Targeting Conserved Pathways as a Strategy for Novel Drug Development: Disabling the Cellular Stress Response
- 1k Downloads
The ability to respond to and cope with stress at a molecular level is essential for cell survival. The stress response is conserved across organisms by the expression of a group of molecular chaperones known as heat shock proteins (HSP). HSP are ubiquitous and highly conserved proteins that regulate cellular protein homeostasis and trafficking under physiological and stressful conditions, including diseases such as cancer and malaria. HSP are good drug targets for the treatment of human diseases, as the significant functional and structural data available suggest that they are essential for cell survival and that, despite conservation across species, there are biophysical and biochemical differences between HSP in normal and disease states that allow HSP to be selectively targeted. In this chapter, we review the international status of this area of research and highlight progress by us and other African researchers towards the characterisation and targeting of HSP from humans and parasites from Plasmodium and Trypanosoma as drug targets.
KeywordsHeat Shock Protein Cancer Stem Cell Molecular Chaperone HSP90 Inhibitor Heat Shock Transcription Factor
4.1 Introduction to the Cellular Stress Response
The biological activity of proteins is dependent on the ability to assume and maintain the appropriate three-dimensional biophysical structure. Despite the fact that the primary amino acid sequence is sufficient to define the three-dimensional structure of a protein , the crowded cellular environment and molecular stress often result in protein misfolding and aggregation. Stress at the biological level can be defined as any stimulus or condition, such as extreme temperature, oxidative radicals, enhanced growth rate and xenobiotics, that perturbs the correct function of the cell, the result of which is the disruption of protein homeostasis. To cope with these conditions, the organism responds by inducing the expression of a series of highly conserved proteins, known as heat shock proteins (HSP), which function as molecular chaperones to overcome the effects of protein misfolding and aggregation. A molecular chaperone is defined as a protein that is capable of interacting with and stabilising non-native protein structures or nascent polypeptides to prevent aggregation and promote the formation of correct, functional conformations . Since the discovery of HSP in heat-treated Drosophila cells in the early 1960s, a significant amount of research has been devoted to the characterisation of the role these proteins play in physiological and stressful processes with the cell [3, 4, 5]. The response to stress is conserved across organisms; the ability to respond to stress involves molecular chaperones that regulate the integrity of cellular proteins. As protein homeostasis is central to all organisms, cellular stress and molecular chaperone function has been linked to a range of human disorders. HSP have been implicated in a range of pathologies that are linked by the core element of cellular stress .
4.2 Heat Shock Proteins as Molecular Chaperones
The ability of HSP to act as molecular chaperones is integral to their protective roles in the cell . Molecular chaperones catalyse the refolding of stress-denatured or nascent polypeptides in addition to regulating the assembly of oligomeric proteins, protein transport and degradation and protein–protein interactions. Molecular chaperones are catalysts in the folding process, in the sense that they do not actively fold proteins, but rather assist folding by preventing non-productive or aggregation reactions and do not constitute part of the final protein structure . HSP may be expressed constitutively or induced by the binding of heat shock transcription factors (HSF) to specific heat shock elements (HSE). Stress induces the phosphorylation of the HSF, leading to the formation of trimeric HSF species that bind the HSE and stimulate expression of the HSP . HSP have been identified in almost all organisms, from bacteria to lower and higher eukaryotes, and show a high level of conservation across species. They are classified and named on the basis of their molecular size, for example, HSP70 will have a molecular weight of approximately 70 kDa. To date, a number of HSP families have been described. These are the HSP100, HSP90, HSP70, HSP60, HSP40 and the small HSP (sHSPs; ~18–43 kDa) families [8, 10]. The link between cell stress, molecular chaperones and human disease has made HSP attractive drug targets. In particular, the focus has been on the development of inhibitors of the HSP90 and HSP70 families of molecular chaperones as drug targets for the treatment of cancer and parasitic diseases.
4.2.1 Heat Shock Protein 90 (HSP90)
In eukaryotic cells, HSP90 species are some of the most abundant molecular chaperones . The human genome encodes five distinct HSP90 genes. Cytosolic HSP90 has two isoforms, the α and β isoforms, which display 85% sequence identity and are encoded for by separate genes that arose from duplication of the original HSP90A gene . HSP90α expression is defined as more inducible than that of HSP90β . HSP90 isoforms are also expressed in the mitochondria (HSP75/TRAP-1), the endoplasmic reticulum (GRP94/GRP96) and under certain conditions, HSP90α, HSP90β and GP96 isoforms have been identified in the plasma membrane or extracellular matrix. The genomes of the parasites Plasmodium falciparum and Trypanosoma cruzi encode 11 and 4 different HSP90 genes, respectively . While both parasites encode cytosolic and ER isoforms of HSP90, they do not encode for the α and β cytosolic isoforms that are found in humans . While eukaryotic HSP90 is essential for cell survival, bacterial HSP90 (HTPG) is dispensable for cell growth. HSP90 is also subject to post-translational modification, including s-nitrosylation, phosphorylation, methylation and acetylation, which may influence its function and cellular localisation [11, 16, 17, 18, 19, 20].
HSP90 functions as a dimer, with each monomer consisting of three highly conserved functional domains, namely an N-terminal domain (25 kDa), a middle (M) domain (35 kDa) and a C-terminal domain (12 kDa) [21, 22]. The N-terminal and M-domain are connected by a charged linker region, which varies in length and amino acid composition according to the species or isoform (and is entirely absent from the prokaryotic HSP90) [23, 24]. The N-terminal domain contains the primary binding site for ATP/ADP , as determined by crystallisation studies on both yeast and human HSP90. This ATP/ADP binding site is the same as that bound by the natural HSP90 inhibitor, geldanamycin , which can also bind to HSP90 from other species, including P. falciparum  and T. cruzi . The C-terminal domain is the site of dimerisation of HSP90  and contains the MEEVD motif which is the primary binding site for the TPR-containing co-chaperones, such as HOP (HSP70/HSP90 organising protein) [30, 31]. The structure of full-length HSP90 has been determined at low resolution by cryo-EM , while each of these functional domains has been crystallised independently.
4.2.2 Heat Shock Protein 70 (HSP70)
The HSP70 molecular chaperone family is composed of members that are essential for correct protein folding, transport of proteins to different sub-cellular compartments and degradation of unstable proteins in the cell . HSP70 isoforms are found in most species. We have demonstrated that even the ancient coelacanth  encodes a functional HSP70, as do protozoan parasites such as P. falciparum and T. cruzi. There are 13 different isoforms of HSP70 in humans, with 6 isoforms in P. falciparum and 12 in T. brucei [35, 36]. HSP70 have three functional domains, the N-terminal ATPase domain (~44 kDa), substrate binding region (~18 kDa) and a variable C-terminal region (~10 kDa) that is the site of specific interactions with other proteins, such as HOP [31, 37]. The ATPase domain of HSP70 forms two distinct, equally sized structural lobes, separated from each other by a deep cleft, which is the binding site of ATP [38, 39]. ATP hydrolysis by the ATPase domain of HSP70s is central to the role of the protein as a chaperone during assisted protein folding . Despite the basic HSP70 structure being conserved in HSP70 from different organisms [36, 39, 41], there is evidence that there are differences between HSP70 from different organisms that may be exploited to selectively inhibit the HSP70 from certain organisms .
4.3 Molecular Chaperones as Drug Targets in Africa
Biological molecules are required to conform to a set of criteria if they are to be considered potential drug targets . Molecular chaperones, and in particular HSP90 and HSP70, are currently considered as bona fide biological targets for drug development as they conform to many of these criteria . Ideally, a drug target should be essential for the development of the disorder, but absent under normal conditions, to allow selective targeting of diseased and non-healthy cells. HSP are important in the development of human disorders including cancer and are conserved across different species of human parasites, probably because they are often vital or essential to survival of the parasite. HSP70 and HSP90 are among the most well-characterised chaperones, with functional and structural data available for different protein isoforms from a range of species. In addition, the naturally occurring inhibitor of HSP90, the ansamycin antibiotic geldanamycin (GA), has facilitated the rational development of HSP90 inhibitors, many of which are analogues of GA (e.g. 17-AAG, 17-DMAG), as well as the search for numerous other natural products that have the potential to be HSP90 inhibitors. The targeting of such highly conserved proteins of human parasites may not resonate with current dogmas of drug discovery, especially in terms of selectivity for the diseased state over the normal condition (such as parasite over human host or cancer cell versus normal cell). However, highly conserved proteins that are essential for cell survival are likely to evolve considerably slower than other less conserved protein families, making them less susceptible to variation under selection pressure, a desirable attribute for any potential drug target. There is also evidence that despite the high levels of conservation, these HSP are not identical and that there are functional and structural differences that may be exploited. In fact, there is even evidence that disease states can generate HSP with different biophysical and biochemical characteristics to their counterparts in normal cells, meaning that the same protein may be biochemically distinct from the normal cell. For example, 17-AAG was found to have a higher affinity for HSP90 in transformed cells compared to normal cells and is currently in clinical trials as an anti-cancer agent . Our research into the identification of inhibitors of HSP90 and HSP70 in protozoan parasites and in cancer is based on our fundamental understanding of these chaperones as a result of comparative studies of the chaperones in humans and parasites. We have identified HSP70 and HSP90 as potential drug targets in the treatment of cancer, malaria and trypanosomiasis and have used the biochemical differences between these proteins to screen potential inhibitors.
4.3.1 Molecular Chaperones of Protozoan Parasites
We and others have begun to scrutinise the chaperone machineries of the causative agents of certain infectious diseases of humans, such as malaria (P. falciparum; reviewed in [36, 45, 46]) and the various trypanosomiases (Trypanosoma brucei and T. cruzi; reviewed in ). These parasites have evolved chaperone systems to cope with the extreme demands of cyclical development through physiologically diverse host and vector systems . HSP70 and HSP90, in particular, are expressed at critical developmental stages and persist at high levels in the infectious stages of these parasites. Therefore, it is not surprising that inhibitors of HSP90 (e.g. GA) have been shown to arrest the growth of these parasites [27, 28, 47]. Therefore, plasmodial HSP90 (PfHSP90) is an obvious anti-plasmodial drug target, and promising pre-clinical tests have been conducted using GA and its derivatives . We are discovering that while there is a high degree of conservation at the sequence level between parasitic and human HSPs, there are major structural and functional differences that suggest that the parasitic HSP can be selectively targeted [36, 49].
We and others have conducted extensive biochemical and cell biological studies on the major cytosolic and inducible plasmodial HSP70, PfHSP70-1. As expected, it is a bona fide chaperone, exhibiting ATPase and protein aggregation suppression activities [50, 51, 52], and capable of functionally substituting for both a bacterial  and a yeast HSP70 . It is also a highly abundant protein expressed at all erythrocytic stages of the P. falciparum life cycle, with increased expression after heat shock [54, 55]. Importantly, there is strong evidence that PfHSP70-1 is relatively thermostable compared to human HSP70 (hHSP70) and capable of functioning optimally at temperatures around 50°C . In addition, there are biochemical differences between PfHSP70-1 and hHSP70 at the level of ATP affinity and ATPase activity, and structural modelling has identified differences at their HSP40-binding sites [51, 57].
HSP70 small molecule modulators with anti-plasmodial activitya
Pyrimidinone–peptoid (e.g. MAL3-101)
Anti-cancer and anti-plasmodial
Pyrimidinone–peptoid (e.g. MAL3-39, DMT3024, DMT2264)
Dihydropyrimidines (e.g. 116-9e)
Anti-cancer and anti-plasmodial
4.3.2 Human Molecular Chaperones in Cancer
HSP90 is known anecdotally as the ‘cancer chaperone’ due to its ability to exclusively mediate the folding and stability of numerous transcription factors and signalling intermediates in vivo. Many of these so-called client proteins of HSP90 are oncogenes that are either mutated or upregulated in a range of cancers. HSP90 maintains its client proteins in an inactive, but easily inducible, state . As these states are often inherently unstable and labile in the absence of substrate, the role of the HSP90 multi-chaperone complex is to enhance client protein stability . Therefore, using HSP90 as an anti-cancer target allows the simultaneous inhibition of multiple signalling pathways . The molecular chaperone activity of HSP90 is regulated by conformational changes which are dependent on two factors. The first is the intrinsic ATPase activity of HSP90, the importance of which is demonstrated by the fact that mutations that result in either a loss of ATP binding or ATP hydrolysis inhibit both the in vitro and in vivo functions of HSP90. This fact is exploited in drug discovery as many of the naturally occurring or synthetic HSP90 inhibitors bind to the ATP binding site within the N terminus of HSP90. The second factor is the association of HSP90 with a range of co-chaperones (HSP70, HSP40, HOP, p23, immunophilins) into a multi-chaperone complex [11, 76]. HSP90, in association with client protein and co-chaperones, is considered ‘complexed’ or activated, conditions under which HSP90 displays an enhanced sensitivity to and binding of anti-HSP90 drugs compared to free/uncomplexed HSP90 [77, 78].
In addition to its potential as a general anti-cancer drug target, HSP90 has recently been considered as a treatment for certain cell subtypes within cancers. Extracellular HSP90 protein is expressed on the surface of a range of cell types, including melanoma, fibrosarcoma, breast adenocarcinoma and neuronal cells [86, 87, 88, 89] and linked to cell migration, metastasis and invasion of cancer cells. Selective inhibition of only surface HSP90 reduced migration of a range of cancer cell types ; this may offer a unique therapeutic opportunity for the treatment of metastasis, particularly since over 90% of deaths from cancer are as a result of cancer spread (metastasis). Similarly, HSP90 is emerging as a potential drug target for the removal of a specific subpopulation of cancer cells, known as cancer stem cells (CSC). The CSC hypothesis describes cancer development and maintenance as being driven by cancer cells that display stem-like properties (reviewed by us in ). Cancer stem cells are cancer cells that, like normal tissue stem cells, are capable of both self-renewal and differentiation, which accounts for tumour heterogeneity. These cells are thought to control tumour development, just as normal tissue stem cells are required for organ development [91, 92, 93]. CSC populations are relevant to cancer drug discovery due to their potential role in metastasis and cancer recurrence, the latter as a result of CSC’s apparent resistance to chemotherapy and radiotherapy [94, 95]. We have linked HSP90 function to stem-cell-associated pathways that may be important in CSC function . We and others have shown that STAT3 is an HSP90 client protein and the LIF/STAT3 pathway is important in self-renewal of embryonic stem cells [97, 98]. The HSP90 inhibitor 17-AAG was effective at inhibiting growth of both glioma cells and glioma cancer stem cells . In addition, cancer stem cells from medulloblastoma were more sensitive to inhibition of the kinase Akt, which requires HSP90 for activity, than the cancer cells from the same tumour mass .
However, inhibition of HSP90 as a drug strategy is often associated with the compensatory upregulation of other molecular chaperones, especially HSP70, in cells treated with HSP90 inhibitors. Inhibition of HSP90 causes activation of HSF-1 that induces the expression of other members of the HSP family [101, 102]. While upregulation of other HSP has been associated with drug resistance to HSP90 inhibitors [103, 104], some researchers have used this fortuitous induction of other chaperones as a strategy to treat protein folding diseases such as Alzheimer’s disease  and as a mechanism to promote cytoprotection . The most effective strategy, with respect to anti-cancer therapy, therefore may be the simultaneous inhibition of HSP90 and HSP70  or even the inhibition of the heat shock response by direct inhibition of HSF-1 . We and others have identified a number of small molecule inhibitors of HSP70 function in parasites, which may also have future potential as co-inhibitors of HSP70 in human cancers.
4.4 Future Prospects for HSP-Directed Drug Discovery in Africa
Despite extensive research into the function of HSP in cancer and parasitic diseases, there are still relatively few anti-HSP compounds that have made it to the market as drugs [109, 110, 111]. The most successful compound at present is 17-AAG, currently in clinical trials as an anti-cancer agent and also with potential as a treatment for parasite diseases demonstrated by in vitro studies. However, there is still great potential for drug discovery in this area, particularly as our fundamental knowledge of the structure and function of different HSP systems grows.
4.4.1 Identification of New Indigenous Compounds That Target HSP
A number of small molecules have been found by us and others to modulate the activity of the conserved HSP70 and HSP90 proteins from different organisms. What is striking is the number of these compounds that are either natural products or analogues of natural products. Natural products are a good source of compounds that are difficult to synthesise by traditional chemical methods and may also serve as the basis for rational drug design. Many natural compounds may exhibit activity as the parent compound and may be of value as scaffolds for the development of novel compounds. The diversity of South African flora and fauna has yielded many natural products that show anti-cancer [84, 112, 113] and anti-plasmodial activity  that have structural similarity to known HSP inhibitors [115, 116]. Therefore, the analysis of the effect of these indigenous compounds has the potential to identify novel modulators of chaperone function.
4.4.2 Evaluation of Co-chaperones as Drug Targets
Molecular chaperones function as part of multi-chaperone complexes, complexes which often include other proteins known as co-chaperones. Co-chaperones are a diverse group of proteins that modulate the function of and promote the interaction between the major molecular chaperones. The most well-known co-chaperones of both the HSP90 and HSP70 families are the HSP40 family and the tetratricopeptide repeat containing co-chaperone HOP. The central role played by co-chaperones in the modulation of the function of other chaperones means that these proteins may also be good drug targets.
HSP40s are the major co-chaperones in the HSP70-assisted protein folding process and act via regulation of the ATPase activity of HSP70. HSP40 are the largest and most diverse group of co-chaperones. This diversity is conserved across organisms; there are 49 HSP40 genes in humans, 43 in P. falciparum and 67 genes in T. brucei [46, 117, 118]. HSP40 proteins are classified as type I, II or III, depending on the degree of conservation of the functional motifs they contain, with respect to Escherichia coli (E. coli) DnaJ . We have recently proposed an additional group, type IV HSP40, which, besides a few isoforms in humans and yeast, appears specific to malaria parasites . The type and number of motifs contained in an HSP40 will govern its function, with most of the HSP70 chaperone interactions being modulated by type I HSP40s. More specific functions are likely to be mediated by type III or type IV HSP40s [120, 121]. HSP40 may prove to be the better chaperone drug target, as unlike HSP90 and HSP70 where the same chaperone is responsible for chaperoning numerous clients, many HSP40 are selective and will only bind a restricted number of proteins. This may allow for the preferential targeting of specific signalling pathways through the targeting of a single HSP40 isoform .
HOP is a co-chaperone that is involved in mediating interactions between HSP70 and HSP90. HOP acts as an adapter between the two chaperones and controls the transfer of protein substrates between the two systems. HOP is essential for the chaperone function of both Hsp70 and Hsp90 as part of the Hsp90 complex. Therefore, inhibition of HOP has the theoretical potential not only to inhibit complexed Hsp90 (the state of Hsp90 observed in cancerous cells) but also to facilitate the simultaneous dual inhibition of both Hsp90 and Hsp70, thereby preventing any redundancy between the two chaperone systems that may occur when these proteins are targeted individually. In addition, it is now apparent that HOP is also expressed by parasites, such as P. falciparum and T. cruzi. We and others are therefore in the process of characterising human and parasitic HOP as future potential drug targets [122, 123, 124, 125].
4.4.3 HSP as Biotechnological Tools in Drug Discovery
In addition to the well-described role as drug targets, there is also the potential to use molecular chaperones as tools in biotechnology. The protein folding and quality control functions of HSP can be harnessed to synthesise other drug targets for drug characterisation studies. The limiting factor for many drug development programmes is the availability of sufficient quantities of active, folded drug targets for use in drug characterisation studies. We have recently published on the use of co-expression of molecular chaperones for the synthesis of putative drug targets . We were able to demonstrate the production of active, folded cyclohydrolase enzyme from P. falciparum for use in in vitro enzymatic assays for inhibitor characterisation. Similar techniques were used to successfully produce large quantities of other malarial drug targets, such as PfDXR [127, 128]. To the best of our knowledge, these publications are the first to demonstrate the use of a homologous chaperone–client system, where the drug target and the chaperone used to refold it are from the same organism (in this case, P. falciparum), for biotechnological production of a drug target. This research is ongoing and has great potential to develop a drug discovery platform technology in Africa.
- 4.Li Z, Srivastava P (2004) Heat-shock proteins. Curr Protoc Immunol Appendix 1:Appendix 1TGoogle Scholar
- 5.Morimoto RI, Kline MP et al (1997) The heat-shock response: regulation and function of heat-shock proteins and molecular chaperones. Essays Biochem 32:17–29Google Scholar
- 8.Fink AL (1999) Chaperone-mediated protein folding. Physiol Rev 79(2):425–449Google Scholar
- 21.Nemoto T, Sato N (1998) Oligomeric forms of the 90-kDa heat shock protein. Biochem J 330(2):989–995Google Scholar
- 97.Kim HL, Cassone M et al (2008) HIF-1alpha and STAT3 client proteins interacting with the cancer chaperone Hsp90: therapeutic considerations. Cancer Biol Ther 7(1):10–14Google Scholar
- 98.Setati MM, Prinsloo E et al (2010) Leukemia inhibitory factor promotes Hsp90 association with STAT3 in mouse embryonic stem cells. IUBMB Life 62(1):61–66Google Scholar
- 101.Ali A, Bharadwaj S et al (1998) HSP90 interacts with and regulates the activity of heat shock factor 1 in Xenopus oocytes. Mol Cell Biol 18(9):4949–4960Google Scholar
- 110.Wang Y, Trepel JB et al (2010) STA-9090, a small-molecule Hsp90 inhibitor for the potential treatment of cancer. Curr Opin Investig Drugs 11(12):1466–1476Google Scholar
- 112.Brauns SC, Dealtry G et al (2005) Caspase-3 activation and induction of PARP cleavage by cyclic dipeptide cyclo(Phe-Pro) in HT-29 cells. Anticancer Res 25(6B):4197–4202Google Scholar
- 113.Brauns SC, Milne P et al (2004) Selected cyclic dipeptides inhibit cancer cell growth and induce apoptosis in HT-29 colon cancer cells. Anticancer Res 24(3a):1713–1719Google Scholar
- 115.Afolayan AF, Bolton JJ et al (2008) Fucoxanthin, tetraprenylated toluquinone and toluhydroquinone metabolites from Sargassum heterophyllum inhibit the in vitro growth of the malaria parasite Plasmodium falciparum. Z Naturforsch C 63(11–12):848–852Google Scholar
- 128.Goble JL, Adendorff MR et al (2010) The malarial drug target Plasmodium falciparum 1-deoxy-D-xylulose-5-phosphate reductoisomerase (PfDXR): development of a 3-D model for identification of novel, structural and functional features and for inhibitor screening. Protein Pept Lett 17(1):109–120CrossRefGoogle Scholar