Expansion of the HSP70 gene family in Tegillarca granosa and expression profiles in response to zinc toxicity

Abstract

Zinc (Zn) is an essential micronutrient in organisms and an abundant element in the Earth's crust. Trace amounts of Zn released from natural sources can enter aquatic ecosystems through weathering and erosion. Zn accumulates in organisms, and when its intracellular concentration exceeds a certain level, it can induce oxidative stress and trigger oxidative stress-mediated heat shock protein (HSP) modulation. HSP70 is the most evolutionarily conserved among the HSP families. Despite extensive research on HSP70 genes in bivalves, the HSP70 gene family of Tegillarca granosa is still poorly characterized. We identified 65 HSP70 genes belonging to 6 families in the T. granosa genome, with 50 HSPa12 and 11 HSPa B2 genes highly expanded. On chromosome 11, 39 HSP70 (60%) genes were identified, and the HSPa12A genes were highly duplicated. A total of 527 and 538 differentially expressed genes were identified in the gills and mantle based on Zn exposure, respectively. The Gene Ontology of cellular anatomical entities was significantly enriched with upregulated differentially expressed genes in the gills and mantle. Eight of the 11 HSPa B2 genes were upregulated in both tissues. Most of the genes identified in both tissues were involved in “protein homeostasis” and “inhibition of apoptosis,” which are associated with the HSP70 family's resistance to extrinsic and intrinsic stress. Hence, this study identified that the HSP70 gene family plays a vital role in the adaptation of aquatic organisms to heavy metal (e.g., Zn) stress in contaminated environments by compiling the different physiological responses to preserve homeostasis.

Introduction

Tegillarca granosa is an important aquaculture bivalve species native to the Indo-Pacific region.1, 2 It is mostly farmed in Suncheon Bay, Jeollanam-do, South Korea, a well-known brackish water habitat situated in the southern part of the Korean peninsula. However, due to pollution from several companies and wastewater in the vicinity of Suncheon Bay, the marine environment has lately been severely contaminated, such that inhabitant marine species are becoming increasingly vulnerable to heavy metal exposure.³

One of the environmental pressures that affect living organisms is heavy metal contamination. When the intracellular level of a metal ion exceeds a particular threshold, it affects the expression of a number of intracellular proteins as well as the important functions of the organism, leading to cell death.⁴ Heavy metals can be found in aquatic environments as dissolved or particulate phases. Recent research on water quality in Suncheon Bay revealed that the quantity of zinc (Zn), as well as other heavy metals, surpassed the threshold effect level at certain stations.⁵ The concentration of Zn, in particular, was identified to be higher than that of other heavy metals, and it was shown to be considerably higher in Suncheon, Jeonnam than in other locations.⁶ According to the analytical results, Zn was identified at an average of 56.7 mg/kg (6.70–466 mg/kg), which was similar to or greater than previous research (Zn: 3.34 to 217 mg/kg).⁶ Furthermore, it was relatively measured higher in oysters and other bivalves, and Zn may concentrate preferentially in the body when compared to other heavy metals.⁷

Zn is an abundant and important element in the Earth's crust, and trace quantities can be discharged into aquatic habitats through weathering and erosion.⁸ In addition to natural processes, large amounts of Zn are released into aquatic ecosystems through mining, smelting, and refining of Zn sulfide ores, flooding and leaching of mine tailings, corrosion of galvanized goods and Zn alloys, and road and agricultural run-off.⁹ Zn pollution in marine habitats from historical and present-day mining operations is a serious problem because these activities generate massive amounts of waste and have long-term and negative effects on formerly healthy aquatic ecosystems.¹⁰ Aquatic species absorb these metals directly from their surroundings through exposed tissues and feeding. These metals can build up in their body tissues, with lethal or sublethal consequences.¹¹ Furthermore, several heavy metal compounds are known to be stress agents, promoting various defensive measures, such as inducing heat shock proteins (HSPs), apoptosis, and autophagy in a dose-dependent and time-dependent way.¹²

HSPs, which were first identified in Drosophila melanogaster salivary gland cells,¹³ are present in the genomes of all living species, from bacteria to humans. Some HSP family members are classified as housekeeping proteins because they perform various functions in protein folding, assembly, secretion, intracellular localization, control, and degradation in unstressed cells.¹⁴ Heat or various environmental stressors trigger the production of HSPs, a vast family of highly conserved molecular chaperons.¹⁵ HSPs comprise around 5% of the total intracellular protein content in the absence of stress (constitutive forms), but this can quickly rise to 15% or more in response to stress.¹⁶ HSPs can be categorized into several families according to their molecular weight, including HSP40, 70-kDa heat shock protein (HSP70), HSP90, HSP110, and small HSPs.¹⁷ HSPs are found in the mitochondria, cytosol, endoplasmic reticulum, and nucleus.¹⁸ The HSP70 gene family is one of the most evolutionarily preserved HSP families in terms of function and structure.19, 17

The three main conserved domains found in HSP70 genes are a 40 kDa N-terminal ATPase domain (NBD), which controls the cooperation with the target protein, an 18 kDa substrate-binding domain (SBD), which identifies the hydrophobic regions in a target protein during the initial stages of target protein folding, and a 10 kDa variable C-terminal domain.19, 20 The NBD is conserved in all HSP70 gene family members, with the exception of a few HSPa12 genes that encode divergent NBDs with unknown nucleotide binding characteristics.²¹ HSP70 activity is dependent on their ATP-regulated capacity to bind with exposed hydrophobic protein surfaces. During the folding of nascent polypeptides, ATP hydrolysis and ADP/ATP exchange are critical processes for substrate binding and HSP70 release.²² The proteins HSPa12A and HSPa12B are distant relatives of the HSP70 family, owing to an unusual ATP-binding domain.²¹

The HSP70 gene family plays critical functions in facilitating protein folding and proteostasis,²³ the degradation of misfolded proteins,²⁴ protein membrane translocation,¹⁸ the avoidance of harmful protein aggregation, as well as other cell-protective biological processes, in addition to exhibiting essential functions to counteract various environmental, physiological, and immunological stresses.25, 26, 16 Due to their organelle-specific and tissue-specific expression, the members of the HSP70 family have a diverse set of biological roles, and occasionally, even overlapping roles.19, 27

Several HSP70 gene family members have been cloned from bivalve species and were shown to be upregulated in response to environmental, physiological, and immunological stresses.28, 1, 2, 16 Analysis has revealed massive expansion of the HSP70 gene family in the oysters Crassostrea gigas and Crassostrea hongkongensis,29, 30 the Sydney rock oyster,³¹ the pearl oyster,³² the scallop,³³ the Manila clam,³⁴ the deep-sea vent mussel, the shallow-water mussel,³⁵ and the blue mussel (Mytilus edulis),37, 36 suggesting that these genes are necessary for adaptation to various environmental stresses in diverse aquatic species residing in highly variable habitats.³⁸ Phylogenetic analyses of bivalves have revealed recent species-specific HSP70 expansions as well as older mixed-species clusters extending back to ancestral bivalve lineages, indicating that natural selection favored the expansion of this particular gene family.³⁹

Based on the high-quality chromosome-level genome of T. granosa,⁴⁰ we identified all members of the HSP70 gene family and also investigated gene expression after Zn exposure in two tissues, with a particular focus on HSP70 genes. The primary goal of this study was to elucidate the importance of the HSP70 gene family in alleviating the impacts of polluted environments, particularly Zn-induced stress, and in maintaining cellular homeostasis.

Results

Whole-transcriptome sequencing results

The gill and mantle in the control group produced 5.52 and 5.05 Gb of raw data, respectively. After exposure to Zn for 12 and 48 h, the gill generated 5.80 and 6.06 Gb of raw data, and the corresponding values for the mantle were 5.12 and 4.95 Gb, respectively (Supplementary Table 1).

Analysis of HSP70 genes

In the T. granosa genome, 65 HSP70 genes were identified. We used known sequences of all HSP70 genes from three different bivalve genomes and conducted a Blast local alignment search tool (BLAST) search to identify any missing or misannotated HSP70 genes. There were 11 HSPa B2 genes, 37 HSPa12A genes, 13 HSPa12B genes, and 1 HSPa13, HSPa14, HSPa4, and HSPa9 gene in T. granosa (Table 1). The HSP70 gene family mainly comprised the HSPa12 genes (50 genes (76.92% of the total number of HSP70 genes)) in T. granosa, which is similar to the HSP70 gene expansions identified in C. gigas³⁰ and Mercenaria mercenaria (Table 1; Hu et al.³⁸). In Scapharca broughtonii, a phylogenetically close species to T. granosa, a total of 92 HSP70 genes were identified, and 61 HSPa12A and 22 HSPa12B genes were identified. It is unknown why S. broughtonii has more HSP70 gene counts than T. granosa; however, 83 HSPa12 (90.21%) genes were substantially duplicated among HSP70 genes, like other bivalves (Table 1). The predicted proteins varied in length from 173 to 944 aa in T. granosa (Supplementary Table 2).

Table 1.

HSPa gene counts in Tegillarca granosa, Scapharca broughtonii, Crassostrea gigas, and Mercenaria mercenaria.

HSP70 genes	Gene count
	T. granosa	S. broughtonii	C. gigas	M. mercenaria
HSPa B2	11	7	5	4
HSPa12A	37	61	56 17	48
HSPa12B	13	22		74
HSPa13	1	1	1	-
HSPa14	1	1	1	1
HSPa17	-	-	-	1
HSPa4	1	-	-	1
HSPa5	-	-	7	1
HSPa8	-	-	-	1
HSPa9	1	-	1	2
Total	65	92	88	133

The HSPa genes were identified in the genomes of four bivalves, and mis-annotated genes were identified using BLAST searches.

From the phylogenetic tree in Figure 1, the T. granosa HSP70 gene family had seven clusters. In particular, the HSPa12 families were highly duplicated, as was the HSPa B2 family. The HSPa9 and HSPa4 genes were clustered separately in T. granosa only, not in S. broughtonii. Conversely, the HSPa13 genes were clustered with the HSPa14 genes, and the HSPa12 subfamily was extensively duplicated across the two species (Figure 1).

Fig. 1.

Phylogenetic tree analysis of HSPa genes in Tegillarca granosa with Scapharca broughtonii. The phylogenetic tree was constructed using the JTT method with 1,000 bootstrap replications, and red and yellow circles indicate the bootstrap of the phylogenetic tree. The HSPa genes in T. granosa were compared with those in S. broughtonii and different clusters in the phylogenetic tree are represented by specific colors: red for HSPa4, yellow for HSPa B2, blue for HSPa14, brown for HSPa13, orange for HSPa9, purple for HSPa12A, and green for HSPa12B. The HSPa genes of T. granosa are colored in red. The Homo sapiens TLR8 (AAF78036.1) gene was used as an outgroup, and it is highlighted in blue. The species were abbreviated as Tg (T. granosa) and Sb (S. broughtonii). Abbreviation used: JTT, Jones–Taylor–Thornton; TLR8, Toll-likereceptor 8.

HSP70 gene duplication and chromosomal distribution

The 65 identified HSP70 genes were irregularly represented across 10 (out of 19) different chromosomes (except chr 1, 2, 3, 5, 7, 9, 12, 18, and 19). Thirty-nine HSP70 (60% of the total number of HSP70 genes) genes were localized on chromosome 11, and the HSPa12A genes were substantially duplicated (Figure 2, Supplementary Table 2). Eleven HSP70 (16.92%) genes were identified as being present on chromosome 16. Among these, eight HSPa B2 genes were predominantly located on chromosome 16. Only two HSP70 genes were identified on chromosome 8, 14, and 15. Only one HSP70 gene was mapped to chromosomes 4, 6, 13, and 17 each (Figure 2, Supplementary Table 2).

Fig. 2.

Chromosomal distribution of Tegillarca granosa HSPa genes. The x-axis represents the chromosome numbers, and the y-axis indicates the length of the chromosome. HSPa genes are widely distributed on 10 different chromosomes, and the sequence names are indicated in various colors according to the type of HSPa genes corresponding to each chromosome.

A synteny analysis of HSP70 genes was performed to determine on which chromosomes the HSP70 genes in T. granosa were duplicated, for example, 39 HSP70 genes were most dispersed on chromosome 11, followed by 11 HSP70 genes on chromosome 16. There was a large physical distance between the HSP70 genes on most chromosomes, but there were exceptions. For example, on chromosome 11, the HSPa12A and HSPa12B genes were separated by a short physical distance (Figure 3, Supplementary Table 3).

Fig. 3.

Gene synteny of HSPa genes in Tegillarca granosa chromosomes. The chromosome numbers are shown on the left side, and gene orientations are represented with right or left arrows. The distances between genes are represented by a double slash, and the length is indicated on the top. Supplementary Table 3 shows the gene abbreviations, which are shown on top of each arrow. The legend indicates which colors represent HSPa genes.

Differentially expressed genes analysis of Zn exposure in gill and mantle tissue

A total of 527 and 538 upregulated and/or downregulated genes were identified in Zn-exposed gill and mantle tissues, respectively. In gills exposed for 12 h, 242 genes were upregulated, and 285 genes were downregulated, respectively. Two hundred thirty-eight and 289 genes were upregulated and downregulated in gills exposed for 48 h. In the mantle exposed for 12 h, 326 genes were identified as upregulated and 212 genes were downregulated, respectively. Three hundred twenty-nine and 209 genes were upregulated and downregulated in the mantle exposed for 48 h (P-value cutoff of ≤0.05).

The upregulated genes in the gills and mantle were used for Gene Ontology (GO) enrichment analysis. Multiple experiment viewer (MeV) analysis was performed to build heat maps of the upregulated genes from the 20 GOs in the Zn-exposed gill and mantle.

Gene expression of HSP70 genes

Among the 121 upregulated genes annotated in the top 20 GOs identified in gill tissue after Zn exposure, 7 HSP70 genes were upregulated (Figure 4(a) and Supplementary Figure 1(a)). In the case of the mantle tissue, 191 genes were upregulated, with 8 HSP70 genes being increased (Figure 4(b) and Supplementary Figure 1(b)).

Fig. 4.

The heatmaps of HSPa genes identified in two tissues. (a, b) The log₂FC scale was provided at the top of the heatmap. The gene expression differences in the gills and mantle after 12 and 48 h of Zn treatment are shown, and the names of the sequences are displayed on the right side of the gene descriptions. Abbreviation used: Zn, Zinc.

According to the results of the differentially expressed genes (DEG) analysis, seven HSPa B2 genes were identified in the gills after 12 and 48 h of Zn exposure. Between 12 and 48 h of exposure, the expression of the upregulated HSPa B2 genes increased by a factor of 4.17 (log₂FC) on average, with 5.67 (log₂FC) being the greatest increase. At 48 h, the highest log₂FC value was 10.94 (Supplementary Table 4). Seven HSPa B2 genes and one HSPa12A gene were upregulated in the mantle. When comparing their expression levels at 24 and 48 h of exposure, the log₂FC value was 3.73 on average, with the largest being 6.07. At 48 h, the highest log₂FC value was 8.85 (Supplementary Table 4). The upregulation of 8 out of a total of 11 HSPa B2 genes was identified in the 2 Zn-exposed tissues.

A domain analysis was conducted on the HSPa B2 genes because of their frequent identification in Zn-exposed gills and mantle tissue. The NBD and SBD of all 11 HSPa B2 genes were identified; however, only 5 genes have the EEVD motif, which relates to having a regulatory role (Figure 5). In chaperones, the EEVD motif refers to a specific protein sequence located at the C-terminus of various HSPs. The EEVD motif in chaperones represents Glu-Glu-Val-Asp, which is recognized by other cellular proteins such as tetratricopeptide repeat domain-containing proteins and helps particular chaperones connect and operate in cellular processes.⁴¹

Fig. 5.

The organization of the conserved domains of HSPa B2 in Tegillarca granosa. HSPa B2 has unique functional domains, including a substrate-binding domain (SBD) and a nucleotide-binding domain (NBD), which includes the ATP-binding cleft domain, as well as two additional domains shared by most family members: the carboxyl-terminal EEVD motif, or chaperone motif. The HSPa B2 genes identified by Zn exposure are indicated with a red color. Abbreviation used: Zn, Zinc.

As well as evaluating HSP70 genes, we also went on to identify other genes associated with HSPs involved in the stress response (Figure 6). The HSP70 gene and transcription factors known as heat shock factors (HSFs), notably HSF1, stay monomeric under unstressed conditions. Various stressors, such as Zn, cause HSF1 trimers to develop, which then bind to heat shock elements in the promoter region of HSP70 family member genes and induce transcription by producing HSP70 genes.⁴²

Fig. 6.

HSP70 has a variety of roles to counteract heavy metal toxicity and other environment stresses. The abbreviations are shown in the figure. HSF1 becomes activated in response to environmental stress, which triggers HSP70 overexpression, causing various mechanisms. (a) Protein homeostasis: HSP70 aids in the unfolding or refolding of misfolded proteins or denatured peptides to create native proteins. (b) Autophagy promotion: HSP70 localization to the lysosomes stabilizes the lysosomal membrane, promoting autophagy. (c) HSP70 cochaperonage: HSP70 functions as a cochaperone for HSP90 and is necessary for the appropriate maturation of HSP90 target proteins. (d) Apoptosis inhibition: HSP70 suppresses both intrinsic and extrinsic apoptotic pathways by blocking APAF1 recruitment to the apoptosome, suppressing AIF expression, and inhibiting the expression of other stress-induced kinases.

Firstly, when Zn stress was exposed, genes related to “protein homeostasis” were identified among various HSP70 roles. Various E3 ubiquitin-protein ligase genes were identified based on the DEGs in two tissues. In the gill and mantle, a total of 13 and 20 genes were identified, respectively (Table 2). The highest upregulated genes in the gill and mantle were E3 ubiquitin-protein ligase TRIM71-like (TRIM71) and E3 ubiquitin-protein ligase UBR4 (UBR4), with a log₂FC value of 7.35 at 12 h and increased to 8.49 at 48 h in the gill and a log₂FC value of 11.12 at 12 h and increased to 11.51 at 48 h in the mantle (Supplementary Tables 5 and 6). In this study, many E3 ubiquitin ligase genes have been identified in both tissues. The genes participate in protein degradation in response to Zn stress, and E3 ubiquitin ligases play a role in regulating numerous cellular functions and maintaining cellular homeostasis.⁴³

Table 2.

The gene count of four different roles in HSP70 genes when exposed to Zn in the gill and mantle.

RNA-sequencing	Gill				Mantle
	12 h		48 h		12 h		48 h
	Up	Down	Up	Down	Up	Down	Up	Down
Protein homeostasis	6	7	8	5	12	8	14	6
Autophagy	2	0	2	0	2	0	2	0
Co-Chaperonage to HSP90	5	2	6	1	5	2	6	1
Inhibition of apoptosis	11	7	11	7	11	9	11	9

The 12 and 48 h treatments are represented by the gene count of upregulated and downregulated genes for each role.

The second role for HSP’s stress response is “lysosomal membrane stabilization” associated with HSP70 genes, and two genes were identified in both tissues (Table 2). In the gill and mantle, lysosome-associated membrane glycoprotein 1 and lysosome membrane protein 2-like genes were identified, and both genes were highly upregulated in 48 h (Supplementary Tables 5 and 6). Because the identical gene was found in both tissues and was highly upregulated after 48 h, it was determined that these genes play an important role in lysosome stability during the stress response.

The third role for HSPs stress response is the “co-chaperonage of HSP90” genes, and a total of seven genes were identified in each tissue (Table 2). HSP70/HSP90 organizing protein 1 (HOP), a co-chaperone of the major chaperones HSP70 and HSP90,⁴⁴ and a dnaJ homolog subfamily B member 4 (DnaJB4), a molecular chaperone,⁴⁵ were identified to have higher expression levels at 48 h than at 12 h in both tissues (Supplementary Tables 5 and 6). As a result of the persistent damage caused by Zn stress, several distinct genes are expected to be involved in protein folding and homeostasis.

The fourth role for HSPs stress response is “inhibition of apoptosis,” and a total of 18 and 20 genes were identified in the gill and mantle, respectively (Table 2). When an apoptosome formation occurs, cytochrome c (cyt-c) is released from the mitochondria and binds to the cytosolic protein apoptosis protease activating factor-1 (apaf-1) before recruiting and activating the dormant pro-caspase-9 and causing caspase activation in the intrinsic pathway of apoptosis.⁴⁶ The apaf-1 gene involved in apoptosome formation has not been identified in T. granosa nor in other bivalves⁴⁷; however, cyt-c and the caspase 3 (CASP3) genes had low log₂FC values in both tissues (Supplementary Tables 5 and 6). Furthermore, HSP70 regulates caspase-dependent programmed cell death by reacting to apoptosis-inducing factor (AIF) and directly preventing AIF-induced chromatin condensation.⁴⁸ In both tissues, the AIF genes showed slight gene expression changes in the log₂FC values at 12 and 48 h (Supplementary Tables 5 and 6). Finally, HSP70 suppresses stress-induced signaling via c-Jun N-terminal kinases (JNKs), p38 mitogen-activated protein kinases (p38 MAPKs), and apoptosis signal-regulating kinase 1 (ASK1), also known as mitogen-activated protein kinase 5 (MAP3K5) to regulate apoptosis in the pre-mitochondrial stage.⁴⁹ In the gill, the ASK1 gene had a slight change in both 12 and 48 h gene expression; instead, JNK-related genes were upregulated in 48 h (Supplementary Table 5). In the mantle, one ASK1 gene was upregulated, and JNK-related genes had a slight gene expression change in both 12 and 48 h (Supplementary Table 6). In Zn-stress, most kinase-signaling inhibitory-associated genes did not demonstrate a significant variation in expression level.

Discussion

Although the role of the HSP70 family members has been studied in a variety of bivalves,33, 50, 30 T. granosa has yet to be the focus of a comprehensive genome-wide study of the HSP70 family. This study here, comprehensively describes the HSP70 gene family in T. granosa and the environmental stress response to acute Zn exposure, thus establishing a solid platform for future research on the function of HSP70 during heavy metal-induced stress.

HSP70 gene family members have been identified in all domains of life, and the number of copies varies between species. HSP70 genes have undergone dramatic expansions in bivalves, in particular the HSPa12 genes which are atypical HSP70 family members.39, 50 In this research, 65 HSP70 genes were identified in T. granosa (Table 1). Bivalves have significantly varied HSP70 gene copy numbers, which is thought to be due to the highly variable environments some of these species live in and that they act as stress regulators with their regulatory variation influencing physiological variation in response to environmental changes.37, 30 Interestingly, 60% of the HSPa12 gene family is situated on chromosome 11, and it has been duplicated (Fig. 2, Fig. 3). Gene duplication has the potential to generate resources for evolutionary innovation,51, 52 with the expansion of the HSP70 gene families in oysters and scallops caused by duplication.33, 50, 30

The HSPa12 gene family belongs to the HSP70 family but is substantially different from the other traditional HSP70 members, with little splice site conservation and few homologous regions. In general, the HSPa12A and 12B paralogs share only 61–65% amino acid similarity within the same species, whereas each paralog is substantially conserved between species (82–83% amino acid similarity; Brocchieri et al.⁵³). They have an unusual ATPase domain, no substrate binding, and tetratrico peptide repeat 1(TPR1) domains (which interact with HSP90) or a ubiquitin-binding peptide.⁵³ Thus, the HSPa12A and 12B genes differ significantly from the classical members, which are tightly involved with the cellular stress response, and even in humans, little is known about the function of these proteins.⁵⁴ However, the genome sequencing of mollusks has revealed a massive expansion of both HSPa12A and HSPa12B in bivalves (Cheng et al.³³; Table 1).

The high number of duplicates is the consequence of a bivalve-specific expansion followed by species-specific tandem duplications.³³ According to Cheng et al.,³³ the bivalve expansion of this divergent HSP70 gene family was significantly selected for and sustained by adaptation to the sessile lifestyle in the constantly changing marine environment with its diverse biotic and abiotic stressors. Indeed, analysis of HSPa12 gene family expression levels in various bivalves has revealed that specific subsets of the gene family are associated with responses to specific stresses, such as temperature and heavy metals in C. gigas and toxic dinoflagellates in Patinopectenyessoensis.33, 30 So far, only one HSPa12A gene has been cloned in Mytilus, with upregulation related to the cadmium response in Mytilusgalloprovincialis,⁵⁵ and in our study, one HSPa12A gene was upregulated in response to Zn stress (Figure 4). According to this research, several HSPa12 gene families show greater changes in expression in response to other stressors, such as heat stress³⁶ than in response to heavy metal stress. The findings of the study show a striking tendency within the HSPa12 gene families, with more variability in expression levels when subjected to heat stress as opposed to heavy metal stress. Further studies on the relationship between various stresses and the HSPa12 gene family will be required in the future to support this data.

T. granosa HSP70 genes were classified into six subfamilies. The majority of HSP70 members were accounted for by 50 HSPa12 family genes, which was similar to previous studies.33, 37, 50, 38, 30 Interestingly, in this study, we identified 11 and 7 HSPa B2 genes in the genomes of T. granosa and S. broughtonii, respectively, which is a larger number than the HSPa B2 genes identified in the genomes of C. gigas and M. mercenaria (Table 1; Zhang et al.³⁰ and Hu et al.³⁸). The precise reasons for the gene expansion of HSPa B2 genes in bivalves are unknown; however, further genomics research is being conducted to provide support. The gills in bivalves are thought to be sensitive to environmental changes, and high expression of the HSPa gene family promotes environmental change response regulation.⁵⁶ HSPa genes, particularly HSPa B2, were shown to be significantly expressed in the gills of Ostrea denselamellosa.⁵⁷ Furthermore, a study found that the HSPa B2 gene was represented as a stress response gene in Physella acuta when exposed to pesticides.⁵⁸ Still little is known about the function of these particular HSP70 family members, however, another research identified that HSPa B2 gene pairs were highly expressed in M. mercenaria in response to heat and severe hypoxia stress,³⁸ suggesting that these genes are affected by environmental stress. Further research on the HSPa B2 genes in invertebrates is needed to better understand their roles and regulatory processes. Specifically, there is a need for additional investigations into the expansion of HSPa B2 genes in bivalves. Targeted research in bivalve species will provide important insights into the evolutionary importance and adaptive responses associated with HSPa B2 gene amplification in this group of invertebrates. This specialized study is critical for explaining the molecular processes and ecological consequences of HSPa B2 genes in bivalves, allowing for a more comprehensive understanding of their function in stress response and adaptation in these species.

In T. granosa GO analysis identified 7 HSPa B2 genes among the 121 and 191 upregulated genes in the gill and mantle, respectively (Supplementary Figure 1). Most HSPa B2 genes showed higher expression at 48 h than at 12 h of Zn exposure, and seven genes were identified to be shared by both tissues (Figure 4). An HSPa B2 domain analysis was conducted, and five genes included the EEVD motif, whereas the NBD and SBD domains were identified in all genes. Of the eight HSPa B2 genes identified by RNA-sequencing (RNA-seq) analysis, only three had the EEVD motif (Figure 5). The KUTeg00000969-RA sequence showed the highest expression change in the gill and mantle, with log₂FC values of 5.67 and 5.64 when comparing 12 and 48 h of Zn exposure (Supplementary Table 4). Chaperones of the HSP70 family interact with various cochaperones through protein–protein interactions, and one essential motif for cochaperone binding is the amino acid sequences (EEVD) motif, which is generally located at the extreme C-terminus of cytoplasmic HSP70s. This motif is noted to connect to cochaperones with tetratricopeptide repeat domains, such as the E3 ubiquitin ligase.⁴⁵ A structurally distinct domain in class B J-domain proteins, like DnaJB4, is also known to interact with the EEVD motif.⁴⁵

HSP70 family members are upregulated in response to numerous stimuli that alter protein folding, such as heat treatment, hazardous substance exposure, ultraviolet (UV) irradiation, and pathogen attack.59, 60 The transcription factor HSF1 is principally responsible for the upregulation, and species-specific variations in HSF1 function have been studied in several taxa, including the fruit fly D. melanogaster⁶¹ and the yeast Saccharomyces cerevisiae,⁶² and the HSF1-dependent transactivation pathway is mainly conserved among eukaryotes. Heavy metals, which are known stress agents, alter the expression of HSP70 family members, promoting diverse defensive measures, such as HSP production, apoptosis, and autophagy, in a dose-dependent and time-dependent way.¹² As a result, HSPs aid in stress recovery by either refolding or degrading proteins, restoring protein homeostasis, and increasing cell survival.

When T. granosa was exposed to Zn, genes associated with protein homeostasis and cochaperones were identified, as well as several genes connected to E3 ubiquitin-protein ligases in the gill and mantle, respectively. Five E3 ubiquitin-protein ligase tripartite motif (TRIM) genes were identified in each tissue (Supplementary Tables 5 and 6). The E3 ubiquitin ligases control the final phase of the enzymatic cascade, which also includes ubiquitin-activating and conjugating enzymes (E1s and E2s). E3 ligases can selectively attach ubiquitin to lysine, serine, threonine, or cysteine residues in specific substrates, and the process of attaching ubiquitin and ubiquitin-like proteins to cellular proteins is known as ubiquitylation, which is important during posttranslational modification.63, 43 As previously stated, the ubiquitin-proteasome degradation pathway is one of the most essential mechanisms for modulating protein expression levels. The TRIM protein family is a wide group of RING-type E3 ligase subfamilies that regulate a variety of cellular processes, including innate immunological responses.⁶⁴ Furthermore, the TRIM family members may not only be potential viral restriction factors, but they may also have anti-viral properties, implying roles in the immune response. As a result, the E3 ubiquitin-protein ligase TRIM genes identified in this study may provide a clue that they will play an important part in a variety of immunological responses.

The genes identified that were related to the cochaperones, including the DnaJ homolog subfamily B member 4 gene were shown to have a higher expression level at 48 h than at 12 h in both tissues (Supplementary Tables 5 and 6). The DnaJ homolog subfamily B member 4gene is inducible by proteotoxic stress, which largely increases its expression.⁶⁵ These studies imply that the EEVD motif has emerged to enable a wide range of protein–protein interactions, and cochaperones may aid in determining whether HSP70-bound proteins are folded or degraded when exposed to Zn.

When chaperones are unable to fix the misfolded protein, they target it for destruction through the ubiquitin-proteasome pathway or lysosome-mediated autophagy. These proteostasis quality control systems limit the buildup of improperly folded proteins, which would otherwise cause them to aggregate and accumulate as inclusions.⁶⁶ The two primary mechanisms for the degradation of intracellular misfolded proteins are the autophagosomal-lysosomal pathway and the ubiquitin-proteasome system.⁶⁷ HSP70 is commonly known as a cytosolic molecular chaperone; however, in pathophysiological states, it also localizes to the luminal side of the endosomal-lysosomal system,⁶⁸ the plasma membrane,69, 70 and the extracellular system.⁷¹ Furthermore, as identified in this study, HSP70 stabilizes lysosomal membranes and protects cells from a variety of stressors.⁷² Lysosomes are widely present in digestive cells and hemocytes in bivalves and have been identified as a specific target for the toxic effects of many pollutants.⁷³ More research in other tissues will be required to confirm the correlation between HSP70 and lysosomal membranes.

HSP70 suppresses apoptosis by inhibiting both intrinsic and extrinsic apoptotic pathways. The intrinsic mitochondrial route is a non-receptor-mediated pathway that receives inputs from a variety of sources, including UV radiation, reactive oxygen species, mitochondrial DNA damage, viral infection, and environmental contaminants.⁷⁴ Not only is the intrinsic pathway the site where antiapoptotic and proapoptotic proteins interact and decide cell fate, but it is also the source of signals that activate caspase activation via multiple pathways. Cyt-c is required for the activation of the initiator caspase-9 (CASP9) in the apoptosome complex. Smac (second mitochondrial-derived activator of cascade) and Omi can both attach to inhibitors of apoptosis (IAPs) and alleviate their inhibitory effects on caspase activation after being released from mitochondria.⁷⁵ It was confirmed in this study that the inhibitory effect on caspase activity was identified through downregulated genes since many IAP genes were upregulated in both gills and mantle tissues (Supplementary Tables 5 and 6). Most genes involved in the intrinsic pathways, including the AIF gene, have been identified; however, the apaf-1 gene, which is critical in apoptosome formation,⁴⁷ has not been identified in the T. granosa genome.

According to a recent study, no direct apaf-1 homologs were identified in C. gigas, nor has an apaf-1 homolog been characterized in other bivalves.⁴⁷ Furthermore, the apaf-1 protein in oysters, which has no sequence ID, lacks the caspase-recruitment domain (CARD) and only has WD40 repeat (WD)-domains for binding cyt-c.⁷⁶ During apoptosome formation in vertebrates, the apaf-1 CARD domain binds to the CASP9 CARD domain. It appears surprising, given that CASP9 plays a crucial role in the intrinsic apoptotic pathway, forming apoptosomes with cyt-c and apaf-1 proteins for further downstream activation of executioner caspases, that the apaf-1 and CASP9 genes have not been formally described in bivalves. Instead, only automatically annotated sequences for CASP9 have been identified in oysters and Manila clams.77, 78 To provide insight into this missing caspase type, more bivalve genomes must be explored for putative CASP9 homologs.⁴⁷ A comprehensive in silico examination of apoptosome-forming proteins across metazoan species revealed that, whereas apoptosomes occur early in metazoan evolution, species-specific or taxon-specific deletion and duplication events may result in distinct forms of apoptosome formation.⁷⁹ Hence, given the lack of apaf-1 in bivalves, it is probable that the bivalve mitochondrial pathways vary from their vertebrate counterparts, which is contrary to what has previously been described.⁸⁰

The extrinsic apoptotic pathway entails the binding of death ligands to cell surface receptors, like the tumor necrosis factor (TNF) receptor. This causes the recruitment of the Fas-associated death domain or TNF receptor-associated death domain to the ASK1 cytosolic end of the receptor, the induction of the death-inducing signaling complex at the plasma membrane, and the activation of the initiator caspases-8. The executioner caspase-3, caspase-7, or caspase-6, which control the final stages of apoptosis, including DNA fragmentation, plasma blebbing, and the proteolysis of important structural and cell cycle proteins, are activated by caspases-8.81, 82 Furthermore, HSP70 genes are unable to influence the extrinsic apoptotic pathway and regulates apoptosis in the pre-mitochondrial stage by inhibiting stress-induced signaling, such as JNKs, p38 MAPKs, and ASK1, or by stabilizing lysosomal membranes.⁴⁹ Most of the genes implicated in this were identified in two tissues, however, the kinase signaling inhibitory related genes did not dramatically show gene expression variation when it had been exposed to Zn.

The massive expansion of the HSP70 gene family, particularly the HSPa 12 and HSPa B2 gene families, was identified in the T. granosa genome. Furthermore, the expression of HSP70 genes in the gill and mantle, tissues highly susceptible to environmental stressors in bivalves, with a focus on heavy metal exposure, reveals intriguing correlations into the HSP70 gene family's multifaceted roles in responding to environmental stresses. The upregulation of HSP70 gene expression in response to Zn stress reveals a critical role in the cellular response to heavy metal exposure, emphasizing the potential utility of HSP70 as a molecular indicator of Zn stress in bivalves. To enhance the depth of our understanding, future research should go beyond the limitations of existing exposure conditions, encompassing a broader range of Zn concentrations and exposure periods. This approach will allow for a more comprehensive understanding of the complex relationship between Zn stress and the HSP70 gene family, which will aid in the refinement of biomonitoring strategies and the development of targeted interventions to mitigate the impact of heavy metal exposure in aquatic ecosystems.

Materials and methods

Sample collection and transcriptome sequencing

A total of 110 adult blood clams (T. granosa) with a shell length of 35.01 ± 0.4 mm and a width of 1.8 ± 0.6 mm (mean ± standard deviation (SD)) were collected in January 2021 from Suncheon Bay, South Korea. The blood clams were acclimatized through a process in an environment similar to Suncheon Bay⁸³ for 1 week in a 5 L tank filled with filtered salt water at a salinity of 30‰ ± 1‰ and a temperature of 25 ± 1 ℃. During the acclimatization stage, the blood clams were maintained under constant light throughout the day and were not fed. Half of the saltwater was withdrawn every three days and replenished with new filtered salt water, and no mortality was recorded throughout the acclimatization period.

Then, 30 blood clams each were randomly moved to one tank with 5 L of filtered salt water (control group) and another tank with 102.6 mg/L of 1,000 ppm (100 mL) purchased Zn standard solution (Daejung Co., South Korea), in which 5 L of salt water was spiked with 1.5 times the standard threshold effect level concentrations identified in Suncheon Bay (experimental group). All blood clams were under the same conditions, apart from the difference in Zn concentration. To minimize the difference between the evaporation of filtered salt water and the Zn concentration, two tanks were covered with a lid, and the experiment was conducted. Sampling time points were set at 0 h (control), 12 h, and 48 h (experimental group), respectively. Three individual blood clams were randomly selected at each sampling point. The gill and mantle tissues were collected, flash-frozen in liquid nitrogen, and kept at −80 °C for total RNA extraction and whole-transcriptome sequencing.

The tissues were ground using liquid nitrogen, and the total RNA was extracted using the TRIzol reagent (Invitrogen, Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s instructions. The total RNA quality and quantity were checked using a Qubit fluorometer (Invitrogen) and a fragment analyzer (Agilent Technologies, Santa Clara, CA, USA), respectively. In consideration of the biological replicate, the total RNA from three individual clams was pooled based on the experimental conditions and using an Illumina TruSeq Standard mRNA Prep Kit (Illumina, San Diego, CA, USA), one pair of paired-end libraries pooled from three individual clams was produced and sequenced on the Illumina NovaSeq 6000 platform (Illumina).

Identification of HSP70 genes

HSP70 genes in the genome of T. granosa were identified using a previously published reference genome (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA670820, accession number: JARBDR000000000; Kim et al.⁴⁰). The methods used to identify the HSP70 gene family include sequence homology and conserved domain analysis. The National Center for Biotechnology Information(NCBI) GenBank and GigaScience databases were used to collect HSP70 gene sequences from three marine bivalves (S. broughtonii (GigaScience: PRJNA521075; Bai et al.⁸⁴), Pecten maximus (NCBI accession: GCF_902652985.1; Kenny et al.⁸⁵), and C. gigas (GCF_902806645.1; Penaloza et al.⁸⁶)). To detect missing and misannotated HSP70 genes, translated basic local alignment search tool for nucleotide (TBLASTN) searches (E-value cutoff of 1E-5) were performed against the T. granosa genome. The HSP70 amino acid sequences from three bivalves served as a query database for a BLASTP search (E-value cutoff of 1E-5) against the T. granosa genome. T. granosa’s putative HSP70 genes were aligned to the NCBI non-redundant database (http://blast.ncbi.nlm.nih.gov) for confirmation and analysis.

Sequence analysis of HSP70 genes

T. granosa and S. broughtonii full-length protein sequences of HSP70 genes⁸⁴ were aligned using the ClustalW algorithm.⁸⁰ A maximum likelihood (mL) phylogenetic tree was constructed using the Jones–Taylor–Thornton model and a bootstrap of 1,000 repetitions in MEGA-X v11.0.⁸⁷ The phylogenetic analysis of the S. broughtonii protein sequences was undertaken to examine the evolutionary relationship of HSP70.

The chromosomal position of the HSP70 genes was identified using the recently published T. granosa reference genome.⁴⁰ BLASTP (E-value cutoff of 1E-10) was used to identify the chromosomal locations, and the results were visualized using TBtools.⁸⁸ The gene synteny of each HSP70 gene was manually classified in the T. granosa genome to identify the tandemly duplicated gene orientations.

DEGs analysis after Zn exposure

The quality of pooled RNA-seq data was assessed using FastQC v0.11.9.⁸⁹ Then, the adapter sequence was removed, and the default parameter of TopHat v2.1.1⁹⁰ was used to align the RNA-seq reads to the T. granosa genome. Cufflinks v2.1.1 with the default parameters was used to assemble the mapped reads into potential transcripts and conduct the final transcriptome assembly. Finally, the default parameters of Cuffdiff of the Cufflinks algorithm were used to compare the gene expression changes between the untreated control group and the Zn-treated group based on the RNA-seq data.⁹⁰ The DEGs were identified using statistically significant criteria of |log₂FC = ≥2 and a P-value ≤0.05. VolcaNoseR was employed to visualize the final DEGs.⁹¹

The final sorted DEGs were used to identify the upregulated genes from the gills and mantle. MeV v4.8.1 was used to create a heatmap of the expression of the selected genes in gill and mantle tissues.⁹² GO enrichment analysis was performed using Blast2GO v6.0.⁹³ Then, the enriched GO terms for biological process, molecular function, and cellular component in each tissue were identified using a two-tailed Fisher's exact test with a P-value of 0.05 as the cutoff for significance in order to account for the effects of multiple testing. Using the enriched GO terms, the top 20 GO terms, with respect to the gene abundance data, were selected and identified. The results based on GO class, gene abundance, and P-values were used to visualize the top 20 GO terms.⁹⁴

Tissue expression of HSP70 genes and domain analysis

HSP70 genes included in 20 GO terms were identified. The expression of the HSP70 genes in gill and mantle tissues was compared at 12 and 48 h, and a heatmap was conducted using MeV v4.8.1.⁹² Heatmaps were created using the complete linkage method, which was optimized for each gene, and an Euclidean distance analysis was performed to find correlations.⁹² The domain analysis of HSPa B2 genes in T. granosa was conducted among the identified HSP70 genes. SMART v9.0⁹⁵ and the NCBI Conserved Domains Database⁹⁶ were used to identify the NBD, SBD, and EEVD motifs of HSPa B2 genes.

Funding and support

This research was supported by a grant from the National Institute of Fisheries Science, Korea (grant no. R2024048) and a grant from the Korea University.

Author contribution

H.P. and J.S.L. conceived the study. J.K, H.J.K., E.C., M.A.J., M.C., and S.C. performed experiments and data analysis. J.K., H.J.K. J.S.L., and H.P. wrote the manuscript. All the authors contributed to writing and editing the manuscript, collating the supplementary information, and preparing the figures. This work has been read and approved by all the authors.

Declarations of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary material

Supplementary material

.

Data availability statement

Data will be made available on request.

• The accession numbers for the SRA data are SRR24656882, SRR24656883, SRR24656884, SRR24656885, SRR23117963, and SRR23117964. (Original data) (NCBI)