Introduction

Moonlighting proteins, also referred to as ‘gene sharing’, refer to a subset of multifunctional proteins in which two or more different functions are performed by one polypeptide chain, and the multiple functions are not a result of splice variants, gene fusions, or multiple isoforms [1]. In addition, they do not include proteins with the same function in multiple locations or protein families in which different members have different functions, if each individual member has only one function. A single protein with multiple functions may seem surprising, but there are actually many cases of proteins that ‘moonlight’.

Examples and mechanisms of combining two functions in one protein

The current examples of moonlighting proteins include enzymes, DNA binding proteins, receptors, transmembrane channels, chaperones and ribosomal proteins (Table 4.1). In general, there are several different methods by which a moonlighting protein can combine two functions within one polypeptide chain. A single protein can have a second function when it moves to a different cellular location; when it is expressed in a different cell type; when it binds a substrate, product, or cofactor; when it interacts with another protein to form a multimer, or when it interacts with a large multiprotein complex. In addition, a few enzymes have two active sites for different substrates (Figure 4.1). The methods are not mutually exclusive and sometimes a combination of methods is employed.

Cellular location: Several cytosolic or nuclear enzymes have a second function outside of the cell.

Introduction

Vasoactive intestinal peptide (VIP), which was originally discovered in the intestine as a 28–amino acid peptide and shown to induce vasodilation, was later found to be a major brain peptide with neuroprotective activities in vivo [1–5]. To exert neuroprotective activity in the brain, VIP requires glial cells that secrete protective proteins such as activity-dependent neurotrophic factor (ADNF [6]). ADNF, isolated by sequential chromatographic methods, was named activity-dependent neurotrophic factor because it protects neurons from death associated with the blockade of electrical activity.

ADNF is a 14-kDa protein, and structure-activity studies have identified femtomolar-active neuroprotective peptides, ADNF-14 (VLGGGSALLRSIPA) [6] and ADNF-9 (SALLRSIPA) [7]. ADNF-9 exhibits protective activity in Alzheimer's disease–related systems (β-amyloid toxicity [7], presenilin 1 mutation [8], apolipoprotein E deficiencies [9] – genes that have been associated with the onset and progression of Alzheimer's disease (AD)). Other studies have indicated protection against oxidative stress via the maintenance of mitochondrial function and a reduction in the accumulation of intracellular reactive oxygen species [10]. In the target neurons, ADNF-9 regulates transcriptional activation associated with neuroprotection (nuclear factor-κB [11]), promotes axonal elongation through transcriptionally regulated cAMP-dependent mechanisms [12] and increases chaperonin 60 (Cpn60/Hsp60) expression, thereby providing cellular protection against the β-amyloid peptide [13].

Longer peptides that include the ADNF-9 sequence (e.g., ADNF-14) activate protein kinase C and mitogen-associated protein kinase kinase and protect developing mouse brain against excitotoxicity [14].

Introduction

Like a Brian Rix farce, in which the characters' identities are continuously changing, the functions of the class of protein known as molecular chaperones has been unfolding continuously over the past decade resulting in substantial confusion. However, like such farces, we are confident that the dénouement will be a complete surprise and will provide a new world picture of the processes with which molecular chaperones are involved. This short chapter aims to introduce the reader to the rapidly changing world of molecular chaperones as an aid to the reading of the rest of the chapters in this volume.

Molecular chaperones are protein folders

Our story starts with a huff and a puff with the study of the response of the polytene chromosomes of Drosophila to various stressors. This revealed novel patterns of specific chromosomal puffs, in response to heat, and a variety of other environmental stresses, representing the transcription of selected genes [1, 2]. The behaviour of cells exposed to various stresses became known as the heat shock response or the cell stress response and we now appreciate the very large number of environmental factors to which cells will respond in this stereotypical manner. The ‘molecularisation’ of the cell stress response occurred in the late 1980s with the pioneering work of Ellis and colleagues [3], who introduced both the concept of protein chaperoning and the term molecular chaperone.

Introduction

Intercellular communications are fundamental for many of the biological processes that are involved in the survival of living organisms, and secretory proteins are among the most important messengers in this network of information. Proteins destined for this function are endowed with a hydrophobic signal peptide which targets them to the endoplasmic reticulum (ER) and are released in the extracellular environment by a ‘classical’ pathway of constitutive or regulated secretion. However, in the early 1990s it became evident that non-classical mechanisms must exist for the secretion of some proteins which, despite their extracellular localisation and function, lack a signal peptide. Indeed, the family of these leaderless secretory proteins continues to grow and comprises proteins that, although apparently unrelated, share both structural and functional features. This chapter will review current hypotheses on the mechanisms underlying non-classical secretion and discuss their implications in the regulation of the inflammatory and immune response. The relevance of non-classical secretion pathways to molecular chaperone biology is also discussed in Chapters 2 and 12.

Leaderless secretory proteins

Secretory mechanisms that are discrete to the classical pathways appear early in evolution. Gram negative bacteria are endowed with many (up to six) types of secretion mechanisms that are, at least in part, independent of the general secretory pathway, the prototype being the haemolysin secretion system [1]. In addition, two pathways of secretion that avoid the ER exist in yeast.

Introduction

Many signalling molecules such as steroid hormone receptors and other receptors, protein kinases and phosphatases are found associated with various types of heat shock proteins, including Hsp90, Hsp70, Hsp40 and other co-chaperones. The functional role of these associations appears to be multi-faceted and the association of signalling proteins with these chaperone cohorts plays a pivotal role in initial folding and maturation of steroid hormone receptors and many kinases (e.g., Src). In addition, association with Hsp90 and its co-chaperones is critical for the stability of signalling proteins, because inhibition of Hsp90 by geldanamycin and other specific inhibitors leads to rapid ubiquitin-dependent degradation of Raf-1, Akt and other kinases that normally associate with Hsp90 [1, 2]. In fact, the anti-cancer activities of Hsp90 inhibitors could be related to the degradation and downregulation of signalling pathways that are controlled by these kinases [1, 3]. In contrast to Hsp90, which protects from degradation, an association with Hsp70 might target these proteins for rapid ubiquitination (usually via a ubiquitin ligase CHIP) followed by proteolysis [4].

In addition to their critical role in folding, maturation and stability of various signalling components, chaperones may be directly involved in regulation of their activities. In fact, it appears that Hsp70 and other chaperones play a regulatory role in the activation of many signalling pathways that are elicited by heat shock and other stresses.

There are multiple members of the Hsp70 protein family.

Predicting the future of the exobiology of molecular chaperones is bound to be risky business: after all, unravelling the intracellular lives of the chaperones has become legendary for its unexpected twists and turns. Can we expect differently for their extracellular capers? I cannot claim the clearest crystal, but I do have a unique perspective on the field from my perch as Editor-in-Chief of the major specialty journal in the field, Cell Stress & Chaperones. I will refer to papers in recent issues that will lead interested readers to other papers in key areas that I believe provide insights into the future as well. Perhaps we can begin to illuminate the crystal ball by listing major unsolved problems and by identifying the disciplines of the investigators that these problems are now attracting into the field.

One of the exciting and renewing aspects of the heat shock field, as it was known historically, has been the succession of colleagues from different disciplines that have entered and moved the field forward. The chance initial finding of the heat shock response in Drosophila by Ritossa in 1962 [1] was pursued by a small group of Drosophila biologists until about 1978 when the response was discovered in a variety of other organisms. Molecular geneticists were attracted to the heat shock genes as models of inducible eukaryotic gene expression, and the field took on a more global interest.

Introduction

The heat shock proteins (Hsps) are a group of highly conserved proteins that have major physiological roles in protein homeostasis [1, 2]. In most cell types, 1–2% of total proteins consist of heat shock proteins even prior to stress, which suggests important roles for these proteins in the biology and physiology of the unstressed cell. These roles particularly concern regulating the folding and unfolding of other proteins. The term ‘heat shock proteins’ was coined because these proteins were first identified on the basis of their increased synthesis following exposure to elevated temperatures [3]. Subsequently, it has been clearly shown that they can be induced following a variety of stressful stimuli. Some heat shock proteins, such as Hsp90 (each heat shock protein is named according to its mass in kilodaltons – see Chapter 1 for more details), are detectable at significant levels in unstressed cells and increase in abundance following a suitable stimulus, whereas others such as Hsp70 exist in both constitutively expressed and inducible forms [4, 5].

The dual role of heat shock proteins in both normal and stressed cells evidently requires the existence of complex regulatory processes which ensure that the correct expression pattern is produced. Indeed, such processes must be operative at the very earliest stages of embryonic development, since the genes encoding Hsp70 and Hsp90 are amongst the first embryonic genes to be transcribed [6, 7].

Introduction

The term molecular chaperone came into general use after the appearance of an article in Nature that suggested it was an appropriate phrase to describe a newly defined intracellular function – the ability of several unrelated protein families to assist the correct folding and assembly/disassembly of other proteins [1]. The identification of the chaperonin family of molecular chaperones in the following year [2] triggered a tidal wave of research in several laboratories aimed at unravelling how the GroEL/GroES chaperones, and later the DnaK/DnaJ chaperones, from Escherichia coli facilitate the folding of newly synthesised polypeptide chains and the refolding of denatured proteins. This wave continues to surge, with the result that much detailed information is available about the structure and function of those families of chaperone that assist protein folding [3].

It is now well established that a subset of proteins requires this chaperone function, not because chaperones provide steric information required for correct folding but because chaperones inhibit side reactions that would otherwise cause some of the chains to form non-functional aggregates. The number of different protein families described as chaperones is now more than 25 – some, but not all, of which are also stress proteins – and there is no slackening in the rate of discovery of new ones. The success of this wave of research has changed the paradigm of protein folding from the earlier view that it is a spontaneous self-assembly process to the current view that it is an assisted self-assembly process [4].

Introduction

Tumour antigens can be broadly classified into four categories: (i) those that are expressed in larger quantities in tumours than their normal counterparts (e.g., tumour-associated carbohydrate antigens) [1], (ii) onco-fetal antigens (e.g., carcinoembryonic antigen) [2], (iii) differentiation antigens (e.g., melanoma differentiation antigen) [3, 4] and (iv) tumour-specific antigens. Tumour antigens in the first three categories could serve as useful markers for diagnostic and prognostic purposes. Although some of these antigens are being used in immunotherapy, none can be called tumour-specific in a true sense. Only the last group includes antigens that are truly specific for tumour cells, in that they contain tumour-specific mutations that are unique for individual tumours such as the tumour-specific point mutation that is found in cyclin-dependent kinase-4. Such a mutation gives rise to a novel antigenic epitope which can be recognised by cytotoxic T lymphocytes (CTLs) [5]. However, for these antigens to be of any value as therapeutic agents, they must be detected in and epitopes isolated from a large range of cancers, and this makes the general use of these antigens difficult.

In the past two decades, evidence has accumulated to support the concept that molecular chaperones or heat shock proteins can be used as a potent source of cancer vaccines [6, 7]. Molecular chaperones, particularly those derived from the Hsp70 and Hsp90 families, are now being tested in the clinical arena for therapeutic efficacy against a range of cancers (Table 18.1).

Introduction

Although for many years the perception has been that mammalian heat shock proteins are intracellular molecules that are only released into the extracellular environment in pathological situations such as necrotic cell death, it is now known that these molecules can be released from a variety of viable (non-necrotic) cell types [1–4]. Moreover, we and a number of others have reported Hsp60 and/or Hsp70 to be present in the peripheral circulation of normal individuals [5–12]. These observations have profound implications for the perceived role of these proteins as pro-inflammatory intercellular ‘danger’ signalling molecules and have prompted a re-evaluation of the functional significance and role(s) of these ubiquitously expressed and highly conserved families of molecules. The reader should refer to Chapter 2, which discusses the intracellular dispositions of molecular chaperones and also touches on the release of heat shock proteins, and Chapter 3, in which novel pathways of protein release are described.

The mechanism(s) leading to the release of heat shock proteins are unknown, as is the source of circulating heat shock proteins in the peripheral circulation and their physiological and pathophysiological role(s). The inverse relationship between levels of circulating Hsp70 and the progression of carotid atherosclerosis [13], or the presence of coronary artery disease (CAD) [14], appears to be inconsistent with the concept that this molecule is a danger signal and an in vitro activator of innate and pro-inflammatory immunity [15].

Discovery of Toll-like receptors

The basic concept of the immune system postulates an ability to discriminate between self and non-self and to free the organism from the latter. Two major contributions advanced the comprehension of the cellular basis of self- versus non-self-discrimination. The first was the hypothesis regarding the expansion of antigen-recognising clones on encounter with a respective antigen, which allowed antigenic specificities of the resulting immune reactions to be explained. The co-stimulatory signal hypothesis represented another essential advancement. It postulated the necessity of a second, antigen-independent signal for lymphocyte activation. Its nature was put into an elegant metaphor of the ‘immunologist's dirty little secret’ [1], referring to substances of microbial origin that should be present concomitant with an antigen to prime an immune response to it.

Of a number of host receptors participating in detection of microbial constituents [2], Toll-like receptors (TLRs) currently represent the most interesting group. Their importance is assumed from the prominent cell activating capacity which they display after engagement with their cognate ligands. The name originates from the Drosophila homologue Toll, which was discovered as a part of the dorsoventral patterning cascade during the developmental larva stage of the fruit fly, and this seminal study established an additional, anti-microbial function for Toll in adult flies [3]. It demonstrated that mutants of the genes in the cassette between the Toll ligand Spätzle down to the IκB homologue Cactus showed a compromised inducibility of the anti-fungal peptide drosomycin upon fungal challenge and consequently succumbed to the infection.

Introduction

This chapter reviews work on the intracellular disposition of a number of molecular chaperones that are generally believed to be localised and function mainly within the mitochondria of eukaryotic cells. However, in recent years, compelling evidence has accumulated from many lines of investigation indicating that several of these mitochondrial (m-) chaperones are also localised and perform important functions at a variety of other sites/compartments within cells (see [1, 2]). The four chaperone proteins that are the subjects of this chapter include the following: (i) the 60-kDa heat shock chaperonin protein (Hsp60, also known as chaperonin 60, Cpn60), which is a major protein in both stressed and unstressed cells and plays an essential role in the proper folding and assembly into oligomeric complexes of other proteins [3–6]; (ii) the 10-kDa heat shock chaperonin (Hsp10 or Cpn10), which is a co-chaperone for Hsp60 in the protein folding process [7]; (iii) the mitochondrial homologue of the major 70-kDa heat shock protein (mHsp70), which plays a central role in the import of various proteins into mitochondria and their proper folding [4, 6]; and (iv) the mitochondrial Hsp90 protein, which was originally identified in mammalian cells as the tumour necrosis factor receptor-associated protein-1 (TRAP-1) [8, 9] and is commonly referred to by this latter name.

All of these proteins are encoded by nuclear genes, and, after translation of their transcripts in the cytosol, their protein products are then imported into mitochondria.

Introduction

The unusual immunogenicity of heat shock proteins (also known as stress proteins) was discovered in studies of the immune response to microbial infection, in which a large proportion of the humoral and cellular immune response to diverse microbial pathogens was found to be specific for pathogen-derived heat shock protein [1]. These studies demonstrated that immune recognition of pathogen-derived heat shock proteins occurs in natural and experimental settings in animals and man. This is discussed in detail in Chapter 16. Immune responses elicited to mycobacterial heat shock proteins have been particularly well studied. In man, recognition of mycobacterial heat shock protein by CD4+ T cells occurs in the context of numerous human leukocyte antigen (HLA) alleles, and epitopes have been identified that are presented by multiple HLA molecules [2]. The promiscuous recognition of mycobacterial heat shock proteins supports their utility as ‘universal’ immunogens for the genetically diverse human population. The immunogenic properties of microbial heat shock proteins have accordingly led to their application in a variety of immunisation formats as prophylactic and therapeutic agents in models of infectious disease and cancer [3]. In these studies, heat shock proteins have been delivered as subunit vaccines, carrier proteins in chemical conjugates, recombinant fusion proteins and DNA expression vectors for induction of humoral and cellular immunity.

To explain the disproportionate focus of the immune response on a small subset of pathogen antigens, heat shock proteins were proposed to act as ‘red flags’ – alerting the immune system to the presence of a foreign invader [4].

Introduction

The heat shock protein (Hsp) 70 family is a collection of evolutionarily conserved, ubiquitous proteins that are either constitutively expressed and/or stress induced and which are nominally defined by their molecular weight (Hsp70, Hsc73, BiP (binding immunoglobulin protein, or glucose regulated protein (grp) 78)). Historically, these proteins have been perceived to function as intracellular molecular chaperones that ensure the correct folding of nascent proteins and are involved in the translocation of proteins and assist in protein degradation through the proteasome [1]. At times of physical or chemical stress, such chaperones are upregulated by the unfolded protein response and provide protection against the accumulation and aggregation of denatured proteins.

In contrast to this long-standing perception, there is now increasing interest in an intercellular signalling role for these proteins and, as a consequence, they have been termed ‘chaperokines’ in light of their cytokine-like qualities [2, 3]. The interaction between heat shock proteins and specific cell surface receptors that signal the release of inflammatory mediators has revealed a link between the innate and adaptive immune response. A wide range of extracellular receptors for human Hsp70 has been identified. These include CD14 [2, 4, 5], Toll-like receptor (TLR) 4, TLR2 [4, 6], CD91 [7, 8] and CD40 [9, 10] on monocytes, and scavenger receptors such as LOX-1 on dendritic cells (DCs) [11]. The role of these receptors is detailed in Chapters 7 and 10.

The last four decades of the 20th century saw the discovery of the heat shock or cell stress response and the identification of the proteins produced by cells in response to adverse environmental conditions. In 1987, the term ‘molecular chaperone’ was coined to describe several unrelated protein families which had the ability to assist the correct folding and assembly/disassembly of other proteins. The past 20 years have seen the elucidation of the structural mechanisms of protein chaperoning by several key molecules including chaperonin (Hsp) 60 and Hsp70 and the realisation that not all molecular chaperones are cell stress proteins and vice versa. The genesis of molecular chaperones was contemporaneous with the identification of these highly conserved proteins as paradoxical immunodominant antigens that appeared to be important in microbial infection and autoimmunity. Indeed, the administration of molecular chaperones such as Hsp60 and Hsp70 was found to inhibit experimental autoimmune disease. By the 1990s, it was realised that correct protein folding was the key to cellular homeostasis and the paradigm that developed was that molecular chaperones were intracellular proteins whose function was to assist in protein folding. The paradoxical immunogenicity and immunomodulatory effects of molecular chaperones remained unexplained.

Another strand of the molecular chaperone story began to develop in the late 1980s and early 1990s with reports of the appearance of certain molecular chaperones on the surface of cells. Later, reports began to appear that molecular chaperones when applied exogenously to cells in culture had effects similar to those of pro-inflammatory cytokines.