was found to be up-regulated in non-responders relative to responders, and other transcripts (ENST00000457634.2 MHC haplotype SSTO, ENST00000452371.2 MHC haplotype DBB and ENST00000455842.2 MHC haplotype APD) were found to be down-regulated in non-responders relative to responders. The discrepancy between the up-regulated and down-regulated state was observed between analyses with correction to T cells and dendritic cell deconvolution fractions, respectively. Interestingly, the up-regulated transcript ENST00000443713.2 MHC haplotype COX was observed only in T cell fraction corrected analysis, and the other three transcripts were down-regulated only in the dendritic cell fraction corrected analysis. The later further proves the complexity of anti-TNF response heterogeneity. TAP2 is a transporter 2, ATP binding cassette subfamily B member and is a mediator of viral and auto-antigen immunity by pumping cytosolic peptides into the endoplasmatic reticulum for MHC-I-mediated antigen presentation [66]. It was also noted that the expression of TAP2 was elevated in inflammatory bowel disease in comparison with the controls [66]. It was also shown that the TAP2 locus may play an important role in Crohn’s disease heterogeneity and steroid responsiveness [68]. Additionally, the TAP2 gene was identified as a member of whole blood gene panel to distinguish Crohn’s disease and ulcerative colitis [69]. On the other hand, TAP2 was also identified as a tissue-related common gene for ulcerative colitis in an integrative study of transcriptome-wide association analysis and mRNA expression profiles [70]. Moreover, TAP2 was one of the identified differentially expressed genes between biopsies from ulcerative colitis patients who were naïve with respect to the treatment and healthy biopsy samples [71].
Legumain (LGMN) gene transcript variant 20 ENST00000555169.1 was found to be down-regulated in non-responders relative to responders. LGMN encodes a cysteine protease with strict specificity for hydrolysis of asparaginyl bonds, and is involved in processing of the bacterial peptides and endogenous proteins for MHC class II presentation in the lysosome and endosome. Expression of LGMN occurs after monocytes differentiate into dendritic cells. In relation to Crohn’s disease, the LGMN gene was previously mentioned in a group of highly variable genes between resident and inflammatory macrophage clusters in a study in which the authors identified a cellular module named GIMATS, which consisted of activated dendritic cells, inflammatory macrophages, activated T cells, IgG plasma cells, activated fibroblasts and endothelial cells [32]. GIMATS module was associated with failure to achieve durable remission without corticosteroid use upon initiation of any anti-TNF therapy [32]. To further strengthen the role of macrophages in the response to anti-TNF therapy, it was shown that CD14+ macrophages play a significant role in anti-TNF refractory patients through heightened IL-23 production, which, in turn, leads to binding to the IL23R bearing on TNFR2+ gut CD4+ T cells and results in subsequent STAT3 activation induction leading to expansion of apoptosis resistant intestinal T cells [31]. However, in contrary to our findings, in the present study we did not observe any statistically significant differentially expression of TNFR2 or IL23R genes. Furthermore, the monocyte to macrophage differentiation-associated gene was previously identified as a key player in the adalimumab response [30]. These findings further support the LGMN gene as an important gene in the anti-TNF response, as it plays a role in a cell subtype associated with the anti-TNF response. Additionally, concerning the findings that both MHC classes I and II contribute to the Crohn’s disease risk, the present study identified both antigen presentation mechanisms to be important in anti-TNF response [67].
Transcript ENST00000583839.1 corresponding to ANAPC11 gene transcript variant 19 was identified as down-regulated in non-responders relative to responders. ANAPC11 is anaphase promoting complex subunit 11 and, to the best of our knowledge, was never before mentioned in relation to inflammatory bowel disease or anti-TNF response. ANAPC11 is a ubiquitination process gene and is associated with dysregulation in innate immunity in malaria [72]. Moreover, ANAPC11 has been found to be a new independent predictive biomarker for colorectal cancer [73].
Transcript ENST00000218758.5 of ACP5 gene transcript variant 2 was also down-regulated in non-responders relative to responders. ACP5 is acid phosphatase 5, tartrate resistant and encodes an iron-containing glycoprotein that catalyzes orthophosphoric monoester to alcohol and orthophosphate. It was identified in a cluster of inflammatory macrophages as a macrophage marker in patients with the GIMATS module of the anti-TNF non-response [32]. ACP5 also regulates cell cycle progression through mitosis and G1, and global hypoacetylation observed in APC mutant yeasts likely leads to chromatin remodeling that can repress expression of genes involved in genomic stability [74].
Integration with genomic data additionally identified 16 variants, which were further functionally annotated and were shown to have a possible impact on gene expression through epigenetic modifications, chromatin remodeling or protein/motif binding alternations. Nevertheless, only three variants have a confirmed eQTL association with aforementioned genes (LGMN and APC5), and others have a confirmed eQTL with neighboring genes in identified gene regions. Thus, we hypothesize that the epigenetic impact of variants may be extended to the identified gene loci. Furthermore, rs2229531 is an evolutionary conserved missense variant in ACP5 gene, which may play a role in chromatin remodeling.
It was also observed that the aforementioned genes, except HLA-DMB, indirectly interacted with each other through a network of auxiliary gene nodes, where the distance between target genes was not more than one node. To further validate these findings, we applied machine learning probability modeling to obtained data, which further confirmed the role of the aforementioned transcripts and the three eQTL variants in response to anti-TNF therapy responsiveness.
The main limitation of the present study is that we did not have paired healthy-inflamed colon tissues available for our analyses. On the other hand, we acknowledge the homogenous cohort, which was naïve with respect to the anti-TNF treatment, and the use of the only one anti-TNF agent, as strengths in our study.

5. Conclusions

Our study used isoform transcript-specific colon tissue transcriptome-wide analysis with subsequent genomic integration, and identified loci, some of which were previously identified to contribute to inflammatory bowel disease susceptibility, as novel loci involved in the adalimumab response in Crohn’s disease patients. As adalimumab is still widely used as a first-line biological treatment, results obtained in the present study may further extent the knowledge of biomarkers for adalimumab treatment response and thus contribute to individualized treatment plans in the evolving field of personalized medicine. Moreover, identified genes aid in better understanding intersections in common molecular pathways of anti-TNF treatment response, and thus may help to discover novel molecular targets for the treatment of Crohn’s disease patients, particularly those who are non-responsive to anti-TNF treatment.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/pharmaceutics14091893/s1, Tables S1–S8. Transcript specific RNA-seq analysis; Tables S9–S13. Gene ontology enrichment analysis; Figure S1. Deconvolution fractions of tissue samples. H: healthy tissue; D: Inflamed tissue; Figure S2. Statistically significant differentially expressed transcripts showing opposite direction in healthy and inflamed tissue in first approach; Figure S3. Statistically significant differentially expressed transcripts showing opposite direction in healthy and inflamed tissue in second approach; Figure S4. Statistically significant differentially expressed transcripts showing opposite direction in healthy and inflamed tissue in third approach; Figure S5. Statistically significant differentially expressed transcripts showing opposite direction in healthy and inflamed tissue in fourth approach; Figures S6–S14. Regional manhattan plot of ACP5, ANAPC11, CTSE, HDLBP, HLA-DMB, HSPA9, LGMN, MACF1, TAP2. Figure S15. Approach and analyses flowchart.

Author Contributions

Conceptualization, M.G. and U.P.; methodology, M.G. and U.P.; software, M.G.; validation, M.G., G.J., M.K., P.S. and U.P.; formal analysis, M.G., G.J. and U.P.; investigation, M.G., G.J., P.S. and U.P.; data curation, M.G. and M.K.; writing—original draft preparation, M.G.; writing—review and editing, G.J., M.K., P.S. and U.P.; visualization, M.G. and G.J.; supervision, P.S. and U.P.; project administration, U.P.; funding acquisition, U.P. All authors have read and agreed to the published version of the manuscript.

Funding

The authors acknowledge the financial support from the Slovenian Research Agency research core funding No. P3-0427 and research grant No. J3-9258.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of Slovenian National Committee for Medical Ethics (KME 80/10/07, 21p/12/07).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study. In the case of minors, written informed consent was obtained from parents and/or legal guardians.

Data Availability Statement

Data are available on request due to privacy restrictions. The data presented in this study are available on request from the corresponding author.

Acknowledgments

The authors would like to thank Boris Gole for support and careful review of the paper.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study, in the collection, analyses, or interpretation of data, in the writing of the manuscript, or in the decision to publish the results.

References

  1. Abraham, C.; Cho, J.H. Inflammatory bowel disease. N. Engl. J. Med. 2009, 361, 2066–2078. [Google Scholar] [CrossRef] [PubMed]
  2. Ordas, I.; Eckmann, L.; Talamini, M.; Baumgart, D.C.; Sandborn, W.J. Ulcerative colitis. Lancet 2012, 380, 1606–1619. [Google Scholar] [CrossRef]
  3. Baumgart, D.C.; Sandborn, W.J. Crohn’s disease. Lancet 2012, 380, 1590–1605. [Google Scholar] [CrossRef]
  4. Molodecky, N.A.; Soon, I.S.; Rabi, D.M.; Ghali, W.A.; Ferris, M.; Chernoff, G.; Benchimol, E.I.; Panaccione, R.; Ghosh, S.; Barkema, H.W.; et al. Increasing incidence and prevalence of the inflammatory bowel diseases with time, based on systematic review. Gastroenterology 2012, 142, 46–54.e42. [Google Scholar] [CrossRef] [PubMed]
  5. Rosenstiel, P.; Sina, C.; Franke, A.; Schreiber, S. Towards a molecular risk map--recent advances on the etiology of inflammatory bowel disease. Semin. Immunol. 2009, 21, 334–345. [Google Scholar] [CrossRef] [PubMed]
  6. Wehkamp, J.; Gotz, M.; Herrlinger, K.; Steurer, W.; Stange, E.F. Inflammatory Bowel Disease. Dtsch. Arztebl. Int. 2016, 113, 72–82. [Google Scholar] [CrossRef]
  7. Duchmann, R.; May, E.; Heike, M.; Knolle, P.; Neurath, M.; Meyer zum Buschenfelde, K.H. T cell specificity and cross reactivity towards enterobacteria, bacteroides, bifidobacterium, and antigens from resident intestinal flora in humans. Gut 1999, 44, 812–818. [Google Scholar] [CrossRef] [PubMed]
  8. Moussata, D.; Goetz, M.; Gloeckner, A.; Kerner, M.; Campbell, B.; Hoffman, A.; Biesterfeld, S.; Flourie, B.; Saurin, J.C.; Galle, P.R.; et al. Confocal laser endomicroscopy is a new imaging modality for recognition of intramucosal bacteria in inflammatory bowel disease in vivo. Gut 2011, 60, 26–33. [Google Scholar] [CrossRef]
  9. Wehkamp, J.; Fellermann, K.; Herrlinger, K.R.; Bevins, C.L.; Stange, E.F. Mechanisms of disease: Defensins in gastrointestinal diseases. Nat. Clin. Pract. Gastroenterol. Hepatol. 2005, 2, 406–415. [Google Scholar] [CrossRef]
  10. Van Dussen, K.L.; Liu, T.C.; Li, D.; Towfic, F.; Modiano, N.; Winter, R.; Haritunians, T.; Taylor, K.D.; Dhall, D.; Targan, S.R.; et al. Genetic variants synthesize to produce paneth cell phenotypes that define subtypes of Crohn’s disease. Gastroenterology 2014, 146, 200–209. [Google Scholar] [CrossRef] [Green Version]
  11. Zundler, S.; Neurath, M.F. Immunopathogenesis of inflammatory bowel diseases: Functional role of T cells and T cell homing. Clin. Exp. Rheumatol. 2015, 33, S19–S28. [Google Scholar] [PubMed]
  12. Danese, S.; Fiorino, G.; Reinisch, W. Review article: Causative factors and the clinical management of patients with Crohn’s disease who lose response to anti-TNF-alpha therapy. Aliment. Pharmacol. Ther. 2011, 34, 1–10. [Google Scholar] [CrossRef]
  13. Van Deventer, S.J. Tumour necrosis factor and Crohn’s disease. Gut 1997, 40, 443–448. [Google Scholar] [CrossRef]
  14. Colombel, J.F.; Sandborn, W.J.; Rutgeerts, P.; Enns, R.; Hanauer, S.B.; Panaccione, R.; Schreiber, S.; Byczkowski, D.; Li, J.; Kent, J.D.; et al. Adalimumab for maintenance of clinical response and remission in patients with Crohn’s disease: The CHARM trial. Gastroenterology 2007, 132, 52–65. [Google Scholar] [CrossRef]
  15. Hanauer, S.B.; Feagan, B.G.; Lichtenstein, G.R.; Mayer, L.F.; Schreiber, S.; Colombel, J.F.; Rachmilewitz, D.; Wolf, D.C.; Olson, A.; Bao, W.; et al. Maintenance infliximab for Crohn’s disease: The ACCENT I randomised trial. Lancet 2002, 359, 1541–1549. [Google Scholar] [CrossRef]
  16. Sandborn, W.J.; Hanauer, S.B.; Rutgeerts, P.; Fedorak, R.N.; Lukas, M.; MacIntosh, D.G.; Panaccione, R.; Wolf, D.; Kent, J.D.; Bittle, B.; et al. Adalimumab for maintenance treatment of Crohn’s disease: Results of the CLASSIC II trial. Gut 2007, 56, 1232–1239. [Google Scholar] [CrossRef] [PubMed]
  17. Sandborn, W.J.; Feagan, B.G.; Rutgeerts, P.; Hanauer, S.; Colombel, J.F.; Sands, B.E.; Lukas, M.; Fedorak, R.N.; Lee, S.; Bressler, B.; et al. Vedolizumab as induction and maintenance therapy for Crohn’s disease. N. Engl. J. Med. 2013, 369, 711–721. [Google Scholar] [CrossRef]
  18. Schoultz, I.; Keita, A.V. Cellular and Molecular Therapeutic Targets in Inflammatory Bowel Disease-Focusing on Intestinal Barrier Function. Cells 2019, 8, 193. [Google Scholar] [CrossRef]
  19. Feagan, B.G.; Sandborn, W.J.; Gasink, C.; Jacobstein, D.; Lang, Y.; Friedman, J.R.; Blank, M.A.; Johanns, J.; Gao, L.L.; Miao, Y.; et al. Ustekinumab as Induction and Maintenance Therapy for Crohn’s Disease. N. Engl. J. Med. 2016, 375, 1946–1960. [Google Scholar] [CrossRef]
  20. Feagan, B.G.; Panes, J.; Ferrante, M.; Kaser, A.; D’Haens, G.R.; Sandborn, W.J.; Louis, E.; Neurath, M.F.; Franchimont, D.; Dewit, O.; et al. Risankizumab in patients with moderate to severe Crohn’s disease: An open-label extension study. Lancet Gastroenterol. Hepatol. 2018, 3, 671–680. [Google Scholar] [CrossRef]
  21. Sands, B.E.; Peyrin-Biroulet, L.; Loftus, E.V., Jr.; Danese, S.; Colombel, J.F.; Toruner, M.; Jonaitis, L.; Abhyankar, B.; Chen, J.; Rogers, R.; et al. Vedolizumab versus Adalimumab for Moderate-to-Severe Ulcerative Colitis. N. Engl. J. Med. 2019, 381, 1215–1226. [Google Scholar] [CrossRef] [PubMed]
  22. Singh, S.; Murad, M.H.; Fumery, M.; Sedano, R.; Jairath, V.; Panaccione, R.; Sandborn, W.J.; Ma, C. Comparative efficacy and safety of biologic therapies for moderate-to-severe Crohn’s disease: A systematic review and network meta-analysis. Lancet Gastroenterol. Hepatol. 2021, 6, 1002–1014. [Google Scholar] [CrossRef]
  23. Dezelak, M.; Repnik, K.; Koder, S.; Ferkolj, I.; Potocnik, U. A Prospective Pharmacogenomic Study of Crohn’s Disease Patients during Routine Therapy with Anti-TNF-alpha Drug Adalimumab: Contribution of ATG5, NFKB1, and CRP Genes to Pharmacodynamic Variability. OMICS A J. Integr. Biol. 2016, 20, 296–309. [Google Scholar] [CrossRef]
  24. Koder, S.; Repnik, K.; Ferkolj, I.; Pernat, C.; Skok, P.; Weersma, R.K.; Potocnik, U. Genetic polymorphism in ATG16L1 gene influences the response to adalimumab in Crohn’s disease patients. Pharmacogenomics 2015, 16, 191–204. [Google Scholar] [CrossRef] [PubMed]
  25. Repnik, K.; Koder, S.; Skok, P.; Ferkolj, I.; Potocnik, U. Transferrin Level Before Treatment and Genetic Polymorphism in HFE Gene as Predictive Markers for Response to Adalimumab in Crohn’s Disease Patients. Biochem. Genet. 2016, 54, 476–486. [Google Scholar] [CrossRef]
  26. Barber, G.E.; Yajnik, V.; Khalili, H.; Giallourakis, C.; Garber, J.; Xavier, R.; Ananthakrishnan, A.N. Genetic Markers Predict Primary Non-Response and Durable Response To Anti-TNF Biologic Therapies in Crohn’s Disease. Am. J. Gastroenterol. 2016, 111, 1816–1822. [Google Scholar] [CrossRef] [PubMed]
  27. Yoon, S.M.; Haritunians, T.; Chhina, S.; Liu, Z.; Yang, S.; Landers, C.; Li, D.; Ye, B.D.; Shih, D.; Vasiliauskas, E.A.; et al. Colonic Phenotypes Are Associated with Poorer Response to Anti-TNF Therapies in Patients with IBD. Inflamm. Bowel Dis. 2017, 23, 1382–1393. [Google Scholar] [CrossRef] [PubMed]
  28. Arijs, I.; De Hertogh, G.; Lemaire, K.; Quintens, R.; Van Lommel, L.; Van Steen, K.; Leemans, P.; Cleynen, I.; Van Assche, G.; Vermeire, S.; et al. Mucosal gene expression of antimicrobial peptides in inflammatory bowel disease before and after first infliximab treatment. PLoS ONE 2009, 4, e7984. [Google Scholar] [CrossRef]
  29. Gorenjak, M.; Repnik, K.; Jezernik, G.; Jurgec, S.; Skok, P.; Potocnik, U. Genetic prediction profile for adalimumab response in Slovenian Crohn’s disease patients. Z. Gastroenterol. 2019, 57, 1218–1225. [Google Scholar] [CrossRef]
  30. Gorenjak, M.; Zupin, M.; Jezernik, G.; Skok, P.; Potocnik, U. Omics data integration identifies ELOVL7 and MMD gene regions as novel loci for adalimumab response in patients with Crohn’s disease. Sci. Rep. 2021, 11, 5449. [Google Scholar] [CrossRef]
  31. Schmitt, H.; Billmeier, U.; Dieterich, W.; Rath, T.; Sonnewald, S.; Reid, S.; Hirschmann, S.; Hildner, K.; Waldner, M.J.; Mudter, J.; et al. Expansion of IL-23 receptor bearing TNFR2+ T cells is associated with molecular resistance to anti-TNF therapy in Crohn’s disease. Gut 2019, 68, 814–828. [Google Scholar] [CrossRef]
  32. Martin, J.C.; Chang, C.; Boschetti, G.; Ungaro, R.; Giri, M.; Grout, J.A.; Gettler, K.; Chuang, L.S.; Nayar, S.; Greenstein, A.J.; et al. Single-Cell Analysis of Crohn’s Disease Lesions Identifies a Pathogenic Cellular Module Associated with Resistance to Anti-TNF Therapy. Cell 2019, 178, 1493–1508.e1420. [Google Scholar] [CrossRef]
  33. Gole, B.; Potocnik, U. Pre-Treatment Biomarkers of Anti-Tumour Necrosis Factor Therapy Response in Crohn’s Disease-A Systematic Review and Gene Ontology Analysis. Cells 2019, 8, 515. [Google Scholar] [CrossRef]
  34. Bai, X.; Liu, W.; Chen, H.; Zuo, T.; Wu, X. Immune Cell Landscaping Reveals Distinct Immune Signatures of Inflammatory Bowel Disease. Front. Immunol. 2022, 13, 861790. [Google Scholar] [CrossRef] [PubMed]
  35. Maaser, C.; Sturm, A.; Vavricka, S.R.; Kucharzik, T.; Fiorino, G.; Annese, V.; Calabrese, E.; Baumgart, D.C.; Bettenworth, D.; Nunes, P.B.; et al. ECCO-ESGAR Guideline for Diagnostic Assessment in IBD Part 1: Initial diagnosis, monitoring of known IBD, detection of complications. J. Crohns Colitis 2019, 13, 144–164. [Google Scholar] [CrossRef] [PubMed]
  36. Panaccione, R.; Loftus, E.V., Jr.; Binion, D.; McHugh, K.; Alam, S.; Chen, N.; Guerette, B.; Mulani, P.; Chao, J. Efficacy and safety of adalimumab in Canadian patients with moderate to severe Crohn’s disease: Results of the Adalimumab in Canadian SubjeCts with ModErate to Severe Crohn’s DiseaSe (ACCESS) trial. Can. J. Gastroenterol. 2011, 25, 419–425. [Google Scholar] [CrossRef] [PubMed]
  37. Hlavaty, T.; Persoons, P.; Vermeire, S.; Ferrante, M.; Pierik, M.; Van Assche, G.; Rutgeerts, P. Evaluation of short-term responsiveness and cutoff values of inflammatory bowel disease questionnaire in Crohn’s disease. Inflamm. Bowel Dis. 2006, 12, 199–204. [Google Scholar] [CrossRef] [PubMed]
  38. Guyatt, G.; Mitchell, A.; Irvine, E.J.; Singer, J.; Williams, N.; Goodacre, R.; Tompkins, C. A new measure of health status for clinical trials in inflammatory bowel disease. Gastroenterology 1989, 96, 804–810. [Google Scholar] [CrossRef]
  39. Andrews, S. FastQC: A Quality Control Tool for High throughput Sequence Data. 2010. Available online: http://www.bioinformatics.babraham.ac.uk/projects/fastqc (accessed on 29 April 2022).
  40. Bolger, A.M.; Lohse, M.; Usadel, B. Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics 2014, 30, 2114–2120. [Google Scholar] [CrossRef] [PubMed]
  41. Bray, N.L.; Pimentel, H.; Melsted, P.; Pachter, L. Near-optimal probabilistic RNA-seq quantification. Nat. Biotechnol. 2016, 34, 525–527. [Google Scholar] [CrossRef]
  42. Hernandez-Pacheco, N.; Gorenjak, M.; Jurgec, S.; Corrales, A.; Jorgensen, A.; Karimi, L.; Vijverberg, S.J.; Berce, V.; Schieck, M.; Acosta-Herrera, M.; et al. Combined analysis of transcriptomic and genetic data for the identification of loci involved in glucocorticosteroid response in asthma. Allergy 2021, 76, 1238–1243. [Google Scholar] [CrossRef] [PubMed]
  43. Newman, A.M.; Liu, C.L.; Green, M.R.; Gentles, A.J.; Feng, W.; Xu, Y.; Hoang, C.D.; Diehn, M.; Alizadeh, A.A. Robust enumeration of cell subsets from tissue expression profiles. Nat. Methods 2015, 12, 453–457. [Google Scholar] [CrossRef] [PubMed]
  44. Prada, C.; Lima, D.; Nakaya, H. MetaVolcanoR: Gene Expression Meta-Analysis Visualization Tool. 2020. Available online: https://www.bioconductor.org/packages/release/bioc/html/MetaVolcanoR.html (accessed on 29 April 2022).
  45. Wu, T.; Hu, E.; Xu, S.; Chen, M.; Guo, P.; Dai, Z.; Feng, T.; Zhou, L.; Tang, W.; Zhan, L.; et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation 2021, 2, 100141. [Google Scholar] [CrossRef]
  46. Machiela, M.J.; Chanock, S.J. LDlink: A web-based application for exploring population-specific haplotype structure and linking correlated alleles of possible functional variants. Bioinformatics 2015, 31, 3555–3557. [Google Scholar] [CrossRef] [PubMed]
  47. Pruim, R.J.; Welch, R.P.; Sanna, S.; Teslovich, T.M.; Chines, P.S.; Gliedt, T.P.; Boehnke, M.; Abecasis, G.R.; Willer, C.J. LocusZoom: Regional visualization of genome-wide association scan results. Bioinformatics 2010, 26, 2336–2337. [Google Scholar] [CrossRef]
  48. Ward, L.D.; Kellis, M. HaploReg: A resource for exploring chromatin states, conservation, and regulatory motif alterations within sets of genetically linked variants. Nucleic Acids Res. 2012, 40, D930–D934. [Google Scholar] [CrossRef]
  49. GTEx Consortium. Erratum: Genetic effects on gene expression across human tissues. Nature 2018, 553, 530. [Google Scholar] [CrossRef]
  50. Hunt, S.E.; McLaren, W.; Gil, L.; Thormann, A.; Schuilenburg, H.; Sheppard, D.; Parton, A.; Armean, I.M.; Trevanion, S.J.; Flicek, P.; et al. Ensembl variation resources. Database 2018, 2018, 119. [Google Scholar] [CrossRef]
  51. Shannon, P.; Markiel, A.; Ozier, O.; Baliga, N.S.; Wang, J.T.; Ramage, D.; Amin, N.; Schwikowski, B.; Ideker, T. Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 2003, 13, 2498–2504. [Google Scholar] [CrossRef]
  52. Bindea, G.; Mlecnik, B.; Hackl, H.; Charoentong, P.; Tosolini, M.; Kirilovsky, A.; Fridman, W.H.; Pages, F.; Trajanoski, Z.; Galon, J. ClueGO: A Cytoscape plug-in to decipher functionally grouped gene ontology and pathway annotation networks. Bioinformatics 2009, 25, 1091–1093. [Google Scholar] [CrossRef] [Green Version]
  53. Oughtred, R.; Stark, C.; Breitkreutz, B.J.; Rust, J.; Boucher, L.; Chang, C.; Kolas, N.; O‘Donnell, L.; Leung, G.; McAdam, R.; et al. The BioGRID interaction database: 2019 update. Nucleic Acids Res. 2019, 47, D529–D541. [Google Scholar] [CrossRef] [PubMed]
  54. Stark, C.; Breitkreutz, B.J.; Reguly, T.; Boucher, L.; Breitkreutz, A.; Tyers, M. BioGRID: A general repository for interaction datasets. Nucleic Acids Res. 2006, 34, D535–D539. [Google Scholar] [CrossRef] [PubMed]
  55. Liaw, A.; Wiener, M. Classification and Regression by randomForest. R. News 2002, 2, 18–22. [Google Scholar]
  56. Robin, X.; Turck, N.; Hainard, A.; Tiberti, N.; Lisacek, F.; Sanchez, J.C.; Muller, M. pROC: An open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinform. 2011, 12, 77. [Google Scholar] [CrossRef]
  57. Suozzi, K.C.; Wu, X.; Fuchs, E. Spectraplakins: Master orchestrators of cytoskeletal dynamics. J. Cell Biol. 2012, 197, 465–475. [Google Scholar] [CrossRef] [PubMed]
  58. Ma, Y.; Yue, J.; Zhang, Y.; Shi, C.; Odenwald, M.; Liang, W.G.; Wei, Q.; Goel, A.; Gou, X.; Zhang, J.; et al. ACF7 regulates inflammatory colitis and intestinal wound response by orchestrating tight junction dynamics. Nat. Commun. 2017, 8, 15375. [Google Scholar] [CrossRef] [PubMed]
  59. Clark, P.M.; Dawany, N.; Dampier, W.; Byers, S.W.; Pestell, R.G.; Tozeren, A. Bioinformatics analysis reveals transcriptome and microRNA signatures and drug repositioning targets for IBD and other autoimmune diseases. Inflamm. Bowel Dis. 2012, 18, 2315–2333. [Google Scholar] [CrossRef]
  60. Zimmermann, P.; Baumann, R. Genes Specifically Regulated in Inflammatory Bowel Disease; Nebion: Zürich, Switzerland, 2015. [Google Scholar]
  61. Ehrhardt, K.; Steck, N.; Kappelhoff, R.; Stein, S.; Rieder, F.; Gordon, I.O.; Boyle, E.C.; Braubach, P.; Overall, C.M.; Finlay, B.B.; et al. Persistent Salmonella enterica Serovar Typhimurium Infection Induces Protease Expression During Intestinal Fibrosis. Inflamm. Bowel Dis. 2019, 25, 1629–1643. [Google Scholar] [CrossRef]
  62. Zinnall, U.; Milek, M.; Minia, I.; Vieira-Vieira, C.H.; Muller, S.; Mastrobuoni, G.; Hazapis, O.G.; Del Giudice, S.; Schwefel, D.; Bley, N.; et al. HDLBP binds ER-targeted mRNAs by multivalent interactions to promote protein synthesis of transmembrane and secreted proteins. Nat. Commun. 2022, 13, 2727. [Google Scholar] [CrossRef]
  63. Kosmas, K.; Filippakis, H.; Khabibullin, D.; Turkiewicz, M.; Lam, H.C.; Yu, J.; Kedersha, N.L.; Anderson, P.J.; Henske, E.P. TSC2 Interacts with HDLBP/Vigilin and Regulates Stress Granule Formation. Mol. Cancer Res. 2021, 19, 1389–1397. [Google Scholar] [CrossRef] [PubMed]
  64. Wilson, R.; Gundamaraju, R.; Vemuri, R.; Angelucci, C.; Geraghty, D.; Gueven, N.; Eri, R. Identification of Key Pro-Survival Proteins in Isolated Colonic Goblet Cells of Winnie, a Murine Model of Spontaneous Colitis. Inflamm. Bowel Dis. 2020, 26, 80–92. [Google Scholar] [CrossRef]
  65. Loren, V.; Garcia-Jaraquemada, A.; Naves, J.E.; Carmona, X.; Manosa, M.; Aransay, A.M.; Lavin, J.L.; Sanchez, I.; Cabre, E.; Manye, J.; et al. ANP32E, a Protein Involved in Steroid-Refractoriness in Ulcerative Colitis, Identified by a Systems Biology Approach. J. Crohns Colitis 2019, 13, 351–361. [Google Scholar] [CrossRef]
  66. Yang, L.; Tang, S.; Baker, S.S.; Arijs, I.; Liu, W.; Alkhouri, R.; Lan, P.; Baker, R.D.; Tang, Z.; Ji, G.; et al. Difference in Pathomechanism Between Crohn’s Disease and Ulcerative Colitis Revealed by Colon Transcriptome. Inflamm. Bowel Dis. 2019, 25, 722–731. [Google Scholar] [CrossRef]
  67. Goyette, P.; Boucher, G.; Mallon, D.; Ellinghaus, E.; Jostins, L.; Huang, H.; Ripke, S.; Gusareva, E.S.; Annese, V.; Hauser, S.L.; et al. High-density mapping of the MHC identifies a shared role for HLA-DRB1*01:03 in inflammatory bowel diseases and heterozygous advantage in ulcerative colitis. Nat. Genet. 2015, 47, 172–179. [Google Scholar] [CrossRef]
  68. Heresbach, D.; Alizadeh, M.; Bretagne, J.F.; Dabadie, A.; Colombel, J.F.; Pagenault, M.; Heresbach-Le Berre, N.; Genetet, B.; Gosselin, M.; Semana, G. TAP gene transporter polymorphism in inflammatory bowel diseases. Scand. J. Gastroenterol. 1997, 32, 1022–1027. [Google Scholar] [CrossRef]
  69. Burakoff, R.; Pabby, V.; Onyewadume, L.; Odze, R.; Adackapara, C.; Wang, W.; Friedman, S.; Hamilton, M.; Korzenik, J.; Levine, J.; et al. Blood-based biomarkers used to predict disease activity in Crohn’s disease and ulcerative colitis. Inflamm. Bowel Dis. 2015, 21, 1132–1140. [Google Scholar] [CrossRef] [PubMed]
  70. Cheng, B.; Liang, X.; Wen, Y.; Li, P.; Zhang, L.; Ma, M.; Cheng, S.; Du, Y.; Liu, L.; Ding, M.; et al. Integrative analysis of transcriptome-wide association study data and messenger RNA expression profiles identified candidate genes and pathways for inflammatory bowel disease. J. Cell Biochem. 2019, 120, 14831–14837. [Google Scholar] [CrossRef]
  71. Taman, H.; Fenton, C.G.; Hensel, I.V.; Anderssen, E.; Florholmen, J.; Paulssen, R.H. Transcriptomic Landscape of Treatment-Naive Ulcerative Colitis. J. Crohns Colitis 2018, 12, 327–336. [Google Scholar] [CrossRef] [PubMed]
  72. Anyona, S.B.; Cheng, Q.; Raballah, E.; Hurwitz, I.; Lambert, C.G.; McMahon, B.H.; Ouma, C.; Perkins, D.J. Ingestion of hemozoin by peripheral blood mononuclear cells alters temporal gene expression of ubiquitination processes. Biochem. Biophys. Rep. 2022, 29, 101207. [Google Scholar] [CrossRef] [PubMed]
  73. Drouet, Y.; Treilleux, I.; Viari, A.; Leon, S.; Devouassoux-Shisheboran, M.; Voirin, N.; de la Fouchardiere, C.; Manship, B.; Puisieux, A.; Lasset, C.; et al. Integrated analysis highlights APC11 protein expression as a likely new independent predictive marker for colorectal cancer. Sci. Rep. 2018, 8, 7386. [Google Scholar] [CrossRef]
  74. Turner, E.L.; Malo, M.E.; Pisclevich, M.G.; Dash, M.D.; Davies, G.F.; Arnason, T.G.; Harkness, T.A. The Saccharomyces cerevisiae anaphase-promoting complex interacts with multiple histone-modifying enzymes to regulate cell cycle progression. Eukaryot. Cell 2010, 9, 1418–1431. [Google Scholar] [CrossRef] [PubMed] [Green Version]
Pharmaceutics 14 01893 g001 550
Figure 1. Volcano plots of RNA-seq transcript specific analyses. (a) First approach; (b) approach corrected for T cells deconvolution fraction; (c) approach corrected for monocytes/macrophages deconvolution fraction; (d) approach corrected for dendritic cells deconvolution fraction. Red dashed lines indicate borders for statistically significant differentially expressed transcripts.
Figure 1. Volcano plots of RNA-seq transcript specific analyses. (a) First approach; (b) approach corrected for T cells deconvolution fraction; (c) approach corrected for monocytes/macrophages deconvolution fraction; (d) approach corrected for dendritic cells deconvolution fraction. Red dashed lines indicate borders for statistically significant differentially expressed transcripts.
Pharmaceutics 14 01893 g001
Pharmaceutics 14 01893 g002 550
Figure 2. BioGRID network of gene interaction. Red nodes present selected genes. Blue nodes are gene interactors.
Figure 2. BioGRID network of gene interaction. Red nodes present selected genes. Blue nodes are gene interactors.
Pharmaceutics 14 01893 g002
Pharmaceutics 14 01893 g003 550
Figure 3. ROC curve analysis for selected variants.
Figure 3. ROC curve analysis for selected variants.
Pharmaceutics 14 01893 g003
Table
Table 1. Gene ontology most significantly enriched terms.
Table 1. Gene ontology most significantly enriched terms.
ApproachGO IDDescriptionq Value
Biological Process
I consistent *GO:0002474Antigen processing and presentation of peptide antigen via MHC class I9.27 × 10−6
I opposite **NANANA
II consistentGO:0048002Antigen processing and presentation of peptide antigen0.00018
II oppositeNANANA
III consistentGO:0002181Cytoplasmic translation4.16 × 10−17
III oppositeGO:0019885Antigen processing and presentation of endogenous peptide antigen via MHC class I0.00014
IV consistentGO:0019885Antigen processing and presentation of endogenous peptide antigen via MHC class I1.13 × 10−05
IV oppositeGO:0039531Regulation of viral-induced cytoplasmic pattern recognition receptor signaling pathway0.0019
Cellular Component
I consistentGO:0071556Integral component of lumenal side of endoplasmic reticulum membrane2.91 × 10−05
I oppositeGO:0030139Endocytic vesicle0.0057
II consistentGO:0042611MHC protein complex0.0015
II oppositeNANANA
III consistentGO:0022626Cytosolic ribosome2.15 × 10−16
III oppositeGO:0071556Integral component of lumenal side of endoplasmic reticulum membrane3.25 × 10−05
IV consistentGO:0071556Integral component of lumenal side of endoplasmic reticulum membrane0.00049
IV oppositeNANANA
Molecular Function
I consistentGO:0003823Antigen binding4.39 × 10−08
I oppositeNANANA
II consistentNANANA
II oppositeNANANA
III consistentGO:0003735Structural constituent of ribosome2.24 × 10−13
III oppositeGO:0042605Peptide antigen binding9.90 × 10−05
IV consistentGO:0042605Peptide antigen binding9.83 × 10−05
IV oppositeGO:0045296Cadherin binding3.44 × 10−05
NA: Enriched terms not applicable; * same direction of isoform gene expression change in non-responders vs. responders in healthy and inflamed tissue; ** opposite direction of isoform gene expression change in non-responders vs. responders in healthy and inflamed tissue.
Table
Table 2. Independent statistically significant variants.
Table 2. Independent statistically significant variants.
LOCUSdbSNPCHRBPp Valueq ValueβSEAF.RESPAF.NON
MACF1rs108886331394871640.00810.03981.01670.384160.445450.22414
MACF1rs354413291395372910.00300.0148−1.40060.472370.127270.31034
CTSErs2839680612062568690.00600.0360−1.31330.477740.118180.31034
HDLBPrs7559059822421856540.00020.0496−2.39730.651550.0545450.24138
HSPA9rs689100751379666430.00540.0294−1.08350.389790.254550.46552
HSPA9rs2838694751379781630.00210.0115−1.42050.462510.181820.41379
HSPA9rs5891426851379878140.00570.0306−1.05780.382370.190910.39655
HLA-DMBrs77618826328591370.00190.0443−1.75560.566050.10.2931
TAP2 *rs77618826328591370.00190.0297−1.75560.566050.10.2931
LGMNrs5853121614930713250.00170.0158−1.66270.528380.0727270.25862
LGMNrs1013216514930918290.00360.03471.41520.486350.881820.7069
LGMNrs1162184314931161240.00070.0071−1.95070.578410.0545450.25862
LGMNrs381483014931181980.00420.0399−1.32420.462120.10.2931
LGMNrs1013793414931939120.00350.03392.37740.814960.227270.034483
ANAPC11rs14494882217797665200.00680.0136−3.20321.18370.00909090.12069
ANAPC11rs2845494717797694660.02140.0428−1.12230.487790.109090.24138
ACP5rs222953119116871950.00110.0392−1.960.600670.0818180.24138
BP: Base pair location on DNA; β: Beta was calculated for minor allele and response; SE: Standard error; AF.RESP: Minor allele frequency responders; AF.NON: Minor allele frequency non-responders; * Same variant present as above.
Table
Table 3. Estimated and calculated transcript eQTLs.
Table 3. Estimated and calculated transcript eQTLs.
GENETRANSCRIPTdbSNPeQTL p Value ALLeQTL p Value HeQTL p Value D
MACF1ENST00000372925.2rs108886330.0840.4850.099
MACF1ENST00000372925.2rs3544132910.1340.119
CTSEENST00000360218.2rs283968060.3150.4390.142
CTSEENST00000581049.1rs283968060.3150.4390.142
HDLBPENST00000310931.4rs755905980.11510.197
HSPA9ENST00000501917.2rs68910070.8050.3190.887
HSPA9ENST00000501917.2rs283869470.0650.1020.331
HSPA9ENST00000501917.2rs589142680.4220.0770.187
HLA-DMBENST00000383231.2rs77618820.4620.2990.339
HLA-DMBENST00000428420.2rs77618820.4620.2990.339
HLA-DMBENST00000440078.2rs77618820.4620.2990.339
TAP2ENST00000443713.2rs77618820.0680.2350.218
TAP2ENST00000452371.2rs77618820.1810.3250.441
TAP2ENST00000455842.2rs77618820.1810.3250.441
TAP2ENST00000457634.2rs77618820.0830.3250.339
LGMNENST00000555169.1rs585312160.0040.0530.052
LGMNENST00000555169.1rs101321650.3450.1340.769
LGMNENST00000555169.1rs116218430.2110.6990.052
LGMNENST00000555169.1rs38148300.3550.6990.065
LGMNENST00000555169.1rs101379340.6370.6170.462
ANAPC11ENST00000583839.1rs1449488220.5010.1340.883
ANAPC11ENST00000583839.1rs284549470.39510.378
ACP5ENST00000218758.5rs22295310.5150.1210.624
H: healthy tissue; D: inflamed tissue.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Gorenjak, M.; Jezernik, G.; Krušič, M.; Skok, P.; Potočnik, U. Identification of Novel Loci Involved in Adalimumab Response in Crohn’s Disease Patients Using Integration of Genome Profiling and Isoform-Level Immune-Cell Deconvoluted Transcriptome Profiling of Colon Tissue. Pharmaceutics 2022, 14, 1893. https://doi.org/10.3390/pharmaceutics14091893

AMA Style

Gorenjak M, Jezernik G, Krušič M, Skok P, Potočnik U. Identification of Novel Loci Involved in Adalimumab Response in Crohn’s Disease Patients Using Integration of Genome Profiling and Isoform-Level Immune-Cell Deconvoluted Transcriptome Profiling of Colon Tissue. Pharmaceutics. 2022; 14(9):1893. https://doi.org/10.3390/pharmaceutics14091893

Chicago/Turabian Style

Gorenjak, Mario, Gregor Jezernik, Martina Krušič, Pavel Skok, and Uroš Potočnik. 2022. "Identification of Novel Loci Involved in Adalimumab Response in Crohn’s Disease Patients Using Integration of Genome Profiling and Isoform-Level Immune-Cell Deconvoluted Transcriptome Profiling of Colon Tissue" Pharmaceutics 14, no. 9: 1893. https://doi.org/10.3390/pharmaceutics14091893

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop