Published online before print August 24, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* INSERM U-517, Faculty of Medicine and Pharmacy, Dijon, France; and
Department of Hematology/Oncology, University Hospital Regensburg, Regensburg, Germany
2Correspondence: INSERM U517, 7 Boulevard Jeanne d'Arc, Faculty of Medicine, Dijon 21079, France. E-mail: cgarrido{at}u-bourgogne.fr
ABSTRACT
Stress or heat shock proteins (HSPs) are the most conserved proteins present in both prokaryotes and eukaryotes. Their expression is induced in response to a wide variety of physiological and environmental insults. These proteins play an essential role as molecular chaperones by assisting the correct folding of nascent and stress-accumulated misfolded proteins, and preventing their aggregation. HSPs have a dual function depending on their intracellular or extracellular location. Intracellular HSPs have a protective function. They allow the cells to survive lethal conditions. Various mechanisms have been proposed to account for the cytoprotective functions of HSPs. Several HSPs have also been demonstrated to directly interact with various components of the tightly regulated programmed cell death machinery, upstream and downstream of the mitochondrial events. On the other hand, extracellular located or membrane-bound HSPs mediate immunological functions. They can elicit an immune response modulated either by the adaptive or innate immune system. This review will focus on HSP27, HSP70, and HSP90. We will discuss the dual role of these HSPs, protective vs. immunogenic properties, making a special emphasis in their utility as targets in cancer therapy.
Key Words: apoptosis • immunogenicity
INTRODUCTION
Stress or heat shock proteins (HSPs) were first discovered in 1962 [1 ] as a set of highly conserved proteins whose expression was induced by different kinds of stress. It has subsequently been shown that most HSPs have strong cytoprotective effects and behave as molecular chaperones for other cellular proteins. Inappropriate activation of signaling pathways could occur during acute or chronic stress as a result of protein misfolding, protein aggregation, or disruption of regulatory complexes. The action of chaperones, through their properties in protein homeostasis, is thought to restore balance. Mammalian HSPs have been classified into five families, according to their molecular size: HSP100, HSP90, HSP70, HSP60, and the small HSPs. Each family of HSPs is composed of members expressed either constitutively or regulated inductively and are targeted to different subcellular compartments. For instance, although HSP90 is constitutively abundantly expressed in the cells, HSP70 and HSP27 are highly induced by different stresses such as heat, oxidative stress, or anticancer drugs. In normal, non stressed cells HSP70 and HSP27 are either not expressed or at very low levels. Once induced, these HSPs, which influence aggregation, transport, and folding of other proteins, also directly modulate the execution of the apoptotic signaling pathway. HSPs can act at multiple points in the apoptotic pathways to ensure that stress-induced damage does not inappropriately trigger cell death. In contrast to normal, nontransformed cells, HSP70 and/or HSP27 basal level is unusually high in cells or tissues from a wide range of tumors. Experimental rodent models have demonstrated that HSP27 and HSP70 increased the tumorigenic potential of cancer cells [2 , 3 ]. The contribution of HSPs to tumorigenesis may be attributed to their pleiotropic activities as molecular chaperones that provide the cancer cell with an opportunity to alter protein activities, in particular, components of the cell cycle machinery, kinases, and other proteins that influence tumor cell growth. HSP70 and HSP27 tumorigenic role has also been associated to their antiapoptotic properties [4 ].
Heat shock proteins can also have an extracellular location. HSP70 has been found externally expressed, bound to the plasma membrane. HSPs like HSP70, HSP90, and Gp96 have been found in the extracellular medium. Several mechanisms may account for the cellular release of HSPs, including the necrosis of the cells. The function of extracellular HSPs is immunogenic through the chaperoning of antigenic peptides. This review deals with the dual function of HSPs: the intracellular cytoprotective/antiapoptotic function, and the extracellular immunogenic function. In both cases, the interest of targeting HSPs in cancer therapy is discussed. We will focus mainly on HSP27, HSP70, and HSP90.
INTRACELLULAR FUNCTION OF HSPS
HSPs as cellular lifeguards: inhibitors of apoptosis
HSP27, a survival protein
HSP27 belongs to the subfamily of small HSPs, a group of proteins that vary in size from 15 to 30 kDa and share sequence homologies and biochemical properties such as phosphorylation and oligomerization. HSP27 can form oligomers up to 1000 kDa. The dimer of HSP27 seems to be the building block for the multimeric complexes. HSP27 oligomerization is a dynamic process that depends on the phosphorylation status of the protein and exposure to stress [5
, 6
]. HSP27 can be phosphorylated at three serine residues, and its dephosphorylation enhances oligomerization. This phosphorylation is a reversible process catalyzed by the MAPKAP kinases 2 and 3 in response to a variety of stresses, including differentiating agents, mitogens, inflammatory cytokines, such as tumor necrosis factor-
(TNF-) and IL-1ß, hydrogen peroxide and other oxidants. HSP27 is expressed in many cell types and tissues, at specific stages of development and differentiation [3
]. HSP27 is an ATP-independent chaperone, its main chaperone function being protection against protein aggregation [7
].
Overexpressed HSP27 protects against apoptotic cell death triggered by various stimuli, including hyperthermia, oxidative stress, staurosporine, ligation of the Fas/Apo-1/CD95 death receptor, and cytotoxic drugs [8 , 9 ]. HSP27 has been shown to interact and inhibit components of both stress- and receptor-induced apoptotic pathways (Fig. 1 ). We have demonstrated that HSP27 could prevent the activation of caspases [10 ]. It does so by directly sequestering cytochrome c when released from the mitochondria into the cytosol [5 ]. The heme group of cytochrome c is necessary but not sufficient for this interaction that involves amino acids 51 and 141 of HSP27 and requires dimerization of the stress protein. At higher HSP27 intracellular levels, the protein has been shown also to interfere with caspase activation upstream of the mitochondria. This effect seems to be related to the ability of HSP27 to stabilize actin microfilaments [11 ]. HSP27 binds to F-actin to prevent disruption of the cytoskeleton resulting from either heat shock, cytochalasin D, and other stresses [12 ]. In L929 murine fibrosarcoma cells exposed to cytochalasin D or staurosporine, overexpression of HSP27 prevents the cytoskeletal disruption and Bid intracellular redistribution that precede cytochrome c release [13 ]. HSP27 has also been shown to inhibit the mitochondrial release of Smac and thereby to confer resistance of multiple myeloma cells to dexamethasone [14 ] (Fig. 1) . HSP27 increases the antioxidant defense of cells by decreasing reactive oxygen species cell content [15 ] and neutralizes the toxic effects of oxidized proteins [16 ]. This latter effect may occur more specifically in neuronal cells, in which HSP27 protective effect does not depend on its interaction with cytochrome c and involves phosphorylated HSP27 [17 ].
 View larger version (36K): [in a new window] |
Figure 1. Modulation of apoptotic pathways by HSPs. At the mitochondrial level, HSP27, through Bid, and HSP70, by inhibiting Bax, inhibit the mitochondrial release of proapoptotic proteins. At the postmitochondrial level, HSP27 binds to cytochrome c, HSP70 and HSP90 bind to Apaf1 in all cases resulting in the inhibition of apoptosome formation and thereby prevention of caspase activation and apoptosis. HSP27 can also interact with and inhibit Daxx apoptotic pathway whereas HSP70 binds to JNK 1, resulting in inhibition of JNK activation. HSP90 interacts with RIP1 kinase and Akt, resulting in both cases in promotion of NF- B-mediated inhibition of apoptosis.
|
in different cancer cell lines could result from an increase in the activity of the survival transcription factor nuclear factor-B (NF-B). This is the result of an increase in the proteasomal degradation of the NF-B inhibitor IB. Under stress conditions, HSP27 increases IB ubiquitination/degradation, which results in an increase in NF-B activity and increased survival [22
]. In conclusion, HSP27 can interact with different partners implicated in the apoptotic process. Although the role of oligomerization/phosphorylation in the antiapoptotic functions of HSP27 remains uncertain, it is believed that this ATP-independent chaperone, for which no cochaperones have been described, seems to modulate its different protective properties by changing its oligomerization pattern (which is regulated by the phosphorylation status of the protein). We have demonstrated in vitro and in vivo that for HSP27 caspase-dependent antiapoptotic effect, large nonphosphorylated oligomers of HSP27 were the active form of the protein [5 ]. In contrast, for HSP27 thermotolerant function, small oligomers of HSP27 were necessary [23 ], and it is in its phosphorylated form that HSP27 directly interacts with Daxx [21 ]. These results suggest that the oligomerization/phosphorylation of the protein alters HSP27 conformation and hence determine its capacity to interact with different apoptotic proteins. Depending on the physiological needs of the cell, HSP27 may be phosphorylated, and the equilibrium between large and small oligomers may shift. As a consequence, the affinity of the protein for one or another subtracts increases, and thereby, the relevance of HSP27's protective effect of one key apoptotic protein with respect to the other may be different depending on the cellular needs to survive.
HSP70: a powerful antiapoptotic protein
The HSP70 family constitutes the most conserved and best studied class of HSPs. Human cells contain several HSP70 family members, including stress-inducible HSP70, constitutively expressed HSP70, mitochondrial HSP75, and GRP78, localized in the endoplasmic reticulum [24
]. Under normal conditions, HSP70 proteins function as ATP-dependent molecular chaperones by assisting the folding of newly synthesized polypeptides, the assembly of multiprotein complexes, and the transport of proteins across cellular membranes [25
]. A panoply of stimuli, including anticancer agents, induces the synthesis of stress-inducible HSP70, which enhances the ability of the cell to survive those otherwise lethal conditions. Gene ablation studies show that HSP70 plays an important role in apoptosis. Recently, a mouse line lacking hsp70.1 and hsp70.3 was generated. These cells are very sensitive to apoptosis induced by a wide range of lethal stimuli [26
]. The testis-specific isoform of HSP70 (hsp70.2) when ablated results in germ cell apoptosis [27
].
Induced HSP70 levels block the apoptotic pathway at different levels (Fig. 1) . HSP70 reduces or blocks caspase activation and suppresses mitochondrial damage and nuclear fragmentation [28 ]. These findings were supported by Li et al. [29 ], who found HSP70 inhibited apoptosis downstream of the release of cytochrome c and upstream of the activation of caspase-3. This antiapoptotic effect was explained by the HSP70-mediated modulation of the apoptosome. Indeed, HSP70 has been demonstrated to directly bind to apoptosis protease-activating factor-1 (Apaf-1), thereby preventing the recruitment of procaspase-9 to the apoptosome [30 ]. The ATPase domain of HSP70 has been described to be necessary for this interaction [31 ]. Other reports have shown that HSP70 interacts with procaspase-3 and procaspase-7 and prevents their maturation, thereby inhibiting the caspase-dependent apoptotic signaling [32 ]. However, these results have been contradicted by a study in which authors demonstrate that the inhibition of caspase-dependent apoptosis by HSP70 results from an inhibition of cytochrome c release from the mitochondria and not from any direct effect of HSP70 in caspase activation. They explain this contradictory result by showing that it is a high salt concentration but not HSP70 that inhibits caspase activation [33 ].
HSP70 can also prevent caspase-independent apoptosis pathways. HSP70 prevents cell death under conditions in which caspase activation does not occur, because of the addition of exogenous caspase inhibitors [34 ]. We have recently observed that overexpression of HSP70 protects Apaf-1–/– cells from apoptotic cell death induced by serum withdrawal [35 ], indicating that the cytochrome c/Apaf-1/caspase was not the sole pathway of the antiapoptotic action of HSP70. Indeed, HSP70 directly binds to apoptosis-inducing factor (AIF) and inhibits AIF-induced chromatin condensation. HSP70 was found to neutralize the apoptogenic effects of AIF in cell-free systems, in intact cells microinjected with recombinant HSP70 and/or AIF protein, as well as in cells transiently transfected with AIF cDNA. Of note, endogenous levels of HSP70 seem to be sufficiently high to control AIF-mediated apoptosis, since down-regulation of HSP70 by an antisense construct sensitized the cells to serum withdrawal and AIF [35 ].
HSP70 can also rescue cells from a later phase of apoptosis than any known survival-enhancing drug or protein. In TNF-induced apoptosis, HSP70 does not preclude the activation of caspase-3 but prevents downstream morphological changes that are characteristic of dying cells like activation of phospholipase A2 and changes in nuclear morphology [36 ]. During the final phases of apoptosis, chromosomal DNA is digested by the DNase CAD (caspase-activated DNase) following activation by caspase-3. The enzymatic activity and proper folding of CAD has been reported to be regulated by HSP70, its cochaperone HSP40, and ICAD, the inhibitor of CAD. ICAD recognizes an intermediate folding state conferred by HSP70-HSP40 [37 ]. It has also been reported in TCR-stimulated T cells that HSP70 binds CAD and enhances its activity [38 ]. Another final target of caspase-3 is the transcription factor GATA-1. We have demonstrated in hematopoietic cells that HSP70 can protect GATA-1 from caspase-3 cleavage. As a consequence, cells do not die by apoptosis but instead differentiate [39 ].
HSP70 has also been shown to act at the premitochondrial stage by inhibiting stress-activated kinases. HSP70 binds to and functions as a natural inhibitory protein of c-Jun N-terminal Kinase (JNK1) [40 ] The ATPase domain of HSP70 was dispensable for this binding [31 ]. HSP70 also appears to affect the Bid-dependent apoptotic pathway. HSP70 is able to inhibit TNF-induced cell death. However, this protective effect is lost in Bid homozygous-deleted MEF cells [31 ]. It has been suggested that HSP70, by inhibiting JNK activation, could affect Bid-dependent cell death, by a mechanism that is not fully clear [41 ]. In apoptosis induced by hyperosmolarity, HSP70 has been found to modulate JNK and ERK phosphorylation [42 ]. Also, at the premitochondrial level, HSP70 has been shown to inhibit p53 [43 ] and to lysosomal membrane permeabilization [44 ]. At the mitochondrial level, HSP70 blocks Bax translocation, preventing mitochondrial outer membrane permeabilization and thereby inhibiting cytochrome c and AIF mitochondrial release [45 ]. This HSP70 function depends on both the chaperone and the ATP hydrolytic domains [46 ].
HSP70 can act at the death receptor level to mediate Bcr-Abl-mediated resistance to apoptosis in human acute leukemia cells. HSP70 binds to the death receptors DR4 and DR5, thereby inhibiting TRAIL-induced assembly and activity of death-inducing signaling complex (DISC) [47 ]. Exposure of hematopoietic cells to TNF induces the activity of the proapoptotic double-stranded RNA-dependent protein kinase (PKR). An inhibitor of PKR is the Fanconi anemia complementation group C gene product (FANCC). HSP70 interacts with the FANCC protein via its ATPase domain and, together with HSP40, inhibits TNF-induced apoptosis through the ternary complex HSP70, FANCC, and PKR [48 , 49 ]. HSP70 has been shown to bind to nonphosphorylated protein kinase C (PKC) via the kinase’s unphosphorylated carboxyl terminus, priming the kinase for rephosphorylation and stabilizing the protein. In a similar manner, HSP70 binds Akt, resulting in its stabilization [50 ].
Another apoptosis regulatory protein interacting with HSP70 is Bag-1. Bag-1 has been reported to function as a cochaperone of HSP70 and simultaneously regulates the activities of proteins such as Bcl-2 and Raf-1. It has been shown that HSP70/Bag-1 regulates Raf-1/ERK kinase and cell growth in response to stress [51 , 52 ]. Whether the HSP70-Bag-1 interaction is important for HSP70-mediated apoptosis regulation is unknown. Finally, HSP70 is also involved in the inhibition of cathepsines, lysosome proteases also involved in apoptosis [53 ].
In conclusion, HSP70 is a decisive negative regulator of the mitochondrial pathway of apoptosis that can block apoptosis at different levels: at a premitochondrial stage by inhibiting stress-inducing signaling, at the mitochondrial stage by preventing mitochondrial membrane permeabilization through the blockage of Bax translocation, and finally, at the postmitochondrial level by interacting with AIF and Apaf-1 (Fig. 1) .
Cyto-protective functions of HSP90
Prominent members of the HSP90 family of proteins are HSP90 and HSP90ß [54
]. These two HSP90 isoforms are essential for the viability of eukaryotic cells. They are rather abundant constitutively, make up 1-2% of cytosolic proteins, and can be further stimulated in their expression level by stress.
HSP90 associates with a number of signaling proteins, including ligand-dependent transcription factors such as steroid receptor [55 ], ligand-independent transcription factors such as MyoD [56 ], tyrosine kinases such as v-Src [57 ], and serine/threonine kinases such as Raf-1 [58 ]. The main chaperone role of HSP90 is to promote the conformational maturation of these receptors and signal-transducing kinases. HSP90, like HSP70 and HSP60, binds ATP and undergoes a conformational change upon ATP binding needed to facilitate the refolding of denatured proteins.
HSP90 overexpression in human leukemic U937 cells can inhibit apoptosis induced by staurosporin and can prevent the activation of caspases in cytosolic extracts treated with cytochrome c [59
]. Pandey et al. [59
] reported that HSP90 inhibited apoptosis as a result of a negative effect on Apaf-1 function. HSP90 directly binds Apaf-1 and inhibits its oligomerization and further recruitment of procaspase-9 (Fig. 1)
. The anti-apoptotic action of HSP90 is also reflected by its capacity to interact with phosphorylated serine/threonine kinase Akt/PKB, a protein that generates a survival signal in response to growth factor stimulation. Phosphorylated Akt can phosphorylate the Bcl-2 family protein Bad and caspase-9 [60
], leading to their inactivation and to cell survival. But Akt has been also shown to phosphorylate IB kinase, which results in promotion of NF-B-mediated inhibition of apoptosis [19
].
HSP90 has also been shown to interact with and stabilize the receptor interacting protein (RIP). Upon ligation of TNFR-1, RIP-1 is recruited to the receptor and promotes the activation of NF-B and JNK. Degradation of RIP-1 in the absence of HSP90 precludes activation of NF-B mediated by tumor necrosis factor- (TNF-) and sensitizes cells to apoptosis [61
]. Another route by which HSP90 can affect NF-B survival activity is via the IKK complex. This complex is composed of two catalytic and one regulatory subunit, and recently, it was determined that HSP90 and Cdc37 were also present, with the association mediated through the kinase domain of the catalytic subunits [62
].
Another pathway of cell survival in which HSP90 can be involved implied p53. It has been demonstrated that p53 represses HSP90 gene expression in UV-irradiated cells [63
]. Other client proteins of HSP90 through which this chaperone could participate in cell survival are the transcription factors Her2 and Hif1a [64
, 65
].
Finally, HSP90 proteins may also protect apoptosis by inhibiting the action of the calcium-dependent proteases calpains. The HSP90 family protein Grp94 was shown to cleave calpain and to protect human neuroblastoma cells from hypoxia/reoxygenation-induced apoptosis involving calpains [66 ].
In conclusion, HSP90 seems to have different molecular partners depending on the cell death stimuli. It should be noted that most studies do not differentiate between the and ß isoforms of HSP90.
Intracellular HSPs expression in tumors
HSPs tumorigenic properties: markers for poor prognosis?
Rat colon cancer cells engineered to express human HSP27 were observed to form more aggressive tumors in syngeneic animals than control cells, and the increase in tumorigenicity correlated with a reduced rate of tumor cell apoptosis [3
]. Overexpressed HSP27 did not increase the tumorigenicity of these rat colon cancer cells, nor did it affect their survival rate in vivo when inoculated into immunodeficient animals, suggesting that HSP27 somehow subverts the tumor-specific immune response. However, HSP27 overexpression was also reported to increase the metastatic potential of human breast cancer cells inoculated into athymic (nude) mice [67
]. Conversely, HSP27 antisense oligonucleotides-induced silencing in prostate tumors enhance apoptosis and delay tumor progression [68
]. Like HSP27, HSP70 has been shown to increase the tumorigenicity of cancer cells in rodent models [2
]. Down-regulation of HSP70 induced cancer cell death or increased their sensibility to die depending on the cellular model. As a result, in experimental models, HSP70 down-regulation strongly decreases tumorigenicity [4
, 69
].
Clinically, in a number of cancers such as breast cancer, ovarian cancer, osteosarcoma, endometrial cancer, and leukemias, an increased level of HSP27, relative to its level in nontransformed cells, has been detected [3 ]. In ovarian tumors, HSP27 expression increases with the stage of the tumor [30 ]. In addition, the pattern of HSP27 phosphorylation in tumor cells is different from that observed in primary nontransformed cells [55 ]. Consequently, the diversity of HSP27 isoforms may also represent a useful tumor marker, as recently demonstrated in human renal cell carcinomas [70 ]. Increased expression of HSP70 has been reported in high-grade malignant tumors such as endometrial cancer, osteosarcoma, and renal cell tumors [71 , 72 ]. HSP70 is very abundant in Bcr-Abl human leukemia cells, and the GATA-RE element found in hsp70 promotor is necessary for this accumulation [73 ]. High expression of HSP70 in breast, endometrial, or gastric cancer has been associated with metastasis, poor prognosis, and resistance to chemotherapy or radiation therapy [74 75 76 ]. HSP70.2, a member of the HSP70 family that is normally expressed only in spermatogenesis, is present at high levels in breast cancer and inhibits the onset of senescence. HSP70, along with PSA, are good tumor markers to identify patients with early-stage prostate cancer [77 ]. However, HSP27 or HSP70 are not universal markers of poor prognosis. Even though HSP70 levels correlate with malignancy in osteosarcoma and renal cell tumors, its expression is paradoxically associated with improved prognosis [71 , 78 ].
Concerning HSP90, this abundant protein is overexpressed in breast tumors, lung cancer, leukemias, and Hodgkin's disease [79 ]. HSP90 has also been shown to be overexpressed in B cell non-Hodgkin lymphomas compared with normal B cells [80 ]. In breast cancer, overexpression of HSP90, like HSP70, correlates with poor prognosis [81 ]. The c-myc proto-oncogene directly activates HSP90 transcription in different tumor cell models [82 ]. HSP90 has a role in facilitating the emergence of polymorphisms and mutations that support the evolution of resistant clones. This chaperone stabilizes the conformations of mutant proteins that arise during transformation such as Bcr-Abl and p53 [83 84 85 ].
Anticancer therapeutic approach: the inhibition of intracellular HSPs
Constitutively high HSP expression is a property of and essential for the survival of most cancers. Neutralizing HSPs is therefore an attractive strategy for anticancer therapy. Up to now, only inhibitors of HSP90 have been available that can be clinically tested. This is the case of the benzoquinone ansamycin antibiotic geldanamycin and its analog 17-AAG (17-allylamino-17-deemethoxygeldanamycin), two drugs that are actually undergoing phase I and II clinical trials for anticancer activity. Pharmacological targeting of HSP90 with specific chemical inhibitors leads to degradation of the client proteins and inhibition of tumor growth trough G1 arrest and activation of apoptosis [86
, 87
]. The fact that geldanamycin and 17-AAG selectively kill cancer cells has been rationalized by assuming that tumor cells, as compared with their normal counterparts, would exhibit a stressed phenotype, with an enhanced dependency on the cytoprotective action of HSP90. In tumors, HSP90 is present entirely in multichaperone complexes with high ATPase activity, whereas the HSP90 form in normal tissues is present in an uncomplexed state [83
, 84
]. These drug inhibitors of HSP90 indirectly target diverse proteins required for malignant cell growth and show considerable promise in the trials [88
, 89
]. However, this novel class of anticancer drugs strongly induces HSP70 [90
], a highly protective protein that strongly reduces the cell death-sensitizing effect provoked by the HSP90 inhibitors. Zaarur et al. knocked out the HSF-1 using a si-RNA in several cancer cells, which become more sensitive to 17-AAG [90
]. Enhancement of 17-AAG activity has also been reported for proteasome inhibitors such as bortezomib. This effect might result from the increase in protein misfolding that is induced by 17-AAG, coupled to the impaired clearance of proteins by the ubiquitin-proteasome pathway [91
, 92
].
An inhibitor of HSP70 would be very useful in cancer therapy alone and in combination with the mentioned inhibitors of HSP90. We and others have extensively reported that HSP70 antisense constructs have chemosensitizing properties and may even kill cancer cell lines (in the context of adenoviral infection) in the absence of additional stimuli [93 , 94 ]. The cytotoxic effect of HSP70 down-modulation is particularly strong in transformed cells yet undetectable in normal, nontransformed cell lines or primary cells [26 ]. Studies in Bcr-Abl human leukemia cells show that HSP70 is a promising therapeutic target for reversing cancer cell drug resistance, probably by its ability to inhibit apoptosis both upstream and downstream of the mitochondria [47 , 73 ]. Unfortunately, thus far, no small molecules that would selectively inhibit cytosolic HSP70 are available. We have recently demonstrated that rationally engineered decoy targets of HSP70-derived from AIF can sensitize cancer cells to apoptosis induction by neutralizing HSP70 function. These AIF-derived peptides all carry the AIF region from aa 150 to aa 228, previously defined as required for HSP70 binding [93 ]. These constructs bind to HSP70 but lack an apoptotic function. Experiments using different cancer cell lines (leukemia, colon cancer, breast cancer, and cervical cancer) demonstrate that certain of these AIF derivative inhibitors of HSP70 strongly increase the sensitivity of cancer cells to chemotherapy in vitro. This effect was merely related to their ability to neutralize endogenous HSP70 because this proapoptotic activity was lost in HSP70-negative cells [26 ]. In vivo, in a syngeneic rat colon cancer cell model and in a mice model of melanoma (B16F10), these inhibitors, called ADD70 (for AIF-derived decoy for HSP70), decreased the size of the tumors for the rats and provoked an important delay in the growth of the mice tumors. In addition, ADD70 sensitizes both the rat colon cancer cells and mouse melanoma cells to the chemotherapeutic agent cisplatin. This ADD70 antitumorigenic effect was only observed in syngeneic animals but not in immunodeficient animals. We have demonstrated that ADD70 antitumorigenic effects are associated with an increase in tumor-infiltrating cytotoxic CD8+ T-cells [90 ]. Therefore, a positive strategy aimed at interfering with HSP70, as opposed to negative strategies based on antisense constructs or iRNA interference, is feasible for chemosensitization, at least in vitro and in vivo in experimental models. The future will tell whether a similar strategy may allow for the chemosensitization of HSP70-expressing human tumors.
Concerning HSP27, phosphorothioate HSP27 antisense oligonucleotides have demonstrated in prostate cancer to enhance apoptosis and delay tumor progression [68
]. Paclitaxel, by inhibiting HSP27 expression, seems to overcome drug resistance to etoposide, colcemid, and vincristine in ovarian and uterine cancer cells in vitro [95
]. Antisense strategies have also demonstrated that lymphomas and multiple myelomas can be sensitized to chemotherapeutic drugs like dexamethasone and the inhibitor of proteasome Velcade (PS-341). In dexamethasone-resistant cell lines, HSP27 is overexpressed. Its down-regulation by siRNA restores the apoptotic response to dexamethasone by triggering caspase activation [96
]. We have demonstrated that HSP27 participates in proteins' ubiquitination/proteasomal degradation and that this effect contributes to its protective functions by enhancing the activity of proteins like NF-B [22
]. Velcade, currently tested in clinical trials involving multiple myelomas, has been shown in vitro to induce apoptosis in several cancer cell lines. HSP27 confers Velcade resistance, and an HSP27 antisense approach sensitizes cells to Velcade-induced apoptosis [14
, 96
]. It is therefore tempting to conclude that a combinational therapy using velcade together with an inhibitor of HSP27 will increase the chemosensitization effect of both products. In conclusion, all of these findings demonstrate an interest in developing novel therapeutic drugs targeting HSPs to improve patient outcome in different cancers.
EXTRACELLULAR FUNCTIONS OF HSPS
HSPs are present in the extracellular space
Apart from their protective roles in the cytosol, HSPs have been found to play key roles in the stimulation of the immune system when located in the extracellular space or on the plasma membrane. In humans, their presence in the serum is associated with stress conditions, including inflammation, bacterial, and viral infections. In vitro, members of the HSP70 and HSP90 families have been detected in the medium of antigen-presenting cells (APCs). The active release of HSP70 from viable tumor cells could be further enhanced after exogenous stress, including proinflammatory cytokines [97
]. Although the immunological role of extracellular and membrane-bound HSPs appears apparent, the mechanism of transport to the plasma membrane, the membrane anchorage, and the export remains enigmatic. Cytosolic HSPs do not contain leader peptides enabling membrane localization. However, transport of other proteins across lipid membranes is one of their major tasks. Regarding these results, it is conceivable to assume that cytosolic HSPs are transported to the plasma membrane in concert with other proteins possessing transmembrane domains that fulfill shuttle functions. Presently, the molecular nature of these associated proteins has not yet been identified. Another possibility for membrane anchorage might be a direct interaction of HSPs with lipid components. DeMaio et al. showed an association of members of the HSP70 family with phosphatidylserine (PS) in PC12 tumor cells [98
]. It can be hypothesized that after binding of HSP70 to PS, a flip-flop mechanism similar to that shown for annexin might facilitate the transport of HSP70 from inside the cell to the outer membrane leaflet. The same group demonstrated that HSP70s have the capacity to induce ion conductance channels in artificial lipid bilayers [99
]. A proteomic profiling of cholesterol-rich membrane microdomains also termed as lipid rafts revealed the presence of signaling and trafficking factors and of members of the HSP70 and HSP90 families [100
, 101
]. Upon stimulation with lipopolysaccharides (LPS), an association of HSP70/HSP90 with chemokine receptors, TLR-4/CD14 clusters in lipid rafts was initiated [102
, 103
]. These data indicate that bacterial recognition by the innate immune system is based on the recruitment of a multimeric receptor complex, including HSPs within lipid rafts.
For Gp96, an endoplasmic reticulum-residing member of the HSP90 family, it was speculated that transport to the plasma membrane is enabled by masking of the endoplasmic reticulum (ER)-retention sequence KDEL [104 ]. An alternative vesicular pathway bypassing the ER-Golgi route was hypothesized for HSP70 [105 ]. Other examples for this non-ER-Golgi route are IL-1-ß, a mediator of inflammation, lacking a signal sequence and basic fibroblast growth factor [106 ]. Both molecules are frequently found in the plasma membrane and in the extracellular space of viable cells.
Necrosis is another mechanism by which HSPs have been demonstrated to be exported into the medium [107 ]. However, because the amount of free HSP70 proteins in the medium is very low, this nonspecific mechanism seems rather unlikely. Our group determined an active release of HSP70 in concert with Bag-4 from viable human colon and pancreatic carcinoma cells in detergent-soluble lipid vesicles. Biochemical and biophysical properties, including density on a sucrose gradient, acetylcholine esterase activity, and protein composition identified them as exosomes. An enrichment of the small GTPase Rab-4 inside and on the exosomal surface of tumor-derived exosomes documented their intracellular transport route from the early endosomal compartment to the plasma membrane, and finally to the extracellular space [108 ]. In line with these results, the group of Febbraio described the secretion of HSP70s from peripheral blood mononuclear cells (PBMNCs) by exosomes [109 ]. However, in contrast to other groups, they exclude the involvement of lipid rafts in the process of HSP70 release from PBMNCs.
Immunological role of extracellular heat shock proteins
Peptide-carrier function
HSPs with molecular weights of 70 and 90 kDa were identified as key regulators of the host's immune system. After cross-presentation of HSP-chaperoned peptides on MHC class I molecules [110
111
112
113
114
], an antigen-specific CD8+ T cell response is initiated in vitro and in tumor mouse models [115
116
117
118
119
120
]. Cross-presentation describes the transfer of exogenous peptides into the MHC class I pathway via an endosomal pathway. HSPs have been found to be important players in the process of cross-presentation of tumor-derived, antigenic peptides. The uptake of HSP-peptide complexes by antigen-presenting cells (APCs) was found to be specific, saturable, and concentration-dependent [110
, 121
, 122
], and thus the existence of HSP-specific receptors was hypothesized [113
]. The heterodimeric TLR2/TLR4 cluster in combination with the lipopolysaccharide (LPS) receptor CD14 were identified as relevant receptor complexes mediating the uptake of HSP70/HSC70 [123
, 124
], and Gp96 [125
].
Depending on the source of HSPs, different functions have been reported. Binding of mycobacterial HSP70 to CD40 was found to mediate a calcium-dependent cell signaling and the release of CC chemokines, proinflammatory cytokines, and nitric oxide [126 , 127 ], whereas mammalian HSP70s were found to facilitate receptor-mediated endocytosis [128 ]. Recently, scavenger receptors, including LOX-1, SR-A were also discussed to play pivotal roles in the uptake of HSP-peptide complexes [128 , 129 ]. Gp96 has been found to compete with ß-2 microglobulin for binding to the receptor CD91 [112 , 130 ]. After uptake of HSP-peptide complexes into APCs, processing and representation of HSP-chaperoned peptides on MHC class I molecules, a CD8+ cytotoxic T lymphocyte (CTL) response is initiated [116 , 118 , 131 ]. A schematic representation of the carrier function of HSPs and its immunological relevance for cross-presentation is shown schematically in Fig. 2 .
 View larger version (32K): [in a new window] |
Figure 2. Carrier function of HSPs. HSP-chaperoned antigenic peptides derived from tumor cells are internalized by endocytosis via HSP receptors (HSP-R; i.e., CD91, CD40, TLR2/4 ± CD14, CD35, LOX-1, SR-A). After uptake and processing, the peptides are represented on MHC class I molecules of the APC and thus induce a CD8+ T cell response against cancer.
|
[133
]. Yet it remains unclear whether HSP70s are released as free soluble proteins or in detergent-soluble membrane vesicles. Exosomes have been discussed as potent export vehicles for HSP70 s from the endosomal compartment into the extracellular space [108
, 109
].
An interaction of peptide-free HSP70 with CD14, a glycosylphosphatidylinositol (GPI)-anchored receptor, and TLR2/4 on APCs initiates the release of proinflammatory cytokines, including TNF-, IL-1ß, IL-12, IL-6, and GM-CSF [123
, 127
, 134
], as shown in Fig. 3
. This process is triggered by the translocation of NF-B into the nucleus [123
]. These cytokines result in a nonspecific stimulation of the innate immune system.
 View larger version (26K): [in a new window] |
Figure 3. Cytokine-inducing effects of HSP70. In the absence of antigenic, tumor-derived peptides, HSPs engage signaling receptors (e.g., TLR2/4 plus CD14) triggering NF-B activation. The APCs release proinflammatory cytokines (e.g., IL-1ß, IL-6, TNF-) leading to a nonspecific stimulation of the innate immune system. GPI, glycosylphosphatidylinositol.
|
Immunostimulation of NK cells
NK cells are important effector cells of the innate immune system. NK effector functions are regulated by a number of killer cell inhibitory and activating receptors. These receptors either belong to the killer cell immunoglobulin-like (KIR), the immunoglobulin-like transcript (ILT), C-type lectin receptor [141
], or the natural cytotoxicity receptor (NCR) families [142
]. Depending on their intracellularly located immunoreceptor tyrosine-based inhibitory or activatory motifs, these receptors mediate NK cell activation or tolerance toward self-antigens [142
, 143
]. According to the "missing self" theory [144
], tumor cells with an altered or missing MHC expression pattern provide ideal targets for the cytolytic attack mediated by NK cells. However, evidence is accumulating that apart from missing self, additional activating signals are necessary to mediate an efficient NK cell activation. Under physiological conditions, nonclassical HLA-E molecules presenting leader peptides of HLA-A, -B, and -C alleles serve as ligands for the inhibitory heteromeric receptor complex CD94/NKG2A. After stress, an HSP60-associated signaling peptide competes with HLA leader peptides for binding to HLA-E. HLA-E/Hsp60-peptide complexes are no longer recognized by the inhibitory receptor complex CD94/NKG2A and thus these stressed cells provide targets for the cytolytic attack of NK cells [145
, 146
]. These data provide a first hint that environmental stress is able to modulate the immune response of the innate immune system.
In line with these findings, we identified HSP70 as a triggering factor for NK cells with a high cell surface density of CD94 [147 , 148 ]. Mapping of the HSP70 sequence revealed that the 14-mer peptide T-K-D-N-N-L-L-G-R-F-E-L-S-G (TKD), derived from HSP70 C-terminal substrate binding domain (aa 450-463), has similar immunostimulatory capacity on NK cells like full-length HSP70 or the isolated C-terminal domain [149 , 150 ]. Incubation of NK cells with cytokines plus soluble HSP70 protein or TKD peptide enhances the cell surface density of activating NK cell receptors, including CD94 [147 ]. Antibody blocking studies also suggested an important role for CD94 in the interaction of NK cells with HSP70 presented on tumor cells [148 ]. Concomitantly, the cytolytic and migratory capacity of resting NK cells was found to be initiated [108 ].
A broad screening program of human tumor biopsies revealed that HSP70 is frequently present on the plasma membrane of colon, lung, pancreas, mammary, head and neck and metastases derived thereof [62 , 151 ]. Also bone marrow-derived leukemic blasts from patients with hematological malignancies, including acute and chronic myeloid leukemia are frequently HSP70 membrane-positive [152 ]. Interestingly, the corresponding normal tissues were always found to be HSP70 membrane-negative. The cell surface density of HSP70 on tumors could be further enhanced by clinically applied reagents, and procedures, including membrane-interactive alkyl-lysophospholipids [153 ], cytostatic drugs, including taxoides and vincristinsulfate [154 ], cyclooxygenase (COX-1/2) inhibitors, acetyl salicyl acid, insulin sensitizers [155 ], hyperthermia [156 ], radiation, and photodynamic therapy [157 ]. As expected, the stress-inducible increased HSP70 surface density correlated with an increased sensitivity toward NK cell-mediated cell killing. These data indicate that an NK cell-based immunotherapy could be combined with standard chemo-/radiotherapeutical approaches.
Apart from HSP70, a variety of other chaperones were found to be present on the plasma membrane of tumor cell lines, as determined by selective cell surface protein profiling by Bong Kyung Shin and co-workers [158 ]. However, predominantly, members of the HSP70 and HSP90 group have the capacity to stimulate the innate immune system when presented on the cell surface [62 , 159 , 160 ].
The mechanism of lysis of HSP70 membrane-positive tumor cells was biochemically characterized as granzyme B (GrB)-mediated but perforin-independent apoptosis [161 ]. We have experimental evidence that, apart from this vesicular-mediated uptake of GrB, which requires perforin, enzymatically active GrB can be taken up through an alternative pathway, which is facilitated by membrane-bound HSP70. Full-length HSP70, as well as the 14-mer Hsp70 peptide TKD, which is exposed to the extracellular milieu of tumor cells [149 ], both have the capacity to bind GrB as determined by affinity chromatography [161 ]. Further, sepharose columns coupled to GrB precipitated HSP70 and HSP27 from cell lysates [162 ]. We have demonstrated that GrB, even in the absence of perforin, is selectively taken up by HSP70 membrane-positive tumor cells and thus causes apoptosis [161 ]. These data led us to the following working hypothesis: in a first step, GrB binds to membrane-bound HSP70 on tumor cells; then HSP70 facilitates uptake of GrB into the tumor cells and thus initiates classical apoptosis in a caspase-dependent manner.
Anticancer therapeutical approaches based on extracellular HSPs
Most HSPs-based immunotherapeutical approaches against cancer exploit their carrier function for immunogenic peptides [163
, 164
]. HSP70 and Gp96 peptide complexes purified from patient-derived tumors were used as a vaccine to treat and prevent cancer. In contrast, HSP preparations derived from normal tissues did not induce an anticancer immune response, as determined in animal models. These results indicated that the immunogenicity in this approach is dependent on the tumor-specific peptides chaperoned by HSPs. Besides their chaperone activity, HSP-peptide complexes are internalized into antigen-presenting cells by a receptor-mediated endocytosis, then they traffic into cellular compartments where the chaperoned peptides are released, processed, and represented on MHC class I molecules. Concomitantly, the receptor-mediated uptake initiates the maturation of antigen-presenting cells. Oncophage, an individualized Gp96-peptide vaccine was tested in a phase II clinical trial in 29 patients with metastatic colorectal cancer who had undergone complete surgical removal [165
, 166
]. After surgery, all patients received two cycles of the Oncophage vaccine. More than half of the patients demonstrated significant immunological responses that correlated with a statistically significant survival benefit. The immune responses induced in the vaccinated patients included CD8 T cells directed against shared tumor antigens. Clinical responders showed a two-year overall survival rate of 100%, compared with 50% for nonresponders, and a disease-free survival rate of 51%, compared with 8% among nonresponders. In summary, vaccination with Gp96 derived from autologous tumors represents a feasible and safe approach to induce an active antitumor response in about half of the patients. These encouraging results initiated phase III clinical trials for metastatic melanoma and kidney cancer and several other clinical studies on lymphoma, pancreatic, and gastric cancers. The results of these studies await final analysis. A scientific issue of the approach might be to elucidate the nature of HSP-chaperoned peptides because after purification of the HSP-peptide complexes form patient-derived tumors, the amount of vaccine is limited.
Our group is exploring the effect of the HSP70-derived peptide TKD in the stimulation of resting NK cells against HSP70 membrane-positive tumors. Membrane-bound HSP70 serves as a tumor-selective target structure, since HSP70 is frequently presented on the plasma membrane of tumors and metastases but not in normal tissues. Incubation of peripheral blood lymphocytes with TKD peptide plus a low dose of interleukin 2 (IL-2) initiates the cytolytic and migratory capacity of NK cells toward HSP70 membrane-positive tumor cells in vitro and in a xenograft tumor mouse model [167 ]. Encouraged by these results, a phase I clinical trial was performed in patients with therapy-refractory, metastasized colorectal and non-small lung cell carcinoma to study the tolerability and feasibility of TKD-activated NK cells. After ex vivo stimulation of the leukapheresis product with IL-2/TKD, the peptide was removed and the activated autologous cells were reinfused intravenously. The procedure was repeated up to six cycles, applying a dose escalation schedule in four patients. Reinfusion of TKD-activated NK cells was well tolerated, feasible, and safe. Ten of twelve patients showed significant immunological responses, including an up-regulated cytolytic activity against HSP70 membrane-positive tumors and an increase in the cell surface density of activatory NK cell receptors, including the C-type lectin receptor CD94, which serves as a surrogate marker for an HSP70 reactivity. Moreover, two of five patients receiving more than four treatment cycles showed clinical responses. This result was not expected since all patients had progressive tumor disease during their last standard chemoradiotherapy [168 ]. The advantage of the approach is the excellent safety profile and the unlimited availability of the synthetic HSP70 peptide TKD, which stimulates NK cells.
CONCLUDING REMARKS
The dual role of HSPs, depending on their intracellular or extracellular location, may be a paradox in cancer therapy. On the one hand, HSPs accumulated in cancer cells and are involved in their survival. Therefore, they are negative factors that must be neutralized to sensitize cancer cells to cell death. Many laboratories are working on the project of developing new inhibitors of HSPs to be tested in clinical trials. However, on the other hand, on the basis of the extracellular immunological role of HSPs, patients are being injected with HSP preparations in order to induce a specific antitumoral immune response able to reject the tumor. So, in that context, HSP has a positive role in cancer therapy. Which is the best anticancer therapy? The solution would be inhibiting intracellular HSPs and increasing extracellular or membrane-bound HSPs.
Both approaches that are now being tested in clinical trials independently should be tested in combination in the future. In the theoretically ideal approach, both therapies should be tried successively: first, the inhibitors of HSPs in association with a classical chemotherapy in order to increase the sensitivity of cancer cells to the cytotoxic drug and second, the HSP-based therapy to boost the immune system and thereby to avoid the apparition of metastasis. In conclusion, the dual function of HSPs, depending on their location, strongly increases the interest of these molecules in cancer therapy.
ACKNOWLEDGEMENTS
This work was supported by grants from the Ligue Nationale Contre le Cancer (and its committees in the Nièvre), l’Association pour la Recherche contre le Cancer, Conseil Regional de Bourgogne, Deutsche Forschungsgemeinschaft (MU 1238/7-2), European Union TRANSEUROPE (QOL-2001-3.1.3), Bundesministerium für Bildung und Forschung, and multimmune GmbH. We gratefully acknowledge the technical assistance of Lydia Rossbacher. M. B. is a recipient of a doctoral fellowship from the "Ministère de l’Education."
FOOTNOTES
1 These authors contributed equally to this work. 
Received March 5, 2006; revised May 24, 2006; accepted May 29, 2006.
REFERENCES
B-crystallin expression-mediated increase in glutathione is essential for the protective activity of these proteins against TNF-induced cell death EMBO J. 15,2695-2706[Medline]B activation by tumour necrosis factor requires the Akt serine-threonine kinase Nature 401,82-85[CrossRef][Medline]B proteasomal degradation Mol. Cell. Biol. 23,5790-5802B activation J. Biol. Chem. 275,10519-10526/aryl hydrocarbon receptor nuclear translocator DNA binding by the 90-kDa heat-shock protein inhibitor radicicol Mol. Pharmacol. 62,975-982This article has been cited by other articles:
 |
|
 |
|
  S. Reuter, P. Bangen, B. Edemir, U. Hillebrand, H. Pavenstadt, S. Heidenreich, and D. Lang The HSP72 stress response of monocytes from patients on haemodialysis is impaired Nephrol. Dial. Transplant., April 1, 2009; (2009) gfp142v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chu, M. Alexiadis, and P. J. Fuller Proteasome Inhibition by Bortezomib Decreases Proliferation and Increases Apoptosis in Ovarian Granulosa Cell Tumors Reproductive Sciences, April 1, 2009; 16(4): 397 - 407. [Abstract] [PDF]
|
|
|
|
 |
|
 D. E. Voth and R. A. Heinzen Sustained Activation of Akt and Erk1/2 Is Required for Coxiella burnetii Antiapoptotic Activity Infect. Immun., January 1, 2009; 77(1): 205 - 213. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. K. McCollum, K. B. Lukasiewicz, C. J. TenEyck, W. L. Lingle, D. O. Toft, and C. Erlichman Cisplatin abrogates the geldanamycin-induced heat shock response Mol. Cancer Ther., October 1, 2008; 7(10): 3256 - 3264. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. K. McCollum, C. J. TenEyck, B. Stensgard, B. W. Morlan, K. V. Ballman, R. B. Jenkins, D. O. Toft, and C. Erlichman P-Glycoprotein-Mediated Resistance to Hsp90-Directed Therapy Is Eclipsed by the Heat Shock Response Cancer Res., September 15, 2008; 68(18): 7419 - 7427. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-F. Cheng, J. Fan, M. Fedesco, S. Guan, Y. Li, B. Bandyopadhyay, A. M. Bright, D. Yerushalmi, M. Liang, M. Chen, et al. Transforming Growth Factor {alpha} (TGF{alpha})-Stimulated Secretion of HSP90{alpha}: Using the Receptor LRP-1/CD91 To Promote Human Skin Cell Migration against a TGF{beta}-Rich Environment during Wound Healing Mol. Cell. Biol., May 15, 2008; 28(10): 3344 - 3358. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Fandrey Erythropoietin Receptors on Tumor Cells: What Do They Mean? Oncologist, May 1, 2008; 13(suppl_3): 16 - 20. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Brierley-Hobson Binding of (-)-epigallocatechin-3-gallate to the Hsp70 ATPase domain may promote apoptosis in colorectal cancer Bioscience Horizons, March 1, 2008; 1(1): 9 - 18. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Faingold, J.-C. Marshall, E. Antecka, S. Di Cesare, A. N. Odashiro, S. Bakalian, B. F. Fernandes, and M. N. Burnier Jr. Immune Expression and Inhibition of Heat Shock Protein 90 in Uveal Melanoma Clin. Cancer Res., February 1, 2008; 14(3): 847 - 855. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. W. Graner, R. I. Cumming, and D. D. Bigner The Heat Shock Response and Chaperones/Heat Shock Proteins in Brain Tumors: Surface Expression, Release, and Possible Immune Consequences J. Neurosci., October 17, 2007; 27(42): 11214 - 11227. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Bianchi DAMPs, PAMPs and alarmins: all we need to know about danger J. Leukoc. Biol., January 1, 2007; 81(1): 1 - 5. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
