Analysis of mass spectrometry data from the secretome of an explant model of articular cartilage exposed to pro-inflammatory and anti-inflammatory stimuli using machine learning

Chondroitin Sulfate/Hyaluronic Acid-Blended Hydrogels Suppress Chondrocyte Inflammation under Pro-Inflammatory Conditions

Osteoarthritis is characterized by enzymatic breakdown of the articular cartilage via the disruption of chondrocyte homeostasis, ultimately resulting in the destruction of the articular surface. Decades of research have highlighted the importance of inflammation in osteoarthritis progression, with inflammatory cytokines shifting resident chondrocytes into a pro-catabolic state. Inflammation can result in poor outcomes for cells implanted for cartilage regeneration. Therefore, a method to promote the growth of new cartilage and protect the implanted cells from the pro-inflammatory cytokines found in the joint space is required. In this study, we fabricate two gel types: polymer network hydrogels composed of chondroitin sulfate and hyaluronic acid, glycosaminoglycans (GAGs) known for their anti-inflammatory and prochondrogenic activity, and interpenetrating networks of GAGs and collagen I. Compared to a collagen-only hydrogel, which does not provide an anti-inflammatory stimulus, chondrocytes in GAG hydrogels result in reduced production of pro-inflammatory cytokines and enzymes as well as preservation of collagen II and aggrecan expression. Overall, GAG-based hydrogels have the potential to promote cartilage regeneration under pro-inflammatory conditions. Further, the data have implications for the use of GAGs to generally support tissue engineering in pro-inflammatory environments.

A human meniscus explant model for studying early events in osteoarthritis development by proteomics

Degenerative meniscus lesions have been associated with both osteoarthritis etiology and its progression. We, therefore, sought to establish a human meniscus ex vivo model to study the meniscal response to cytokine treatment using a proteomics approach. Lateral menisci were obtained from five knee-healthy donors. The meniscal body was cut into vertical slices and further divided into an inner (avascular) and outer region. Explants were either left untreated (controls) or stimulated with cytokines. Medium changes were conducted every 3 days up to Day 21 and liquid chromatography-mass spectrometry was performed at all the time points for the identification and quantification of proteins. Mixed-effect linear regression models were used for statistical analysis to estimate the effect of treatments versus control on protein abundance. Treatment by IL1ß increased release of cytokines such as interleukins, chemokines, and matrix metalloproteinases but a limited catabolic effect in healthy human menisci explants. Further, we observed an increased release of matrix proteins (collagens, integrins, prolargin, tenascin) in response to oncostatin M (OSM) + tumor necrosis factor (TNF) and TNF+interleukin-6 (IL6) + sIL6R treatments, and analysis of semitryptic peptides provided additional evidence of increased catabolic effects in response to these treatments. The induced activation of catabolic processes may play a role in osteoarthritis development.

In Vitro Models and Proteomics in Osteoarthritis Research

This review summarizes and exemplifies the current understanding of osteoarthritis in vitro models and describes their relevance for new insights in the future of osteoarthritis research. Our friend and highly appreciated colleague, Prof. Alan Grodzinsky has contributed greatly to the understanding of joint tissue biology and cartilage biomechanics. He frequently utilizes in vitro models and cartilage explant cultures, and recent work also includes proteomics studies. This review is dedicated to honor his 75-year birthday and will focus on recent proteomic in vitro studies related to osteoarthritis, and within this topic highlight some of his contributions to the field.

The application of machine learning in early diagnosis of osteoarthritis: a narrative review

Osteoarthritis (OA) is the commonest musculoskeletal disease worldwide, with an increasing prevalence due to aging. It causes joint pain and disability, decreased quality of life, and a huge burden on healthcare services for society. However, the current main diagnostic methods are not suitable for early diagnosing patients of OA. The use of machine learning (ML) in OA diagnosis has increased dramatically in the past few years. Hence, in this review article, we describe the research progress in the application of ML in the early diagnosis of OA, discuss the current trends and limitations of ML approaches, and propose future research priorities to apply the tools in the field of OA. Accurate ML-based predictive models with imaging techniques that are sensitive to early changes in OA ahead of the emergence of clinical features are expected to address the current dilemma. The diagnostic ability of the fusion model that combines multidimensional information makes patient-specific early diagnosis and prognosis estimation of OA possible in the future.

Machine Learning in Rheumatic Diseases

With advances in information technology, the demand for using data science to enhance healthcare and disease management is rapidly increasing. Among these technologies, machine learning (ML) has become ubiquitous and indispensable for solving complex problems in many scientific fields, including medical science. ML allows the development of guidelines and framing of the evaluation system for complex diseases based on massive data. In the analysis of rheumatic diseases, which are chronic and remarkably heterogeneous, ML can be anticipated to be extremely helpful in deciphering and revealing the inherent interrelationships in disease development and progression, which can further enhance the overall understanding of the disease, optimize patients' stratification, calibrate therapeutic strategies, and predict prognosis and outcomes. In this review, the basics of ML, its potential clinical applications in rheumatology, together with its strengths and limitations are summarized.

Chondrocyte secretome: a source of novel insights and exploratory biomarkers of osteoarthritis

The extracellular matrix (ECM) of articular cartilage is comprised of complex networks of proteins and glycoproteins, all of which are expressed by its resident cell, the chondrocyte. Cartilage is a unique tissue given its complexity and ability to resist repeated load and deformation. The mechanisms by which articular cartilage maintains its integrity throughout our lifetime is not fully understood, however there are numerous regulatory pathways known to govern ECM turnover in response to mechanical stimuli. To further our understanding of this field, we envision that proteomic analysis of the secretome will provide information on how the chondrocyte remodels the surrounding ECM in response to load, in addition to providing information on the metabolic state of the cell. In this review, we attempt to summarize the recent mass spectrometry-based proteomic discoveries in healthy and diseased cartilage and chondrocytes, to facilitate the discovery of novel biomarkers linked to degenerative pathologies, such as osteoarthritis (OA).

Label-free relative quantification of secreted proteins as a non-invasive method for the quality control of chondrogenesis in bioengineered substitutes for cartilage repair

Cartilage tissue engineering is making progress, but the competing available strategies still leave room for improvement and consensual overviews regarding the best combinations of scaffolds and cell sources are limited by the capacity to compare them directly. In addition, because most strategies involve autologous cell transfer, once these are optimized, the resulting implants require individual quality control prior to grafting in order to emphasize patient-to-patient differential responsiveness to engineering processes. Here, cartilage substitutes prepared from human mesenchymal stem cells undergoing chondrogenic differentiation within distinct scaffolds were used as pilot samples to investigate the pertinence of a novel method with the aim of characterizing the implants. The limits and advantages of analysing, by label-free liquid chromatography-coupled matrix-assisted laser desorption and ionization (LC-MALDI) mass spectrometry, the secreted proteome released into culture medium by engineered cartilage tissues were investigated and compared with more classically used methods for biomaterial characterization. This method did not require sacrificing the biomaterials and robustly evidenced their chondrogenic statuses. In more detail, the method highlighted differences between batches prepared from distinct donors. It was adapted to distinct scaffolds and allowed a comparison of the influence of individual engineering steps, such as growth factor combinations and oxygen tension. Finally, it evidenced subtle changes between replicate substitutes within a series, thereby distinguishing the least and most accomplished ones. We conclude that relative quantification of secreted proteins through label-free LC-MALDI will be useful, not only to orientate engineering methodologies, but also to ultimately provide non-invasive quality control of engineered tissue substitutes for the repair of cartilage and possibly other connective tissues.

Functional networks inference from rule-based machine learning models

BACKGROUND

Functional networks play an important role in the analysis of biological processes and systems. The inference of these networks from high-throughput (-omics) data is an area of intense research. So far, the similarity-based inference paradigm (e.g. gene co-expression) has been the most popular approach. It assumes a functional relationship between genes which are expressed at similar levels across different samples. An alternative to this paradigm is the inference of relationships from the structure of machine learning models. These models are able to capture complex relationships between variables, that often are different/complementary to the similarity-based methods.

RESULTS

We propose a protocol to infer functional networks from machine learning models, called FuNeL. It assumes, that genes used together within a rule-based machine learning model to classify the samples, might also be functionally related at a biological level. The protocol is first tested on synthetic datasets and then evaluated on a test suite of 8 real-world datasets related to human cancer. The networks inferred from the real-world data are compared against gene co-expression networks of equal size, generated with 3 different methods. The comparison is performed from two different points of view. We analyse the enriched biological terms in the set of network nodes and the relationships between known disease-associated genes in a context of the network topology. The comparison confirms both the biological relevance and the complementary character of the knowledge captured by the FuNeL networks in relation to similarity-based methods and demonstrates its potential to identify known disease associations as core elements of the network. Finally, using a prostate cancer dataset as a case study, we confirm that the biological knowledge captured by our method is relevant to the disease and consistent with the specialised literature and with an independent dataset not used in the inference process.

AVAILABILITY

The implementation of our network inference protocol is available at: http://ico2s.org/software/funel.html.

Applying computation biology and "big data" to develop multiplex diagnostics for complex chronic diseases such as osteoarthritis

The data explosion in the last decade is revolutionizing diagnostics research and the healthcare industry, offering both opportunities and challenges. These high-throughput “omics” techniques have generated more scientific data in the last few years than in the entire history of mankind. Here we present a brief summary of how “big data” have influenced early diagnosis of complex diseases. We will also review some of the most commonly used “omics” techniques and their applications in diagnostics. Finally, we will discuss the issues brought by these new techniques when translating laboratory discoveries to clinical practice.

Aging-related inflammation in osteoarthritis

It is well accepted that aging is an important contributing factor to the development of osteoarthritis (OA). The mechanisms responsible appear to be multifactorial and may include an age-related pro-inflammatory state that has been termed "inflamm-aging." Age-related inflammation can be both systemic and local. Systemic inflammation can be promoted by aging changes in adipose tissue that result in increased production of cytokines such as interleukin (IL)-6 and tumor necrosis factor-α (TNFα). Numerous studies have shown an age-related increase in blood levels of IL-6 that has been associated with decreased physical function and frailty. Importantly, higher levels of IL-6 have been associated with an increased risk of knee OA progression. However, knockout of IL-6 in male mice resulted in worse age-related OA rather than less OA. Joint tissue cells, including chondrocytes and meniscal cells, as well as the neighboring infrapatellar fat in the knee joint, can be a local source of inflammatory mediators that increase with age and contribute to OA. An increased production of pro-inflammatory mediators that include cytokines and chemokines, as well as matrix-degrading enzymes important in joint tissue destruction, can be the result of cell senescence and the development of the senescence-associated secretory phenotype (SASP). Further studies are needed to better understand the basis for inflamm-aging and its role in OA with the hope that this work will lead to new interventions targeting inflammation to reduce not only joint tissue destruction but also pain and disability in older adults with OA.

A machine learning heuristic to identify biologically relevant and minimal biomarker panels from omics data

BACKGROUND

Investigations into novel biomarkers using omics techniques generate large amounts of data. Due to their size and numbers of attributes, these data are suitable for analysis with machine learning methods. A key component of typical machine learning pipelines for omics data is feature selection, which is used to reduce the raw high-dimensional data into a tractable number of features. Feature selection needs to balance the objective of using as few features as possible, while maintaining high predictive power. This balance is crucial when the goal of data analysis is the identification of highly accurate but small panels of biomarkers with potential clinical utility. In this paper we propose a heuristic for the selection of very small feature subsets, via an iterative feature elimination process that is guided by rule-based machine learning, called RGIFE (Rule-guided Iterative Feature Elimination). We use this heuristic to identify putative biomarkers of osteoarthritis (OA), articular cartilage degradation and synovial inflammation, using both proteomic and transcriptomic datasets.

RESULTS AND DISCUSSION

Our RGIFE heuristic increased the classification accuracies achieved for all datasets when no feature selection is used, and performed well in a comparison with other feature selection methods. Using this method the datasets were reduced to a smaller number of genes or proteins, including those known to be relevant to OA, cartilage degradation and joint inflammation. The results have shown the RGIFE feature reduction method to be suitable for analysing both proteomic and transcriptomics data. Methods that generate large 'omics' datasets are increasingly being used in the area of rheumatology.

CONCLUSIONS

Feature reduction methods are advantageous for the analysis of omics data in the field of rheumatology, as the applications of such techniques are likely to result in improvements in diagnosis, treatment and drug discovery.

A computational method to differentiate normal individuals, osteoarthritis and rheumatoid arthritis patients using serum biomarkers

The objective of this study was to develop a method for categorizing normal individuals (normal, n = 100) as well as patients with osteoarthritis (OA, n = 100), and rheumatoid arthritis (RA, n = 100) based on a panel of inflammatory cytokines expressed in serum samples. Two panels of inflammatory proteins were used as training sets in the construction of two separate artificial neural networks (ANNs). The first training set consisted of all proteins (38 in total) and the second consisted of only the significantly different proteins expressed (12 in total) between at least two patient groups. Both ANNs obtained high levels of sensitivity and specificity, with the first and second ANN each diagnosing 100% of test set patients correctly. These results were then verified by re-investigating the entire dataset using a decision tree algorithm. We show that ANNs can be used for the accurate differentiation between serum samples of patients with OA, a diagnosed RA patient comparator cohort and normal/control cohort. Using neural network and systems biology approaches to manage large datasets derived from high-throughput proteomics should be further explored and considered for diagnosing diseases with complex pathologies.

Proteomic Analysis of Engineered Cartilage

Tissue engineering holds promise for the treatment of damaged and diseased tissues, especially for those tissues that do not undergo repair and regeneration readily in situ. Many techniques are available for cell and tissue culturing and differentiation of chondrocytes using a variety of cell types, differentiation methods, and scaffolds. In each case, it is critical to demonstrate the cellular phenotype and tissue composition, with particular attention to the extracellular matrix molecules that play a structural role and that contribute to the mechanical properties of the resulting tissue construct. Mass spectrometry provides an ideal analytical method with which to characterize the full spectrum of proteins produced by tissue-engineered cartilage. Using normal cartilage tissue as a standard, tissue-engineered cartilage can be optimized according to the entire proteome. Proteomic analysis is a complementary approach to biochemical, immunohistochemical, and mechanical testing of cartilage constructs. Proteomics is applicable as an analysis approach to most cartilage constructs generated from a variety of cellular sources including primary chondrocytes, mesenchymal stem cells from bone marrow, adipose tissue, induced pluripotent stem cells, and embryonic stem cells. Additionally, proteomics can be used to optimize novel scaffolds and bioreactor applications, yielding cartilage tissue with the proteomic profile of natural cartilage.

Osteoarthritis year in review 2014: we need more biochemical biomarkers in qualification phase

The current diagnosis of osteoarthritis (OA) relies on the description of pain symptoms, affected joint stiffness, and radiography used as the reference technique for determining the grade of joint destruction. Limitations of the presently available diagnostic tests have provided an impetus for the substantial increase in interest in finding new specific biological markers for cartilage degradation to facilitate the early diagnosis of joint destruction, evaluate disease progression and improve disease prognosis. Biomarkers for OA are also useful for drug development, treatment monitoring, and as a basis for personalized evidence-based action plans. This review summarizes 29 manuscripts published during 2013 with a focus on soluble biochemical biomarkers, primarily those utilizing proteomic, metabolomics, lipidomic and imaging mass spectrometry technologies.

Quantitative proteomic profiling of human articular cartilage degradation in osteoarthritis

Osteoarthritis (OA) is the most common rheumatic pathology and is characterized primarily by articular cartilage degradation. Despite its high prevalence, there is no effective therapy to slow disease progression or regenerate the damaged tissue. Therefore, new diagnostic and monitoring tests for OA are urgently needed, which would also promote the development of alternative therapeutic strategies. In the present study, we have performed an iTRAQ-based quantitative proteomic analysis of secretomes from healthy human articular cartilage explants, comparing their protein profile to those from unwounded (early disease) and wounded (advanced disease) zones of osteoarthritic tissue. This strategy allowed us to identify a panel of 76 proteins that are distinctively released by the diseased tissue. Clustering analysis allowed the classification of proteins according to their different profile of release from cartilage. Among these proteins, the altered release of osteoprotegerin (decreased in OA) and periostin (increased in OA), both involved in bone remodelling processes, was verified in further analyses. Moreover, periostin was also increased in the synovial fluid of OA patients. Altogether, the present work provides a novel insight into the mechanisms of human cartilage degradation and a number of new cartilage-characteristic proteins with possible biomarker value for early diagnosis and prognosis of OA.

Hard Data Analytics Problems Make for Better Data Analysis Algorithms: Bioinformatics as an Example

Data mining and knowledge discovery techniques have greatly progressed in the last decade. They are now able to handle larger and larger datasets, process heterogeneous information, integrate complex metadata, and extract and visualize new knowledge. Often these advances were driven by new challenges arising from real-world domains, with biology and biotechnology a prime source of diverse and hard (e.g., high volume, high throughput, high variety, and high noise) data analytics problems. The aim of this article is to show the broad spectrum of data mining tasks and challenges present in biological data, and how these challenges have driven us over the years to design new data mining and knowledge discovery procedures for biodata. This is illustrated with the help of two kinds of case studies. The first kind is focused on the field of protein structure prediction, where we have contributed in several areas: by designing, through regression, functions that can distinguish between good and bad models of a protein's predicted structure; by creating new measures to characterize aspects of a protein's structure associated with individual positions in a protein's sequence, measures containing information that might be useful for protein structure prediction; and by creating accurate estimators of these structural aspects. The second kind of case study is focused on omics data analytics, a class of biological data characterized for having extremely high dimensionalities. Our methods were able not only to generate very accurate classification models, but also to discover new biological knowledge that was later ratified by experimentalists. Finally, we describe several strategies to tightly integrate knowledge extraction and data mining in order to create a new class of biodata mining algorithms that can natively embrace the complexity of biological data, efficiently generate accurate information in the form of classification/regression models, and extract valuable new knowledge. Thus, a complete data-to-information-to-knowledge pipeline is presented.