1. Introduction
Phospholipase A2 (PLA2) is a ubiquitous intracellular enzyme that functions in the excision of fatty acids from the sn-2 position of structural membrane lipids. This enzyme plays an important role in the degradation and metabolism of fatty acids. Another significant function of this enzyme activity is to supply arachidonic acid, which is stored in the intracellular side of membrane phospholipids, for the synthesis of eicosanoids. However, a soluble, circulating serum form of this enzyme (sPLA2) has been identified [1] , which is thought to impart extensive immune function [2] [3] [4] . A role for sPLA2 has been implicated in innate immunity [2] and soluble PLA2 has been identified as a potent antibacterial component of tears in mammalian systems [3] [5] . The sPLA2 in the peripheral blood is thought to cleave fatty acids from the membranes of microbes, thus compromising pathogen membrane integrity.
There are currently 23 extant members of the family Crocodylia. Four genera (Alligator, Caiman, Paleosuchus, and Melanosuchus), including eight species, comprise the Alligatoridae. The Crocodylidae are represented by three genera (Crocodylus, Osteoleamus, and Mecistops), which include 13 species. A third clade, Gavialidae, has two monophyletic members (Gavialis gangeticus and Tomistoma schlegelii). Phylogenetic divergence of these taxa is evident in molecular data (Dessauer et al., 2002) [6] and morphological data [7] . Temporally, the Alligatoridae is thought to have diverged from Crocodylidae, approximately 80 million years ago [8] [9] . Recent studies in our laboratory showed that the antibacterial properties of serum of the 23 members of Crocodylia exhibited distinctive differences among broad phylogenetic lineages (Merchant et al., 2006) [10] .
This study was conducted to compare the differences in PLA2 activities of all 23 extant crocodilian species against nine species of bacteria, with the hypothesis that PLA2 activity is more similar among more closely related taxa. It should be noted that when this study was conducted, Crocodylus suchus had not been split from Crocodylus nitolticus, and thus the results of this study do not reflect this relatively new development in crocodilian species descriptions.
2. Materials and Methods
Chemicals and biochemicals―4,4-difluoro-5,7-dimethyl-4-bora-3a,4a-diaza-sinda- cene-3-hexadecanoic acid (BODIPY™ FL C16) was purchased from Invitrogen (Carlsbad, CA, USA). Calcium chloride and trizma HCl were purchased from Sigma Chemical Co. (St. Louis, MO, USA).
Treatment of animals?Blood was collected from the spinal vein [11] [12] using six-mL syringes and 3.81 cm 21 ga needles. All animals were feeding normally and were apparently healthy upon inspection. The samples were allowed to clot overnight at ambient temperature. The serum was separated and stored at −20˚C in poly propylene tubes until ready for use. The PLA2 activities are stable for at least three months at −20˚C when stored in a non frost-free freezer (data not shown).
Bacterial cultures―One mL cultures of each bacterial species were grown overnight at 37˚C in nutrient broth. Each culture was used to inoculate 100 mL of sterile nutrient broth. These cultures were incubated for 24 h in the presence of 100 μg of BODIPY™ FL C16 (dissolved in 100 μL of DMSO). The bacteria were centrifuged at 8000 ×g for 15 min at ambient temperature, the cultures were decanted, and the bacteria were resuspended in 10 mL of assay buffer (1 mM Ca2+, 100 mM tris-HCl, pH 7.4).
PLA2 assay―The serum from each crocodilian species was combined such that a single value for each species could be obtained. However, prior to combination, the activity of each individual was determined to ensure that fluctuations in individual PLA2 activities were not radically altering the average for a species. Fifty μL of serum from each crocodilian species was incubated with 600 μL of assay buffer and 100 μL of each bacterial species (BODIPY-labeled) for 60 min at ambient temperature. The reactions were centrifuged at 16,000 ×g for 5 min and the clear supernatants were removed to one mL plastic cuvettes. The fluorescent intensity of each reaction supernatant was measured at an excitation λ of 500 nm and an emission λ of 510 nm (excitation and emission slits = 1 nm) in a Horiba Jobin Yvon Fluoromax™-4 fluorimeter. The PLA2 activities of each crocodilian species were measured using a single bacterial preparation for each microbial species so that the relative activities were directly comparable without standardization. Previous results from our laboratory have shown that the production of fluorescent product is asymptotic with respect to time when 50 μL of serum is used in 750 μL total assay volume [13] .
Statistics and controls―Each sample was analyzed in at least duplicate. The result from each crocodilian species’ activity against each bacterial species was compared to all others using Pearson’s correlation, thus generating a similarity index for each comparison [14] . In addition, each crocodilian genus was compared to all others using a Pearson correlation. The immune function of the Alligatoroidae, Crocodylidae and Gavialidae were compared via ANOVA using Duncan’s post hoc comparisons to obtain the statistical level of significance for each comparison [15] .
3. Results
Analysis of the PLA2 innate immunological data collected from each crocodilian species indicated the similarities between members of the three extant clades of crocodilians (Figure 1). PLA2 activity towards different bacteria species differed among crocodilians at the family level (Alligatorids, Crocodylids, and Gavialids). Based on the PLA2 activities of the sera of each crocodilian species, Duncan’s multiple range comparisons revealed a statistical difference between the Alligatoroidae and Crocodylidae (p < 0.001). Likewise, the Gavialidae were also distinguishable from the Alligatoridae (p < 0.01). However, there were no statistically discernable differences between the Gavialidae and Crocodylidae (p > 0.05). The relationships based on these PLA2 activities reflect similar associations observed when innate immune activity was used as an indicator (Merchant et al., 2006) [10] , and are very similar to the relations noted by other investigators using genetic similarity matrices [16] [17] and albumin protein sequence analyses [18] .
Figure 1. Pearson’s correlation comparing the phospholipase A2 activity of all eight genera of extant crocodylians. These results highlight the similarities of PLA2 activities of the alligatorids, the differences between the alligatoroids and crocodylids, and the strong similarities of the gavialids.
PLA2 activities were very similar among genera of the Alligatoroidae. The only aberrant association among this lineage was the low correlation of PLA2 activities between Melanosuchus and Caiman (p = 61.3). In addition, member of the genus Crocodylus shared similar PLA2 activities with the Mecistops cataphractus, and Gavialis gangeticus, and moderately high similarity with Tomistoma shlegelii. Of interest was low similarity in PLA2 activity between the dwarf crocodile (Osteolaemus) and Crocodylids. The PLA2 activities of Osteolaemus were very divergent from nearly every other crocodilian species, with few exceptions (Table 1, Figure 1 and Figure 2), which is very similar to results previously reported when antibacterial studies of all crocodilian species were compared [10] . This is a result that was not predicted and mimicked by its sister species, Mecistops cataphractus. Additionally, serum PLA2 activity of Tomistoma showed a much higher correlation with Gavialis and two Caiman than with Crocodylidae. Comparisons in PLA2 activities of all 23 extant crocodilian species are displayed in Figure 2.
4. Discussion
Several recent studies have highlighted various components of the innate immune systems of crocodilians. For instance, serum complement activities have been described for the American alligator [19] , the freshwater and saltwater crocodiles [20] , the broad-snouted caiman [21] , and the American crocodile [22] . In addition, dipeptidyl peptidase IV activity has been characterized in the American alligator [23] and the American crocodile [24] . Furthermore, several crocodilian species have been shown to express serum PLA2 activities (Merchant
Figure 2. Pearson’s correlation comparing the phospholipase A2 activity of all 23 species of extant crocodylians toward 10 species of diverse bacteria. The correlations highlight the relations of phospholipase A2 activity specificities toward different bacterial species between the individual species of crocodylians.
et al., 2008, Merchant et al., 2009c, Nevalainen et al., 2009) [25] [26] [27] [28] . Merchant et al. [10] showed that the antibacterial activities of all 23 species of crocodilian correlated with molecular and morphological hypotheses of crocodilian diversification. In this study, we have employed the PLA2 assay to determine the sn-2 fatty acid excision activities of all extant crocodilian species against nine different bacterial species.
The results from our analyses indicate that the three families of crocodilians are distinguishable by their PLA2 activity profiles (Figure 1). In general, the PLA2 activities among members of the Family Crocodylidae were more disparate compared to those within the Alligatoridae and Gavialidae (Figure 1). This was also noted when serum complement immune activities were compared by Merchant et al. [10] . This observation, potentially, is an artifact of relatively high species richness and biogeographic breadth within Crocodylidae compared to Alligatoridae and Gavialidae, allowing or necessitating greater diversification in enzymatic activity. The members of the Family Crocodylidae are more diverse, being comprised of 14 species compared to eight Alligatoridae and two Gavialidae. The Crocodylidae are also found in a much wider geographical distribution, living on five continents, compared to three for the Alligatoridae and two for the Gavialidae), occupy more types of habitats (broad range of salinities, environments, etc.), and thus are potentially exposed to a broader spectrum of microbial
Table 1. PLA2 activities of serum from all 23 species of extant crocodilian species were measured against nine different bacterial species. A = Escherichia coliform, B = Providencia stuartii, C = Staphylococcus aureus, D = Streptococcus pyrogens, E = Streptococcus faecalis, F = Shigella flexneri, G = Salmonella typhimurium, H = Enterobacter cloacae, I = Klebsiella oxytoca. The PLA2 activities are expressed as 0 - 106 = +, 1 × 106 - 2 × 106 = ++, 2 × 106 - 3 × 106 = +++, 3 × 106 - 4 × 106 = ++++, 4 × 106 - 5 × 106 = +++++, and 5 × 106 - 6 × 106 = ++++++.
Table 2. Amino acid sequence identities between representatives of the three families of crocodylians, Alligatoridae, Crocodylidae, and Gavialidae.
flora. Therefore, the contrasting range of PLA2 activities in the Crocodylidae, compared to the Alligatoridae and Gavialidae, is not surprising.
The PLA2 activities are more similar among species within the same family than between species in different families (Table 1, Figure 2). The same general trend is true at the level of genus aside from two notable aberrations: Melanosuchus and Oesteolamus. Interestingly, Osteolaemus exhibited PLA2 activities that were more similar to those of many of the Alligatoridae than the Crocodylidae (Table 1, Figure 2). Likewise, although Mecistops PLA2 activities correlate strongly with C. rhombifer, C. siamensis, and C. novaguinae, they also exhibit high similarity with members of the Family Alligatoridae (Ca. yacare, latirostris, and crocodilus, and P. palpebrosus and A. sinensis, Table 1, Figure 2). Additionally, and with regards to the similarities between PLA2 activity presented here and existing phylogenies, PLA2 activity in Tomistoma more closely resembles Gavilais than it does any genus within Crocodylidae. This finding is consistent with morphological [7] and molecular [29] phylogenetic hypotheses recovering a sister relationship between Tomistoma and Gavialis. However, the molecular phylogeny of Brochu and Densmore [30] did not recover this sister relationship. Therefore, the PLA2 activity presented here could be homologous between Tomistoma and Gavialis. In addition, this conclusion supports the findings of phylogenetic linkages of antibacterial immunological activities [10] and amino acid sequences of, and immunoreactivity to, specific proteins [18] . Furthermore, the PLA2 activities of C. niloticus are also interesting from an evolutionary perspective. One may hypothesize that PLA2 activity in C. niloticus would resemble its sister clade (New World Crocodylus) [16] [29] . However, the PLA2 activities of C. niloticus are very unique among most members of its genus, and do not resemble neither New World nor Old World Crocodylus.
The positive relationship between species-specific PLA2 activities and the evolutionary relationships among those taxa indicates potential for immunological homology. Such an interpretation should be approached with caution, given the complex ability of the immune system to acclimatize via exposure, and other ecological influences on activity that mask relevant relatedness by descent of immunological traits. In defense of this notion, PLA2 acts as an antimicrobial component in innate immunity, defined solely by the transcription of coding genes and limiting the ecological modification of activity from exposure history. Applicable is the work of Nakashima et al. [31] , whose analysis recovered rapid evolution in the nucleotide sequence of protein-coding regions of PLA2 isozyme genes between the venom glands of two closely related viper species (Trimeresurus). Therefore, the taxonomic resemblance of PLA2 activity is likely more evident of immunological homology than the general serum antimicrobial activity among members of Crocodylia [10] . Nevertheless, the PLA2 activity documented here reflects phylogenetic lineage relatedness to a great degree, indicating the potential for lineage- specific conservation in PLA2 function based on coded structure. The amino acid sequences of the sPLA2 for representatives (American alligator, estuarine crocodile, and gharial) of the three crocodylian Families (Alligatoridae, Crocodylidae, and Gavialidae), are quite divergent compared to other protein sequences that we have analyzed between these groups (Table 2). The sequence identities of 86.2 to 91.9% are dissimilar enough to provide different PLA2 activity profiles toward the phospholipids of different bacterial species. In comparison, these same crocodilian species share 95.5% - 97.0% amino acid identity in their serum complement C3 proteins (Merchant et al., 2016), and 96.5% - 97.8% amino acid identity in nuclear factor kB transcription factor (Merchant, unpublished data) are much higher than in the more divergent PLA2 proteins. The diversity in sPLA2 amino acid sequences between these crocodilians may reflect plasticity in the evolution of genes that code for proteins with important roles in immunological defenses. The optimization of immunological traits on existing phylogenies to explore immunological character evolution may be a worthy endeavor.
5. Conclusion
In conclusion, PLA2 activity among extant crocodilians shows high taxonomic similarity based on Pearson correlation indices. Such results are likely indicative of relation by descent in the genetic underpinning of enzymatic operation. Additionally, regardless of phylogenetic hypotheses, PLA2 activity appears to show a fair amount of convergence and independent evolution that makes for an interesting exercise in character evolution on proposed phylogenetic trees. More work need to be conducted concerning lineage-dependent shifts in PLA2 activity. This would elucidate the “aberrations” found here and help decipher whether mutation or selective processes may account for activities that diverge from phylogenetic hypotheses.
Acknowledgements
The authors wish to acknowledge the assistance of John Brueggen, of the St. Augustine Alligator Farm Zoological Park in St. Augustine, FL, USA, for the collection of blood from captive crocodilians. We would also like to thank Dr. Kent Vliet, of the Department of Zoology at University of Florida, for help in collecting blood samples. We also thank Mr. Gordon Henley, of the Ellen Trout Zoo in Lufkin, TX, USA, for access to animals. This work was supported by a Mc- Neese State University College of Science Endowed Professorship grant awarded to Mark Merchant. All of the work conducted in this study was approved by the McNeese State University Animal Care and Use Committee.