Content area
Prolonged marine heat waves are a major manifestation of climate change, threatening marine biodiversity. Using the Mytilus galloprovincialis mussel as a sentinel organism, we investigated the impact of experimental, prolonged high temperatures as expected in 2050 on byssus-producing organ. We demonstrated that 30 days exposure at 28 °C, alters molecular and functional properties of mussel foot. Results revealed severe downregulation in the synthesis of the proximal, distal and non-gradient pre-collagen byssus components. The stressed mussels produced less than half the number of byssus filaments compared to the control. Byssus threads showed reduced mechanical resistance to traction, with a 55% decrease in maximum force. At the organism’s health status level, a long-term collapse in heat shock protein 70 subspecies expression suggests energy depletion. This is confirmed by a reduction of the hepatosomatic index during exposure and by the physiological impairment with clearance rate reduction and loss of air exposure tolerance. These effects compromise the mussel’s substrate adhesion, increasing the risk of detachment but also threaten bivalve survival. The potential consequences at the ecological level could be severe, including a reduction in coastal biodiversity, in the ecosystem services provided by these animals, as well as risks to the sustainability of Mediterranean mussel farming.
Introduction
The Mediterranean Sea is the second fastest warming region in the world, with temperatures rising 20% faster than the global ocean average. Between 1987 and 2017, the Adriatic Sea, the northernmost arm of the Mediterranean sea, showed a consistent increase in surface temperature (SST), with areas reaching 30 °C in 20241. Moreover, projections under high-emission scenarios suggest that SSTs may further increase by around 1.5 °C by 20502. The Northern Adriatic in particular has been identified as a hotspot, due to the increasing frequency and duration of marine heat waves (MHWs) affecting this geographical area. The impact of MHWs on bivalve populations is devastating, leading to mass mortality, disease outbreaks, and economic losses3. Looking at the current temperatures reached, the MHW of 2022, for example, recorded average SST anomalies for the North Adriatic as high as + 2 °C, with peaks of + 2.5 °C to around + 3 °C in late July4. In this period, there was a mortality of 100% of local wild Mytilus galloprovincialis beds5. Responses to thermal stress depend on heat intensity and duration. In case of acute stress, such as in the study by Xu et al. (2023), where Mytilus coruscus was exposed to 33 °C for 24 h, the mussels increased SOD (superoxide dismutase), ACP (acid phosphatase), and LZM (lysozyme) activity in hepatopancreas, with weakened digestive function and an overexpression of heat shock protein genes6. When high temperatures persist for longer periods, as observed in Mytella strigata exposed to a simulated heat wave of 27 °C for 5 days, the mussels produced significantly lowered number, length, and diameter of byssal threads, with impairments in the adhesion capacity7. A study by Ducker et al. (2023) showed that Perna viridis, was able to recover and maintain its vital functions after 21 days at 30 °C8, suggesting that this mussel may develop a thermal tolerance. Nevertheless, the effects of more prolonged thermal stress, closer to the future climate conditions, on M. galloprovincialis remain largely unexplored. As demonstrated by Galli et al.9, heat waves in the Northern Adriatic Sea will remain below 28 °C. However, they will persist for up to 30 days, which will wreck ecosystems and cause diffuse mortality among thermally sensitive bivalve mollusks. To simulate the possible mechanisms underlying MHW-related mass mortality and identify key factors for bivalve production under the forecasted scenario, we carried out the first study in experimentally controlled MHW-like conditions on the Mytilus galloprovincialis mussel, a representative sentinel organism which generates significant income for this region. Mussels were experimentally exposed to an increased water temperature to mimic the MHWs predicted for 2041–2050 in the Northern Adriatic Sea (28 °C for 30 days), and their physiological, molecular and structural responses evaluated.
Results and discussion
The analysis of biometric parameters highlighted a significant reduction in hepatosomatic index, a stress index related to the role of the hepatopancreas as a storage site for metabolic reserves to be used during periods of high metabolic demand10, suggesting the occurrence of a metabolic switch in exposed mussels. A constant reduction of this index was observed between days 3 and 30 (Supplementary Figure S1C), associated with an unaltered pulp yield (Supplementary Figure S1A,B) and with a reduction in clearance rate from day 5 of exposure onwards (Supplementary Figure S1D), possibly indicating an energy-saving strategy. From day 3 of exposure onwards, however, we observed a severe reduction in resistance to air (Supplementary Figure S1E), a parameter widely used as an ecophysiological indicator of resilience to environmental stressors such as temperature11. Finally, an increase in hemolymphatic electrolyte concentration in exposed mussels (Supplementary Figure S1F–J) suggests an increase in anaerobic metabolism12.
To investigate the molecular changes triggered by the exposure to 28 °C water temperature, a whole transcriptomic analysis was performed on the mussel foot at days 3 and 30 of exposure. The foot was selected due to its production of byssus, a bundle of fibres with an essential role in anchoring mussels, thus impacting the animals’ ability to thrive by their resistance to physical stimuli13. To date, the mussel genome has not been annotated, therefore results are generated by a de novo transcriptome assembly to reconstruct the transcript sequences without a reference genome. The complete dataset of the RNAseq analysis has been deposited in a public repository (Web link to the data: https://amsacta.unibo.it/id/eprint/8262/, accession number: 8262).
The expression pattern in the mussel foot was affected by both the MHW-like exposure and its length (Fig. 1A; Supplementary Figure S2). In fact, the principal component analysis, which highlights the main molecular differences within groups, showed significant differences between control and exposed groups (Fig. 1B), with a further time-dependent difference in the exposed groups only (Supplementary Figures S3 and S4).
[See PDF for image]
Fig. 1
Whole transcriptome analysis. (A,B) Clusterization of all groups using a supervised (A) and unsupervised (B) approach. (C–E) Pathway enrichment analysis using GO—Molecular Functions (C), Biological Process (D), and Cellular Component (E). (F) Analysis of the HSP family genes at 3 and 30 days, shown as differences between exposed and non-exposed groups. (G–I) Gene expression analysis of preCOL-P (G), preCOL-D (H), and preCOL-NG (I) genes by qPCR at all considered time points, compared to the results obtained by RNAseq. Data are reported as mean value ± standard deviation. The asterisks denote a statistically significant difference between the groups (*p < 0.05; **p < 0.01; ***p < 0.001).
Pathway enrichment analysis identified different pathways involved mainly in the gene ontology groups “Molecular Function” (Fig. 1C), “Biological Process” (Fig. 1D), and “Cellular Component” (Fig. 1E), which are differentially regulated depending on the MHW-like exposure and its length.
Various heat shock protein (HSP) transcripts in particular, including several HSP70 subspecies, were overexpressed after 3 days of experimental exposure, returning to baseline after 30 days (Fig. 1F). Described in both prokaryotic and eukaryotic organisms (including aquatic organisms), HSPs are well-known molecular chaperones, playing a crucial role in cellular responses to external stimuli, such as adaptation to elevated temperatures14. These results are in line with the acute bivalve stress response to extreme temperature15, consisting of a rapid increase in HSP expression levels associated with a conformational rearrangement to expose substrate-binding sites10. Increased HSP70 levels have been described in different bivalve tissues in response to environmental stress16, suggesting that the acute upregulation of these proteins supports a stress-related energy switch17. However, this acute molecular compensatory response is inadequate to sustain bivalve physiology over the long-term, ultimately causing a depletion of the animals’ energy reserves, as suggested by the stress hepatosomatic index.
To further investigate the possible mechanisms of MHW-related mass mortality, we analyzed the production of byssus collagen. This natural proteinaceous filament is produced by a specific gland in the mussel foot, and its main components are collagen18,19 and mussel foot proteins (MGFPS 1–6). HSPs also act as chaperones for collagen synthesis and assembly20,21.
We found that three main byssus components (pre-COL-P, preCOL-D, and preCOL-NG) were altered, as indicated by RNA sequencing. To validate these results, we used qPCR testing six different housekeeping genes, finally selecting GAPDH for the relative quantification of the gene expression (Supplementary Figure S5). In this analysis, additional timepoints (days 5 and 10) were included. Both molecular analyses showed a strong upregulation of preCOL-P expression at 3 days (Mann–Whitney test, p = 0.0087) followed by a downregulation of all three genes at 30 days (Mann–Whitney test, p = 0.0152) (Supplementary Figure S6), suggesting a collapse of the mussels’ capability to synthetize the byssus under MHW-like exposure (Fig. 1G–I). This was confirmed by a reduction in preCOL-D (Mann–Whitney test, p = 0.0411) and preCOL-NG (Mann–Whitney test, p = 0.0411) gene expression at 30 days of exposure. Notably, byssus composition and structure are known to be affected by chemical and physical changes in sea water parameters, including temperature22.
We then investigated the morphological, chemical and biometric characteristics of the byssus produced by mussels after 30 days of experimental MHW-like exposure. The adhesion plaques appeared damaged on the edges in the exposed group (Fig. 2A 1–4), but the byssus maintained its external thread surface morphology (Fig. 2A5–8). The elemental composition of the whole thread (Supplementary Figure S7), which is essential for the proper mechanical function of the byssus, showed normal expression of MGFPs (Supplementary Figure S8). Byssus filament count was significantly lower in the exposed group (exposed vs. CTR, mean ± SD, 22 ± 8 vs58 ± 23; unpaired Student’s t-test, p = 0.0006), although the exposure did not alter thread dimension (length: mm; exposed vs. CTR, mean ± SD, 26.70 ± 5.69 vs. 23.09 ± 6.96; unpaired Student’s t-test, p = 0.8067; diameter: µm; exposed vs. CTR, mean ± SD, 114 ± 31.50 vs. 122.13 ± 39.86; unpaired Student’s t-test, p = 0.6351) (Fig. 2B). A significant loss in mechanical performance was observed in the byssus thread of the exposed group, with a 55% loss of maximum force (N; unpaired Student’s t-test, p = 0.0044) and 40% loss of maximum stress (Mpa; unpaired Student’s t-test, p = 0.0176) (Figs. 2E,F; 2S D,E). This could be due to collagen softening23 or to a change in its molecular organization, leading to a premature collapse. However, no differences were observed when testing byssus deformation at maximum force (%; unpaired Student’s t-test, p = 0.1218) or elastic modulus (Mpa; unpaired Student’s t-test, p = 0.4807) (Fig. 2C,D). These results are consistent with the lack of difference between exposed and control groups in adhesion tightness (maximum stress of plaque: Mpa; unpaired Student’s t-test, p = 0.0730) and probability of detachment from surfaces (maximum force of plaque: %; unpaired Student’s t-test, p = 0.1824) (Fig. 2E,F). The reduced number of filaments and weakening of the threads increases the byssus vulnerability to external mechanical stimuli, favoring mussel detachment and death.
[See PDF for image]
Fig. 2
Evaluation of byssus performance. (A) Scanning electron microscopy images at 30 days of the freeze-dried byssus control (1, 3) and exposed plaque (2, 4) (scale bars are 100 µm (left), 10 µm (right)), control proximal and distal regions of threads (scale bars 10 µm) (5, 6), exposed proximal and distal regions of threads (scale bars 10 µm) (7, 8). (B) Count of thread number per group at 30 days, with measurements of length (mm) and diameter (µm) taken at the terminal portion of the proximal region. (C–F) Tensile strength tests on the byssus following 30 days of exposure in the control and exposed mussels, evaluating the deformation of the thread and plaque (C), the elastic modulus of the thread (D), maximum stress (E) and maximum force (F) for byssal thread resistance and plaque adhesion.
The differences between M. galloprovincialis and other bivalves (such as the Pacific oyster M. gigas) exposed to the same experimental conditions3 show M. galloprovincialis to be a good candidate for the monitoring of temperature-related environmental changes.
Conclusion
This study showed that Marine heatwaves are a major stressor for mussels, with significant impacts on physiological animal status and byssus properties. Consequently, these changes could compromise adhesion and increase mortality risk. Mussel mass mortality is a crucial issue for both the preservation of coastal communities and for the production-related economy24. Following the MHW which hit the central Adriatic Sea in the summer of 2022, the percentage of substrate covered by mussels saw an astounding fall from 81.6 to 0%25. The climate change prediction models simulated in this study indicate that weaking of mussel byssus and physiological status could lead to mass mortalities with the potential to jeopardize mussel colonies in other European regions, with prospective economic losses of 1.5 billion euros/year. A partial recovery might be achieved through the adaptive selection of heat-resilient species26 or the targeted selection of restoration sites, with possible changes in the distribution range of mussels27.
Methods
Animals and experimental design
Mytilus galloprovincialis mussels were purchased from a local farm in the Adriatic Sea off Cesenatico (Italy), 5 km from the coast, at a depth of 3 m, and transferred to the laboratory of Aquaculture at the Department of Veterinary Medical Sciences of the University of Bologna. At the beginning of the experiment, the animals were 1 year old, with an average body weight of 21.4 ± 0.1 g, and a shell length of 65 ± 1 mm. A total of 264 animals were randomly distributed into twelve 70-L aquaria filled with recirculating natural Adriatic seawater at a salinity of 28–30 ppt, a temperature of 20.0 ± 0.5 °C, and a pH of 8.0 ± 0.1. The mussels were fed daily with a maintenance ration of live microalgae (Isochrysis galbana), consisting of 3% of the mean dry-meat weight of the animals in dry weight of algae feed per day. Following 1 week of acclimatization, the byssus threads were cut to monitor byssus production during the thermal stress induced by the experiment23, and the animals divided into two groups of six aquaria each. The control group was maintained at a temperature of 20.0 ± 0.5 °C, while the MHW-like-exposed group (hereinafter “exposed group”) was brought to 28.0 ± 0.5 °C by warming the water at a rate of 0.5 °C h−1 with heaters (Askoll AA230004). The water temperatures were kept constant for 30 days using heat exchangers (ECOB-5.0), and seawater parameters checked daily (temperature and pH with a multiple parameters probe Orion Star A221 portable meter, and salinity with portable refractometer ATC).
Biometric indices
Six mussels from each group (control and exposed) were sampled at each experimental timepoint (3, 5, 10 and 30 days after MHW-like exposure) to analyze the biometric indices. The animals were individually weighed using a precision scale (SAUTER Re 2012 Precision Electronic Weighing Scales, METTLER TOLEDO Instruments), then the soft body was separated with a scalpel from the shell to weigh the valves and wet meat, and calculate the meat yield (MY = wet soft tissue weight (g)/ whole specimen weight (g)* 100)28 and condition index (CI = dry soft tissue weight (g)/ dry valve weight (g)) *100)29. At each timepoint, the hepatopancreas of the sampled animals was dissected and weighed to calculate the hepatosomatic index (HSI = hepatopancreatic weight (g)/ whole specimen weight (g) * 100)30.
Physiological parameters
At each timepoint, an additional three mussels per group were chosen at random to measure the clearance rate (CR = (lnC0 – lnC1) V / t, where C0 and C1 were the algal cell concentrations at the beginning and end of each time increment t = 0.5 h). To measure the CR, mussels were individually placed in beakers containing 1 L of aquarium seawater (filtered through a mechanical sponge filter, biofilter and UV lights) and a defined quantity of algal cells. 10–15 ml aliquots were then sampled from each beaker at 30 min intervals over a 2-h period to measure the variation in algal cell concentration due to filtration by the mussels31. The algal concentration was monitored using a Bürker cell counting chamber by inverted microscope (Nikon Eclipse TS100). Six mussels per group were sacrificed at each timepoint to collect 0.5–1.0 mL of hemolymph from the adductor muscles, using a 1 mL syringe with 25-gauge needle. The hemolymph was then centrifuged for 3 min at 1000×g32, and the supernatants analyzed using a automated chemistry analyzer (AU 480; Olympus/Beckman Coulter, Brea, CA, United States) to assess total protein content (Total Protein OSR6632 kit, Beckman Coulter), aspartate transaminase (AST) activity (AST OSR6109 kit, Beckman Coulter) and concentrations of glucose (Glucose OSR6121 kit, Beckman Coulter), phosphate (Inorganic Phosphorus OSR6122 kit, Beckman Coulter), magnesium (Magnesium OSR6189 kit, Beckman Coulter), calcium (Calcium Arsenazo OSR60117 kit, Beckman Coulter), sodium (66,317 ISE Low Serum Standard; 66,316 ISE High Serum Standard kit, Beckman Coulter), potassium (66,317 ISE Low Serum Standard; 66,316 ISE High Serum Standard kit, Beckman Coulter) and chloride (66,317 ISE Low Serum Standard; 66,316 ISE High Serum Standard kit, Beckman Coulter). Ammonia concentration was measured using a point-of-care analyzer (FUJI DRI-CHEM NX10 N, Tokyo, Japan). The resistance to air exposure test was carried out at each timepoint, selecting six animals for each experimental group. As described by Eertman e de Zwaan in 1993, animals were placed in an airtight container at 18 °C and 80–90% humidity, and survival assessed daily until 100% mortality was reached. The LT50 value was recorded, indicating the time in which 50% of the tested bivalves died. The mussels were considered dead when found with their valves open and unresponsive to external stimuli33.
Characterization of byssus properties
At the beginning of the experimental procedures, 10 mussels per group were stripped of their byssus and individually placed on 18 × 18 cm-squares of polypropylene (PP) film coating the bottom of the aquaria to simulate attachment to the PP socks used in farming, and to evaluate the quality of the newly-synthesized byssus under thermal stress. At the final 30-day timepoint, all mussels attached to PP squares were removed from the supports, cutting the byssus as close to the body as feasible (ideally the proximal region of the stem) to preserve the threads attached to the film. These samples were then stored in a sealed, airtight sterile box containing sterile aquarium seawater at 4 °C and salinity 28–30 ppt until analysis, and no mold formation was observed. Preserving the sample in sterile seawater prevents alterations in the material due to compaction from drying and therefore partial restoration of swelling during rehydration34.
Individual thread count was performed by naked eye, including only intact fibers attached to the stem or support.
The actual thickness of each hydrated sample was measured using a SM-LUX POL microscope fitted with a 5.0 MP digital camera (Motic Moticam 5 +) for image capture. Diameters were measured at the terminal portion of the proximal region, with ImageJ used to analyze the images, while thread length was measured using a caliber (± 0.05 mm). Only intact threads attached to the stem or to the support were used to measure thickness and length.
Filament morphology was assessed using scanning electron microscopy (SEM). Wet filaments were washed with ultrapure water (Milli-Q), freeze-dried using a FreeZone 1 freeze dryer (Labconco Corp., Kansas City, MO), glued to carbon tape, and coated with 20 nm of gold prior to imaging. SEM images were acquired using a LEO 1530 FEG (Zeiss) at a tension of 5 kV and 30 µm aperture. Analyses were carried out on three samples from different animals for each experimental group.
Elemental analysis was performed using Inductively Coupled Plasma Mass Spectrometry (ICP-MS). About 0.1 g of dry byssus was rinsed three times with ultrapure water (Milli-Q) to remove seawater residues, cut into 3 mm pieces with a ceramic knife on a glass plate, dried at 40 °C and weighed. Digestion was performed by adding 8 mL of nitric acid (HNO3, superpure grade, 67–70%, Merck, Darmstadt, Germany) and 2 mL of hydrogen peroxide (H2O2, Suprapur, 30%, Merck, Darmstadt, Germany) to each sample. Digested samples were then mineralized under microwave heating (Tmax of 220 °C, peak power of 1800 Watt; two cycles of 20 min each) using a Milestone ETHOS UP. After digestion, samples were filtered (Whatman® Quantitative Filter Paper: Grade 42), diluted to a final volume of 30 mL with ultrapure water (Milli-Q) and the total inorganic composition analyzed using an Agilent Technologies 7900 ICM-MS equipped with a MicroMist nebulizer and a double-pass quartz spray chamber. The analyses were performed using an RF Power of 1550 W, RF Matching of 1.8 V, plasma gas flow of 15.0 L/min, nebulizer gas flow of 1.05 L/min, collision gas flow rate (He) of 5.0 mL/min, and auxiliary gas flow of 0.9 L/min. ICP-MS was used to determine the amount of silver (Ag), aluminum (Al), arsenic (As), boron (B), barium (Ba), beryllium (Be), cadmium (Cd), cobalt (Co), chromium (Cr), copper (Cu), iron (Fe), mercury (Hg), magnesium (Mn), molybdenum (Mo), nickel (Ni), lead (Pb), antimony (Sb), tin (Sn), strontium (Sr), titanium (Ti), thallium (Tl), vanadium (V) and zinc (Zn) in the samples. External calibration was performed using a CPA chem Certified Reference Material (ISO 9001, ISO 17,025, ISO 17,034) ICP multi-element standard solution with 30 components. Both the control sample and procedural blank were analyzed, and the reference material NCS DC8102a used to determine the following recovery: Cu 95%, Zn 102%, Fe 86%, Mn 91%, Cd 103%, Pb 95%, Ni 99%, Cr 95%, Co 94%. For both control and exposed samples, the byssus of at least two different animals was used to compose a single ICP-MS sample, and two samples analyzed for both control and exposed groups.
To test the tensile strength of the byssus, we cut the threads proximally close to the stem, then removed the support together with the portion around the plaque. We then immersed the byssus and support in the filtered aquarium seawater (0.2 µm PTFE syringe filter) and stored them at 4 °C. To measure the adhesion between the thread and support plaque and ensure optimal alignment of the byssus thread, we designed a custom apparatus consisting of a plastic anchor device glued to the support film on the byssus-free side, and a piece of sandpaper glued to the proximal end of the byssus threads (Supplementary Fig. S7B). To measure the tensile strength of the byssus thread, we glued each end of the byssus filaments to a piece of sandpaper to prevent slipping (Supplementary Fig. S7B). We chose cyano-acrylate glue for these tests, and carried out preliminary tests to ensure that ruptures occurred on the thread itself and not at the attachment between the thread and the sandpaper.
To ensure that the samples remained wet for the duration of the test (around 30 s), they were kept immersed in seawater microfiltered with a 0.2 µm PTFE syringe filter until testing.
All mechanical tests were performed using an Anton Paar MCR 702 TwinDrive Dynamic Mechanical Analysis (DMA) instrument fitted with two solid rectangular fixtures. The samples were placed between the fixtures and a preload of 0.001 N was applied. We carried out a monotonic quasi-static uniaxial traction test, applying a constant crosshead speed of 1 mm min − 1 until rupture, consisting either of plaque detachment or thread failure, depending on the test. For both adhesion and tensile strength tests, we measured the maximum force (MF) and tensile strain (σmax), and maximum displacement and strain (εmax) respectively. Based on the stress–strain curve of the tensile test, we measured the elastic modulus of the threads, calculating the slope of the tangent to origin.
Biochemical analysis
Total RNA extraction
We performed total RNA extraction on the basal portion of the foot tissues (50–100 mg), collected from six mussels per group at 3, 5, 10 and 30 days after exposure. Tissues were immersed in RNAlater (10 μL/mg) to preserve the integrity of the nucleic acids, and stored at − 80 °C until analysis. We chose the foot because it extends outside the shell, where it directly interacts with the external environment and performs several crucial functions, producing byssus threads for attachment to substrates, and enabling locomotion, determining where the mussel will adhere and influencing interspecific clumping patterns. Total RNA was extracted by RNeasy® Plus Universal Mini Kit (Qiagen) according to the manufacturer’s instructions. In brief, the tissues were removed from RNAlater, placed in clean tubes containing 900 μL QIAzol, and mechanically homogenized. To remove the genomic DNA and separate the RNA, 100 μL of gDNA Eliminator solution and 180 μL of chloroform were added to the lysate, shaken for 15 s and then centrifuged at 12,000×g at 4 °C for 15 min. Following centrifugation, 600 μL of the separated aqueous phases containing the RNA were loaded into an automatic RNA extraction QIAcube (Qiagen).
Whole transcriptome analysis
Whole transcriptome analysis by RNA seq was performed at 3 and 30 days, while PCR/qPCR analysis was performed for all the examined timepoints (3, 5, 10, 30 days). Tissues were homogenized and total RNA extracted using the Qiagen RNAeasy mini kit. The same samples were used for both methods (day 3 and 30).
RNAseq was performed by Macrogen Europe® using gene ontology analysis on a phylogenetic species with a whole genome sequenced in the databases. A GO-pathway enrichment analysis was also performed to identify common pathways between differentially expressed genes (DEGs).
For RNA sequencing, we used a paired-ends strategy, selecting a read length of 101 bp. The library was produced by the Illumina platform, using the TruSeq Stranded mRNA library prep kit and the TruSeq stranded mRNA Reference Guide #1,000,000,040,498 v00.
De novo transcriptome assembly was performed to reconstruct the transcript sequences without a reference genome sequence. Assembled contigs are shown as the expressed transcripts for the species. Contigs were merged to non-redundant unique transcripts as long as possible and clustered into “unigenes” (i.e., unique genes) with a minimum length of 200 bp.
To annotate the clustered unigenes, we blasted them against the Gene Ontology (GO), UniProt, NCBI non-redundant Protein (NR), Pfam, EggNOG, NCBI Nucleotide (NT) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases35, using the BLASTX, BLASTN and blastx algorithms. We also predicted the open reading frames (ORFs) to identify protein-coding regions within the unigenes. The unigenes were processed for read alignment, and their abundance extracted as read count from the alignment.
Gene expression analysis by qPCR
To validate the RNAseq data, we performed qPCR analysis on the same tissues. Extracted RNA was used for the reverse transcription reaction, using the iScript cDNA Synthesis Kit (Bio-Rad) according to the manufacturer’s instructions. Semiquantitative analysis of the gene expression was performed using real-time qPCR (CFX 96 machine; Bio-Rad), loading 10 ng of cDNA per analysis, using Sso Advanced Universal SYBR Green Supermix (Bio-Rad).
As a first step, we tested seven different housekeeping genes (Table 1) and analyzed three different pre-collagens of the byssus: preCOL-D, preCOL-P, preCOL-NG (Table 2). We also used the classic PCR method to analyze two isoforms of the Mytilus galloprovincialis foot protein 3, MGFP3-A and B (Table 2).
Table 1. Primer sequences for M. galloprovincialis housekeeping genes36.
Gene | Forward primer | Reverse primer |
|---|---|---|
TPM | 5′-GATGCTGAAAATCGTGCAAC-3′ | 5′-CGGTCTACTTCTTTTTGCAACTT-3′ |
RPS4 | 5′-TGGGTTATCGAGGGCGTAG-3′ | 5′-TCCCTTAGTTTGTTGAGGACCTG-3′ |
18S | 5′-CCTGGAAAGGTCGGGTAAC-3′ | 5′-AATTACAAGCCCCAATCCCTA-3′ |
18S-L33448 | 5′-CATTAGTCAAGAACGAAAGTCAGAG-3′ | 5′-GCCTGCCGAGTCATTGAAG-3′ |
GAPDH | 5′-AGGAATGGCCTTCAGGG-3′ | 5′-TCAGATGCTGCTTTAATGGCTG-3′ |
EF1 | 5′-CCTCCCACCATCAAGACCTA-3′ | 5′-GGCTGGAGCAAAGGTAACAA-3′ |
Table 2. Primer sequences for M. galloprovincialis real-time and standard PCR.
Gene | Forward primer | Reverse primer | Refs. |
|---|---|---|---|
preCOL-D | 5′-AAGGGAATCCAAGGAAGCCG-3′ | 5′-TTACCGTCTGGGCCAGTTTC-3′ | (reference sequence) https://www.ncbi.nlm.nih.gov/nuccore/21105302 |
preCOL-P | 5′-AGAAGGAAAAGCCGGACCAG-3′ | 5′-GGTCCTTGGGCTCCTCTTTC-3′ | (reference sequence) https://www.ncbi.nlm.nih.gov/nuccore/21105300 |
preCOL-NG | 5′-GGTGAACAAGGAGACCTCGG-3′ | 5′-CTGGTCCTTCTTTGCCGACT-3′ | (reference sequence) https://www.ncbi.nlm.nih.gov/nuccore/AF448524 |
MGFP3-A | 5′- AAGTCTCAGTAATCAGCACCTTTC -3′ | 5′- ATTGACAGTTTACTGATGTCTGTA -3′ | (reference paper)37 |
MGFP3-B | 5′- AAGTCTCAGTAATCAGCACCTTTC -3′ | 5′- TCGTTTCCATACAGATAGGTAATA -3′ | (reference paper)37 |
Amplification conditions were as follows: a denaturation step of 3 min at 98 °C, followed by 40 cycles of 10 s at 95 °C and 1 min at 60 °C, followed by a melting curve analysis consisting of 80 cycles of 10 s, with an increase of 0.5 °C between each cycle from 65 to 95 °C. A single peak was observed on the melting curve, confirming the specificity of the reaction. GAPDH was selected as housekeeping gene for the analysis (Supplementary Fig. S5).
For MGFP3-A and MGFP3-B genes, we used a standard PCR protocol to quantify expression levels37. The sequences of the forward and reverse primers used in PCR were specific for M. galloprovincialis, synthesized by Integrated DNA Technologies, and are listed in Table 2.
Amplification conditions for standard PCR were: a denaturation step of 3 min at 98 °C, followed by 50 cycles of denaturation at 94 °C for 30 s, annealing at 55 °C for 30 s, elongation at 70 °C, and a final dissociation step at 95 °C for 1 min.
Following the amplification reaction, 10 μL of PCR product was analyzed using the Qiaxcel Advanced System (QIAGEN) and visualized using the QIAxcel DNA High Resolution Cartridge (2,759,315) using the following conditions: QX Alignment Marker 15 bp/3 kb (175,040,454), a standard QX size marker 100 bp-2.5 kb at 5 ng/μl concentration (175,037,497), and the OM500 method. The cartridge was calibrated using the QX Intensity Calibration Marker (175,041) according to the manufacturer’s instructions.
The PCR product sizes (286 bp for MGFP3-A and 313 bp for MGFP3-B) were estimated and quantified using QIAxcel ScreenGel Software 1.6.0.
Statistical analyses
All statistical analyses were conducted using GraphPad Prism (v. 10).
The data pertaining to thread number, length, diameter, elemental composition, maximum force, deformation, maximum byssus thread and plaque stress, and Young’s modulus of thread were checked for the normality Shapiro–Wilk test. Then the length values were analyzed using the Mann–Whitney test, the other parameters were analyzed by unpaired Student’s t-test. The mRNA expression data from qPCR for HSP70, preCOL-D, preCOL-P, preCOL-NG, MGFP3-A and MGFP3-B were analyzed using the Mann–Whitney test, and the data from the exposed group normalized on the time-related controls to assess the effect of temperature during exposure.
Author contributions
A. D. M.: conceptualization, methodology, investigation, data curation, writing—original draft preparation. V. A. B.: conceptualization, methodology, investigation, data curation, writing—original draft preparation. C. Q.: data curation, writing—original draft preparation. V. B.: investigation, data curation. L. C.: conceptualization, supervision, project administration. L. G.: conceptualization, visualization, supervision, project administration. D. M.: conceptualization, methodology, investigation, data curation, writing—original draft preparation. G. F.: conceptualization, supervision. G. G.: methodology, investigation, data curation, writing—original draft preparation. N. G.: methodology, investigation. F. D.: methodology, visualization, supervision, writing—original draft preparation. M. G. F.: methodology, investigation, data curation, writing—original draft preparation. P. P. G.: visualization, supervision. L. P.: conceptualization, visualization, supervision. A. B.: conceptualization, visualization, supervision, project administration, funding acquisition. All authors reviewed the manuscript.
Data availability
The complete dataset of the RNAseq analysis performed during the current study is available in AMS Act repository. Web link to the data: https://amsacta.unibo.it/id/eprint/8262/. Accession number to the datasets: 8262.
Declarations
Competing interests
The authors declare no competing interests.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
1. Tojčić, I; Denamiel, C; Vilibić, I. Kilometer-scale trends and variability of the Adriatic present climate (1987–2017). Clim. Dyn.; 2023; 61,
2. Moulin, A. et al. Projections of the Adriatic wave conditions under climate changes. Front. Climate6, (2024).
3. De Marco, A et al. Prolonged heat waves reduce the condition index and alter the molecular parameters in the pacific oyster Crassostrea gigas. Fish Shellfish Immunol.; 2023; 133, 108518. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36610607]
4. Marullo, S et al. Record-breaking persistence of the 2022/23 marine heatwave in the Mediterranean Sea. Environ. Res. Lett.; 2023; 18, 114041.2023ERL..18k4041M
5. Bracchetti, L et al. Mass mortality event of mediterranean mussels (Mytilusgalloprovincialis) in the middle adriatic: Potential implications of the climate crisis for marine ecosystems. Diversity; 2024; [DOI: https://dx.doi.org/10.3390/d16030130]
6. Xu, L et al. Insights into the response in digestive gland of Mytilus coruscus under heat stress Using TMT-based proteomics. Animals; 2023; 12, 2248.
7. Xu, X. et al. Heatwaves hinder mussel invasion by weakening byssus production. Front. Mar. Sci.10, (2023).
8. Ducker, J et al. Mussels show capacity for persistence under, and recovery from, marine heatwaves. Mar. Biol.; 2023; 170, 120.
9. Galli, G., Solidoro, C. & Lovato, T. Marine heat waves hazard 3D maps and the risk for low motility organisms in a warming Mediterranean sea.Front. Mar. Sci.4, (2017).
10. Cartier, S et al. Use of an index based on the blue mussel (Mytilusedulis and Mytilustrossulus) digestive gland weight to assess the nutritional quality of mussel farm sites. Aquaculture; 2004; 241,
11. Monari, M et al. Effects of high temperatures on functional responses of haemocytes in the clam Chameleagallina. Fish Shellfish Immunol.; 2007; 22,
12. Dowd, WW; Somero, GN. Behavior and survival of Mytilus congeners following episodes of elevated body temperature in air and seawater. J. Exp. Biol.; 2013; 216,
13. McCartney, MA. Structure, function and parallel evolution of the bivalve byssus, with insights from proteomes and the zebra mussel genome. Philos. Trans. R. Soc. Lond. B Biol. Sci.; 1825; 2021,
14. Jeyachandran, S. et al. A review on the involvement of heat shock proteins (Extrinsic Chaperones) in response to stress conditions in aquatic organisms. Antioxidants (Basel)12(7), (2023).
15. Gill, LT; Kennedy, JR; Marshall, KE. Proteostasis in ice: The role of heat shock proteins and ubiquitin in the freeze tolerance of the intertidal mussel Mytilus trossulus. J. Comp. Physiol. B; 2023; 193,
16. Fabbri, E; Valbonesi, P; Franzellitti, S. HSP expression in bivalves. ISJ-Invertebrate Surv. J.; 2008; 5, pp. 135-161.
17. Jeffrey, JD et al. Responses to elevated CO(2) exposure in a freshwater mussel Fusconaia flava. J. Comp. Physiol. B; 2017; 187,
18. Filippidi, E et al. The microscopic network structure of mussel (Mytilus) adhesive plaques. J. R. Soc. Interface; 2015; 12,
19. Sagert, J; Waite, JH. Hyperunstable matrix proteins in the byssus of Mytilus galloprovincialis. J. Exp. Biol.; 2009; 212,
20. García, R et al. Extracellular heat shock protein 90 binding to TGFβ receptor I participates in TGFβ-mediated collagen production in myocardial fibroblasts. Cell Signal; 2016; 28,
21. Abd El-Fattah, EE; Zakaria, AY. Targeting HSP47 and HSP70: Promising therapeutic approaches in liver fibrosis management. J. Transl. Med.; 2022; 20,
22. Silverman, HG; Roberto, FF. Understanding marine mussel adhesion. Mar. Biotechnol. (NY); 2007; 9,
23. Li, YF et al. Near-future levels of ocean temperature weaken the byssus production and performance of the mussel Mytiluscoruscus. Sci. Total Environ.; 2020; 733, 139347.1:CAS:528:DC%2BB3cXpvFamu7o%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32446082]
24. Fujii, K. K. et al. The thermal stability of the collagen triple helix is tuned according to the environmental temperature. Int. J. Mol. Sci.23(4), (2022).
25. Moeser, GM; Leba, H; Carrington, E. Seasonal influence of wave action on thread production in Mytilus edulis. J. Exp. Biol.; 2006; 209,
26. Ericson, JA et al. Differential responses of selectively bred mussels (Perna canaliculus) to heat stress-survival, immunology, gene expression and microbiome diversity. Front. Physiol.; 2023; 14, 1265879. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38425477]
27. Benjamin, ED et al. Aerial exposure and critical temperatures limit the survival of restored intertidal mussels. Restor. Ecol.; 2024; 32,
28. Chelyadina, NS et al. Comparative characteristics of indices to assess the quality of mussel production by an example of cultivated Mytilus galloprovincialis (Crimea, the Black Sea). Turk. J. Fish. Aquat. Sci.; 2019; 19, pp. 719-726.
29. Lassoued, J et al. The Mediterranean mussel Mytilus galloprovincialis: Responses to climate change scenarios as a function of the original habitat. Conserv. Physiol.; 2021; 9,
30. Kopecka, J et al. Measurements of biomarker levels in flounder (Platichthysflesus) and blue mussel (Mytilustrossulus) from the Gulf of Gdańsk (southern Baltic). Mar. Pollut. Bull.; 2006; 53,
31. Romano, C et al. Effect of the presence of the shore crab, Carcinus maenas, on burrowing behaviour and clearance rate of the common cockle Cerastoderma edule. Mar. Biol.; 2011; 158, pp. 2685-2694.
32. Gustafson, LL et al. Reference ranges for hemolymph chemistries from Elliptio omplanate of North Carolina. Dis. Aquatic; 2005; 65, pp. 167-176.
33. Eertman, RHM et al. “Survival in air” of the blue mussel Mytilusedulis L. as a sensitive response to pollution-induced environmental stress. J. Exp. Mar. Biol. Ecol.; 1993; 170,
34. Hagenau, A et al. Mussel collagen molecules with silk-like domains as load-bearing elements in distal byssal threads. J. Struct. Biol.; 2011; 175,
35. Kanehisa, M. et al. KEGG: Biological systems database as a model of the real world. Nucleic Acids Res. 53(D672–D677), (2025).
36. Prego-Faraldo, M. V., Martínez, L. & Méndez, J. RNA-Seq analysis for assessing the early response to DSP toxins in Mytilus galloprovincialis digestive gland and gill. Toxins (Basel)10(10), (2018).
37. Inoue, K et al. Cloning, sequencing and sites of expression of genes for the hydroxyarginine-containing adhesive-plaque protein of the mussel Mytilus galloprovincialis. Eur. J. Biochem.; 1996; 239,
© The Author(s) 2025. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.