73,52 €
The book provides an updated panorama of the functional relevance of molecular chaperones in the proper folding of client factors, protein-protein interactions, the regulation of key biological functions, the development of ligand-based structural complex
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 566
Veröffentlichungsjahr: 2018
This is an agreement between you and Bentham Science Publishers Ltd. Please read this License Agreement carefully before using the ebook/echapter/ejournal (“Work”). Your use of the Work constitutes your agreement to the terms and conditions set forth in this License Agreement. If you do not agree to these terms and conditions then you should not use the Work.
Bentham Science Publishers agrees to grant you a non-exclusive, non-transferable limited license to use the Work subject to and in accordance with the following terms and conditions. This License Agreement is for non-library, personal use only. For a library / institutional / multi user license in respect of the Work, please contact: [email protected].
3. The unauthorised use or distribution of copyrighted or other proprietary content is illegal and could subject you to liability for substantial money damages. You will be liable for any damage resulting from your misuse of the Work or any violation of this License Agreement, including any infringement by you of copyrights or proprietary rights.
Bentham Science Publishers does not guarantee that the information in the Work is error-free, or warrant that it will meet your requirements or that access to the Work will be uninterrupted or error-free. The Work is provided "as is" without warranty of any kind, either express or implied or statutory, including, without limitation, implied warranties of merchantability and fitness for a particular purpose. The entire risk as to the results and performance of the Work is assumed by you. No responsibility is assumed by Bentham Science Publishers, its staff, editors and/or authors for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products instruction, advertisements or ideas contained in the Work.
In no event will Bentham Science Publishers, its staff, editors and/or authors, be liable for any damages, including, without limitation, special, incidental and/or consequential damages and/or damages for lost data and/or profits arising out of (whether directly or indirectly) the use or inability to use the Work. The entire liability of Bentham Science Publishers shall be limited to the amount actually paid by you for the Work.
Bentham Science Publishers Ltd. Executive Suite Y - 2 PO Box 7917, Saif Zone Sharjah, U.A.E. Email: [email protected]
From a social perspective, the word chaperone refers to a matron who used to accompany young people in public, especially ladies, and supervise them at a social gathering to ensure proper behavior. Similarly, those proteins that assist others in their proper folding and biological functions are also referred to as chaperones. During the late ‘70s, it was coined the term ‘molecular chaperones’ to make reference to the ability of nucleoplasmin to prevent the aggregation of histones with DNA during the assembly of nucleosomes. As a consequence, this nomenclature was extended to all proteins able to mediate the post-translational assembly of protein complexes. Although the primary concept of molecular chaperone was related to its ability to ensure the correct folding of newly synthesized peptides and refolding of stress-denatured proteins, it should be noted that chaperones are also involved in essential and more sophisticated functions such as promoting the correct assembly of oligomeric complexes. One of the most remarkable examples for this special feature is the ability of a particular subfamily of molecular chaperones, the heat-shock proteins, to assist the proper assembly of steroid receptors with chaperones and co-chaperones. This important feature permits the binding of steroids to activate isoform of the receptor, which functions as a ligand-dependent transcription factor.
The term ‘heat-shock protein’ stems from the original observation that heat-stress greatly enhances the production of this particular class of molecular chaperones. This means that all heat-shock proteins are molecular chaperones, but not all molecular chaperones are necessarily heat-shock proteins. Temperature is not the only stimulus able to induce heat-shock proteins. Upon the onset of several environmental types of stress or due to the exposure to damaging and extreme insults, the cells increase dramatically the production of molecular chaperones, which play prominent roles in many of the most basic cellular processes by stabilizing unfolded or misfolded peptides, giving the cell time to repair or re-synthesize damaged proteins. In addition to commanding the proper folding of a factor exposed to an environmental injury, many chaperones are also related to other key functions such as enzyme activity, cytoskeletal architecture, nuclear organization, protein trafficking, transcriptional regulation, epigenetic alterations of gene expression and, even more intriguingly, heritable alterations in chromatin state.
The biological relevance of molecular chaperones during the modern times was discovered during the early 1960s when the Italian scientist Ferruccio Ritossa was studying nucleic acid synthesis in puffs of Drosophila salivary glands. A colleague accidentally changed the temperature of the cell incubator and an incredible transcriptional activity of new chromosomal puffs was evidenced. The induction of these proteins is one of the most important manifestations of environmentally induced changes in gene expression.
The whole proteome of the cell is successfully maintained thanks to the assistance of molecular chaperones. In addition, the subcellular localization, local concentration, and biological activity of each protein must be strictly regulated in response to both intrinsic and environmental stimuli. The recently coined portmanteau word proteostasis describes this equilibrated state of the healthy proteome balance, whereas the term proteostasis network refers to the group of cellular events and factors involved in proteostasis maintenance. Failures of proteostasis regulation are responsible for a number of diseases as well as for the deleterious consequences of physiologic processes such as ageing. Since molecular chaperones play a key role in the maintenance of this proteostasis network, they became potential pharmacological targets to preserve that proteostatic function and to improve the biology of the cells by enhancing certain activities (or preventing others). In this regard, several endeavors are currently focused in targeting Hsp90 and some of its cochaperones such as high molecular weight immunophilins and p23. Currently, this is being tested as an exciting alternative for molecular-based therapies, particularly in both malignant and neurodegenerative diseases.
In this book entitled Role of molecular chaperones in structural folding, biological functions and drug interactions of client proteins, several aspects of the biology of these key proteins have been addressed with the purpose of providing an updated overview of the field. The major aim is to present a broad spectrum of the molecular mechanisms of action of several molecular chaperones. Understanding these mechanisms will permit to focus on the design of small molecules able to regulate such functions in the complex cellular milieu affecting the proteostasis network in diseases characterized by aberrant protein folding. Above all, I wish to acknowledge the valuable viewpoint of all contributing authors and hope that this assemblage of perspectives will be a valuable resource for researchers in this and other related fields. Also, I hope that the high enthusiasm showed by all our contributors to make this endeavor possible will be appreciated by the readers. Finally, I must express my greatest thanks to the editorial for the encouraging support to face this endeavor.
The HSP70 family of molecular chaperones plays a significant role in cancer. Notably, the inducible protein Hsp72 becomes expressed in many cancers, often to high levels and underlies escape of tumor cells from senescence and increased tumor initiation and metastasis. Examination of database suggests that mutation within the ORFs of HSP70 genes in cancer is relatively rare suggesting a requirement for intact function. At the molecular level, Hsp72 is thought to chaperone key proteins in tumorigenesis and permit their accumulation in the malignant cell. In addition, an important role for Hsp72 in RNA metabolism is emerging, indicating mechanisms potentially involving the RNA binding protein HuR. The existence of multiple HSP70 pseudogenes may also be important for future studies of long non-coding RNA (lncRNA) regulation through this family of chaperones. As the significance of this family of chaperones in cancer emerges, small molecule inhibitors have been developed as future potential cancer pharmaceuticals. We discuss the targeting of individual HSP70 families at key functional domains in the proteins.
The seventy kilodalton heat shock protein (HSP70) family is found in all cellular organisms [1]. From the beginning of cellular life, these proteins have permitted protein folding in the crowded intracellular environment and have come to the aid
of stressed cells dealing with proteotoxic insults [1]. Over 20 years ago, Ciocca and co-workers studying clinical biopsy material noticed that human cancers contained elevated levels of HSPs and this finding has been confirmed in many subsequent studies [2-4]. The rationale behind these increases in HSP70 levels was not immediately apparent, although it was suspected that this change in expression might be related to alterations in the folding status of the malignant cell. Indeed, investigators studying the potential role in cancer of another member of the HSP family, Hsp90 had noticed a large increase in this chaperone in many cancer types and concluded that malignant cells were “addicted to chaperones” [5]. Withdrawal of the folding power of Hsp90 by exposure to specific drugs led to widespread denaturation and degradation of oncogenic proteins. Indeed targeting Hsp90 has become a major endeavor in cancer therapeutics [6]. Not unreasonably it could be suggested that HSP70 might be the next candidate for drug therapy based on its elevated levels and tumor dependence.
Most cellular organisms contain a relatively large (over 10 member) HSP70 family [7]. In this review, we will discuss HSP70 family members shown to be located in the cytoplasm and nucleus, including the essential protein Hsc70, HspA1L and the inducible proteins Hsp72 and Hsp70B'.
The common structure of HSP70 family proteins includes an N-terminal ATPase domain or nucleotide binding domain (AD / NBD, 45k), a protease active short linker (8 aa), substrate binding domain/ peptide binding domain (SBD / PBD, 15k) and C-terminal helical lid region (10k) [8], [9]. The C-terminal lid region of HSP70 also contains a tetrapeptide EEVD motif at the extreme C-terminus [9] also found in HSP90, that recognizes the tetratricopeptide repeat (TPR) motif found in a range of TPR domain proteins including the Hop co-chaperone /adaptor protein [10, 11].
The human HSP70 gene family is characterized by considerable diversity, apparently resulting from multiple duplications and retrotranspositions of the highly expressed gene HSPA8/HSC70/HSP73 [[12]]. Indeed, Brocchieri et al identified forty-seven Hsp70 sequences including seventeen genes and thirty pseudogenes in the human genome. The N-terminal AD was conserved at least partially in the majority of the genes while the SBD was more commonly lost homologies. Structurally well-conserved and functional HSP70 family members include HSC70/HSPA8, Hsp72/HSPA1A, Hsp72/HSPA1B, HSPA1L/HSP70T/ HSP70-hom, HSP70B'/HSPA6, HSP70B/HSPA7 (possibly a pseudogene) and HSPA2, shown in phylogenetic tree analyses [12] (Table 1).
Clearly there are numerous pseudogenes of HSP70 so that interpretation of experiments in this field should likely take this into account. For example, qRT-PCR may detect both coding HSP70 and HSP70 pseudogene RNAs. In addition, one antibody might detect a member of the HSP70 family but might also associate with smaller peptides translated from the same mRNA, or larger proteins such as Hsp110 or Grp170. Therefore, interpretation of western blotting data should be carefully considered.
Hsc70 (heat shock cognate protein 70) is expressed most abundantly among the HSP70 family proteins. This protein is encoded by the HSPA8 gene and is moderately inducible by stress. Numerous pseudogenes of HSPA8/HSC70 have been found [12] and some of these potentially might be expressed and function as lncRNA (Fig. 1) [13, 14]. For example, the HSC70 pseudogene ncRNA could possibly play role in Hsc70 expression by assuming a decoy / sponge function or antagonizing miRNA that target HSC70/HSPA8.
HSPA1A and HSPA1B are the major inducible-type HSP70 genes and together encode Hsp72. The homology between these two genes is high and it is difficult to distinguish them at the mRNA and protein levels. In addition, HSPA1L/HSP70T/HSP70-hom is also homologous to HSPA1A and to HSPA1B and all three of these genes are located close together in both human and mouse genomes [12]. HSPA1A and HSPA1L are located on human chromosome 6p21.33 in plus and minus directions, respectively. HSPA1B is also located close by in chromosome 6p21.32, in the plus strand. Four pseudogenes of HSP70 were also found on chromosome 6 [12]. The physical location of these genes in the chromosome suggests the possibility of functional interactions.
Fig. (1)) HSP70 transcripts and proteins derived from HSP70 genes in co-translational folding, as non-coding RNA (ncRNA) and as origins of smaller HSP70-related polypeptides.Two more HSP70 genes, HSPA6 and HSPA7 encode the Hsp70B' and Hsp70B proteins, respectively, (although HSPA7/HSP70B could be a pseudogene [12]). These two HSP70 genes exhibit strong homology and are found only in primates, suggesting the possibility that their function could be compensated for by HspA1A/HspA1B/HspA1L in rodents, or that Hsp70B and Hsp70B' have primate-specific roles. The HSPA6 and HSPA7 mRNAs are minimally expressed in normal tissues but are strongly induced by stress and in cancer cells [15]. High-level expression of HSP70B' was correlated with elevated histone H3 lysine 4 trimethyl (H3K4me3), a marker of active transcription in the malignant prostate cancer cell line PC-3 (27% of input HSP70B' gene was coprecipitated with H3K4me3 in PC-3 cells, T. Eguchi & SK Calderwood, in preparation). Interestingly, it was shown that the HSPA6/HSP70B' product could be secreted from macrophages upon stimulation by LDL [16]. This property may indeed not be restricted to macrophages, as we recently found that Hsp70B' to be secreted from breast cancer cell line MDA-MB-231 after heat shock (T. Eguchi & SK Calderwood, unpublished). Thus extracellular Hsp70B' might play a role in human cancer [17].
Finally, the minor HSP70 family member, HspA2 was reported to localized in the nuclei, nucleoli and centrosomes of heat shocked cancer cells [18]. HspA2 appears to be a unique HSP70 family protein and is encoded by neither of the Hsp72 genes. HSPA2 may be involved in the etiology of cancers including hepatocellular carcinoma, NSCLC and pancreatic cancer [19-21].
We have shown that, in the mouse mammary tumor virus (MMTV) model, in which breast cancer develops spontaneously, hsp72 gene inactivation delayed tumor initiation and inhibited metastasis through down-regulation of the met oncogene [22]. We have confirmed in RNA-seq experiments that hspa1a and hspa1b are minimally expressed in the hsp72 KO MMTV mice, and we are attempting to determine if hsp70L1/hspa1L is deleted at physical and/or functional levels in these mice.
HSP70 proteins are thought to be the primary chaperone for many polypeptides, encountering elongating proteins on ribosomes and then handing on such proteins to Hsp90 to complete folding [23]. HSP70 protein interaction with ribosomes and with ribonucleoprotein (RNP) has also recently emerged as an important theme. Hundley et al reported that two HSP70 proteins (Ssb and Ssz1) and one of the DnaJ/Hsp40 class (Zuo1/zuotin) of co-chaperones reside on the yeast ribosome, forming a stable heterodimer named the ribosome-associated complex (RAC), and that Ssb could be crosslinked to nascent polypeptide chains on ribosomes [24]. Otto et al subsequently reported that the human homolog of the J-domain protein MPP11/DNJC1/Hsp40 and its mouse homolog MIDA1/DnaJC1 formed a stable complex with Hsp70L1 and became associated with ribosomes [25]. Alternatively, it was suggested that HSP70 Ssb function might be limited to the protection of nascent polypeptide from aggregation until downstream chaperones take over and actively fold their substrates. It was recently demonstrated that deletion of Ssb leads to widespread aggregation of nascent polypeptides [26]. Ubiquitin E3 Ligases can, alternatively, associate with ribosomes and become involved in poly-ubiquitination and degradation of the nascent polypeptides in budding yeast [27, 28] and this system is also found in mammalian cells. We recently analyzed the RNA-seq transcriptome in hsp72 KO MMTV and control MMTV mice. The quantity of ribosomal RNA was largely reduced in the hsp72 KO mammary tumor compared to the control mammary tumor (T Eguchi, B Lang & SK Calderwood, unpublished data). Concurrently expression patterns of miRNA were also altered in hsp72 KO mammary tumors. These data indicated that hsp72/hspa1a and the closely related gene hspa1L could be involved in ribosomal RNA stability, co-translational folding, the formation and function of processing bodies (P-bodies / GW-bodies) and stress granules (Fig. 2) [24, 29-35]. These are key structures involved in miRNA and messenger ribonucleoprotein (mRNP) processing. Moreover, Hsp72 appears to be important in cancer stem cell (CSC) function. Over the last decade, the notion that tumors are maintained by their own stem cells has gained ground [36]. Such CSC are characterized by self-renewal ability, reversible differentiation-dedifferentiation or EMT, stress resistances (radio- or chemo- resistances) and supported through interactions with niche cells through ligand-receptor binding or extracellular matrices (ECM) [37, 38].
Fig. (2)) A model for HSP70 control of messenger ribonucleoproteins (mRNP).To determine how cell stress might be involved in the properties of CSCs, we recently stimulated CSC-abundant breast cancer cell line MDA-MB-231 with heat shock, and found that shorter polypeptides were detected after two min heat shock (Eguchi T and Calderwood SK, unpublished data). These findings regarding rapid induction of shorter polypeptides is reminiscent of the results of a HSP70 deficiency study [26]. It is known that cell stress induces processing bodies (P-bodies) and stress granules that incorporate miRNA and mRNP aggregates through which the mRNA is either degraded or recruited to ribosomes for translational reinitiation (Fig. 2). Adenyl uridylated (AU)-rich element (ARE) binding proteins recognize AUUUA penta-ribonucleotide within U-rich sequences in the P-bodies. In addition, the Hu protein family including HuR, HuA, HuB, HuC, and HuD bind to the ARE and stabilize the mRNA, while other AU-rich element binding proteins including AUF1, TTP, BRF1, TIA-1, TIAR and KSRP destabilize the mRNA oppositely.
Indeed, one of the mRNA stabilizing protein HuR appeared to stabilize β-catenin (CTNNB1) mRNA under the control of activated HSF1 (pS326) and mTOR in breast cancer stem cell-rich MDA-MB-231 cells [39]. This β-catenin-producing system under the control of mTOR-HSF1 axis could be a key pathway in CSC oncology [39]. Activating mutations in the human CTNNB1 gene and mutations in other genes in the Wnt-β-catenin-TCF signal have been frequently reported. Importantly β-catenin has been recognized a stem cell maintenance factor [36]. Therefore, under the control of the HSF1-HuR axis, active production of β-catenin and HSPs might be a key system in cancer initiation and maintenance.
The Cancer Genome Atlas (TCGA) is a national effort to molecularly characterize every form of cancer and tumor type. Initiated in 2005 by the National Cancer Institute (NCI) and the National Human Genome Research Institute (NHGRI), TCGA aims to quantitatively characterize a representative number of tumor types at the molecular level through genome sequencing, promoter methylation analysis, relative mRNA expression levels and eventual proteomic profiling [40]. This information is made publicly available as it is processed, published and released to the cancer research community. Several sites and institutes host and distribute TCGA data such as cBioPortal at Memorial Sloan-Kettering Cancer Center that was used here to profile the HSP70 family of molecular chaperones [41-44] (Fig. 3). Cross-cancer analysis of the genomic alterations of the 13 HSP70 family members in humans shows that bladder cancer (BLCA) at 32.3%, cutaneous melanoma (SKCM) at 30.9%, lung adenocarcinoma (LUAD) at 29.7%, stomach adenocarcinoma (STAD) at 28.6% and liver hepatocellular carcinoma (LIHC) at 25.4% have the highest rates of HSP70 family genomic alteration (Fig. 3).
Analysis of mRNA levels in 23 tumor types indicates that the HSP70 family is altered in a considerable number of tumor samples. Levels of mRNA expression may be influenced by a number of variables at the epigenetic, transcriptional and post-transcriptional phases of gene expression. In 22 of the 23 tumor types profiled, HSP70 family mRNA levels were over-expressed in 25% of all tumors sampled at a standard deviation greater than 2 (Table 2, blue columns). Tumor types that least over-expressed any of the HSP70 homologs were colon and rectal carcinoma (COAD) at 22% and ovarian serous cystadenocarcinoma (OV) at 26%. Overall this means almost one in four tumor samples exhibited an over-expression of at least one HSP70 homolog. Moreover, in 8 of the 23 tumor types analyzed, an HSP70 homolog was over-expressed in 40% of tumors sampled. These included BLCA at 43%, LUAD at 43%, LIHC at 46%, kidney clear cell carcinoma (KIRC) at 45%, and chromophobe renal cell carcinoma (KICH) at 42%. Currently, three tumor types over-express an HSP70 homolog in 50% of tumor samples. These include diffuse large B-cell lymphoma (DLBC), uterine carcinoma, and adrenocortical carcinoma (ACC). The increased percentage of mRNA over-expression in these tumor types suggests that the HSP70 family may be especially essential towards maintaining cellular proteostasis within the cancer cell.
Fig. (3))Frequency Of Hsp70 Homolog Alterations Across Tumor Types And Profiled By The Cancer Genome Atlas (TCGA) [40]. Each tumor type or cell line collection listed along the x-axis were analyzed for Hsp70 family gene copy number alterations (CNA, red bars for amplification and blue bars for deletion) and open reading frame mutations (green bars). Multiple alterations are indicated by gray bars.Analysis of the open reading frame mutations in each tumor type confirms that the HSP70 family is highly conserved and essential. Only 6 of 22 tumor types have greater than 10% mutational alterations of any HSP70 homolog (Table 2 green columns). DLBC currently does not have mutational data. Moreover, acute myeloid leukemia (AML), papillary thyroid carcinoma (THCA) and KICH have less than 1% mutational alterations. Stomach adenocarcinoma (STAD) and cutaneous melanoma (SKCM) have the highest HSP70 homolog alterations at 17% of all tumor samples. Finally, throughout all tumor types analyzed, the HSP70 homolog genes HSPA1L, HSP9A/mortalin and HSPA14/Hsp70L1 are found to be the most altered by either mRNA over-expression or mutation while the putative pseudogene HSPA7/HSP70B was found to be the least altered. These observations may indicate the significance of these HSP70 homologs in maintaining tumor malignancy.
Hsp72 was not observed in unstressed normal tissues, but was seen in histological studies of human tumors including breast, endometrial, lung, and prostate [2, 3, 45]. Often the expression of this protein was correlated with increased tumor cell proliferation and metastasis to lymph nodes as well as weakened responses to chemotherapeutic agents. Transgenic mice expressing human Hsp72 at high levels developed multiple myeloma [46]. In addition, forced (transgenic) expression of Hsp72 caused development of tumors in nude mice [47]. This HSP70 family member was also found on the plasma membrane of some tumor cells and could also be secreted to the extracellular space from cancerous cells [3]. Abundant expression of Hsp72 in cancer cells was found to correlate with histological grade in such malignancies, suggesting its significance as a cancer marker or target for therapy [48-50].
In their landmark review, Hanahan and Weinberg suggested a number of hallmarks, key phenotypic characteristics that are associated with a wide spectrum of cancers [51]. We will examine the potential role of Hsp72 in some of these hallmarks.
Ability to evade programmed cell death is a characteristic of many cancers [51]. Indeed, it was thought initially that Hsp72’s anti-apoptotic role played an important role in tumor development [52]. The anti-apoptotic role of Hsp72 might involve suppression of the pro-apoptotic c-Jun kinase pathway, which in turn might be important for tumor development. In addition, Hsp72 was thought to allow cancer cells to escape apoptotic cell death mediated by hypoxia, serum starvation, TNF, or FAS [53]. Overall, it was thought that depletion of this protein might cause rapid death of cancer cells whereas its overexpression prevented death. However, it was demonstrated subsequently that Hsp72 depletion was not associated with apoptotic death in prostate cancer cells such as PC-3 and Du-145 [54]. In addition, increased Hsp72 expression in cancer cells could not prevent intracellular pro-apoptotic signaling [54].
In contrast to apoptosis, Sherman et al. showed that the knockdown of Hsp72 could instead lead to senescence in a variety of cancerous cell lines such as MCF10A cells transformed by the Her2 oncogene, but not in untransformed epithelial cells [55]. The same group described mechanisms by which Hsp72 controlled senescence [56, 57]. They deduced that senescence in Her2 positive cancer cells was regulated by Hsp72 through the CDK inhibitory protein, p21 from studies using WT and Hsp72 knockdown cells and p21 KO cells [55]. Later it was shown that Hsp72-controlled senescence was regulated by the protein survivin and that forced expression of this protein almost completely reversed Hsp72 depletion-mediated senescence [55]. However, expression of survivin only could partially reverse Hsp72 depletion-mediated senescence in non-malignant MCF10A cells that have a fully functional p21 pathway, suggesting an important role for p21 in this effect. These effects of p21 and survivin on Hsp72 depletion-mediated senescence depended on the presence of functional p53 [55]. In Her2 positive, p53 mutant cells, Hsp72 depletion did not lead to upregulation of p21 but downregulated survivin, suggesting that survivin function is not regulated by p53 expression in these cells [58]. Thus, targeting Hsp72 in Her2 positive cancer cells would be an interesting approach in the anticancer drug development field as this chaperone plays an essential role in Her2-induced tumorigenesis in mice [55]. In addition, Hsp72 can also suppress senescence by modulating activity of the oncogenic co-chaperone Bag3 [59-61].
It was recently shown that aggressive triple negative breast cancer cells showed an increased expression of Hsp72. This protein was shown to promote survival in these cells through the mediation of the kinases Akt/PKB and PKC as well as the suppression of apoptotic signaling mentioned earlier [62, 63]. A role for Hsp72 in tumor metastasis was also confirmed earlier by many researchers. The metastatic properties of Hsp72 were thought to be regulated by 14-3-3 proteins as well as the activities of protein kinases and phosphatases in triple negative breast, cancer cells [64]. The implication of hsp72 genes in the initiation of cancer and metastasis was confirmed recently in a spontaneous mammary cancer model. Mice expressing the polyomavirus middle T oncogene under control of the mouse mammary tumor virus (MMTV) long terminal repeat developed spontaneous mammary carcinoma with high rate of metastasis to lung [65, 66]. Using this model, it was shown that hsp70 was required for robust initiation of tumors, maintenance of cancer stem cell populations, dissemination of cancer cells and metastasis. Teng et al, had demonstrated a mechanism through which Hsp72 might be involved in actin polymerization and that in turn might be responsible for the increased invasive and metastatic properties [67]. They showed WASF3, an actin polymerizing protein to be bound and stabilized by Hsp72, a process required for cellular movement and metastatic properties [67]. We were able to show that inhibiting Hsp72 activity is associated with loss of WASF3 (A. Murshid & SK Calderwood, unpublished) further suggesting its role as a chaperone in metastasis. Hsp72 was also shown to regulate the MET oncogene, a cell surface receptor which is involved in tumorigenesis and metastasis; inactivation of hsp72 genes led to loss of cells with stem cell markers and depleted the levels of active phospho-c-MET [22, 68].
Hsp72 also appears to play role in yet another of the hallmarks- ability to induce angiogenesis, through its influence on the primary sensor of tumor cell hypoxia- the transcription factor HIF1α [69]. HIF1 is regulated at the level of protein stability and increased concentrations of both Hsp72 and Hsp90 were required for its stabilization and accumulation in cancer cells [69]. In addition to its role in tumorigenesis, HIF1α is involved in the regulation of expression of other genes required for survival during stressful conditions such as hypoxia. Recently it has been reported that the transcription factor that mediates Hsp72 expression, HSF1 regulates a subset of HIF1α-regulated genes involved in tumor progression and also a set of cancer related miRNAs (Let-7, MiR-1991 or Mir-125B) [70].This effect might involve the RNA binding protein, HuR (a major regulator of translation, promote HIF-1α translation) which is also overexpressed in cancer and has strong correlation with cancer progression through HIF1α [71, 72]. Thus, a complex mechanism involving HSF1, Hsp72, HuR and HIF-1 regulation in tumor angiogenesis might be suggested.
The tumorigenic functions of HSP70 family members have therefore established them as attractive therapeutic targets for many human cancers. While research aimed at identifying HSP70 inhibitory molecules has accelerated in recent years, the development of such inhibitors is considered to be at an early stage and is a highly attractive prospect for future therapeutic developments.
The expression of HSP70 family members and their relationship with patient outcome varies between different HSP70 paralogs and within specific human cancer types [73-76]. Both pre-clinical and clinical studies have demonstrated the Hsp72 protein to be a promising target for numerous cancer types [22, 55, 76]. Targeting Hsp72 in cancer treatment fulfills important fundamental features of a promising therapeutic strategy. Firstly, Hsp72 exhibits divergent expression between cancer and normal tissue, with Hsp72 being expressed at low levels in normal tissue while being amplified in many human cancers (Fig. 3) [73]. Furthermore, reduced Hsp72 activity is well tolerated by non-transformed cells as demonstrated by numerous in vitro studies and loss of Hsp72 is also well tolerated by hsp70-/- mice [77, 78]. An additional feature of Hsp72 inhibition is that tumor toxicity is not dampened by activation of a counter-survival response, as is the case for HSR activation upon Hsp90 inhibition [79]
Ablation of Hsp72 expression has been shown to be toxic in many cancer models studied to date, while other certain cancer models have required dual inhibition of Hsp72 and Hsp73 for anti-cancer effects to be observed [78-80]. These studies highlighted a differential importance of HSP70 family members to the tumorigenicity of specific cancers and dually, the importance of identifying which HSP70 paralogs are subject to inhibition by a given HSP70 inhibitory molecule. Addressing these considerations may ultimately enable beneficial application of HSP70 inhibitors for cancer treatment i.e. using an inhibitor that targets the HSP70 paralogs known to be important for the cancer type being treated. Inhibitors of HSP70 family members identified to date that exhibit anti-cancer activity are listed in Fig. (4). Each of these molecules were shown to function by perturbing HSP70: (1) substrate binding (2) ATPase activity, or (3) interactions with regulatory proteins [81].
For cancer types where inhibition of individual HSP70 paralogs is desirable, targeting the SBD may prove a conducive strategy due to the relatively low degree of conservation shared within this region between HSP70 family members [81, 82]. As discussed above, HSP70 stabilizes numerous oncogenic client substrates through interaction with its SBD. HSP70 inhibitory molecules that target the SBD such as 2-phenylethynesulfonamide (PES) perturb substrate interaction with the SBD and lead to substrate de-stabilization and/or degradation [83, 84]. PES was suggested to target both Hsp72 and Hsp73 in a non-specific ‘detergent-like’ manner [80, 85]. Targeting a peptide-binding cleft within the SBD is a strategy that has previously been utilized with promising chemosensitization activity [86]. It was demonstrated that ectopic expression of short peptide sequences of the HSP70-interacting region of apoptosis-inducing factor (AIF) to compete with AIF sequestration by HSP70 and thereby permit AIF nuclear entry and promote apoptosis [86, 87]. While exogenous delivery of small peptide inhibitors to tumors remains a formidable challenge, the approach taken by Schmitt et al., may provide a foundation for the design and/or identification of HSP70 inhibitors through peptide mimicry.
Fig. (4))Inhibitors of Hsp 70 with anti-cancer activity. Molecules with both anti-cancer and HSP 70-inhibitory activity are shown and grouped by the functional region of HSP70 primarily targeted by the given molecule.Inhibition of the ATPase activity of HSP70 proteins prevents substrate protein release and this promotes the holding of substrates by the chaperone and/or degradation via the CHIP-proteasome pathway [80, 83]. Consistent with the tumorigenic role of HSP70 in cancer, inhibitors of ATPase activity showed anti-tumor effects in various cancer models both in vitro and in vivo [88-90]. Inhibitors of HSP70 ATPase activity such as VER155008 and MKT-077 act upon more than one family member. VER155008 has anti-ATPase activity towards Hsp72, Hsp73 and Grp78 and MKT-077 interacts with the mitochondrial Hsp70 protein encoded by HSPA9 as well as Hsp72 [89-91]. Thus, it is possible to inhibit multiple HSP70 species through one molecule that targets ATPase activity, and this approach might prove beneficial in contexts where multiple HSP70 paralogs have pro-tumorigenic functions in the same cancer type. In addition, Macias et al., demonstrated that inhibitors of the HSP70 ATPase domain could also possess specific affinity for single members of the family, further highlighting that the spectrum of HSP70 targets of a given ATPase inhibitor might be molecule specific [91].
Challenges exist, however, in targeting the ATPase domain of HSP70 proteins as a therapeutic approach. For example, the extreme affinity of the HSP70 NBD for ATP sets a high task for achieving competitive inhibition of this binding pocket [81]. In addition, the ATPase domain of HSP70 shares homology with similar structures in unrelated proteins and thus specificity may also prove problematic. For example MKT-077 was also reported to interact with F-actin in NIH-3T3 fibroblasts [81, 92]. In a recent study that utilized a structure-based modeling approach, five ‘druggable’ sites across HSP70 were identified [93]. Among these, the HSP70 inhibitor, YK5, was designed to form a covalent bond with the Cys267 residue of human Hsp72 and Hsp73 [93]. YK5 inhibited HSP70 refolding and ATPase activity while not directly competing for ATP binding. Targeting the Cys267 residue to form a covalent inhibitory interaction was a fine example of how intelligent molecule design could overcome some of the inherent challenges of drugging HSP70 such as its strong affinity to ATP [93].
The pleiotropic functions of HSP70 are facilitated by a large number of accessory proteins including nuclear exchange factors (NEFs), the DnaJ (HSP40) family of co-chaperones and proteins that contain a tetratricopeptide (TPR) domain that interact with the EEVD region at the C-terminal tail of Hsp70 [81]. Perturbing the interaction of HSP70 with accessory proteins is an anti-cancer strategy with promising potential, as indicated by a number of recent pre-clinical studies. For example, the inhibitor, YM-1, was shown to disrupt HSP70 complex formation with the NEF, Bag3. Inhibition of this interaction was associated with selective toxicity to transformed breast cancer cell lines over non-transformed cell lines. YM-1 was also shown to inhibit mammary and melanoma xenograft tumor growth in vivo [60]. A number of HSP70 inhibitors, including myricetin and MAL3-101 targeted the interaction of HSP70 with HSP40 members and had toxic properties in human cancer models including human multiple myeloma and pancreatic cancer cell lines [94, 96].
Inhibition of TPR protein interactions with the EEVD domain of HSP70 members is an avenue yet to be extensively investigated as an anti-cancer strategy [81]. The inhibitor, 15-deoxyspergualin, was shown to interact with the EEVD domain of HSP70. However limited efforts have been made to further investigate anti-cancer properties of 15-deoxyspergualin, since a phase II clinical trial of its use for metastatic breast cancer reported neuromuscular side effects with no benefit for disease [97, 98]. Whilst the outcome of disrupting interactions between HSP70 and accessory proteins may be highly specific to the given interaction, these studies have indicated that at least a selection of these HSP70-accessory protein relationships can indeed be targeted to achieve anti-cancer effects. These studies highlighted the prospect for more targets to emerge as the roles of HSP70-accessory protein interactions are characterized within different cancer contexts.
Although Hsp72 is important in the etiology of cancer, the mechanisms behind its involvement in the disease are still in flux. Indeed, it is possible that the chaperone may play distinct roles according to the dominant driver oncogenes in individual malignancies. At the molecular level, the canonical function of Hsp72 is to chaperone key proteins in tumorigenesis and permit their accumulation in the malignant cell. However, in addition, important roles for Hsp72 in RNA metabolism are emerging and may be significant. The existence of multiple HSP70 pseudogenes may also be important for future studies of potential lncRNA regulation of this family of chaperones. As the significance of this family of chaperones in cancer emerges, small molecule inhibitors are undergoing development as future potential cancer pharmaceuticals and this approach may emerge as an important “second front” in the war on cancer chaperones.
Not applicable.
The author declares no conflict of interest, financial or otherwise.
Declared none.