Innate immune function in the diamondback terrapin (Malaclemys terrapin)

The innate immune system serves as a critical first line of defense in reptiles, particularly under the energetic constraints of ectothermy. This study characterizes complement-mediated innate immune responses in the diamondback terrapin (Malaclemys terrapin), an estuarine turtle species facing increasing environmental and disease pressures. We evaluated the antibacterial capacity of terrapin plasma against Aeromonas hydrophila, Vibrio vulnificus, Staphylococcus aureus, and Streptococcus pyogenes and assessed complement activity through sheep red blood cell (SRBC) hemolysis assays. Plasma showed strong, concentration-dependent antibacterial activity, with complete inhibition of S. aureus at 100% plasma and substantial inhibition of the other species at ≥25%. Hemolytic activity was rapid, occurring primarily within 10 min, and peaked at 25°C–30°C. Complement activity was abolished by heat, protease, and EDTA treatments and restored with divalent cations, indicating a heat-labile, protease-sensitive mechanism consistent with the alternative complement pathway. Mannan affinity chromatography and SDS-PAGE revealed two major lectins (~25.8 and ~33.4 kDa), consistent with known complement-associated lectins in other emydid turtles. These findings demonstrate that diamondback terrapins possess robust, complement-mediated innate immune responses, potentially shaped by evolutionary exposure to estuarine pathogens like Vibrio spp., and support their use in comparative eco-immunological research and conservation health assessments.

The vertebrate immune system comprises innate and acquired components that function synergistically to recognize, contain, and eliminate pathogens. Innate immunity represents the first line of defense and provides rapid, non-specific protection through a variety of physical, cellular, and humoral mechanisms, including phagocytic cells, antimicrobial peptides, lysozyme, and the complement system (Janeway et al., 2001; Zimmerman et al., 2010). Among these, the complement system is a critical effector of innate immune function, promoting pathogen lysis, opsonization, inflammation, and immune complex clearance (Killick et al., 2018; Rios and Zimmerman, 2015).

Complement proteins are activated via three major pathways: the classical pathway, triggered by antigen–antibody complexes; the lectin pathway, initiated by mannose-binding lectins recognizing microbial carbohydrates; and the alternative pathway, which is constitutively active and amplified in the presence of antigenic pathogen surfaces (Law and Dodds, 1997; Petersen et al., 2001). While complement activity has been well-characterized in certain reptilian taxa such as crocodilians and select chelonians (Merchant et al., 2005; Baker and Merchant, 2018), data remain limited for many freshwater turtles, particularly within the family Emydidae.

Diamondback terrapins (Malaclemys terrapin) are estuarine turtles endemic to the Atlantic and Gulf of Mexico coasts of North America, occupying brackish environments where they encounter a wide array of microbial threats and environmental stressors, including salinity fluctuations, pollution, and increasing anthropogenic habitat disturbance (Butler et al., 2006; Harden et al., 2009). Populations are in decline across much of their range due to habitat loss, road mortality, and incidental capture in commercial fishing gear (Seigel and Gibbons, 1995; Gibbons et al., 2001). Despite their conservation concern and exposure to unique environmental pressures, little is known about the immune defenses of this species, including the role of plasma antibacterial and complement-mediated responses.

Studies in other emydid turtles, including box turtles (Terrapene spp.), have demonstrated robust plasma antibacterial activity that is temperature- and concentration-dependent and likely mediated by the complement system (Adamovicz et al., 2020a; Zimmerman et al., 2010). Mechanistic investigations have revealed variation in complement activation pathways even among closely related species, with ornate box turtles (Terrapene ornata) utilizing the alternative pathway, while eastern box turtles (Terrapene carolina) appear to rely on the lectin pathway (Adamovicz et al., 2020a, 2020b). Whether such diversity in complement activation exists in other emydids, such as the diamondback terrapin, remains unknown.

This study focuses on characterizing the innate immune response of the diamondback terrapin. We quantified plasma antibacterial capacity using live bacterial cultures and evaluated complement-mediated activity using a sheep red blood cell hemolysis assay. We hypothesized that terrapin plasma would exhibit concentration- and temperature-dependent antibacterial activity and complement-mediated lysis of red blood cells, consistent with patterns observed in other chelonians. Elucidating the mechanisms of innate immunity in this species may contribute to our understanding of disease susceptibility and inform conservation and management strategies for this imperiled estuarine turtle.

Blood collection

We captured turtles using non-baited fyke nets set in suitable estuarine habitat at Texas Point National Wildlife Refuge during the months of March–July in 2023. Water temperatures varied from lows of 17.4°C in March to highs of 35.6°C in July. Salinities ranged from 10 and 26 ppt during the months of capture.

Traps were checked every 24 h, and whole blood was collected from the subcarapacial sinus using a 22-gauge needle and a 3-mL syringe and immediately transferred to a 3-mL heparinized tube. We centrifuged the blood at 2,500× g for 10 min, and plasma was collected and stored at −20°C until further use. We pooled plasma from 37 individual turtles to use in all experiments. The sample contained plasma from 20 females (138–202 mm carapace length) and 17 males (108–136 mm carapace length). We used pooled plasma in order to provide results that represent an average immune response for this population. These reference values can be used in future studies to compare the magnitude of immune response between different populations, sexes, stage classes, or individuals of different disease states.

All work involving live animals was conducted with approval from the McNeese State University IACUC (protocol #11142022).

Antibacterial assays

We purchased nutrient broth from ISC Bioexpress (Kaysville, UT). Lyophilized bacterial strains were purchased from the American Type Culture Collection (ATCC; Washington, DC). The following ATCC-registered strains were used: Aeromonas hydrophila (35654), Vibrio vulnificus (27562), Staphylococcus aureus (6538), and Streptococcus pyrogenes (19615). We maintained bacteria on nutrient agar slants at 5°C. The day before the experiment, we inoculated a nutrient broth liquid culture from the slant using a sterile cotton swab and incubated it at 37°C overnight. To determine the efficiency of diamondback terrapin plasma at killing the four tested bacterial species, we combined 200 µL of plasma with 400 µL of inoculated broth and transferred 150 µL of bacterial culture to a 96-well plate and measured the optical density at 430 nm using a Benchmark Plus microtiter plate reader (Bio-Rad, Hercules, CA, USA). We mixed plasma with sterile saline and tested six dilutions (0%, 10%, 25%, 50%, 75%, and 100% plasma) at six different time points—0, 3, 6, 9, 12, and 24 h—after mixing. All assays were run in sextuplicate.

Serum complement activity assays

We measured serum complement activity of the pooled diamondback terrapin plasma using sheep red blood cell (SRBC) hemolysis assays. We exposed pooled plasma samples to 1% SRBCs (v/v) to determine the hemolytic capacity of plasma components. Volume-dependent assays were conducted by incubating 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 100% plasma, diluted with unbuffered saline, with SRBCs at ambient temperature (25°C) for 30 min. The samples were centrifuged at 2,500× g and 150 μL of the resulting supernatants were transferred to a 96-well plate. We determined the optical density of each sample at 540 nm using a Benchmark Plus (Bio-Rad, Hercules, CA, USA) microtiter plate reader. All assays were conducted in quadruplicate.

We determined the kinetic characteristics of SRBC hemolysis for diamondback terrapin plasma by incubation of 3.0 mL of plasma, 2.0 mL of unbuffered saline, and 5.0 mL of 1% SRBCs. Plasma dilution was chosen based on the results of the volume-dependent assays described above. Aliquots (250 μL) were removed at different time points (0, 2, 5, 10, 15, 20, 30, 60 min). We centrifuged the samples at 2,500× g and 150 μL of the resulting supernatants were transferred to a 96-well plate. We determined the optical density of each sample at 540 nm using a Benchmark Plus (Bio-Rad, Hercules, CA, USA) microtiter plate reader. All assays were conducted in quadruplicate.

To determine the effects of temperature on the hemolytic activity, we acclimated plasma samples diluted with unbuffered saline at 5° increments from 5°C to 40°C for 10 min. We then exposed the plasma to SRBCs for 30 min and centrifuged at 2,500× g. The supernatants were transferred to a 96-well plate, and the OD 540 was measured as above. To determine the effects of other protease inhibitors, plasma was treated with 0.005 volumes of 1 M of EDTA or 0.02 volumes of 1 M of methylamine or salicylaldoxime. These solutions were exposed to 1% SRBCs (v/v) for 30 min at ambient temperature and then centrifuged at 2,500× g. The resulting supernatants were transferred to a 96-well plate, and the optical densities were determined as described above. Other similar treatments included 5 mM of EDTA plus the addition of 20 mM of CaCl 2 or MgCl 2.

Mannan affinity chromatography

To look for the presence of lectins that would activate the lectin-mediated immune pathway, we used mannan affinity chromatography to selectively isolate lectin proteins from diamondback terrapin plasma. We equilibrated a mannose-agarose suspension (3 mL) with 3 mL of mixing buffer (10 mM of Tris–HCl, pH 7.8, 1.25 M of NaCl) and allowed it to settle into a column prepared from a Pasteur pipette plugged with cotton. Five milliliters of AST or CST plasma was mixed with 5 mL of loading buffer (20 mM of Tris–HCl, pH 7.8, 2.5 M of NaCl). The diluted plasma was allowed to filter through the column and the column was washed with 10 volumes of loading buffer. We eluted the proteins using 5 mL of elution buffer (10 mM of Tris–HCl, pH 7.8, 1.25 M of NaCl, 2 mM of EDTA), and the resulting eluates were transferred to 1.0-kDa centrifugal concentrator tubes (Centriprep, Millipore Corp., Billerica, MA, USA) and centrifuged at 7,500× g to concentrate. When the samples reached minimal volume, they were desalted by the addition of 10 mL of distilled water and centrifuged at 7,500× g for 2 h.

SDS-PAGE analysis —The desalted eluates (20 μL) were mixed with 20 μL of SDS sample loading buffer (Bio-Rad, Hercules, CA, USA), incubated at 95°C for 5 min, cooled on ice for 5 min, and then centrifuged at 16,000× g for 30 s. We loaded samples (35 μL) and 4 μL of PageRuler Plus™ prestained protein markers (Fisher Scientific, Chicago, IL, USA) onto 4%–20% polyacrylamide precast gradient gels (Bio-Rad, Hercules, CA, USA) and resolved at 20 V/cm for 90 min. The gels were stained for 10 min with Coomassie blue R-250 and destained in a solution of 45% H 2 0, 45% MeOH, and 10% glacial AcOH (v/v) for 3 h.

Mechanistic studies

To investigate the mechanism for hemolysis of SRBCs by diamondback terrapin plasma, we exposed plasma to known protein inhibitors before repeating the SRBC hemolysis assays. We treated 300 μL of plasma with each of the following: 2 units of protease derived from Streptomyces griseus (Sigma Aldrich, St. Louis, MO, USA), heated to 56°C in a water bath for 30 min, 5 mM of EDTA and 5 mM of EDTA plus either 20 mM of Ca 2+ or 20 mM of Mg 2+ to probe the role of divalent metal ions, and 20 mM of methylamine or 20 mM of salicylaldoxime (Sigma-Aldrich, St. Louis, MO, USA) to probe the role of the alternative mechanism in SRCB hemolysis. Treated samples were incubated with 2% SRBCs for 30 min at ambient temperature. The samples were centrifuged at 2,500× g and 150 μL of the resulting supernatants were transferred to a 96-well plate. We determined the optical density of each sample at 540 nm. All assays were conducted in quadruplicate.

Statistics and controls

All experiments were conducted in quadruplicate to enable a valid statistical evaluation of the results. The results are presented as the percentage of maximum hemolysis as compared to a positive control (1% SRBCs) that was completely lysed with detergent and represent the means ± standard deviation for four independent determinations. The results were subjected to one-way analysis of variance (ANOVA) using Scheffe’s post hoc comparisons. All statistics were generated using SPSS version 31.0.0.

We found that exposure to diamondback terrapin plasma in concentrations greater than 25% substantially inhibited bacterial growth for all species except S. aureus, which was fully inhibited at 100% plasma but showed a more measured concentration-dependent profile (Figure 1). Kinetic profiles were similar for A. hydrophila, V. vulnificus, and S. pyogenes and showed some bacterial growth between 3 and 12 h following exposure to plasma, but very little additional growth after 12 h (Figure 1). Staphylococcus aureus again showed a different pattern with growth at all plasma concentrations except 100% increasing between 6 and 9 h following exposure and continuing to grow until the 24-h mark (Figure 1).

A one-way ANOVA was used to assess the concentration-dependent hemolysis of SRBCs by serum complement. Plasma volume had a highly significant effect on percent maximum hemolysis (F 10, 22 = 1,683.1, p < 0.0001). Scheffé’s post hoc comparisons showed that hemolysis at 0%–20% complement did not differ significantly, indicating minimal complement-mediated lysis at low concentrations. Hemolysis increased significantly beginning at 30% plasma (p < 0.05 vs. 0%–20%), with progressively higher lysis at 40%–70%. Concentrations from 90% to 100% did not differ significantly from each other, consistent with near saturation of complement activity. Overall, these results indicate a strongly concentration-dependent complement response, with a marked transition between 20% and 40% complement and near-maximal hemolytic activity above 70%.

For each complement concentration, a one-way ANOVA was used to test the effect of incubation time (0–60 min) on percent maximum hemolysis of sheep red blood cells (n = 4 wells per time point). The relationship between terrapin plasma dilution and SRBC hemolysis showed very little hemolytic activity between 10% and 30% plasma volume, but a sharp increase from 40% to 70% and a slower increase between 80% and 100% plasma (Figure 2). Time had a highly significant effect at 40% (F 7, 24 = 686.3, p < 0.0001), 50% (F 7, 24 = 2,860.6, p < 0.0001), and 60% complement (F 7, 24 = 6,037.2, p < 0.0001). Scheffé’s post hoc comparisons indicated that at 40% and 50% complement, hemolysis at 10–60 min was significantly greater than at 0–5 min, whereas at 60% complement, significant increases were already evident by 5 min, with progressive increases at all later time points (p < 0.05 for all significant comparisons). The kinetic curve for SRBC hemolysis shows that most hemolysis occurs within 10 min of exposure of SRBCs to terrapin plasma. Rapid hemolytic activity in all plasma concentrations tested occurred between 0 and 10 min, followed by a slightly increasing plateau between 20 and 60 min (Figure 3).

A one-way ANOVA was used to examine the effect of temperature (5°C–40°C) on the ability of turtle plasma to lyse SRBCs. Temperature had a highly significant effect on percent maximum hemolysis (F 7, 16 = 664.19, p < 0.0001). Scheffé’s post hoc tests indicated that hemolysis at 5°C and 10°C did not differ significantly and remained near zero. Hemolytic activity increased significantly at 15°C (p = 0.022) and rose sharply between 20°C (p < 0.0001) and 30°C (p < 0.0001), where peak hemolysis occurred. Hemolysis at 25°C and 30°C did not differ significantly from each other (p = 0.999) but was significantly higher than at all lower temperatures. Above 30°C, hemolysis declined significantly; values at 35°C and 40°C were significantly lower than those at 25°C–30°C (p < 0.0001). Overall, turtle complement activity exhibited a strong temperature dependence, with maximal activity centered near 25°C–30°C (Figure 4).

We found that the hemolytic activity of plasma from diamondback terrapins is heat-labile and sensitive to treatment with a generic protease. When we treated plasma with mild heat (56°C for 30 min), it reduced SRBC hemolysis from over 98% to less than 1% (p < 0.0001, Table 1). Similarly, when we treated plasma with protease (p < 0.0001) or 5 mM of EDTA (p < 0.0001), hemolytic activity was almost completely inhibited. We could restore hemolytic activity by the addition of Mg 2+ (p = 0.0026) or Ca 2+ (p = 0.0449). However, treatment of plasma with 20 mM of methylamine (p = 0.093) or salicylaldoxime (p = 0.008) resulted in little or no effect on SRBC hemolysis of plasma derived from terrapins (Table 1).

Mannan affinity chromatography followed by SDS-PAGE analysis of the eluate showed that diamondback terrapins express high levels of two proteins that interact with mannan, labeled as lectin 1 and lectin 2, with approximate M r values of 25.8 and 33.4 kDa (Figure 5). As this method isolates only lectin proteins, the two very strong bands seen on the gel must be lectins, and they match the M r range of mannan-binding lectins seen in other turtle species (Baker et al., 2019). Other bands are most likely multimers of lectin 1 and lectin 2 based on their size, as we have also described in previous studies. Also present is a small minor band that has not yet been characterized (Figure 5).

In reptiles, including turtles, the innate immune system plays a particularly vital role due to the energetic constraints associated with ectothermy and limited metabolic capacity. Unlike endotherms, reptiles have lower and more variable metabolic rates, which constrain the energy available for sustaining prolonged or adaptive immune responses (Zimmerman et al., 2010). As a result, reptiles rely more heavily on fast-acting, energetically efficient innate defenses—such as phagocytosis, antimicrobial peptides, and the complement system—to combat pathogens (Merchant et al., 2003; Zimmerman et al., 2013). This reliance is further emphasized during periods of environmental stress, when energy must be prioritized for thermoregulation or reproduction, potentially at the expense of slower adaptive immunity (Demas et al., 2011).

Diamondback terrapins are unique among turtles because they prefer brackish water habitats. Thus, we tested bacterial species that were more likely to be important in these types of habitats. We were particularly interested in the response of the terrapin immune system to bacteria of the genus Vibrio, as these bacteria pose significant health risks. Opportunistic Vibrio species such as V. vulnificus and V. parahaemolyticus are known to cause systemic infections in marine and estuarine organisms and have been isolated from both healthy and diseased turtles (Ebani 2023; Zavala-Norzagaray et al., 2015). These bacteria can exploit immunocompromised hosts or injuries, leading to septicemia, shell lesions, and high mortality, especially under elevated water temperatures that promote Vibrio proliferation (Ebani, 2023) and compromise terrapin complement-mediated immunity (Figure 4). As climate change continues to increase global sea surface temperatures, the geographic range and virulence of Vibrio spp. are expected to expand, potentially heightening disease pressure on vulnerable turtle populations (Vezzulli et al., 2016). Our results showed that terrapin plasma performs very well when challenged with V. vulnificus, possibly as a result of their long co-evolutionary history, as seen in sea turtles (Nash and Ryan, 2023). Future studies could be done to confirm this interpretation.

Diamondback terrapin plasma showed a robust response to challenge with SRBCs, which are a useful surrogate for pathogens that can be encountered in their environment. Direct comparison of the immune function of different species can be difficult, as not all studies use the same methodology, but general comparisons can be made. Hemolysis of SRBCs by diamondback terrapin plasma has a similar temperature profile to that seen in eastern and ornate box turtles (Adamovicz et al., 2020a) and common and alligator snapping turtles (Baker et al., 2019). Plasma from diamondback terrapins performs faster than that of alligator snapping turtles or eastern box turtles, but is similar to that of common snapping turtles and ornate box turtles (Baker et al., 2019; Adamovicz et al., 2020a).

Vertebrates produce lectins that recognize and bind mannan oligosaccharides that are commonly present on the external surfaces of bacteria, activating the lectin-mediated complement pathway (Kilpatrick, 2002; Drommett et al., 2006). The binding of these lectins to bacterial surface mannan residues activates mannan-binding lectin-associated serine proteases (MASPs), which then cleave complement C4 and C2 to generate a C3 convertase capable of activating C3 (Matsushita et al., 2013). The lectins we identified from diamondback terrapin plasma correspond very closely to those identified using MALDI-TOF analysis in alligator snapping turtles (31.0 and 35.8 kDa; Baker et al., 2019) and eastern box turtles (25.6 and 34.7 kDa; Adamovicz et al., 2020a). This makes sense given the shared evolutionary history of turtles. Diamondback terrapins and eastern box turtles are both in the Emydidae family, and they share more similarity in their lectins than they do with snapping turtles, which are in the family Chelydridae. The presence of these calcium-dependent mannan-binding proteins in the plasma of terrapins strongly suggests the presence of an active lectin-mediated complement mechanism.

The observed sensitivity of SRBC hemolysis to mild heat treatment (Table 1) is an indication that the activity is due to the presence of serum complement proteins. Complement proteins are known to be particularly heat-labile and can be inactivated at moderately low temperatures (56°C; Jasin, 1972). These results also indicate that the hemolysis is not due to the presence of antimicrobial peptides, as these molecules are extremely heat-resistant and require much higher temperatures for inactivation (Andaluz-Mejia et al., 2022). In addition, proteins such as lysozyme and disulfide-stabilized phospholipase A2 that exhibit immunological properties do not exhibit such low thermal instabilities (Venkataramani et al., 2013; Singh and Chang, 2004).

Serum complement activity requires Mg 2+ and Ca 2+ (Mayer et al., 1946). Our experiments with hemolytic inhibitors indicate that the hemolytic activity observed in diamondback terrapin plasma is partially the result of the alternative mechanism of complement activation, which is inhibited by EDTA because Mg 2+ ions are required for the interaction of C3b with factor B (Gotze et al., 1977). We found that hemolytic activity was not inhibited by low concentrations of methylamine, which is a known inhibitor of the classical complement cascade (Gosrski and Howard, 1980; Sim and Sim, 1981). This provides evidence that the classical pathway, which involves antibody interactions with bacterial surface antigens, is not involved in the hemolysis of SRBCs. However, it is likely that the plasma of terrapins does not contain antibodies to SRBC antigens, and thus, this is not surprising. However, it is highly likely that terrapin plasma contains antibodies directed toward antigens expressed by V. vulnificus and A. hydrophila, and thus, the antibacterial activities toward these bacterial species were partially due to classical complement activity. This could potentially provide an explanation of why the antibacterial activities against V. vulnificus and A. hydrophila, bacterial species with which terrapins have coexisted for millions of years, are stronger than those against S. aureus and S. pyogenes, which are primarily human pathogens to which wild terrapins probably have not had exposure. The hemolysis of SRBCs was inhibited by salicyladoxime, which acts as a nucleophile that inactivates the internal thioester of the C3b and C4b proteins required for the alternative mechanism of complement activation (Dodds and Sim, 1984). These biochemical inhibitors provide further evidence for the involvement of the alternative pathway of complement activation in the observed antibacterial activities (Figure 1). Furthermore, the fact that the addition of EDTA did not completely eliminate complement activity and the addition of Ca 2+ did not completely restore all of the complement activity (Table 1) indicates that other mechanism(s), such as lectin-mediated complement activity, may be at play.

Our findings substantiate that diamondback terrapin plasma exhibits potent, rapid, and concentration-dependent antibacterial and hemolytic activity, mediated by heat-labile, protease-sensitive complement mechanisms with optimal function in mid-temperature ranges. These traits echo immune patterns across chelonian taxa, reinforcing the complement system’s pivotal role in reptilian innate defense and laying groundwork for future eco-immunological and conservation applications.