Protein function microarrays for customised systems-oriented proteome analysis

Identification of autoantibodies as potential non-invasive biomarkers for intrahepatic cholangiocarcinoma

Intrahepatic cholangiocarcinoma (iCCA) presents a challenging diagnosis due to its nonspecific early clinical manifestations, often resulting in late-stage detection and high mortality. Diagnosing iCCA is further complicated by its limited accuracy, often necessitating multiple invasive procedures for precise identification. Despite carbohydrate antigen 19-9 (CA19-9) having been investigated and employed for iCCA diagnosis, it demonstrates modest diagnostic performance. Consequently, the identification of novel biomarkers with improved sensitivity and specificity remains an imperative yet formidable task. Autoantibodies, as early indicators of the immune response against cancer, offer a promising avenue for enhancing diagnostic accuracy. Our study aimed to identify non-invasive blood-based autoantibody biomarkers capable of distinguishing iCCA patients from healthy individuals (CTRs). We profiled autoantibodies in 26 serum samples (16 iCCAs and 10 CTRs) using protein microarrays containing 1622 functional proteins. Leveraging machine learning techniques, we identified a signature composed of three autoantibody biomarkers (NDE1, PYCR1, and VIM) in conjunction with CA19-9 for iCCA detection. This combined signature demonstrated superior diagnostic performance with an AUC of 96.9%, outperforming CA19-9 alone (AUC: 83.8%). These results suggest the potential of autoantibody biomarkers to develop a complementary non-invasive diagnostic utility for routine iCCA screening.

Integrating circulating T follicular memory cells and autoantibody repertoires for characterization of autoimmune disorders

Introduction

Autoimmune diseases are heterogeneous and often lack specific or sensitive diagnostic tests. Increased percentages of CD4+CXCR5+PD1+ circulating T follicular helper (cTfh) cells and skewed distributions of cTfh subtypes have been associated with autoimmunity. However, cTfh cell percentages can normalize with immunomodulatory treatment despite persistent disease activity, indicating the need for identifying additional cellular and/or serologic features correlating with autoimmunity.

Methods

The cohort included 50 controls and 56 patients with autoimmune cytopenias, gastrointestinal, pulmonary, and/or neurologic autoimmune disease. Flow cytometry was used to measure CD4+CXCR5+ T cell subsets expressing the chemokine receptors CXCR3 and/or CCR6: CXCR3+CCR6- Type 1, CXCR3-CCR6- Type 2, CXCR3+CCR6+ Type 1/17, and CXCR3- CCR6+ Type 17 T cells. IgG and IgA autoantibodies were quantified using a microarray featuring 1616 full-length, conformationally intact protein antigens. The 97.5th percentile in the control cohort defined normal limits for T cell subset percentages and total number (burden) of autoantibodies.

Results

This study focused on CD4+CXCR5+ T cells because CXCR5 upregulation occurs after cognate T-B cell interactions characteristic of autoimmune diseases. We refer to these cells as circulating T follicular memory (cTfm) cells to acknowledge the dynamic nature of antigen-experienced CXCR5+ T cells, which encompass progenitors of cTfh or Tfh cells as well as early effector memory T cells that have not yet lost CXCR5. Compared to controls, 57.1% of patients had increased CXCR5+CXCR3+CCR6+ cTfm1/17 and 25% had increased CXCR5+CXCR3-CCR6+ cTfm17 cell percentages. Patients had significantly more diverse IgG and IgA autoantibodies than controls and 44.6% had an increased burden of autoantibodies of either isotype. Unsupervised autoantibody clustering identified three clusters of patients with IgG autoantibody profiles distinct from those of controls, enriched for patients with active autoimmunity and monogenic diseases. An increased percentage of cTfm17 cells was most closely associated with an increased burden of high-titer IgG and IgA autoantibodies. A composite measure integrating increased cTfm1/17, cTfm17, and high-titer IgG and/or IgA autoantibodies had 91.1% sensitivity and 90.9% specificity for identifying patients with autoimmunity. Percentages of cTfm1/17 and cTfm17 percentages and numbers of high-titer autoantibodies in patients receiving immunomodulatory treatment did not differ from those in untreated patients, thus suggesting that measurements of cTfm can complement measurements of other cellular markers affected by treatment.

Conclusions

This study highlights two new approaches for assessing autoimmunity: measuring CD4+CXCR5+ cTfm subsets as well as total burden of autoantibodies. Our findings suggest that these approaches are particularly relevant to patients with rare autoimmune disorders for whom target antigens and prognosis are often unknown.

Peritoneal autoantibody profiling identifies p53 as an autoantibody target in endometriosis

OBJECTIVE

To map the peritoneal autoantibody (AAb) landscape in women with endometriosis.

DESIGN

Case-control laboratory study.

SETTING

Academic medical and research units.

PATIENT(S)

Women who presented with or without endometriosis.

INTERVENTION(S)

None.

MAIN OUTCOME MEASURE(S)

Using native-conformation and citrullinated modified protein arrays, proteome-wide analysis of AAbs against 1,623 proteins were profiled in peritoneal fluids (PFs) of 25 women with endometriosis and 25 women without endometriosis.

RESULT(S)

In women with endometriosis, the median number of AAbs detected was 4, including AAbs that targeted autoantigens involved in implantation, B-cell activation/development, and aberrant migration and mitogenicity. Forty-six percent of women with endometriosis have ≥5 peritoneal AAbs. Conversely, in women without endometriosis, the median number of detected AAbs was 1. Autoantibodies recognizing tumor suppressor protein p53 were the most commonly detected AAbs, being present in 35% of women with endometriosis, and p53 AAb was associated with a monocyte/macrophage-like PF cytokine signature. Further investigation of the global reactivity of AAbs against citrullinated PF antigens by peptidylarginine deiminase enzymes 1, 2, and 6 revealed anticitrullinated p53 as the only AAb target elevated and citrullinated by all 3 peptidylarginine deiminase isotypes. Furthermore, unsupervised hierarchical clustering and integrative pathway analysis revealed that 60% of women with endometriosis-associated infertility were positive for AAbs, which are involved in platelet-derived growth factor, transforming growth factor-β, RAC1/PAK1/p38/MMP2 signaling, LAT2/NTAL/LAB-mediated calcium mobilization, and integrin-mediated cell adhesion.

CONCLUSION(S)

Together, our data identify peritoneal autoimmunity in a significant subset of women with endometriosis, with implications on infertility and disease pathophysiology. In these patients, p53 was identified as the most frequent PF AAb target, which was present in both the native and citrullinated forms.

Identification of novel autoantibodies in ascites of relapsed paclitaxel-resistant gastric cancer with peritoneal metastasis using immunome protein microarrays and proteomics

BACKGROUND

Gastric cancer (GC) patients with peritoneal metastasis usually have extremely poor prognosis. Intraperitoneal infusion of paclitaxel (PTX) provides an effective treatment, but relapse and PTX-resistance are unavoidable disadvantages, and it is difficult to monitor the occurrence of PTX-resistance.

OBJECTIVE

The aim of this study was to explore novel autoantibodies in the ascites of individuals with relapsed PTX-resistant GC with peritoneal metastasis.

METHODS

Ascites samples were collected before PTX infusion and after the relapse in 3 GC patients. To determine the expression of significantly changed proteins, we performed autoantibody profiling with immunome protein microarrays and tandem mass tag (TMT) quantitative proteomics, and then, the overlapping proteins were selected.

RESULTS

Thirty-eight autoantibodies that were differentially expressed between the ascites in the untreated group and relapsed PTX-resistant group were identified. For confirmation of the results, TMT quantitative proteomics was performed, and 842 dysregulated proteins were identified. Four proteins, TPM3, EFHD2, KRT19 and vimentin, overlapped between these two assays.

CONCLUSIONS

Our results first revealed that TPM3, EFHD2, KRT19 and vimentin were novel autoantibodies in the ascites of relapsed PTX-resistant GC patients. These autoantibodies may be used as potential biomarkers to monitor the occurrence of PTX-resistance.

A Novel Method to Identify Autoantibodies against Putative Target Proteins in Serum from beta-Thalassemia Major: A Pilot Study

Abnormal immune reactivity in patients with beta-thalassemia (beta-thal) major can be associated with poor prognosis. Immunome protein-array analysis represents a powerful approach to identify novel biomarkers. The Sengenics Immunome Protein Array platform was used for high-throughput quantification of autoantibodies in 12 serum samples collected from nine beta-thal major patients and three non-thalassemia controls, which were run together with two pooled normal sera (Sengenics Internal QC samples). To obtain more accurate and reliable results, the evaluation of the biological relevance of the shortlisted biomarkers was analyzed using an Open Target Platform online database. Elevated autoantibodies directed against 23 autoantigens on the immunome array were identified and analyzed using a penetrance fold change-based bioinformatics method. Understanding the autoantibody profile of beta-thal major patients would help to further understand the pathogenesis of the disease. The identified autoantigens may serve as potential biomarkers for the prognosis of beta-thal major.

Advances in the development of human protein microarrays

INTRODUCTION

High-content protein microarrays in principle enable the functional interrogation of the human proteome in a broad range of applications, including biomarker discovery, profiling of immune responses, identification of enzyme substrates, and quantifying protein-small molecule, protein-protein and protein-DNA/RNA interactions. As with other microarrays, the underlying proteomic platforms are under active technological development and a range of different protein microarrays are now commercially available. However, deciphering the differences between these platforms to identify the most suitable protein microarray for the specific research question is not always straightforward. Areas covered: This review provides an overview of the technological basis, applications and limitations of some of the most commonly used full-length, recombinant protein and protein fragment microarray platforms, including ProtoArray Human Protein Microarrays, HuProt Human Proteome Microarrays, Human Protein Atlas Protein Fragment Arrays, Nucleic Acid Programmable Arrays and Immunome Protein Arrays. Expert commentary: The choice of appropriate protein microarray platform depends on the specific biological application in hand, with both more focused, lower density and higher density arrays having distinct advantages. Full-length protein arrays offer advantages in biomarker discovery profiling applications, although care is required in ensuring that the protein production and array fabrication methodology is compatible with the required downstream functionality.

PMA: Protein Microarray Analyser, a user-friendly tool for data processing and normalization

Objective

Protein microarrays provide a high-throughput platform to measure protein interactions and associated functions, and can aid in the discovery of cancer biomarkers. The resulting protein microarray data can however be subject to systematic bias and noise, thus requiring a robust data processing, normalization and analysis pipeline to ensure high quality and robust results. To date, a comprehensive data processing pipeline is yet to be developed. Furthermore, a lack of analysis consistency is evident amongst different research groups, thereby impeding collaborative data consolidation and comparison. Thus, we sought to develop an accessible data processing tool using methods that are generalizable to the protein microarray field and which can be adapted to individual array layouts with minimal software engineering expertise.

Results

We developed an improved version of a previously developed pipeline of protein microarray data processing and implemented it as an open source software tool, with particular focus on widening its use and applicability. The Protein Microarray Analyser software presented here includes the following tools: (1) neighbourhood background correction, (2) net intensity correction, (3) user-defined noise threshold, (4) user-defined CV threshold amongst replicates and (5) assay controls, (6) composite ‘pin-to-pin’ normalization amongst sub-arrays, and (7) ‘array-to-array’ normalization amongst whole arrays.

Electronic supplementary material

The online version of this article (10.1186/s13104-018-3266-0) contains supplementary material, which is available to authorized users.

Novel potential serological prostate cancer biomarkers using CT100+ cancer antigen microarray platform in a multi-cultural South African cohort

There is a growing need for high throughput diagnostic tools for early diagnosis and treatment monitoring of prostate cancer (PCa) in Africa. The role of cancer-testis antigens (CTAs) in PCa in men of African descent is poorly researched. Hence, we aimed to elucidate the role of 123 Tumour Associated Antigens (TAAs) using antigen microarray platform in blood samples (N = 67) from a South African PCa, Benign prostatic hyperplasia (BPH) and disease control (DC) cohort. Linear (fold-over-cutoff) and differential expression quantitation of autoantibody signal intensities were performed. Molecular signatures of candidate PCa antigen biomarkers were identified and analyzed for ethnic group variation. Potential cancer diagnostic and immunotherapeutic inferences were drawn. We identified a total of 41 potential diagnostic/therapeutic antigen biomarkers for PCa. By linear quantitation, four antigens, GAGE1, ROPN1, SPANXA1 and PRKCZ were found to have higher autoantibody titres in PCa serum as compared with BPH where MAGEB1 and PRKCZ were highly expressed. Also, p53 S15A and p53 S46A were found highly expressed in the disease control group. Statistical analysis by differential expression revealed twenty-four antigens as upregulated in PCa samples, while 11 were downregulated in comparison to BPH and DC (FDR = 0.01). FGFR2, COL6A1and CALM1 were verifiable biomarkers of PCa analysis using urinary shotgun proteomics. Functional pathway annotation of identified biomarkers revealed similar enrichment both at genomic and proteomic level and ethnic variations were observed. Cancer antigen arrays are emerging useful in potential diagnostic and immunotherapeutic antigen biomarker discovery.

Specific Capture of Peptide-Receptive Major Histocompatibility Complex Class I Molecules by Antibody Micropatterns Allows for a Novel Peptide-Binding Assay in Live Cells

Binding assays with fluorescently labeled ligands and recombinant receptor proteins are commonly performed in 2D arrays. But many cell surface receptors only function in their native membrane environment and/or in a specific conformation, such as they appear on the surface of live cells. Thus, receptors on live cells should be used for ligand binding assays. Here, it is shown that antibodies preprinted on a glass surface can be used to specifically array a peptide receptor of the immune system, i.e., the major histocompatibility complex class I molecule H-2K^b , into a defined pattern on the surface of live cells. Monoclonal antibodies make it feasible to capture a distinct subpopulation of H-2K^b and hold it at the cell surface. This patterned receptor enables a novel peptide-binding assay, in which the specific binding of a fluorescently labeled index peptide is visualized by microscopy. Measurements of ligand binding to captured cell surface receptors in defined confirmations apply to many problems in cell biology and thus represent a promising tool in the field of biosensors.

Development of a novel, quantitative protein microarray platform for the multiplexed serological analysis of autoantibodies to cancer-testis antigens

The cancer-testis antigens are a group of unrelated proteins aberrantly expressed in various cancers in adult somatic tissues. This aberrant expression can trigger spontaneous immune responses, a phenomenon exploited for the development of disease markers and therapeutic vaccines. However, expression levels often vary amongst patients presenting the same cancer type, and these antigens are therefore unlikely to be individually viable as diagnostic or prognostic markers. Nevertheless, patterns of antigen expression may provide correlates of specific cancer types and disease progression. Herein, we describe the development of a novel, readily customizable cancer-testis antigen microarray platform together with robust bioinformatics tools, with which to quantify anti-cancer testis antigen autoantibody profiles in patient sera. By exploiting the high affinity between autoantibodies and tumor antigens, we achieved linearity of response and an autoantibody quantitation limit in the pg/mL range-equating to a million-fold serum dilution. By using oriented attachment of folded, recombinant antigens and a polyethylene glycol microarray surface coating, we attained minimal non-specific antibody binding. Unlike other proteomics methods, which typically use lower affinity interactions between monoclonal antibodies and tumor antigens for detection, the high sensitivity and specificity realized using our autoantibody-based approach may facilitate the development of better cancer biomarkers, as well as potentially enabling pre-symptomatic diagnosis. We illustrated the usage of our platform by monitoring the response of a melanoma patient cohort to an experimental therapeutic NY-ESO-1-based cancer vaccine; inter alia, we found evidence of determinant spreading in individual patients, as well as differential CT antigen expression and epitope usage.