Matrix Proteases in Health and Disease

Niels Behrendt

• Publisher: Wiley-VCH
• Year: 2012
• Print ISBN: 9783527329915
• eBook ISBN: 9783527649341
• Edition: 1
• Topic: Scienze biologiche
• Subtopic: Biologia molecolare

Presenting a comprehensive overview of the multifaceted field of proteases in the extracellular matrix environment, this reference focuses on the recently elucidated functions of complex proteolytic systems in physiological and pathological tissue remodeling. The proteases treated include both serine proteases such as plasminogen activators and TTSPs, metalloproteases such as MMPs and ADAMS and cysteine protease cathepsins. The text specifically addresses the role of extracellular proteases in cancer cell invasion, stroke and infectious diseases, describing the basic biochemistry behind these disease states, as well as therapeutic strategies based on protease inhibition.

With its trans-disciplinary scope, this reference bridges the gap between fundamental research and biomedical and pharmaceutical application, making this required reading for basic and applied scientists in the molecular life sciences.

Chapter 1

Matrix Proteases and the Degradome

Clara Soria-Valles, Carlos López-Otín, and Ana Gutiérrez-Fernández

1.1 Introduction

Proteases are defined as enzymes that have the ability to perform the hydrolysis of peptide bonds. Owing to this characteristic, proteases were initially described as nonspecific enzymes of protein catabolism, participating in processes such as tissue destruction or degradation of dietary proteins. More recently, a better understanding of their functions has allowed consideration of proteases as enzymes that perform highly specific reactions and take part in multiple biological processes such as DNA replication and transcription, cell proliferation, differentiation and migration, tissue morphogenesis and remodeling, heat shock and unfolded protein responses, neurogenesis, angiogenesis, ovulation, fertilization, wound repair, stem cell mobilization, coagulation, immunity, inflammation, senescence, autophagy, apoptosis, and necrosis [1]. According to the essential roles performed by proteases in all living organisms, alterations in their proteolytic activities may lead to important pathologies such as arthritis, cardiovascular alterations, neurodegenerative disorders, progeroid syndromes, and cancer [2, 3].

The biochemical reaction catalyzed by all proteases consists in the hydrolysis of a peptide bond through the nucleophilic attack at the carbonyl group. However, the way this reaction is performed differs between specific proteases. This characteristic feature has allowed the establishment of six different catalytic classes of proteases according to the group performing the nucleophilic attack: aspartic, metallo, cysteine, serine, and threonine proteases, as well as the most recently described group of glutamic proteases, which has only been found in some species of fungi and bacteria. In the case of aspartic, glutamic, and metalloproteases, a polarized water molecule located in the active center acts directly as a nucleophile, while in the other three classes, the reactive element is a hydroxyl (serine and threonine) or sulfhydryl (cysteine) group from the corresponding catalytic core [4]. Within each class, proteases can be further subdivided into different families and clans according to sequence conservation and three-dimensional structure similarities.

The diversity and complexity of proteases have made necessary the introduction of concepts and tools for their global analysis and characterization. Thus, the term degradome defines the complete set of protease genes expressed by a cell, tissue, or organism at a specific moment or circumstance [5]. Likewise, the degradome of a certain protease is the complete substrate repertoire of that specific enzyme. The ability to explore the complexity of proteases in an organism has been catalyzed by the impressive advances in the sequencing and annotation of complete genomes. In fact, since the completion in 1995 of the genome of Haemophilus influenzae (1 830 140 bp and 1740 genes) [6], the number of available genome sequences has continuously increased at a rapid pace. Nowadays, the genome sequence for most model organisms as well as several vertebrate species and thousands of microorganisms is publicly available [7–13]. This genomic progress has made the study of degradomes more accessible to the scientific community. The information obtained from the analysis of the degradome of a certain organism constitutes an important tool for the comprehension of biological and pathological processes and could be the key to find out new ways to diagnose, treat, or prevent human pathologies.

In this chapter, we discuss available bioinformatic tools for the construction and analysis of degradomes and present an overview of characteristic features and evolutionary aspects of several degradomes of biomedical interest. We also discuss human diseases of proteolysis and, finally, introduce the different classes of proteases with ability to degrade the extracellular matrix (ECM), which is the topic of this book.

1.2 Bioinformatic Tools for the Analysis of Complex Degradomes

The development of novel molecular technologies has significantly reduced the time and cost of generating a genome sequence. However, important parts of the subsequent analysis, including the annotation of functional elements in the genome and the integration of this information into biological processes, are still a difficult task [14]. The complexity of genomic information, in which the coding sequence is interrupted by the presence of large introns, or the existence of numerous pseudogenes with high sequence identity to bona fide genes, hampers the use of straightforward bioinformatic approaches for both the reliable identification of genes and the prediction of protein structure and function. Therefore, although many genes can be directly annotated by using bioinformatic approaches, current tools have limitations in distinguishing genes from pseudogenes. Accordingly, manually supervised annotation is still the most common method to annotate complex sequences such as the human genome and degradome [15]. Another consideration when dealing with homologous sequences, as in the case of protease-coding genes, is the importance of distinguishing between different types of gene relationships. Thus, homologous genes can be classified as either orthologous genes, which have originated as a result of a speciation event derived from a single ancestral gene in the last common ancestor of two given species, or paralogous genes, which originate as a result of a duplication event within the same genome [16]. Orthologous genes generally maintain the same ancestral function, while paralogous genes tend to evolve and acquire novel functions.

One of the main aims of genome sequencing studies is the comparison of the complement of genes between different species. Using this approach, it is possible to identify the presence of novel genes in one species, or the loss of other genes in another species, providing clues about the molecular mechanisms underlying some of the physiological differences between them. Despite the fact that genomes are sequences of nucleic acids, the comparison of the protease repertoire between two genomes is generally performed using protein sequences, as they are more conserved than DNA sequences, and, therefore, they are more informative and the searches are more sensitive [15]. Nucleotide sequences are only compared when analyzing noncoding regions or to determine evolutionary parameters using protein-coding regions.

Another issue to take into account for gene prediction when using available genome sequences is the possibility of having assembly artifacts or sequencing errors in the analyzed sequence [15]. Thus, assembly artifacts usually lead to the collapse of clusters of highly similar genes into an artificial single copy gene. Owing to the fact that about 20% of protease genes are located in clusters [17], careful examination of those regions and additional experimental approaches should be performed to correctly annotate protease clusters. On the other hand, the presence of sequencing errors could lead to the annotation of real genes as pseudogenes, although detailed examination of sequence traces or resequencing of specific regions can solve these problems. Pseudogenes are nonfunctional copies of a gene and depending on the structure and mechanism of generation we can distinguish two main types: conventional pseudogenes and processed pseudogenes. Conventional pseudogenes usually originate from a functional copy of a gene that has been inactivated by mutation and afterwards will degenerate through the accumulation of new mutations. By contrast, processed pseudogenes are derived from the mRNA copy of a gene that is retrotransposed into the genome. The creation of pseudogenes has been a major mechanism in the evolution of the mammalian degradomes [18]; therefore, their study is important to get a global picture when comparing different genomes.

Numerous databases, web pages, and programs are freely available and constitute valuable tools for the study and comparison of complete genomes. Owing to the extension of these resources, we briefly introduce some remarkable tools and databases that can be useful for the genomic study of proteases:

Degradome database (http://degradome.uniovi.es/): resource containing manually annotated information about all proteases and protease inhibitor genes from humans, the chimpanzee, mouse, and rat organized in catalytic classes and families. This database also provides a catalog of human hereditary diseases of proteolysis or degradomopathies and additional information about protease structures, ancillary domains present in proteases, and differences between mammalian degradomes [3, 17, 19, 20].

MEROPS (http://merops.sanger.ac.uk): comprehensive annotation of proteases and inhibitors in all sequenced organisms. Proteases are classified at the protein domain level and those that are statistically and significantly similar in amino acid sequences are grouped in a family. If the families share a common ancestor, they are grouped in a clan. Protease inhibitors are classified in the same way [21, 22].

Other resources for extracting protease information are Ensembl (http://www.ensembl.org), which developed a software for maintaining an automatic annotation and analysis on selected eukaryotic genomes; InterPro (http://www.ebi.ac.uk/interpro), which includes tools for predicting functional sites and protein domains [23] and alignment search tools such as BLAST (http://www.ncbi.nlm.nih.gov/BLAST) and BLAT (http://genome.ucsc.edu). However, in some cases, these methodologies are unable to identify distantly related protease homologs because of the high divergence of their sequences. In these situations, it is necessary to use more sensitive approaches, such as the hidden Markov models or position-specific score matrix, which are probabilistic models that collect information of specific positions from multiple sequence alignments and apply this information for the recognition of protein or DNA sequences in the genome [15]. In this regard, an important tool when working with different protease sequences is the usage of multiple sequence alignment methods developed under the principle of hierarchical clustering. Clustal (http://www.ebi.ac.uk/clustal) is one of the most commonly used methods for hierarchical multiple alignment [24].

1.3 Evolution of Mammalian Degradomes

The availability of complete sequences from several mammalian genomes opens the possibility of performing comparative studies of degradomes between species. This might lead to the identification of either highly conserved elements or genetic differences occurring during mammalian evolution and contribute to clarify the molecular basis of biological pathways and pathologies involving proteolytic systems [25, 26].

1.3.1 Human Degradome

Shortly after the completion of the human genome sequence draft, we performed an exhaustive bioinformatic analysis to try to find out new human protease-coding genes with sequence similarity to proteases already described in other organisms [3, 20]. By using this methodology, 570 proteases and protease-related genes, as well as 150 protease inhibitor genes, were identified, representing more than 2% of the total genes in the human genome. Interestingly, a total of 93 of these protease-related genes encode functional proteins that are catalytically inactive because of substitutions in one or more residues critical to their proteolytic activity. These nonprotease homologs may regulate the activation of other proteases or their access to substrates or inhibitors, although their precise contribution to human biology is still poorly known [1].

Human proteases are divided into 5 catalytic classes (aspartic, metallo, serine, cysteine, and threonine proteases) and 68 families (Figure 1.1). The most abundant classes are metalloproteases (191 members), serine proteases (178 members), and cysteine proteases (153 members). Threonine and aspartic proteases are more specialized classes and, therefore, they only have 27 and 21 members, respectively. From the 68 families of proteolytic enzymes, the most densely populated is the S01 family of serine proteases that includes enzymes of interest in the context of the present book such as plasmin, uPA (urokinase plasminogen activator), tPA (tissue-type plasminogen activator), and TTSPs (type II transmembrane serine proteases). Other representative families in the human degradome are the C01 family of cysteine proteases, mainly formed by cathepsins; the M10 metalloproteases containing more than 20 different matrix metalloproteinases (MMPs) and the M12 metalloproteases that include ADAMs (a disintegrin with metalloprotease domains) and ADAMTSs (ADAMs with thrombospondin repeats). Although the function of many of these proteases is still unknown, most of them are highly conserved between humans and other mammals, indicating that they appeared...

Table of contents

• Cover
• Related Titles
• Title Page
• Copyright
• Preface
• List of Contributors
• Chapter : Introduction
• Chapter 1: Matrix Proteases and the Degradome
• Chapter 2: The Plasminogen Activation System in Normal Tissue Remodeling
• Chapter 3: Physiological Functions of Membrane-Type Metalloproteases
• Chapter 4: Bone Remodeling: Cathepsin K in Collagen Turnover
• Chapter 5: Type-II Transmembrane Serine Proteases: Physiological Functions and Pathological Aspects
• Chapter 6: Plasminogen Activators in Ischemic Stroke
• Chapter 7: Bacterial Abuse of Mammalian Extracellular Proteases during Tissue Invasion and Infection
• Chapter 8: Experimental Approaches for Understanding the Role of Matrix Metalloproteinases in Cancer Invasion
• Chapter 9: Plasminogen Activators and Their Inhibitors in Cancer
• Chapter 10: Protease Nexin-1 – a Serpin with a Possible Proinvasive Role in Cancer
• Chapter 11: Secreted Cysteine Cathepsins – Versatile Players in Extracellular Proteolysis
• Chapter 12: ADAMs in Cancer
• Chapter 13: Urokinase-Type Plasminogen Activator, Its Receptor and Inhibitor as Biomarkers in Cancer
• Chapter 14: Clinical Relevance of MMP and TIMP Measurements in Cancer Tissue
• Chapter 15: New Prospects for Matrix MetalloproteinaseTargeting in Cancer Therapy
• Index