2020 journal article

Caspases from scleractinian coral show unique regulatory features

Journal of Biological Chemistry, 295(43), 14578–14591.

By: S. Shrestha*, J. Tung*, R. Grinshpon n, P. Swartz n, P. Hamilton n, B. Dimos*, L. Mydlarz*, A. Clark*

co-author countries: United States of America 🇺🇸
author keywords: caspase; cysteine protease; allosteric regulation; substrate specificity; coral apoptosis; coral immunity; functional divergence; substrate selection; CARD-caspase; apoptosis
MeSH headings : Amino Acid Sequence; Animals; Anthozoa / chemistry; Anthozoa / cytology; Anthozoa / enzymology; Anthozoa / metabolism; Apoptosis; Caspases / chemistry; Caspases / metabolism; Coral Reefs; Crystallography, X-Ray; Enzyme Activation; Humans; Models, Molecular; Protein Conformation; Protein Domains; Sequence Alignment; Substrate Specificity
Source: ORCID
Added: January 14, 2021

Coral reefs are experiencing precipitous declines around the globe with coral diseases and temperature-induced bleaching being primary drivers of these declines. Regulation of apoptotic cell death is an important component in the coral stress response. Although cnidaria are known to contain complex apoptotic signaling pathways, similar to those in vertebrates, the mechanisms leading to cell death are largely unexplored. We identified and characterized two caspases each from Orbicella faveolata, a disease-sensitive reef-building coral, and Porites astreoides, a disease-resistant reef-building coral. The caspases are predicted homologs of the human executioner caspases-3 and -7, but OfCasp3a (Orbicella faveolata caspase-3a) and PaCasp7a (Porites astreoides caspase-7a), which we show to be DXXDases, contain an N-terminal caspase activation/recruitment domain (CARD) similar to human initiator/inflammatory caspases. OfCasp3b (Orbicella faveolata caspase-3b) and PaCasp3 (Porites astreoides caspase-3), which we show to be VXXDases, have short pro-domains, like human executioner caspases. Our biochemical analyses suggest a mechanism in coral which differs from that of humans, where the CARD-containing DXXDase is activated on death platforms but the protease does not directly activate the VXXDase. The first X-ray crystal structure of a coral caspase, of PaCasp7a determined at 1.57 Å resolution, reveals a conserved fold and an N-terminal peptide bound near the active site that may serve as a regulatory exosite. The binding pocket has been observed in initiator caspases of other species. These results suggest mechanisms for the evolution of substrate selection while maintaining common activation mechanisms of CARD-mediated dimerization. Coral reefs are experiencing precipitous declines around the globe with coral diseases and temperature-induced bleaching being primary drivers of these declines. Regulation of apoptotic cell death is an important component in the coral stress response. Although cnidaria are known to contain complex apoptotic signaling pathways, similar to those in vertebrates, the mechanisms leading to cell death are largely unexplored. We identified and characterized two caspases each from Orbicella faveolata, a disease-sensitive reef-building coral, and Porites astreoides, a disease-resistant reef-building coral. The caspases are predicted homologs of the human executioner caspases-3 and -7, but OfCasp3a (Orbicella faveolata caspase-3a) and PaCasp7a (Porites astreoides caspase-7a), which we show to be DXXDases, contain an N-terminal caspase activation/recruitment domain (CARD) similar to human initiator/inflammatory caspases. OfCasp3b (Orbicella faveolata caspase-3b) and PaCasp3 (Porites astreoides caspase-3), which we show to be VXXDases, have short pro-domains, like human executioner caspases. Our biochemical analyses suggest a mechanism in coral which differs from that of humans, where the CARD-containing DXXDase is activated on death platforms but the protease does not directly activate the VXXDase. The first X-ray crystal structure of a coral caspase, of PaCasp7a determined at 1.57 Å resolution, reveals a conserved fold and an N-terminal peptide bound near the active site that may serve as a regulatory exosite. The binding pocket has been observed in initiator caspases of other species. These results suggest mechanisms for the evolution of substrate selection while maintaining common activation mechanisms of CARD-mediated dimerization. Apoptotic cell death is thought to be a unique characteristic of metazoans, although its evolutionary origins are unclear. Although caspases from human cells and model organisms such as Caenorhabditis elegans and Drosophila have been well-studied both biochemically and structurally (1Song Z. McCall K. Steller H. DCP-1, a Drosophila cell death protease essential for development.Science. 1997; 275 (8999799): 536-54010.1126/science.275.5299.536Crossref PubMed Scopus (251) Google Scholar, 2Dorstyn L. Mills K. Lazebnik Y. Kumar S. The two cytochrome c species, DC3 and DC4, are not required for caspase activation and apoptosis in Drosophila cells.J. Cell Biol. 2004; 167 (15533997): 405-41010.1083/jcb.200408054Crossref PubMed Scopus (99) Google Scholar, 3Ellis H.M. Horvitz H.R. Genetic control of programmed cell death in the nematode C. elegans.Cell. 1986; 44 (3955651): 817-82910.1016/0092-8674(86)90004-8Abstract Full Text PDF PubMed Scopus (1353) Google Scholar, 4Schwartz H.T. Horvitz H.R. The C. elegans protein CEH-30 protects male-specific neurons from apoptosis independently of the Bcl-2 homolog CED-9.Genes Dev. 2007; 21 (18056428): 3181-319410.1101/gad.1607007Crossref PubMed Scopus (58) Google Scholar, 5Shaham S. Horvitz H.R. Developing Caenorhabditis elegans neurons may contain both cell-death protective and killer activities.Genes Dev. 1996; 10 (8598288): 578-59110.1101/gad.10.5.578Crossref PubMed Scopus (203) Google Scholar, 6Yuan J. Horvitz H.R. The Caenorhabditis elegans genes ced-3 and ced-4 act cell autonomously to cause programmed cell death.Dev. Biol. 1990; 138 (2307287): 33-4110.1016/0012-1606(90)90174-HCrossref PubMed Scopus (444) Google Scholar), little is known about caspase activity and regulation in other basal species (7Ramirez M.L.G. Salvesen G.S. A primer on caspase mechanisms.Semin. Cell Dev. Biol. 2018; 82 (29329946): 79-8510.1016/j.semcdb.2018.01.002Crossref PubMed Scopus (80) Google Scholar). C. elegans (3Ellis H.M. Horvitz H.R. Genetic control of programmed cell death in the nematode C. elegans.Cell. 1986; 44 (3955651): 817-82910.1016/0092-8674(86)90004-8Abstract Full Text PDF PubMed Scopus (1353) Google Scholar, 6Yuan J. Horvitz H.R. The Caenorhabditis elegans genes ced-3 and ced-4 act cell autonomously to cause programmed cell death.Dev. Biol. 1990; 138 (2307287): 33-4110.1016/0012-1606(90)90174-HCrossref PubMed Scopus (444) Google Scholar) and Drosophila (8Mydlarz L.D. Fuess L.E. Mann W. Pinzón J.H. Gochfeld D.J. Cnidarian immunity: From genomes to phenomes.in: Goffredo S. Dubinsky Z. The Cnidaria, Past, Present and Future. Springer International Publishing, New York2016: 441-466Crossref Scopus (25) Google Scholar) were some of the first invertebrate caspases to be characterized, but they have proved to be poor models for studying the evolution of the vertebrate apoptotic network as their networks utilize fewer caspases and regulatory proteins compared with higher eukaryotes. C. elegans, for example, utilizes only one effector caspase (CED-3), which also bears a CARD-motif necessary for its activation (9Irmler M. Hofmann K. Vaux D. Tschopp J. Direct physical interaction between the Caenorhabditis elegans “death proteins” CED-3 and CED-4.FEBS Lett. 1997; 406 (9109415): 189-19010.1016/S0014-5793(97)00271-8Crossref PubMed Scopus (72) Google Scholar). Moreover, cytochrome c is not involved in the formation of the apoptosome in Drosophila, indicating that this organism lacks the intrinsic pathway found in humans (2Dorstyn L. Mills K. Lazebnik Y. Kumar S. The two cytochrome c species, DC3 and DC4, are not required for caspase activation and apoptosis in Drosophila cells.J. Cell Biol. 2004; 167 (15533997): 405-41010.1083/jcb.200408054Crossref PubMed Scopus (99) Google Scholar). In contrast, it now appears that vertebrates have retained many characteristics of the apoptotic machinery found in sponges, sea anemone, and coral (10Tchernov D. Kvitt H. Haramaty L. Bibby T.S. Gorbunov M.Y. Rosenfeld H. Falkowski P.G. Apoptosis and the selective survival of host animals following thermal bleaching in zooxanthellate corals.Proc. Natl. Acad. Sci. U. S. A. 2011; 108 (21636790): 9905-990910.1073/pnas.1106924108Crossref PubMed Scopus (126) Google Scholar, 11Wiens M. Krasko A. Perovic S. Müller W.E.G. Caspase-mediated apoptosis in sponges: Cloning and function of the phylogenetic oldest apoptotic proteases from Metazoa.Biochim. Biophys. Acta. 2003; 1593 (12581862): 179-18910.1016/s0167-4889(02)00388-9Crossref PubMed Scopus (58) Google Scholar, 12Furla P. Sabourault C. Zucchini N. Allemand D. Courtiade J. Richier S. Oxidative stress and apoptotic events during thermal stress in the symbiotic sea anemone, Anemonia viridis.FEBS J. 2006; 273 (16907933): 4186-419810.1111/j.1742-4658.2006.05414.xCrossref PubMed Scopus (93) Google Scholar). Genomic studies of cnidarians, the sister group to the bilateria, revealed many genes that were previously thought to have been vertebrate innovations, demonstrating that the extensive gene loss in C. elegans and in Drosophila resulted in apoptotic pathways that do not reflect the characteristics of ancestral metazoans (13Salvesen G.S. Walsh C.M. Functions of caspase 8: The identified and the mysterious.Semin. Immunol. 2014; 26 (24856110): 246-25210.1016/j.smim.2014.03.005Crossref PubMed Scopus (98) Google Scholar, 14Kortschak R.D. Samuel G. Saint R. Miller D.J. EST analysis of the cnidarian Acropora millepora reveals extensive gene loss and rapid sequence divergence in the model invertebrates.Curr. Biol. 2003; 13 (14680636): 2190-219510.1016/j.cub.2003.11.030Abstract Full Text Full Text PDF PubMed Scopus (286) Google Scholar). Basal metazoans, which appear to have a full complement of apoptotic signaling molecules, may therefore be more relevant to the evolutionary pathways of vertebrate apoptotic networks. Cnidarians including reef-building corals from the genus Scleractinia are ecologically important organisms that are on the decline (15Hughes T.P. Barnes M.L. Bellwood D.R. Cinner J.E. Cumming G.S. Jackson J.B.C. Kleypas J. Van De Leemput I.A. Lough J.M. Morrison T.H. Palumbi S.R. Van Nes E.H. Scheffer M. Coral reefs in the Anthropocene.Nature. 2017; 546 (28569801): 82-9010.1038/nature22901Crossref PubMed Scopus (942) Google Scholar) and cell death has been indicated to be important in these processes (16Dunn S.R. Schnitzler C.E. Weis V.M. Apoptosis and autophagy as mechanisms of dinoflagellate symbiont release during cnidarian bleaching: Every which way you lose.Proc. R. Soc. B Biol. Sci. 2007; 274 (17925275): 3079-308510.1098/rspb.2007.0711Crossref PubMed Scopus (128) Google Scholar). The two primary drivers of coral declines are marine diseases affecting reef-building corals (17Maynard J. Van Hooidonk R. Eakin C.M. Puotinen M. Garren M. Williams G. Heron S.F. Lamb J. Weil E. Willis B. Harvell C.D. Projections of climate conditions that increase coral disease susceptibility and pathogen abundance and virulence.Nature Clim. Change. 2015; 5: 688-69410.1038/nclimate2625Crossref Scopus (167) Google Scholar) as well as temperature-induced loss of the coral's symbiont known as bleaching (15Hughes T.P. Barnes M.L. Bellwood D.R. Cinner J.E. Cumming G.S. Jackson J.B.C. Kleypas J. Van De Leemput I.A. Lough J.M. Morrison T.H. Palumbi S.R. Van Nes E.H. Scheffer M. Coral reefs in the Anthropocene.Nature. 2017; 546 (28569801): 82-9010.1038/nature22901Crossref PubMed Scopus (942) Google Scholar). Coral possess an innate immune system that both defends the animals against pathogenic organisms and also serves as general stress responses (18Mansfield K.M. Gilmore T.D. Innate immunity and cnidarian-Symbiodiniaceae mutualism.Dev. Comp. Immunol. 2019; 90 (30268783): 199-20910.1016/j.dci.2018.09.020Crossref PubMed Scopus (32) Google Scholar). Therefore, the coral immune system is critical in the response of these organisms to both coral diseases and bleaching. Activation of the innate immune system activates apoptotic pathways (19Fuess L.E. Pinzón C J.H. Weil E. Grinshpon R.D. Mydlarz L.D. Life or death: Disease-tolerant coral species activate autophagy following immune challenge.Proc. R. Soc. Biol. Sci. 2017; 284 (28592676): 2017077110.1098/rspb.2017.0771Crossref PubMed Scopus (45) Google Scholar); however, to date very few functional studies have been performed to characterize caspase structure and subsequent function in corals (20Palmer C.V. Traylor-Knowles N. Towards an integrated network of coral immune mechanisms.Proc. R. Soc. B Biol. Sci. 2012; 279 (22896649): 4106-411410.1098/rspb.2012.1477Crossref PubMed Scopus (88) Google Scholar). There are several examples pointing to the importance of apoptotic pathways and caspases in coral survival to both disease and temperature stress. An increase in expression of apoptosis-related genes was detected in a diseased Caribbean soft coral resulting in a visible inflammatory response (black-melanized appearance) (21Fuess L.E. Mann W.T. Jinks L.R. Brinkhuis V. Mydlarz L.D. Transcriptional analyses provide new insight into the late-stage immune response of a diseased Caribbean coral.R. Soc. Open Sci. 2018; 5 (29892394): 17206210.1098/rsos.172062Crossref PubMed Scopus (21) Google Scholar). Also, executioner caspase genes were up-regulated in the branching coral Acropora infected with white band disease (22Libro S. Kaluziak S.T. Vollmer S.V. RNA-seq profiles of immune related genes in the staghorn coral Acropora cervicornis infected with white band disease.PLoS One. 2013; 8 (24278460): e8182110.1371/journal.pone.0081821Crossref PubMed Scopus (89) Google Scholar). Several studies have gleaned important insights into coral apoptosis post-temperature stress by demonstrating that corals activate cell death responses following expulsion of their algal symbiont (19Fuess L.E. Pinzón C J.H. Weil E. Grinshpon R.D. Mydlarz L.D. Life or death: Disease-tolerant coral species activate autophagy following immune challenge.Proc. R. Soc. Biol. Sci. 2017; 284 (28592676): 2017077110.1098/rspb.2017.0771Crossref PubMed Scopus (45) Google Scholar, 23Kaniewska P. Campbell P.R. Kline D.I. Rodriguez-Lanetty M. Miller D.J. Dove S. Hoegh-Guldberg O. Major cellular and physiological impacts of ocean acidification on a reef building coral.PLoS One. 2012; 7 (22509341): e3465910.1371/journal.pone.0034659Crossref PubMed Scopus (183) Google Scholar, 24Dunn S.R. Thomason J.C. Le Tissier M.D.A. Bythell J.C. Heat stress induces different forms of cell death in sea anemones and their endosymbiotic algae depending on temperature and duration.Cell Death Differ. 2004; 11 (15286684): 1213-122210.1038/sj.cdd.4401484Crossref PubMed Scopus (136) Google Scholar, 25Dunn S.R. Bythell J.C. Le Tissier M.D. Burnett W.J. Thomason J.C. Programmed cell death and cell necrosis activity during hyperthermic stress-induced bleaching of the symbiotic sea anemone Aiptasia sp.J. Exp. Mar. Biol. Ecol. 2002; 272: 29-5310.1016/S0022-0981(02)00036-9Crossref Scopus (129) Google Scholar). Specifically, the anti-apoptotic protein Bcl-2 in Acropora millepora is up-regulated during temperature stress (26Pernice M. Dunn S.R. Miard T. Dufour S. Dove S. Hoegh-Guldberg O. Regulation of apoptotic mediators reveals dynamic responses to thermal stress in the reef building coral Acropora millepora.PLoS One. 2011; 6 (21283671): e1609510.1371/journal.pone.0016095Crossref PubMed Scopus (64) Google Scholar), indicating that this species likely has an intrinsic apoptosis mechanism as well as mechanisms to regulate this process. Interestingly, it was shown that application of caspase inhibitors can prevent the death of bleached coral (10Tchernov D. Kvitt H. Haramaty L. Bibby T.S. Gorbunov M.Y. Rosenfeld H. Falkowski P.G. Apoptosis and the selective survival of host animals following thermal bleaching in zooxanthellate corals.Proc. Natl. Acad. Sci. U. S. A. 2011; 108 (21636790): 9905-990910.1073/pnas.1106924108Crossref PubMed Scopus (126) Google Scholar). Collectively, the data show the potential for complex apoptotic signaling pathways in coral but data on activation and control mechanisms, and how they compare with those in vertebrates, are lacking because of a dearth of biochemical characterization. To gain insight into caspase activity and regulation in coral, we expressed and characterized two caspases each from two species of Caribbean reef-building corals, Orbicella faveolata and Porites astreoides. The two coral species are found on opposite ends of the stress-tolerance spectrum where the disease-susceptible O. faveolata activates caspase-mediated apoptotic pathways upon immune challenge, whereas the disease-tolerant P. astreoides activates an adaptive autophagic response (19Fuess L.E. Pinzón C J.H. Weil E. Grinshpon R.D. Mydlarz L.D. Life or death: Disease-tolerant coral species activate autophagy following immune challenge.Proc. R. Soc. Biol. Sci. 2017; 284 (28592676): 2017077110.1098/rspb.2017.0771Crossref PubMed Scopus (45) Google Scholar). These findings indicate that understanding the apoptotic machinery in corals likely has significant implication in understanding species stress tolerance. In this investigation we describe the structural composition of each species' caspase repertoire, and we use these data to functionally characterize both initiator and effector caspases from both species. Two proteins referred to as PaCasp7a and OfCasp3a based on sequences similarity to human caspases contain CARD motifs at the N terminus, an unusual combination that has not been observed in caspases-3 or -7 enzymes from higher eukaryotes, and indeed these proteins function as initiator caspases. Additionally, two proteins PaCasp3 and OfCasp3b show canonical caspase-3/7 structural organization, with short pro-domains, and possess effector caspase function. We describe the first biochemical characterization of coral caspases and show that the PaCasp3 and OfCasp3b enzymes are not activated directly by the CARD-containing PaCasp7a and OfCasp3a, respectively. We also report the first X-ray crystal structure of a coral caspase, that of PaCasp7a determined at 1.57 Å resolution, which reveals an N-terminal peptide bound near the active site that may serve as a regulatory exosite. Overall, we find support for complex apoptotic mechanisms in these early metazoans, where the cellular machinery for both intrinsic and extrinsic apoptosis has ancient evolutionary origins. We examined seven caspase genes from O. faveolata based on sequences obtained from previous transcriptomic and genomic data (Fig. S1 and Table S1) (19Fuess L.E. Pinzón C J.H. Weil E. Grinshpon R.D. Mydlarz L.D. Life or death: Disease-tolerant coral species activate autophagy following immune challenge.Proc. R. Soc. Biol. Sci. 2017; 284 (28592676): 2017077110.1098/rspb.2017.0771Crossref PubMed Scopus (45) Google Scholar). The caspases were named based on the E-value from BLAST as well as the sequence similarity to the human orthologs. Results from examining the sequence homology and domain organization suggest that three of the caspases are apoptotic initiators and four are apoptotic effectors in O. faveolata (Fig. 1A). The sequence identities of the seven caspases compared with most human caspases are low, only ∼35% (Table 1), so it is difficult to determine the nature of each coral caspase based solely on sequence comparisons with human orthologs. In addition, two caspases from O. faveolata contain an N-terminal caspase activation and recruitment domain (CARD) motif, similar to those in HsCasp2 and HsCasp9, and one caspase contains tandem death effector domain (DED) motifs, similar to that found in HsCasp8 (Fig. 1A). The remaining four proteins show domain organization similar to the human effector caspases, with short pro-domains (Fig. 1A).Table 1Protein sequence identity/similarity (%) with human caspasesHsCasp3HsCasp7HsCasp6HsCasp2HsCasp8HsCasp10HsCasp9OfCasp737/5438/5232/4928/4334/5033/5334/50OfCasp3c36/5835/5635/5232/4837/5337/5433/49OfCasp3b35/6032/5733/5232/4937/5533/5334/50OfCasp3a47/6945/6538/5429/4839/5439/5528/46OfCasp237/5341/5535/4633/5234/5235/5232/48OfCasp8a39/5639/5334/4835/5332/5830/4934/51OfCasp8b33/5631/5031/4932/5235/5134/5232/46PaCasp336/5837/5936/5633/4937/5434/5435/50PaCasp7a43/6544/6036/5328/4637/5337/5328/44PaCasp7b38/5437/5234/4929/4631/4830/5033/48PaCasp239/5338/5233/4833/5035/5335/5232/49 Open table in a new tab In the case of P. astreoides, four caspase sequences consisted of two initiator-like caspases (called PaCasp7a and PaCasp2) and two effector-like caspases (called PaCasp7b and PaCasp3) (Fig. 1A and Fig. S1). Similar to the results for O. faveolata, the caspase sequences from P. astreoides also have only ∼35% identity with human caspases, regardless of comparisons to initiator or effector caspases (Table 1). The sequences from the two coral species displayed much higher identity to putative homologs in the other coral species. For example, PaCasp7a has a 77% sequence identity with OfCasp3a, whereas PaCasp3 has 71 and 73% sequence identity, respectively, with OfCasp3b and OfCasp3c. Likewise, PaCasp2 demonstrates 76% sequence identity with OfCasp2, and PaCasp7b shares 60% identity with OfCasp7 (Fig. 1B). A phylogenetic analysis of cnidarian and vertebrate caspases demonstrated that cnidarian caspases cluster in separate groups (Fig. 2A). All of the short pro-domain caspases, including PaCasp3 and OfCasp3b, cluster together between vertebrate effector (caspases-3/7) and initiator (caspases-8/10) caspases. Interestingly, the comparative genomics and phylogenetic analyses suggest that short cnidarian caspases, that is, those lacking a CARD or DED, share a common ancestor with vertebrate effector caspases-3 and -7 and with initiator caspases-8 and -10 (Fig. 2A). Homologs of caspase-8 in coral share the same clade with vertebrate caspases-8 and -10 and the CARD-containing OfCasp2 and PaCasp2 clustered with vertebrate caspase-2. With the exceptions of OfCasp2 and PaCasp2, the other CARD-containing coral caspases cluster with OfCasp3a and PaCasp7a and segregate into a different clade, although they share a common ancestor with vertebrate caspases-2 and -9. We analyzed the CARD motifs of cnidarian caspases independently of the protease domains and compared them to the CARD motifs of vertebrate caspases-2 and -9 as well as that of caspase-2 and RIPK1 domain containing adaptor with death domain (CRADD) motifs, which recruit caspase-2 to the PIDDosome (27Park H.H. Structural features of caspase-activating complexes.Int. J. Mol. Sci. 2012; 13 (22606010): 4807-481810.3390/ijms13044807Crossref PubMed Scopus (57) Google Scholar) (Fig. 2B). The CARD motifs of coral caspases-3 and -7 cluster together but are more closely related to the CARD of caspase-2 than those of caspase-9 or CRADD. Based on this analysis, there appear to be many CARD-containing caspase-3–like proteins in cnidaria. At present, it is not clear why CARD-containing caspase-3–like proteins provide an advantage for coral development and/or symbiosis because the animals also contain initiator caspases that presumably activate the short pro-domain effector caspases. CARD-containing caspase-3–like proteins are rarely observed in vertebrate effector caspases. Fish-specific caspases have been found, such as the CARD-containing caspase-8 for example (28Sakamaki K. Satou Y. Caspases: Evolutionary aspects of their functions in vertebrates.J. Fish Biol. 2009; 74 (20735596): 727-75310.1111/j.1095-8649.2009.02184.xCrossref PubMed Scopus (78) Google Scholar), but caspase-2 is, at present, the only characterized DXXDase with a CARD. We chose two caspases from each species to characterize further, based on the sequence comparisons with human effector caspases-3, -6, or -7. In the case of O. faveolata, we chose two caspase-3–like proteins that showed 47 and 35% sequence identity, respectively, with HsCasp3, and we named the two proteins OfCasp3a and OfCasp3b, respectively (Fig. 1A and Table 1). Interestingly, despite predicted similarity to HsCasp3, OfCasp3a also has an N-terminal CARD motif. One caspase from P. astreoides demonstrated the highest sequence identity with HsCasp7 (44%) and was named PaCasp7a, even though it also contains a CARD motif (Fig. 1A and Table 1). The second protein from P. astreoides showed similar sequence identity to human caspases-3, -6, -7, and -8 (36–37%) (Fig. 1A and Table 1), but the protein does not have a DED motif like caspase-8 and the domain organization is more similar to that of caspase-3. Consequently, we named the protein PaCasp3. Overall, the low sequence identity between the vertebrate and invertebrate caspases show that the classification is somewhat arbitrary without further biochemical characterizations of the proteins. Together, the phylogenetic analysis shows that the caspases from P. astreoides and O. faveolata have relatively low sequence identity (∼40%) to mammalian caspases as well as other vertebrate families, but the proteins had much higher sequence identities to caspases from other cnidarian species, such as Pocillopora damicornis, Stylophora pistillata, and Nematostella vectensis. An analysis of the coral caspase sequences shows that the proteins contain all of the conserved features that define a caspase. For example, each protein contains the catalytic dyad, histidine (CP-075) and cysteine (CP-117) (Fig. 3), where “CP” refers to the common position defined previously for caspases (29Grinshpon R.D. Williford A. Titus-McQuillan J. Clay Clark A. The CaspBase: A curated database for evolutionary biochemical studies of caspase functional divergence and ancestral sequence inference.Protein Sci. 2018; 27 (30076665): 1857-187010.1002/pro.3494Crossref PubMed Scopus (10) Google Scholar). The conserved sequence that contains the catalytic histidine (CP-115)-QACRG-(CP-119) is found in the four coral caspases, although PaCasp7a and OfCasp3a contain QACQG as in human caspase-8. One of the most highly variable regions, the intersubunit linker (IL) is the same length in OfCasp3b and PaCasp3 compared with that of HsCasp3, whereas those of PaCasp7a and OfCasp3a have one and two amino acids fewer than HsCasp3, respectively (Fig. 3). We examined the four coral caspases by size exclusion chromatography because CARD-containing human caspases are monomers or mixtures of weak protomer-dimer (30Clark A.C. Caspase allostery and conformational selection.Chem. Rev. 2016; 116 (26750439): 6666-670610.1021/acs.chemrev.5b00540Crossref PubMed Scopus (44) Google Scholar). Because the IL of the procaspase monomer is cleaved during activation, the protomer is defined as a single unit that contains a large and small subunit and a single active site. Thus, the dimer consists of two protomers, or is more formally considered a dimer of heterodimers. The data show that the CARD containing coral caspases, PaCasp7a and OfCasp3a, elute in a single peak with molecular mass of 42.6 and 44 kDa, respectively. The sizes are larger than that of a protomer but smaller than a dimer (Fig. S2 and Table S5), suggesting that the proteins form weak dimers similar to the human initiator caspases. In contrast, the short pro-domain containing caspases, PaCasp3 and OfCasp3b, are dimers similar to the human effector caspases, with molecular mass of 64.5 and 69.2 kDa, respectively (Fig. S3 and Table S5). We also determined the mass of the large and small subunits by MS. Caspase zymogens are cleaved in the IL, and the N-terminal CARD or pro-domain is removed during activation (30Clark A.C. Caspase allostery and conformational selection.Chem. Rev. 2016; 116 (26750439): 6666-670610.1021/acs.chemrev.5b00540Crossref PubMed Scopus (44) Google Scholar). The proteins also autoprocess during overexpression in Escherichia coli. The molecular size of the large and small subunits of each caspase, determined by MS, are shown in Table S5. When compared with the sequences for each protein (Fig. 3), the data show that OfCasp3a and PaCasp7a are cleaved in the intersubunit linker after (CP-127)-DVTD-(CP-130), whereas OfCasp3b and PaCasp3 are cleaved after (CP-127)-VESD-(CP-130). The actual amino acid positions, in addition to the common position number, are shown in Fig. 1A, and the cleavage sites are indicated by the arrow in Fig. 3. In addition, the first 20 or 31 amino acids, respectively, in the pro-domains of OfCasp3b and PaCasp3 are removed following cleavage after VIGD (Asp20) (OfCasp3b) or SSTD (Asp31) (PaCasp3). The CARD motifs of OfCasp3a and of PaCasp7a are removed following cleavage after DEAD (Asp123) and DQAD (Asp119), respectively (Figs. 1A and 3). We note that there are potentially other cleavage sites in the CARD motifs, but in our assays the CARD motif was completely removed. We characterized the substrate specificity for each of the four coral caspases using substrate-phage display assays, as described previously (31Tucker M.B. MacKenzie S.H. Maciag J.J. Dirscherl Ackerman H. Swartz P. Yoder J.A. Hamilton P.T. Clay Clark A. Phage display and structural studies reveal plasticity in substrate specificity of caspase-3a from zebrafish.Protein Sci. 2016; 25 (27577093): 2076-208810.1002/pro.3032Crossref PubMed Scopus (10) Google Scholar). In these assays, we utilize two substrate-phage libraries that determine the P5-P1′ substrate preferences, with either aspartate fixed at the P1 position (P5-XXXXDX-P1′) or random (called 6×), and the results were the same for both libraries. The data show that PaCasp7a and OfCasp3a have Group II specificity, with a preference for aspartate in the P4 position (DXXDase) (Fig. 4, A and B). In contrast, PaCasp3 and OfCasp3b prefer valine in the P4 position (VXXDase) (Fig. 4, C and D), which is defined as Group III specificity like HsCasp6. The activities of PaCasp7a and of OfCasp3a were also examined using DEVD-AFC and VEID-AFC substrates. In all cases, however, the activity against the tetrapeptide substrates was very low because of Km values >500 μm, so we could not reliably determine the steady-state catalytic parameters kcat or Km from the small peptide activity assays. In caspases, the Km is thought to correlate with substrate binding (KD), so the high Km suggests poor binding of the small peptide. Because of the low activity in small peptide assays, we tested the coral caspases for their ability to hydrolyze full-length human procaspases-3 and -6, which were made catalytically inactive