1000/1000
Hot
Most Recent
The Heat Shock Protein B8 (HSPB8) is a small chaperone involved in chaperone-assisted selective autophagy (CASA). CASA promotes the selective degradation of proteins to counteract cell stress such as tumor-induced stress. HSPB8 is also involved in (i) the cell division machinery regulating chromosome segregation and cell cycle arrest in the G0/G1 phase and (ii) inflammation regulating dendritic cell maturation and cytokine production. HSPB8 expression and role are tumor-specific, showing a dual and opposite role.
Human cells have different fine-tuned systems that act as defense mechanisms against a large number of different environmental stresses. Among these, one identified in the 1960s by Ferruccio Ritossa is the heat shock response (HSR), an extraordinary mechanism shared by procaryotes and eukaryotes which responds to and counteracts several potentially harmful cell stresses. In fact, the HSR is now identified more generally as a stress response, being triggered in different ways by a variety of stressful conditions affecting the cells. The stress response may arise outside (i.e., heat shock) or inside (i.e., protein misfolding) the cells, and it is based on the interplay between the nucleus and organelles such as mitochondria, the endoplasmic reticulum, autophagosomes, and lysosomes as well as membrane-less organelles such as P bodies and stress granules (SGs). The stress response is based on several cellular processes such as gene expression, protein synthesis, and protein and organelle degradation and it is involved in aging and in many human diseases such as neurodegeneration, inflammation, obesity, diabetes, autoimmune diseases, atherosclerosis, and cancer [1][2].
The stress response, including the HSR, is based on a rapid and transient gene expression mechanism that controls the expression of molecular chaperones: the heat shock proteins (HSPs). The six molecular chaperone families are ubiquitous and broadly conserved; even if not all of them are induced by heat shock, they are, indeed, differentially controlled by different stresses. Chaperones were originally grouped based on their apparent molecular mass, while now, they are classified mainly by their functions in the folding processes (HSP110s, HSP90s, HSP70s, HSP60s, HSP40s, and small heat shock proteins (sHSPs)), but there are similarities between the two nomenclatures; in fact, based on the HUGO Gene Nomenclature Committee, the new classification recently adopted is as follows: HSPB (sHSP), DNAJ (HSP40), HSPD (HSP60), HSPA (HSP70), HSPC (HSP90), and HSPH (HSP110) [3], and reflects both the functions and the sizes of the members of each subfamily.
The human HSPB8 is a member of the HSPBs which is a ubiquitous family of ATP-independent stress proteins, whose activity is mediated by other ATP-dependent chaperones (i.e., HSPAs), defining whether HSPBs clients can be refolded or degraded [4][5]. HSPBs are able to bind unfolded/misfolded substrates and subsequently new and larger oligomers are formed, preventing irreversible client aggregation and allowing ATP-dependent chaperones recognition to assist protein refolding [6]. HSPBs contribute to protein quality control (PQC) and work to prevent protein aggregation and to generate a pool of non-native proteins that can be rapidly folded [7]. The human genome encodes 10 HSPBs named HSPB1 through HSPB10, with an apparent molecular mass of 12–43 kDa [8]. Despite their name, the expression of HSPBs can be dependently or independently regulated by heat shock transcription factors (HSFs). In fact, it relies on the combinatory effects of many transcription factors. The basal expression or stress inducibility of HSPBs is therefore regulated by different cis-elements localized in the HSPB regulatory regions [9].
A common key feature of HSPBs is the alpha-crystallin domain (ACD), which refers to the best-known family member, the eye lens protein alpha-crystallin [10]. The ACD is flanked by a variable N-terminal domain (NTD) and a short C-terminal domain (CTD), which show high heterogeneity both in sequence and size among the HSPBs. All the three domains are involved in determining the quaternary structure adopted by the HSPBs. Indeed, while the ACD mediates the dimerization of HSPBs, the ACD flanking regions affect HSPBs oligomerization. Since the NTD and CTD highly differ among the HSPBs, the size and composition of HSPB oligomers can vary, ranging from mainly dimers to 600 kDa hetero-oligomers [11][12][13]. A dynamic association/dissociation has been suggested as a primary regulator of HSPB functions and it is often induced by protein phosphorylation [14].
HSPB8, also named small stress protein-like protein (sHSP22), protein kinase H11(H11), E2-induced gene 1 protein (E2IG1), or alpha-crystallin C chain (CRYAC), is different from other HSPBs, since it is preferentially found as monomers and homodimers, even if the protein can also interact with other HSPBs forming heterodimers [15][16][17]. In addition, like HSPB5 and HSPB6, HSPB8 represents an “atypical” member of the HSPB family because in mammalian cells, it can form stable complexes with the Bcl2-associated athanogene 3 (BAG3), which is thought to be the obligate partner of HSPB8 [18][19].
In basal conditions, the heart and the skeletal muscle are the two most noteworthy tissues in which HSPB8 is expressed at relatively high levels. However, these tissues also express many other HSPBs, such as HSPB1, HSPB2, HSPB3, HSPB5, HSPB6, and HSPB7, possibly because during life, muscle tissues are subjected to several stressful conditions (including prolonged mechanical stress, hypoxia, damaged protein exposure). HSPB8 is also highly expressed in the brain, where it can be found both in neurons and in glial cells. Neuronal HSPB8 expression appears to be particularly relevant as a neuroprotective mechanism; for example, under stressful conditions, such as proteotoxic stresses, motoneurons express a very high level of HSPB8 in an attempt to protect themselves from damage [20]. Lower HSPB8 expression has been observed in the prostate, placenta, lungs, kidneys, and skin, while the ovaries, testes, liver, pancreas, and spleen appear to be devoid of HSPB8 [21]. As with many other HSPs, HSPB8 expression can be triggered by a variety of stresses, including starvation, proteasome inhibition, misfolded proteins accumulation, nuclear factor-Kappa B (NF-κB) activation, autophagy activation, microtubule destabilization, and occasionally heat shock [20][22][23][24][25][26].
HSPB8 (Figure 1A) is composed of 196 amino acids with an apparent molecular mass of 21.6 kDa. Its ACD is located close to the CTD approximately between amino acids 86 and 176. The short CTD interacts with aberrantly exposed hydrophobic regions of the substrate, preventing its aggregation during chaperone activity. Different from other members of the HSPB family (HSPB1, HSPB2, HSPB4, and HSPB5), HSPB8 does not contain the conserved I/V-X-I/V motif responsible for oligomeric assembly [27][28]. Indeed, this motif interacts with the neighboring ACD in a pocket between strands β4 and β8, acting as a bridge between the dimers. Instead, a hydrophobic pocket mediates the HSPB8 binding to two I/V-X-I/V domains, namely, IPV domains, present in the BAG3 protein [29], thus explaining the preferential binding of HSPB8 to BAG3 rather than HSPB8 assembly into large oligomers. In this interaction, two molecules of HSPB8 complement one BAG3 molecule, forming a functional chaperone complex (see below).
Figure 1. Graphical representation of human small heat shock proteins (HSPBs) structure and features. (A) Schematic diagram of the domains within human HSPBs. NTD and CTD indicate the variable N- and C-terminal domains, respectively; ACD indicates the conserved alpha-crystallin domain. Red boxes and purple boxes indicate the I/V-X-I/V motifs and the RLFDQxFG conserved sequence, respectively. Amino acid length is indicated on the right. (B) HSPB8 schematic structure, post-translational modifications, and mutations. Post-translational modification residues are reported: phosphorylation sites are indicated in yellow, predicted myristoylation sites in orange, and a SUMOylation site in light blue. Red stars indicate HSPB8 mutations reported in the literature.
The NTD is an intrinsically disordered region enriched in Pro and Arg residues, which is highly susceptible to proteolysis [30][31]. The NTD contains the conserved RLFDQxFG motif that was shown to play an essential role in the interaction with other HSPs [32].
HSPBs can undergo post-translational modifications (Figure 1B). HSPB8 has several putative phosphorylation sites, but only Ser-14, Ser-24, Ser-27, Ser-57, Thr-63, Thr-87, and Tyr-118 phosphorylation has been described so far [33][34]. Ser-14 and Thr-63 phosphorylation was found to be mediated by protein kinase C (PKC), while p44 mitogen-activated protein kinase phosphorylates Ser-27 and Thr-87 [33][35]. Moreover, Ser-24, Ser-27, and Thr-87 are potential sites of ERK1 phosphorylation. Phosphorylation of Ser-24, Ser-27, and Thr-87 was associated with an increase in HSPB8 dimerization and a reduction in its degradation. Instead, phosphorylation of Ser-24 and Ser-27 decreases HSPB8 chaperone activity, while phosphorylation of Thr-87 increases HSPB8 chaperone activity [36]. A similar activity has been linked to Ser-57 phosphorylation mediated by cyclic AMP-dependent protein kinase (PKA) [37]. While HSPB8 phosphorylation has been evaluated through in vitro assays, its role in disease has not been elucidated.
On the other hand, HSPB8 SUMOylation at Lys-106 in MCF-7 cancer cells has been recently described. Noteworthy, HSPB8 SUMOylation is promoted by HSPB1-mediated recruitment of SUMO-2/3 enzymes. This leads to an increase in HSPB8 expression, which favors breast cancer progression (see below) [38].
Despite the fact that it is unknown whether HSPB8 is myristoylated, and that its functional significance is unknown, HSPB8 has two predicted myristoylation sites at residues Gly-62 and Gly-132 that may enhance its membrane-binding potential [39].
Few mutations have been described in the HSPB8 gene and are associated with disease (Figure 1B). The first identified mutation in the HSPB8 gene in cancer cells causes a W51C substitution in the protein sequence and was reported in one melanoma line with an important transforming potential. The mutant W51C HSPB8 acquires a predicted different secondary structure characterized by seven additional β-turns [40] and, in this way, is capable of altering HSPB8 protein–protein interaction, conferring a dominant anti-apoptotic activity to the mutated protein [40]; this greatly differs from the typical pro-apoptotic activity of wtHSPB8 in melanoma cells. Another HSPB8 anti-apoptotic mutation is the missense mutation P173H, found in human melanoma cell line MeWo. P173H HSPB8 loses its propensity to bind the transforming growth factor-beta-activated kinase 1 (TAK1), an interaction required to induce apoptosis [41], and it is thought to confer 5-Aza-deoxy-cytidine resistance to these melanoma cells.
The majority of HSPB8 mutations have been identified in patients with Charcot–Marie–Tooth (CMT) type 2L [42][43], myopathy, and distal hereditary motor neuropathy (dHMN) type IIa [44][45][46]. In particular, mutations of the conserved lysine at position 141 (K141N, K141E, K141T, and K141M), which are the most frequently described, affect HSPB8 dimerization and BAG3 interaction, causing HSPB8 toxic aggregation. Two other new HSPB8 mutations that have been identified in dHMN patients are P90L and N138T, but their role in disease needs further characterization [47]. Instead, the recently described frame shift mutation pP173Sfs*43, caused by the duplication c.515dupC in the HSPB8 gene, has been associated with autosomal dominant rimmed vacuolar myopathy. Moreover, the deletion c.508–509delCA (pQ170Gfs*45) and the duplication c.577–580dupGTCA (pT194Sfs*23) in the HSPB8 gene have been described as responsible for adult-onset axial and distal myopathy and proximal limb-girdle rimmed vacuolar myopathy, respectively [48][49]. These mutations are predicted to be translated in HSPB8 variants with an elongated C-terminal domain [46]. However, since no elongated HSPB8 has been detected in samples derived from patients, due to mRNA decay or protein instability, a haploinsufficiency mechanism has been hypothesized to explain the disease-related phenotype [46][48].
HSPB8 is differentially expressed and displays a dual and opposite role depending upon the type of tumor considered. In fact, in some tumors, HSPB8 promotes tumor growth (e.g., in BC [77,87], lung cancer [88], multiple myeloma [89], ovarian cancer [90], and gastric cancer [91,92]), while in other tumors (e.g., in melanoma [41,93,94], leukemia [39,40,92,95], glioblastoma [69,96], hepatocarcinoma [97,98,99], and prostate cancer (PC) [40,100,101,102]). Therefore, both HSPB8 inducers and repressors could be of interest against cancer, depending upon the specific tumor to be treated.
Estrogens and selective estrogen receptor modulators (SERMs) are the first molecules identified as HSPB8 inducers [50][51][52][53][54]. Some SERMs, such as tamoxifen and 3β-diol (which is an endogenous androgenic derivative characterized by a potent estrogenic activity mediated by ERbeta [55]) induce HSPB8 expression in MCF-7 cells [50], while other SERMs, such as raloxifen and genistein, are not effective [50]. Notably, tamoxifen induces apoptosis and autophagy in BC cells [56], and it is largely used in BC treatment [57]. Unfortunately, BC cells often become resistant to tamoxifen, giving rise to more aggressive tumors. As mentioned before, HSPB8 appears to be directly involved in the acquisition of tamoxifen resistance [58][54]. HSPB8-related effects seem to be mediated by the kinase LMTK3, an upstream regulator of HSPB8 expression, that was identified as a mediator of tamoxifen resistance in BT-474 BC cells [59]. Since these adverse HSPB8 effects in tamR cells prevent patients from benefiting from tamoxifen treatment, an alternative approach has been based on AZD8055, a mTOR kinase inhibitor, which downregulates HSPB8 expression in tamR BC cells. This treatment decreases HSPB8 levels and positively correlates with reduced cell proliferation [54], indicating that AZD8055 is promising for the therapy of tamR BC tumors. Further, the anti-inflammatory statin atorvastatin (ATV) was identified as a negative modulator of HSPB8, both in in vitro and in vivo atherosclerotic models [60][61], but so far has not been tested in BC.
Not only estrogens, but also the other female steroid hormones, progestins, have been identified as HSPB8 inducers. In a gene expression profiling analysis performed on T47D BC cells treated with the synthetic progestin R5020, HSPB8 was found to be upregulated, particularly during the G2/M phase [62]. Notably, progestin-mediated HSPB8 gene transcription is regulated by a complex containing SP1, cyclin D1, and phospho-Ser345 PR, which binds with REs and SP1 DNA-binding motifs located in the promoter region of the HSPB8 gene [62].
Other potent HSPB8 inducers are several proteasome inhibitors [63], some of which are currently used as therapeutic options against cancer, due to their ability to activate autophagy [64]. In fact, when the ubiquitin-proteasome pathway is inhibited, cells upregulate autophagic proteins, including those of the CASA complex. Specifically, the proteasome inhibitors MG132, Velcade, and lactacystin upregulate HSPB8 and its co-chaperone BAG3, both at transcriptional and protein levels, in different cell types [20][65][63][66]. As mentioned above, Velcade is used for the treatment of different types of tumors, but, unfortunately, cancer cells may develop drug resistance. Similar to the case of tamoxifen, in Velcade acquired resistance, HSPB8 plays a crucial role [67]. For example, in myeloma cells, HSPB8 is robustly overexpressed in Velcade-resistant cells, where it confers a selective resistance to the drug [67].
The fact that HSPB8 expression increases when cells preferentially activate autophagy correlates with data showing that autophagy inducers may upregulate HSPB8 expression. An example is the disaccharide trehalose that activates autophagy via the lysosomal-mediated TFEB pathway [68][69]. By activating autophagy, trehalose assists the removal of several toxic misfolded proteins [68][70][71][72], and, not surprisingly, trehalose has been proven as a therapeutic approach in in vitro and in vivo cancer models [73][74][75]. Of note, still related to PQC modulation, HSPB8 and BAG3 expression has been found to respond to some proteotoxic stress via the activation of the NF-κB transcription factor [22][76]. It is known that NF-κB regulates the expression of a hundred genes, involved in inflammation, immunity, proliferation, and cell death; thus, it is difficult to foresee how and if the modulation of HSPB8 expression via NF-κB may be therapeutically useful.
Other HSPB8 inducers are the compounds able to mount the stress response mediated by the activation of HSF1, the key master regulator of HSP family gene expression. Among these are the HSPs inducer geranylgeranylacetone (GGA) [77][78] and the potent HSPs inducer multitarget small molecule N-((5-(3-(1-benzylpiperidin-4-yl)propoxy)-1-methyl-1H-indol-2-yl)methyl)-N-methylprop-2-yn-1-amine (ASS234) [79]. Both molecules have been successfully tested on AD models [80][81][82][83][84][85], exerting a protective role mediated by HSPs in general, and especially by HSPB8. Interestingly, GGA has been proven to induce apoptosis in some tumors, including melanoma [86][87], and suggested as a potential therapeutic agent for these tumors. Recalling the HSPB8 pro-apoptotic activity in melanoma cells, it would be interesting to investigate if HSPB8 may be involved in the effects observed after GGA treatment.
Using a high-throughput screening approach to find small molecules for the treatment of misfolded protein diseases, we identified two other drugs capable of positively modulating HSPB8 gene expression: colchicine and doxorubicin [24]. The two molecules both increased HSPB8 expression in a neuroblastoma ALS cell model, reducing the accumulation of toxic misfolded proteins. Colchicine is a well-known microtubule destabilizing agent, used for the treatment of gout, Mediterranean fever, Bechet’s disease, and recurrent pericarditis, and is now under investigation in a phase II clinical trial for ALS [88]. Moreover, colchicine is widely used as a research tool for the study of microtubule dynamics. Microtubules are an important target for anticancer agents and drugs that disrupt their assembly/disassembly are widely used in chemotherapy. Unfortunately, colchicine is poorly used in clinics because of its toxicity and the development of multi-drug resistance, but several drugs targeting the colchicine-binding site have recently been developed and under trials [89]. Further, the anthracycline antibiotic doxorubicin is known for its ability to kill cancer cells and commonly used to treat many forms of cancer, including BC [90]. Furthermore, in the case of doxorubicin, cancer cells may develop drug resistance [91]. Considering the negative role exerted by HSPB8 in BC cells, we can postulate that HSPB8 may play a role in acquiring resistance, and therefore this aspect in doxorubicin treatment should be carefully considered.