Introduction
The Caribbean Sea has more than 60 species of scleractinian corals some of which began to decline in coral cover and, abundance in the early 1980s producing significant ecological shifts in coral reefs (Bruno & Selig, Reference Bruno and Selig2007; Leggat et al., Reference Leggat, Hoegh-Guldberg, Dove and Yellowlees2007; Weil et al., Reference Weil, Croquer and Urreiztieta2009a; Alvarez-Filip et al., Reference Alvarez-Filip, Paddack, Collen, Robertson and Côté2015; García-Urueña & Garzón-Machado, Reference García-Urueña and Garzón-Machado2020). The Caribbean Sea is famous for its lush coral reefs, one of the most diverse and valuable marine ecosystems on earth; however, these ecosystems have continued declining over the past four decades, more specifically during and after the thermal anomalies (heat stress) events of 1998, 2005–06, 2010–11, 2015 and 2017 (Aronson et al., Reference Aronson, Precht, Toscano and Koltes2002; Bruno et al., Reference Bruno, Selig, Casey, Page, Willis, Harvell, Sweatman and Melendy2007; Eakin et al., Reference Eakin, Morgan, Heron, Smith, Liu, Alvarez-Filip, Baca, Bartels, Bastidas, Bouchon, Brandt, Bruckner, Bunkley-Williams, Cameron, Causey, Chiappone, Christensen, Crabbe, Day, de la Guardia, Díaz-Pulido, DiResta, Gil-Agudelo, Gilliam, Ginsburg, Gore, Guzmán, Hendee, Hernández-Delgado, Husain, Jeffrey, Jones, Jordán-Dahlgren, Kaufman, Kline, Kramer, Lang, Lirman, Mallela, Manfrino, Maréchal, Marks, Mihaly, Miller, Mueller, Muller, Orozco Toro, Oxenford, Ponce-Taylor, Quinn, Ritchie, Rodríguez, Ramírez, Romano, Samhouri, Sánchez, Schmahl, Shank, Skirving, Steiner, Villamizar, Walsh, Walter, Weil, Williams, Roberson and Yusuf2010; Middlebrook et al., Reference Middlebrook, Anthony, Hoegh-Guldberg and Dove2010, Weil & Rogers, Reference Weil and Rogers2011; Burge et al., Reference Burge, Mark Eakin, Friedman, Froelich, Hershberger, Hofmann, Petes, Prager, Weil, Willis, Ford and Harvell2014; Pratchett et al., Reference Pratchett, Anderson, Hoogenboom, Widman, Baird, Pandolfi, Edmunds and Lough2015; Hughes et al., Reference Hughes, Anderson, Connolly, Heron, Kerry, Lough, Baird, Baum, Berumen, Bridge, Claar, Eakin, Gilmour, Graham, Harrison, Hobbs, Hoey, Hoogenboom, Lowe, McCulloch, Pandolfi, Pratchett, Schoepf, Torda and Wilson2018; Muñiz-Castillo et al., Reference Muñiz-Castillo, Rivera-Sosa, Chollett, Eakin, Andrade-Gómez, McField and Arias-González2019). A global warming of 1.5°C above pre-industrial levels (Intergovernmental Panel on Climate Change -IPCC 2018) could result in the loss of 70–90 % of coral reefs with even further losses (up to 99%) if temperature increases by 2ºC.
There is little doubt that reef-building corals will experience higher sea temperatures in the span of the next few decades. However, they also experience large fluctuations in sea temperatures and other environmental variables during diel and seasonal cycles, which may play a role in shaping their resilience or vulnerability to long-term climatic changes. Seasonal variation in Puerto Rico (Rios-Jara, Reference Rios-Jara1998) as manifested in sea temperatures is important environmental variable for reef building corals; the highest temperatures been recorded during the hot wet season (September and October, >30ºC) are conducting to thermal stress (Toth et al., Reference Toth, Stathakopoulos, Kuffner, Ruzicka, Colella and Shinn2019; Caballero-Aragón et al., Reference Caballero-Aragón, Perera-Valderrama, Rey-Villiers, González-Méndez and Armenteros2020; Cramer et al., Reference Cramer, Jackson, Donovan, Greenstein, Korpanty, Cook and Pandolfi2020), leading to physiological disruption, bleaching and diseases (Cróquer & Weil, Reference Cróquer and Weil2009; Skirving et al., Reference Skirving, Heron, Marsh, Liu, JL, Geiger and Eakin2019; Roach et al., Reference Roach, Dilworth, Martin, Jones, Quinn and Drury2020).
The Caribbean Elkhorn coral Acropora palmata (Lamarck 1816) is one of the most ecologically and geologically dominant reef-building species (Bruno & Selig Reference Bruno and Selig2007; Pratchett et al., Reference Pratchett, Anderson, Hoogenboom, Widman, Baird, Pandolfi, Edmunds and Lough2015; van Woesik & Randall, Reference van Woesik and Randall2017; Hagedorn et al., Reference Hagedorn, Page, O’Neil, Flores, Tichy, Chamberland, Lager, Zuchowicz, Lohr, Blackburn, Vardi, Moore, Moore, Vermeij and Marhaver2018) persisting since late Pliocene (McNeill et al., Reference McNeill, Budd and Borne1997; Irwin et al., Reference Irwin, Greer, Humston, Devlin-Durante, Cabe, Lescinsky, Wirth, Allen Curran and Baums2017; Mudge et al., Reference Mudge, Alves, Figueroa-Zavala and Bruno2019). This implies, that they have survived and thrived through many substantial environmental changes and possibly providing some hope for its resilience to the anticipated climate change (Hughes et al., Reference Hughes, Anderson, Connolly, Heron, Kerry, Lough, Baird, Baum, Berumen, Bridge, Claar, Eakin, Gilmour, Graham, Harrison, Hobbs, Hoey, Hoogenboom, Lowe, McCulloch, Pandolfi, Pratchett, Schoepf, Torda and Wilson2018; García-Urueña & Garzón-Machado, Reference García-Urueña and Garzón-Machado2020). However, its dominance in shallow waters (0–5 m), makes it highly vulnerable to prolonged exposed to high temperatures and ultraviolet radiation, especially during the warmer months (June–September) (Soto-Santiago & Weil, Reference Soto-Santiago and Weil2012; Reed et al., Reference Reed, Reibold, Cavaleri, Alonso-Rodríguez, Berberich and Wood2020; Barrera et al., Reference Barrera, Acevedo, Amador, Marzan, Adams and Paz-Bailey2023). This thermal and UV stress represents pose significant challenges to A. palmata and other shallow-water reef-building corals, often leading to physiological stress that manifest as bleaching, disease outbreaks, or mortality (Sutherland & Ritchie, Reference Sutherland and Ritchie2004; Bruckner & Bruckner, Reference Bruckner and Bruckner2006; Weil et al., Reference Weil, Croquer and Urreiztieta2009a; Eakin et al., Reference Eakin, Morgan, Heron, Smith, Liu, Alvarez-Filip, Baca, Bartels, Bastidas, Bouchon, Brandt, Bruckner, Bunkley-Williams, Cameron, Causey, Chiappone, Christensen, Crabbe, Day, de la Guardia, Díaz-Pulido, DiResta, Gil-Agudelo, Gilliam, Ginsburg, Gore, Guzmán, Hendee, Hernández-Delgado, Husain, Jeffrey, Jones, Jordán-Dahlgren, Kaufman, Kline, Kramer, Lang, Lirman, Mallela, Manfrino, Maréchal, Marks, Mihaly, Miller, Mueller, Muller, Orozco Toro, Oxenford, Ponce-Taylor, Quinn, Ritchie, Rodríguez, Ramírez, Romano, Samhouri, Sánchez, Schmahl, Shank, Skirving, Steiner, Villamizar, Walsh, Walter, Weil, Williams, Roberson and Yusuf2010; Rosic et al., Reference Rosic, Kaniewska, Chan, Ling, Edwards, Dove and Hoegh-Guldberg2014; Montilla et al., Reference Montilla, Ramos, García and Cróquer2016; Ricaurte et al., Reference Ricaurte, Schizas, Ciborowski and Boukli2016; Cramer et al., Reference Cramer, Jackson, Donovan, Greenstein, Korpanty, Cook and Pandolfi2020; Cróquer et al., Reference Cróquer, Weil and Rogers2021). Among these threats, elevated sea temperatures are particularly critical, as they disrupt the physiological processes required for coral health and resilience (Sutherland & Ritchie, Reference Sutherland and Ritchie2004; Bruckner & Bruckner, Reference Bruckner and Bruckner2006; Weil et al., Reference Weil, Cróquer and Urreiztieta2009b; Eakin et al., Reference Eakin, Morgan, Heron, Smith, Liu, Alvarez-Filip, Baca, Bartels, Bastidas, Bouchon, Brandt, Bruckner, Bunkley-Williams, Cameron, Causey, Chiappone, Christensen, Crabbe, Day, de la Guardia, Díaz-Pulido, DiResta, Gil-Agudelo, Gilliam, Ginsburg, Gore, Guzmán, Hendee, Hernández-Delgado, Husain, Jeffrey, Jones, Jordán-Dahlgren, Kaufman, Kline, Kramer, Lang, Lirman, Mallela, Manfrino, Maréchal, Marks, Mihaly, Miller, Mueller, Muller, Orozco Toro, Oxenford, Ponce-Taylor, Quinn, Ritchie, Rodríguez, Ramírez, Romano, Samhouri, Sánchez, Schmahl, Shank, Skirving, Steiner, Villamizar, Walsh, Walter, Weil, Williams, Roberson and Yusuf2010; Rosic et al., Reference Rosic, Kaniewska, Chan, Ling, Edwards, Dove and Hoegh-Guldberg2014; Montilla et al., Reference Montilla, Ramos, García and Cróquer2016; Ricaurte et al., Reference Ricaurte, Schizas, Ciborowski and Boukli2016; Cramer et al., Reference Cramer, Jackson, Donovan, Greenstein, Korpanty, Cook and Pandolfi2020; Cróquer et al., Reference Cróquer, Weil and Rogers2021).
Proteomics can provide an overall view of protein functions affected by climate change (Tomanek, Reference Tomanek2014), and their application to marine organisms has increased in recent years (Drake et al., Reference Drake, Mass, Haramaty, Zelzion, Bhattacharya and Falkowski2013; Weston et al., Reference Weston, Dunlap, Beltran, Starcevic, Hranueli, Ward and Long2015; Ricaurte et al., Reference Ricaurte, Schizas, Ciborowski and Boukli2016).
To evaluate the effect of seasonal temperature fluctuations on protein abundance, the proteomic profile of A. palmata colonies was examined during two seasons in a tropical setting. We assessed protein abundance patterns during the cool dry (January–February) and the hot wet season (September–October) in the Natural Reserve of La Parguera, southwest coast of Puerto Rico. Here, we predicted that colonies collected during the hot wet seasons of 2014 and 2015 will exhibit altered differential proteoform abundances as compared to those during the cool dry seasons. This study provides new insights into the protein profile of A. palmata and elucidates a metabolic pathway for acclimation in response to seasonal temperature variations and environmental changes.
Materials and methods
Figure 4 summarizes the methodological workflow, outlining the key procedures employed in the proteomic analysis.
Study area and site selection for sampling
The study was carried out at the inner-shelf reef Enrique (17.57.15 N, 67.02.55 W) and at the mid-shelf reef San Cristobal (17.56.30 N, 67.04.45 W) (Figure 1) during the months of January–February and September–October of 2014–15, representing the cool dry and hot wet seasons, respectively. Coral sampling was conducted once per season, with colonies of Acropora palmata collected near the laboratories of Magueyes Island in the Natural Reserve of La Parguera, on the southwest coast of Puerto Rico.
Map of La Parguera natural reserve indicating the study sites at the inner-shelf reef Enrique (A) and the mid-shelf reef San Cristobal (B) in Puerto Rico.
The coral colonies were sampled around 9 AM, with permission No. O-VS-PVS15-SJ-00668-22042014, issued by the Department of Natural and Environmental Resources of Puerto Rico. At the time of collection, a stainless-steel nail was inserted near each colony, and a plastic tag with a unique identification number was attached. Colonies were sampled at depths ranging from 1 to 2 m, ensuring a minimum separation distance of 3 m between them to avoid sampling genetic clones. A fragment approximately 3 × 3 cm was extracted from each colony using a chisel and hammer. The samples were then flash-frozen with dry ice, placed in plastic bags, and transported to the laboratory. Upon arrival, each fragment was divided into smaller pieces of approximately 2 cm2. Three of these pieces were processed immediately for protein extraction, while the remaining pieces were stored at −80ºC for subsequent RNA extraction. Sampling began in the cool dry season of 2014 and continued with the same colonies resampled during the hot wet season of 2014. This process was repeated in both seasons. In total, 72 biological samples were collected over 2 years from 6 colonies per reef, with each colony was sampled across 4 sampling periods, as detailed in Supplementary Figures S1 and S2. Monthly average sea surface temperatures (SSTs) were measured in situ using a Hobo-pro, covering both the cool dry (January–February) and hot wet (September–October) seasons for 2014 and 2015 (Figure 2). After temperature measurements were completed, the next step involved protein extraction, as previously described (Ricaurte et al., Reference Ricaurte, Schizas, Ciborowski and Boukli2016). Briefly, each fragment was placed in a mortar to be pulverized in 1 ml of rehydration buffer (9.5 M urea, 2% CHAPS, 1% DTT). The homogenized coral tissue was placed in a 1.5 ml tube and lysed under ice sonication, followed by slow stirring at room temperature for 1 hour. Samples were centrifuged at 12,000 rpm for 30 minutes at 4ºC, and the supernatant, specifically the upper layer free of zooxanthellae, was carefully transferred to a new 1.5 ml Eppendorf tube. A small aliquot of the supernatant was examined under a microscope to confirm the effective removal of symbionts prior to proceeding with RNA extraction. Protein concentration was estimated using the Bradford method (Bradford, Reference Bradford1976), following the manufacturer’s recommendations (Invitrogen, Carlsbad, CA, USA).
Monthly average sea surface temperatures (SST) for the cool dry season (Jan–Feb) and the hot wet season (Sep–Oct) at the inner-shelf reef Enrique (A) and in the mid-shelf reef San Cristobal (B) in La Parguera, Puerto Rico, for 2014 and 2015.
High-resolution two-dimensional gel electrophoresis (2D-GE)
A total of 72 2D-GE analyses were performed, corresponding to the 72 biological samples collected across the study period. For each analysis, 200 μg of protein from A. palmata were then mixed with rehydration buffer (Bio-Rad) and incubated overnight on 11 cm immobilized pH gradient (IPG) strips (pH 3–10). The strips were subsequently subjected to first-dimension isoelectric focusing (IEF) using a Bio-Rad protein IEF cell (50 μA/strip) for a total of 20,000 Vh, Following established protocols described by Boukli et al., (Reference Boukli, Shetty, Cubano, Ricaurte, Coelho-Dos-Reis, Nickens, Shah, Talal, Philip and Jain2012); Ricaurte et al., (Reference Ricaurte, Schizas, Ciborowski and Boukli2016, Reference Ricaurte, Schizas, Weil, Ciborowski and Boukli2024). After completion of the run, strips were reduced and alkylated for 2 × 15 min each in equilibration buffer I (6 mol L−1 urea, 2% SDS, 0.05 mol L−1 Tris–HCl, pH 8.8, 50% glycerol, and 2% [w/v] dithiothreitol [DTT]) followed by equilibration buffer II (6 mol L−1 urea, 2% SDS 0.05 mol L−1, Tris–HCl, pH 8.8, 50% glycerol and 2.5% [w/v] iodoacetamide). The second-dimension separation was performed with a 4–20% gradient SDS-polyacrylamide gel, to increase the range of molecular weights and sharpening bands (15 mA/gel for 15 min, then 30 mA/gel). The SDS-polyacrylamide gels were then stained with BioSafe Coomassie blue G-250 dye (Bio-Rad) for visualization using a Bio-Rad imaging system (Versa Doc Model 1000) (Figure 3).
Two-dimensional gel electrophoresis displaying differentially abundant proteoforms in Acropora palmata during cool dry season (A) and hot wet season (B) at the inner-shelf reef Enrique, and during Cool Dry Season (C) and hot wet season (D) at the inner-shelf reef San Cristobal, respectively. Proteoforms analysed by MALDI-TOF-TOF (1–32) are highlighted with red and green circles. The top arrow from 3–10 (pi) represents the isoelectric point. Upward arrows indicate differentially abundant proteoforms, downward arrows indicate differentially less abundant proteoforms and yellow squares denote common proteoforms. Molecular weight (MW) is giving in kiloDaltons (kDa).
Integrated gel image analysis and proteoform in-gel digestion for advance characterization using MALDI-TOF/TOF mass spectrometry
The PD Quest software version 8.0.1 (BioRad) identified spots with a > 2-fold change relative to the control, representing differentially abundant proteoforms during the cool dry season (Supplementary Figures S1 and S2). From these, 32 proteoforms, with diverse molecular weights and isoelectric points, were selected for excision from the 2D gel. These spots were manually excised from Coomassie blue stained gels using a sterile scalpel and kept in sterile microcentrifuge tubes. The gel fragments were distained with 40% methanol/10% acetic acid, reduced with 10 mmol L−1 DTT in 50 mmol L−1 ammonium bicarbonate (AB), then alkylated with 55 mmol L−1 iodoacetamide in 50 mmol L−1 AB. The resulting fragments were rinsed with 50 mmol L−1 AB and 50% acetonitrile (ACN) and dried under a stream of nitrogen. Subsequently, the samples were digested overnight at 37°C with 5% (w/w) trypsin proteomics grade (Sigma-Aldrich). The resulting digest was mixed with an equal volume of saturated cyano-hydroxycinnamic acid prepared in 50% ACN/0.1% trifluoroacetic acid. Finally, half of the mixture from each spot was applied to a MALDI target plate.
Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF/TOF MS)
Mass spectrometry (MS) was conducted using a matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF/TOF) 4800 mass spectrometer obtained from Applied Biosystems (Foster City, CA, USA) at the Durham Research Center, University of Nebraska Medical Center (Omaha, Nebraska). In-gel digested spots from A. palmata were directly applied onto a MALDI plate and co-crystallized with α-cyano-4-hydroxycinnamic acid (CHCA) matrix. The samples were dried at room temperature. Spectra acquisition and processing utilized the 4000 series Explorer software in positive reflectron mode with fixed laser fluency, low mass gate, and delayed extraction. External calibration employed a mixture of five external standards. Peptide masses were collected in 50-spectrum intervals within the range of 800–3000 Da, and MS spectra were generated from 1000 laser shots. MS/MS spectra were acquired in positive mode 1-kV, with 1000 shots summed in increments of 50. Database searches for the identified proteoforms from A. palmata were conducted using the Mascot database available in December 2014. Additionally, the ProteinPilot search engine was used with parameters applied against the Swiss-Prot databank (www.expasy.org), accessed on 14 January 2014. To further validate the data, a Protein BLAST analysis was performed, comparing the identified proteins against sequences from other Acropora species. The results showed high sequence homology (∼99%), supporting the accuracy of the protein identifications. Identified proteoforms can be retrieved using their accession numbers via the NCBI Protein database (https://www.ncbi.nlm.nih.gov/protein). All accession numbers and corresponding protein descriptions are listed in Table 2 and Supplementary Table S3.
Search parameters included carbamidomethylation for cysteines, oxidation for methionines as variable modifications, allowed for one missed tryptic cleavage, and set mass accuracy tolerances at 100 ppm for precursors and 0.5 Da for fragments.
RNA extraction
Samples previously stored at −80ºC, corresponding to each season, were retrieved for RNA extraction. The frozen samples were macerated in a mortar and transferred to small beakers for processing. The macerated tissue was then washed with 1 ml of TRIzol® for 1 hour on a gyratory shaker. After washing, the samples were centrifuged for 3 minutes at 13,000 × g at 4ºC, and the upper supernatant phase was carefully collected to reduce the presence of zooxanthellae. To confirm the absence of symbionts, a small aliquot was examined under a microscope before continuing with RNA extraction. The sample was then washed once more with TRIzol®. The resulting solution was then divided into 1 ml working aliquots. RNA concentrations were estimated by excitable fluorescence at 485 nm (RiboGreen, Molecular Probes, Portland, OR, USA), with yields ranging from 10 to 100 μg per sample. Integrity of the RNA was confirmed by electrophoresis of an aliquot (1%) of each sample on a 1% formaldehyde agarose gel (Sambrook Fritsch, Reference Sambrook, Sambrook and Fritsch1989). Finally, the total RNA underwent further purification through DNase I digestion followed by phenol/CHCl3 extraction (Message Clean®, GenHunter, Nashville, TN, USA).
Primer design and RT-PCR validation
Five primer pairs were designed using the Primer3 software (v.0.4.0) to target genes chosen based on their expression and cellular function in A. palmata: cytochrome P450, pax C, minicollagen, acropsin, and caspase 8 (Table 1). Additionally, three candidate housekeeping genes (HKGs), ribokinase, calcium-dependent protein kinase, and actin were evaluated based on their cellular functions and expected expression stability in A. palmata. Among them, actin was selected as the reference gene due to its consistent and detectable expression across all samples during preliminary evaluations, making it suitable for accurate normalization of gene expression data under varying temperature conditions. This selection informed all subsequent qRT-PCR analyses. Using arbitrary fold change (FC) thresholds of >2 and significance p-values of <0.02, the analysis then focused on proteoforms that exhibited substantial variation across reefs and seasons. Reverse transcription of the RNA (2 μg) was carried out with the iScriptTM cDNA Synthesis Kit following the manufacturer’s recommendations (Biorad). Quantitative real-time PCR (qRT-PCR) was then performed on a CFX96 TouchTM Real Time System (Biorad) in triplicates using 20 μL reactions containing 2 μL of cDNA SYBR Green PCR Master Mix (Qiagen) and 30 pmol L−1 of each primer. The qPCR conditions comprised 40 cycles of denaturation at 98ºC for 15 s, annealing and extension at 60ºC for 30 s, preceded by an initial denaturation step at 95ºC for 30 s. The delta–delta Ct method was applied, using actin as the most stable reference gene under temperature variation. All data analysis was performed using InfoStat (version 2014) and R-Studio (version 4.3) software (Supplementary Table S2, Figure S3).
Candidate reference genes and primer sequences used for qRT-PCR in Acropora palmata, designed with primer 3 software
Statistical analysis
To assess whether the data met the assumptions of normality and homogeneity of variances, a Shapiro–Wilk test was performed, revealing a significant deviation from normality (p < 0.0001). Likewise, Levene’s test indicated heterogeneity of variances among reefs (p < 0.0063) (Supplementary Table S1). The data were separated by protein spots, and each spot was individually tested using the Shapiro–Wilk test; while some followed a normal distribution, others did not. Attempts to normalize the data using log10 and square root transformations were unsuccessful.
Given these violations of parametric assumptions, a PERMANOVA (Permutational Multivariate Analysis of Variance) was employed to compare mean protein spot densities between the cool dry and hot wet season groups. The analysis was carried out in R-Studio (version 2024.12.0 + 467) using the adonis2() function from the vegan package, with a Bray–Curtis distance matrix and 999 permutations. A combination of fold change (FC) thresholds >2 and significance levels (p < 0.02) were used to identify differentially abundant proteoforms with meaningful biological variation across reefs and seasons. Only proteoforms with p < 0.05 were retained for further analysis.
To better visualize group-level variation, we applied a non-metric multidimensional scaling (NMDS) analysis based on the same Bray–Curtis matrix. The resulting ordination plots (Supplementary Figure S4) illustrate spatial separation in protein spot densities between seasons and reef sites, in agreement with PERMANOVA results.
To assess whether these differences were due to genuine biological separation or differences in group dispersion, we used the betadisper() function to evaluate the homogeneity of multivariate dispersions. The test revealed significant dispersion differences for both season and reef factors (season: p = 0.003; reef: p = 0.011), suggesting that the observed seasonal differences in PERMANOVA may be partially driven by dispersion variability. In contrast, reef-level differences appeared more distinct in the NMDS plot, reflecting consistent shifts in protein abundance between sites. These complementary statistical outcomes are summarized in Supplementary Table S1 and Figure S4.
Results
Field and experimental data
From 2014 to 2015, within the inner-shelf reef Enrique, sea surface temperatures (SSTs) ranged from 26.7ºC in the cool dry season to 29.8ºC in the hot wet season (Figures 1 and 2A). Similarly, in the mid-shelf reef San Cristobal, SSTs were comparable, varying between 26.6ºC and 30.0ºC in both respective seasons (Figures 1 and 2B).
Proteomic profiling of Acropora palmata colonies from cool dry and hot wet seasons
The A. palmata colonies sampled from the inner-shelf reef Enrique in 2014 displayed 326 differentially abundant proteoforms, 155 less abundant proteoforms, and 32 proteoforms common to both seasons. In 2015, 315 differentially abundant proteoforms, 131 less abundant proteoforms, and 32 proteoforms as in 2015 were identified (Figure 3 and Figure 5).
Schematic overview and workflow of the proteomics analysis in Acropora palmata conducted during two seasons in the natural reserve of La Parguera, located on the southwest coast of Puerto Rico.
Venn diagram of differentially abundant proteoforms in Acropora palmata across different seasons and reefs: (A) Cool Dry Season (CDS) vs. Hot Wet Season (HWS) in the inner-shelf reef Enrique. (B) CDS vs. HWS in the mid-shelf reef San Cristobal. (C) CDS inEnrique reef vs. San Cristobal Reef. (D) HWS in Enrique reef vs. San Cristobal rref during 2014 at La Parguera, Puerto Rico.
PERMANOVA results based on Bray–Curtis dissimilarity revealed significant differences in proteoform abundance between seasons (p = 0.003) and reefs (p = 0.011). However, the betadisper analysis indicated heterogeneity of dispersion for both grouping factors (season: p = 0.003; reef: p = 0.011). This suggests that the significance observed for seasonal differences may, in part, be driven by dispersion variability, whereas the clearer separation observed between reefs in the NMDS plot supports a more consistent shift in proteoform abundance (Supplementary Figure S4).
Taken together, these multivariate results highlight consistent patterns in proteoform abundance across environmental gradients, particularly between reef sites. To better understand the biological implications of these changes, the differentially abundant proteoforms in A. palmata were classified into 10 categories according to their main biological function associated with temperature stress (Figure 7). While these proteoforms may have multiple roles, our focus lies on their contributions to distinct functional categories that collectively support the proposed response pathway (Figure 6).
Schematic representation of the metabolic pathway impacted by the hot wet season in Acropora palmata. The most abundant proteoforms, identified through MS and validated by RT-PCR, are highlighted with up arrows, while the less abundant proteoforms are indicated with down arrows. Blue arrows represent the proteoforms at the inner-shelf reef Enrique, while red arrows represent those at the mid-shelf reef San Cristobal.
Bar graph showing the percentage of differentially abundant proteoforms categorized by biological function in Acropora palmata at Enrique reef, La Parguera, Puerto Rico. The blue bars represent the Cool Dry Season (CDS, EN), the yellow bars represent the Hot Wet Season (HWS, EN) at the inner-shelf reef Enrique. Green bars represent the Cool Dry Season (CDS, SC), and red bars represent the Hot Wet Season (HWS, SC) at the mid-shelf reef San Cristobal.
Regarding the mid-shelf reef San Cristobal, A. palmata colonies displayed 557 differentially abundant proteoforms, with 449 less abundant proteoforms, and 32 proteoforms common to both seasons. In 2015, 638 differentially abundant proteoforms, 458 less abundant proteoforms, and the 32 common proteoforms were identified (Figures 3 and 5). The differentially abundant proteoforms in A. palmata were classified into 10 categories according to their main biological function associated with temperature stress (Figure 7). While these proteoforms may have multiple functions, our focus lies on their roles within various functions that collectively contribute to the proposed pathway (Figure 6).
Among the more abundant proteoforms, functions associated with stress response (e.g., heat shock transcription factor and glutathione S-transferase), cell signaling (e.g., minicollagen and alpha i G protein), and cell cycle (e.g., calmodulin) were identified. Conversely, the less abundant proteoforms included those related to cytoskeletal binding (e.g., gelsolin and beta-tubulin), UV response (e.g., green fluorescent protein and ATP synthase F0 subunit 6), immunity (e.g., apextrin and ATP synthase subunit 8), transcription factors (e.g., pax C), oxidative phosphorylation (e.g., cytochrome oxidase subunit 1 and cytochrome P450), and apoptosis (e.g., caspase 8).
Regarding less abundant proteoform functions, all are implicated in immunity (e.g., apextrin and ATP synthase subunit 8), while 83% are associated with UV-related responses and oxidative phosphorylation (e.g., green fluorescent protein and cytochrome oxidase subunit 1), cytoskeletal binding, transcription factors, and apoptosis (e.g., actin, pax C, and caspase). Additionally, 75% are linked to these functions (Table 2, Figure 7).
Differentially abundant proteoforms in Acropora palmata. pI/MW1: isoelectric point and molecular mass; P2: number of peptides; MS/MS3: mass spectrometry score; EN STD cd/hw4: standard deviation for the inner-shelf enrique during the cool dry/hot wet seasons; SC STD cd/hw5: standard deviation for the mid-shelf San cristobal during the cool dry/hot wet seasons; EN cd/hw fc6: fold-change for Enrique reef during the cool dry/hot wet seasons, SC cd/hw fc7: fold-change for San cristobal reef during the cool dry/hot wet season; EN cd/hw/SC cd/hw abd8: proteoform abundance. ▲: more abundant proteoforms; ▼: less abundant proteoforms. Symbols in the combined EN cd/hw/SC cd/hw column denote values specific to either the inner-shelf enrique or the mid-shelf san cristobal reefs, as indicated in the corresponding rows
RT-PCR for A. palmata
Among the three reference genes listed in Table 1, ribokinase and calcium-dependent protein kinase were not detected in A. palmata samples (Supplementary Figure S2). Consequently, actin was selected as the reference gene to normalize gene expression. qPCR data for five genes (Pax C, minicollagen, cytochrome P450, acropsin 3, and caspase 8) were transformed and normalized using coefficient of variation (CV) analysis and the pairwise ΔCt method. The relative expression levels (ΔΔCt values) revealed that minicollagen exhibited the highest expression during the Hot Wet Season (HWS) at the inner-shelf reef Enrique. Acropsin and caspase 8 showed lower relative expression, while cytochrome P450 and minicollagen were significantly upregulated during the same period. At the mid-shelf reef San Cristobal, expression patterns shifted: caspase 8 was notably upregulated during HWS, while cytochrome P450 and minicollagen exhibited strong downregulation during the Cool Dry Season (CDS). Pax C showed relatively low and stable expression across both reefs and seasons (Figure 8).
Delta-Delta Cycle threshold (ΔΔCt) values of five candidate genes in Acropora palmata, normalized to actin expression. (A) Inner-shelf reef Enrique. (B) Mid-shelf reef San Cristobal. The bar chart displays relative gene expression changes, where ΔΔCt values above zero indicate higher gene expression (greater abundance), and values below zero indicate lower gene expression (reduced abundance). Ct represents the number of amplification cycles required for the fluorescence signal to reach the detection threshold in qPCR. Vertical error bars indicate the standard deviation of ΔΔCt values. Each bar represents the average ΔΔCt value for each gene across seasons in the respective reef. CDS = Cool Dry Season; HWS = Hot Wet Season.
Discussion
The proteomic response of A. palmata to temperature fluctuations between cool dry and hot wet seasons was significant, with a high percentage of stress-related proteoforms being more abundant. Specifically, 66% of these proteoforms were detected in higher abundance in Enrique and 55% in San Cristobal, two reefs located 3.5 km apart (Figures 1 and 5). Among these, NADH ubiquinone, glutathione S-transferase, cysteine, activin 1, and heat-shock proteins (HSPs) were identified as more abundant, all of which are associated with survival, recovery, and metabolism (Sunagawa et al., Reference Sunagawa, DeSalvo, Voolstra, Reyes-Bermudez and Medina2009; Montilla et al., Reference Montilla, Ramos, García and Cróquer2016; Seveso et al., Reference Seveso, Arrigoni, Montano, Maggioni, Orlandi, Berumen, Galli and Vai2020).
During September and October of 2014 and 2015, sea surface temperatures in La Parguera, Puerto Rico, reached their highest levels for those years. The elevated abundance of these proteoforms aligns with the proteomic profile of an organism experiencing cellular and oxidative stress, likely induced by elevated ambient water temperatures.
Fewer than 33% of the proteoforms of A. palmata from both reefs (Table 2), including beta tubulin, gelsolin, and myosin, exhibited lower abundance compared to the cool dry season. These proteoforms play a crucial role in cellular processes they interact closely with heat-shock proteins and immunophilins, which are essential for cellular repair and protection during thermal stress in corals (Polato et al., Reference Polato, Voolstra, Schnetzer, DeSalvo, Randall, Szmant, Medina and Baums2010; Rosic et al., Reference Rosic, Pernice, Dunn, Dove and Hoegh-Guldberg2010; Tambutté et al., Reference Tambutté, Holcomb, Ferrier-Pagès, Reynaud, Tambutté, Zoccola and Allemand2011; Jin et al., Reference Jin, Lundgren, Lutz, Raina, Howells, Paley, Willis and van Oppen2016). The lower abundance of these proteoforms is associated with the activation of oxidative metabolism and increase in ATP and NADPH production, which has been linked to a reduction of glucose transporters in corals (Jones et al., Reference Jones, Larkum, Schreiber and Hoegh-Guldberg1998; Voolstra et al., Reference Voolstra, Schnetzer, Peshkin, Randall, Szmant and Medina2009; Rosic et al., Reference Rosic, Kaniewska, Chan, Ling, Edwards, Dove and Hoegh-Guldberg2014; Kaniewska et al., Reference Kaniewska, Chan, Kline, Ling, Rosic, Edwards, Hoegh-Guldberg and Dove2015). Additionally, the abundance levels of caspase 8 and ectonucleoside triphosphate diphosphohydrolase, both associated with pro-apoptotic pathways, were reduced in Enrique reef (1.40 and 2.71-fold, respectively). This decrease suggests that A. palmata’s might leverage its capacity to resist heat stress and by minimizing apoptotic signal triggered by thermal stress, thereby adapting to the warming and fluctuating ocean environment (Gibbin et al., Reference Gibbin, Krueger, Putnam, Barott, Bodin, Gates and Meibom2018; Howells et al., Reference Howells, Dunshea, McParland, Vaughan, Heron, Pratchett, Burt and Bauman2018; DeCarlo et al., Reference DeCarlo, Harrison, Gajdzik, Alaguarda, Rodolfo-Metalpa, D’Olivo, Liu, Patalwala and McCulloch2019; Kubicek et al., Reference Kubicek, Breckling, Hoegh-Guldberg and Reuter2019). In contrast, caspase 8 was more abundant in the San Cristobal reef (1.35-fold), suggesting that A. palmata in this location may have already acclimated, as no signs of stress-induced bleaching were observed. Meanwhile, acropsin 3, alpha q G protein, and integrin beta proteoforms, which are associated with photoreception, light-guided behaviours, and immune and metabolic functions (Mason et al., Reference Mason, Schmale, Gibbs, Miller, Wang, Levay, Shestopalov and Slepak2012; Ricaurte et al., Reference Ricaurte, Schizas, Ciborowski and Boukli2016), exhibited only a slight decrease in abundance, with a reduction of 19%. This suggest that these proteoforms may still function as photoreceptors, potentially aiding in light-guided behaviours such as adjusting polyp orientation under light stress (Plaza et al., Reference Plaza, De Jong, Gehring and Miller2003; Roth et al., Reference Roth, Latz, Goericke and Deheyn2010; Harrison, Reference Harrison, Zvy and Noga2011; Hilton et al., Reference Hilton, Brady, Spaho and Vize2012; Hemond et al., Reference Hemond, Kaluziak and Vollmer2014; Iluz & Dubinsky, Reference Iluz and Dubinsky2015; Wuitchik et al., Reference Wuitchik, Wang, Pells, Karimi, Ward and Vize2019). Changes in light conditions are thus managed by these proteoforms, helping corals adapt to light effects (Gates et al., Reference Gates, Baghdasarian and Muscatine1992; Miller et al., Reference Miller, Hemmrich, Ball, Hayward, Khalturin, Funayama, Agata and Bosch2007; DeSalvo et al., Reference DeSalvo, Voolstra, Sunagawa, a, Stillman, a, Szmant a and Medina2008). We suggest that under elevated temperatures, A. palmata may utilize energy conservation as a strategy to sustain essential functions such as calcification and respiration (Jones et al., Reference Jones, Larkum, Schreiber and Hoegh-Guldberg1998; Banaszak & Lesser, Reference Banaszak and Lesser2009; Mayfield et al., Reference Mayfield, Wang, Tang, Fan, Hsiao, Tsai and Chen2011; Bove et al., Reference Bove, Umbanhowar and Castillo2020). Additionally, proteoforms such as apextrin and ATP synthase subunit 8 were downregulated by 12%. In marine organisms, these proteins are vital to specialized physiological functions such as carbon-concentrating mechanisms that enhance photosynthesis in the symbiotic Symbiodiniaceae, which is subsequently transferred to the coral (Muscatine et al., Reference Muscatine, Masuda and Burnap1979; Drake et al., Reference Drake, Mass, Haramaty, Zelzion, Bhattacharya and Falkowski2013; Tresguerres, Reference Tresguerres2016). Notably, apextrin proteins are predominantly found in marine invertebrates (Miller et al., Reference Miller, Hemmrich, Ball, Hayward, Khalturin, Funayama, Agata and Bosch2007). The C-terminal region of apextrins is highly conserved and has been identified as the apextrin C-terminal domain (Li et al., Reference Li, Huang, Ye, Lu, Cheng, Chen and Chen2014). Apextrin has been implicated in embryonic development (Miller et al., Reference Miller, Hemmrich, Ball, Hayward, Khalturin, Funayama, Agata and Bosch2007) and immune defense across cnidarians, including Hydra and reef-building corals. These findings suggest that apextrin proteins could play a critical role in various biological processes of A. palmata, including development and immune response, as observed across different marine invertebrate species (Puill-Stephan et al., Reference Puill-Stephan, Seneca, Miller, van Oppen and Willis2012).
Moreover, proteoforms such as acropsin 3, alpha q G protein and beta integrin were more abundant in the inner-shelf reef Enrique but less abundant in the mid-shelf reef San Cristobal. These proteoforms, known for their roles in photoreception, suggest that since A. palmata colonies in shallow waters, which are exposed to higher levels of UV light compared to other coral species, may require a greater response from chaperone proteins. This includes photoprotective pigments such as green fluorescent proteins (GFP) and Heat Shock Proteins (HSPs) (Salih et al., Reference Salih, Larkum, Cox, Kühl and Hoegh-Guldberg2000; Alieva et al., Reference Alieva, Konzen, Field, Meleshkevitch, Hunt, Beltran-Ramirez, Miller, Wiedenmann, Salih and Matz2008; Scucchia et al., Reference Scucchia, Nativ, Neder, Goodbody-Gringley and Mass2020). GFP and ATP synthase F0 subunit 6 were also less abundant. These proteins are recognized for their diverse functions, mainly playing a role in energy metabolism, and serve as natural sunscreens to absorb and dissipate solar ultraviolet radiation (Mydlarz et al., Reference Mydlarz, McGinty and Harvell2010). These proteins function potentially as a defense mechanism and protect the coral colony against oxidative stress, through their ability to neutralize Reactive Oxygen Species (ROS) (Glynn, Reference Glynn1993; Roth et al., Reference Roth, Goericke and Deheyn2012; Granados-Cifuentes et al., Reference Granados-Cifuentes, Bellantuono, Ridgway, Hoegh-Guldberg and Rodriguez-Lanetty2013). A shared environmental factor driving the stress response of A. palmata across these two coastal reefs was identified as the elevated water temperatures during hot wet season, with increases of 3.0–3.5ºC in September and October in 2014 and 2015 compared to the average temperatures of other months. The proteomic response of A. palmata was characterized by a higher protein abundance of stress markers including heat-shock transcription factor and activin 1 protein, among others. The differences in fold-changes of proteoforms related to stress responses between the two reefs likely reflect A. palmata’s capacity to adapt to changing environments. This adaptability is apparent in Enrique reef, where the increased abundance of HSPs suggests a varying degree of vulnerability in A. palmata species in response to climate change.
Notable examples of differentially abundant proteoforms across both reefs include heat shock protein, H16.2 glutathione S-transferase, NADH oxidoreductase, and cysteine. In Enrique reef acropsin, integrin alpha and pax c were the most abundant and, in San Cristobal caspase 8, was the only stress proteoform more abundant.
The less abundant proteoforms in both reefs included small cysteine-rich protein 6, which is associated with the abundance of HSPs, serving as a molecular protective mechanism against thermal stress (Carballo-Bolaños et al., Reference Carballo-Bolaños, Soto and Chen2019). The lower abundance of this proteoform supports DeSalvo et al.’s (Reference DeSalvo, Voolstra, Sunagawa, a, Stillman, a, Szmant a and Medina2008) findings, which suggests that small cysteine-rich protein 6 is linked to the assembly of HSPs that help induce the expulsion of algal endosymbionts, thus reducing thermal stress on the coral (DeSalvo et al., Reference DeSalvo, Voolstra, Sunagawa, a, Stillman, a, Szmant a and Medina2008; Weston et al., Reference Weston, Dunlap, Beltran, Starcevic, Hranueli, Ward and Long2015). This observation suggests a potential reduction in the coral’s protective mechanism. In addition to cysteine-rich protein 6, other proteoforms with lower abundance included cytochrome oxidase subunit I, cytochrome b, apextrin, ATP synthase F0, and ectonucleoside triphosphate diphosphohydrolase2-like, which are involved in UV-related processes and cytoskeletal binding. Notably, in the Enrique reef, caspase 8 was the only proteoform found to be less abundant. In contrast, in the San Cristobal reef, Pax C and cytochrome P450 were uniquely less abundant compared to Enrique reef, indicating reef-specific differences in protein abundance related to environmental stressors.
We propose a proteomic model with the differentially abundant proteoforms identified in our study. Previous studies have shown that excess ultraviolet light can contribute to temperature increases, leading to damage in cellular components such as DNA, mitochondria, endoplasmic reticulum, ribosomes, and the cell wall, among others (Takahashi & Murata, Reference Takahashi and Murata2008; Naviaux, Reference Naviaux2018). In the proposed model, a positive balance was observed in A. palmata with differentially abundant proteoforms when the temperature increased during the hot wet season. This led to the activation of receptor proteoforms in A. palmata, particularly in the ectoderm and endoderm, such as mini collagen and collagen, which are known for their roles in defense and calcification (Wirshing & Baker, Reference Wirshing and Baker2014). These proteoforms were less abundant, which may indicate that the adhesion processes were maintained during temperature stress promoting the electron transport from the mitochondria to the cytoplasm, for example, the NADPH oxidase activates the cytochrome P450 protein only in colonies of the mid-shelf reef San Cristobal contrary to the colonies in the inner-shelf reef Enrique where they were not abundant. This is an indicator that ROS is creating an imbalance and/or reduction of the defense mechanism and oxidative stress processes in A. palmata (Rosic et al., Reference Rosic, Pernice, Dunn, Dove and Hoegh-Guldberg2010).
Additionally, proteoforms involved in the cell cycle, such as calmodulin, have been shown to play a role in biomineralization (Takeuchi et al., Reference Takeuchi, Yamada, Shinzato, Sawada and Satoh2016; Malik et al., Reference Malik, Einbinder, Martinez, Tchernov, Haviv, Almuly, Zaslansky, Polishchuk, Pokroy, Stolarski and Mass2020). The higher abundance of calmodulin in both reefs suggests that calmodulin’s enhanced ability to bind calcium makes it a crucial regulator in extracellular calcium formation, working in conjunction with acropsin, alpha i G, and pax C proteins (Weston et al., Reference Weston, Dunlap, Beltran, Starcevic, Hranueli, Ward and Long2015). These three last proteoforms were more abundant in colonies of the inner-shelf reef Enrique and less abundant in the mid-shelf reef San Cristobal, suggesting that photoreception mechanisms in reefs closer to the coast are more sensitive to light compared to those in outer reefs. Similarly, the expression of the caspase 8 protein varies between the two reefs, being less abundant in the inner-shelf reef Enrique, and more abundant in San Cristobal, indicating that A. palmata mediates the apoptotic response in the outer reefs, while the inner reef colonies may be inhibiting this apoptotic process. This suggests that A. palmata colonies may acclimate to heat stress by dealing with apoptotic processes, and this acclimation, aligns with the results of our proteomics analysis.
In alignment with these findings, the changes in proteoform abundance identified by the 2D-GE LC-MS/MS analysis in Acropora palmata were further analyzed by assessing the mRNA levels of selected genes. The results revealed that some genes, such as caspase 8 and minicollagen, exhibited expression levels in reef Enrique that were consistent with the abundance trends of their respective proteoforms, while others showed discrepancies. Similarly, in San Cristobal reef, caspase 8, cytochrome P450, and minicollagen displayed abundances aligned with their proteoform trends. These differences likely arise from the multifaceted regulation of gene expression, encompassing transcriptional, post-transcriptional, and translational processes. Figure 6 illustrates these relationships, underscoring the importance of integrating proteomic and gene expression analyses to gain a deeper understanding of A. palmata’s responses to environmental stressors.
Conclusion
This study represents the first investigation of seasonal temperature-driven proteoform profile in A. palmata under natural conditions in Puerto Rico. The combined results suggest that A. palmata activates metabolic pathways and synthesizes high levels of stress-response proteoforms to counteract the adverse effects of elevated temperatures during the hot wet season in southwest Puerto Rico, reflecting an acclimation mechanism to thermal stress. The ability of A. palmata to withstand temperature fluctuations of 3.0–3.5°C, near or beyond its upper thermal tolerance limits (Thomas et al., Reference Thomas, Rose, Bay, López, Morikawa, Ruiz-Jones and Palumbi2018; Klepac & Barshis, Reference Klepac and Barshis2020; Takeuchi et al., Reference Takeuchi, Mizuguchi, Ishibashi, Takayama and Yamashiro2023) indicates not only the health of the colonies but also their capacity to adapt to challenging environmental conditions. This adaptability is further supported by proteomic data, which revealed an increase in more abundant proteins as the dominant trend in protein abundance, highlighting the role of proteoforms related to energy metabolism and stress responses in optimizing photosynthetic efficiency and facilitating acclimation to seasonal temperature variations. Our findings demonstrate the effectiveness of proteomic approaches in identifying early-response proteoform markers and molecular processes involved in A. palmata’s acclimation, laying the foundation for future studies on other marine organisms. These results indicate that A. palmata modulates its proteoform profiles in response to seasonal temperature fluctuations, suggesting active acclimation mechanisms.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0025315425100052.
Acknowledgements
The authors would like to thank Mónica Fernández De Soto from the Genomic Laboratory at North Carolina State University, and Andrés Salcedo from the Department of Entomology and Plant Pathology at North Carolina State University, Raleigh, USA, for their technical assistance with gene expression analysis. Special thanks to Raymond U. Infante Rosa for his help with statistical analysis using R. Funding for the proteomics analyses was provided by the Seed Money from the College of Arts & Sciences, UPRM, and Sea Grant Puerto Rico Grant FRS-535265.
Author contributions
Conceptualization, M.R., N.V.S., E.F.W., and N.M.B.; methodology, M.R. and P.C.; investigation, M.R.; resources, N.V.S. and N.M.B.; writing – original draft preparation, M.R., N.V.S., and N.M.B.; writing – review and editing, N.M.B.; supervision, N.M.B. and N.V.S. All authors have read and agreed to the published version of the manuscript.
Funding
This research was funded by the Seed Money, College of Arts & Sciences, University of Puerto Rico at Mayaguez; the Sea Grant Puerto Rico Grant FRS-535265; the Central Caribbean University (UCC) Biomedical Proteomics Facility, and the US Department of Education: Title V Grant Award # P031S200104 (Nawal M. Boukli).
Competing interest
The author(s) declare no competing interests.
Data availability statement
Any data from this manuscript can be requested and is available upon request to CCD.
