• View in gallery
    Figure 1—

    Photographs of results of western blot analysis (A) and mean values of the densitometric analysis (arbitrary units) of western blot data (B) representing degrees of expression of various proteins involved in COX signaling pathway in noninflamed (day 0; D0) and inflamed (day 7; D7) skin tissue specimens from 6 ball pythons (Python regius). A—Numbers 1 through 6 represent individual snakes. B—Densitometric readings were normalized to those of actin. Error bars represent 95% confidence intervals. *Value differs significantly (P ≤ 0.05) between inflamed and noninflamed tissues by paired t test. Weak expression of COX-2 was detected with both types of antibodies (N-terminal [N-20] and C-terminal [C-20]) in skin tissue, and no NF-κB expression was detected in any tissue specimen. Six specimens (1/snake) were assessed for all proteins except COX-2, for which only 5 specimens were available because of artifact smearing on the western blot membrane for 1 specimen.

  • View in gallery
    Figure 2—

    Photographs of results of western blot analysis (A) and mean values of densitometric analysis (arbitrary units) of western blot data (B) representing degrees of expression of various proteins involved in the COX signaling pathway in noninflamed (day 0; D0) and inflamed (day 7; D7) muscle tissue specimens from 6 ball pythons. Actin was used as a control protein. Weak expression of COX-2 was detected when C-terminal COX-2 antibody was used, although better expression was detected when N-terminal COX-2 antibody was used. Again, no NF-κB expression was detected in any tissue specimen. See Figure 1 for remainder of key.

  • View in gallery
    Figure 3—

    Photographs of results of western blot analysis of skin (A) and muscle (B) specimens from the snakes in Figures 1 and 2 after application of blocking peptide (BP) specifically designed to prevent binding of N-terminal COX-2 antibody to COX-2 protein. Cell lysates from human squamous cell carcinoma cells (hSCC-25) and canine transitional cell carcinoma cells (K9TCCLilly) were used as positive control samples for COX-2 expression.

  • 1. Read MR. Evaluation of the use of anesthesia and analgesia in reptiles. J Am Vet Med Assoc 2004; 224: 547552.

  • 2. Olesen MG, Bertelsen MF, Perry SF, et al. Effects of preoperative administration of butorphanol or meloxicam on physiologic responses to surgery in ball pythons. J Am Vet Med Assoc 2008; 233: 18831888.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3. Vane JR, Botting RM. Mechanism of action of nonsteroidal antiinflammatory drugs. Am J Med 1998; 104: 2S8S.

  • 4. Eberhart CE, Coffey RJ, Radhika A, et al. Up-regulation of cyclooxygenase 2 gene expression in human colorectal adenomas and adenocarcinomas. Gastroenterology 1994; 107: 11831188.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 5. Dannenberg AJ, Lippman SM, Mann JR, et al. Cyclooxygenase-2 and epidermal growth factor receptor: pharmacologic targets for chemoprevention. J Clin Oncol 2005; 23: 254266.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6. Datta SR, Brunet A, Greenberg ME. Cellular survival: a play in three AKTs. Genes Dev 1999; 13: 29052927.

  • 7. Doddareddy MR, Rawling T, Ammit AJ. Targeting mitogen-activated protein kinase phosphatase-1 (MKP-1): structure-based design of MKP-1 inhibitors and upregulators. Curr Med Chem 2012; 19: 163173.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8. Laveti D, Kumar M, Hemalatha R, et al. Anti-inflammatory treatments for chronic diseases: a review. Inflamm Allergy Drug Targets 2013; 12: 349361.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9. Marnett LJ. The COXIB experience: a look in the rearview mirror. Annu Rev Pharmacol Toxicol 2009; 49: 265290.

  • 10. Duncan A. Reptile and amphibian analgesia. In: Miller RE, Fowler M, eds. Fowler's zoo and wild animal medicine. St Louis: WB Saunders, 2012: 247253.

    • Search Google Scholar
    • Export Citation
  • 11. Sladky KK, Mans C. Clinical analgesia in reptiles. J Exot Pet Med 2012; 21: 158167.

  • 12. Reed DW, Bradshaw WS, Xie W, et al. In vivo and in vitro expression of a non-mammalian cyclooxygenase-1. Prostaglandins 1996; 52: 269284.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 13. Seebacher F, Franklin CE. Prostaglandins are important in thermoregulation of a reptile ( Pogona vitticeps). Proc R Soc Lond B Biol Sci 2003; 270: S50S53.

    • Search Google Scholar
    • Export Citation
  • 14. Royal LW, Lascelles BDX, Lewbart GA, et al. Evaluation of cyclooxygenase protein expression in traumatized versus normal tissues from eastern box turtles ( Terrapene carolina carolina). J Zoo Wildl Med 2012; 43: 289295.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15. Smith DA, Barker IK. Healing of cutaneous wounds in the common garter snake ( Thamnophis sirtalis). Can J Vet Res 1988; 52: 111119.

    • Search Google Scholar
    • Export Citation
  • 16. National Research Council. Guide for the care and use of laboratory animals. Washington, DC: National Academies Press, 2011;4149.

  • 17. Hodshon RT, Sura PA, Schumacher JP, et al. Comparison of firstintention healing of carbon dioxide laser, 4.0-MHz radiosurgery, and scalpel incisions in ball pythons (Python regius). Am J Vet Res 2013; 74: 499508.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18. Cekanova M, Uddin MJ, Bartges JW, et al. Molecular imaging of cyclooxygenase-2 in canine transitional cell carcinomas in vitro and in vivo. Cancer Prev Res (Phila Pa) 2013; 6: 466476.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 19. Cekanova M, Uddin MJ, Legendre AM, et al. Single-dose safety and pharmacokinetic evaluation of fluorocoxib A: pilot study of novel cyclooxygenase-2-targeted optical imaging agent in a canine model. J Biomed Opt 2012; 17: 116002.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20. Rathore K, Cekanova M. Animal model of naturally occurring bladder cancer: characterization of four new canine transitional cell carcinoma cell lines. BMC Cancer 2014; 14: 465.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21. Tuttle AD, Papich M, Lewbart GA, et al. Pharmacokinetics of ketoprofen in the green iguana (Iguana iguana) following single intravenous and intramuscular injections. J Zoo Wildl Med 2006; 37: 567570.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22. Divers SJ, Papich M, McBride M, et al. Pharmacokinetics of meloxicam following intravenous and oral administration in green iguanas ( Iguana iguana). Am J Vet Res 2010; 71: 12771283.

    • Crossref
    • Search Google Scholar
    • Export Citation

Advertisement

Evaluation of the role of the cyclooxygenase signaling pathway during inflammation in skin and muscle tissues of ball pythons (Python regius)

Ryan A. SadlerDepartment of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37996.

Search for other papers by Ryan A. Sadler in
Current site
Google Scholar
PubMed
Close
 DVM
,
Juergen P. SchumacherDepartment of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37996.

Search for other papers by Juergen P. Schumacher in
Current site
Google Scholar
PubMed
Close
 Dr Med Vet
,
Kusum RathoreDepartment of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37996.

Search for other papers by Kusum Rathore in
Current site
Google Scholar
PubMed
Close
 PhD
,
Kim M. NewkirkDepartment of Biomedical and Diagnostic Sciences, College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37996.

Search for other papers by Kim M. Newkirk in
Current site
Google Scholar
PubMed
Close
 DVM, PhD
,
Grayson ColeDepartment of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37996.

Search for other papers by Grayson Cole in
Current site
Google Scholar
PubMed
Close
 DVM
,
Rachel SeibertDepartment of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37996.

Search for other papers by Rachel Seibert in
Current site
Google Scholar
PubMed
Close
 DVM
, and
Maria CekanovaDepartment of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37996.

Search for other papers by Maria Cekanova in
Current site
Google Scholar
PubMed
Close
 RNDR, PhD

Abstract

OBJECTIVE To determine degrees of production of cyclooxygenase (COX)-1 and -2 and other mediators of inflammation in noninflamed and inflamed skin and muscle tissues in ball pythons (Python regius).

ANIMALS 6 healthy adult male ball pythons.

PROCEDURES Biopsy specimens of noninflamed skin and muscle tissue were collected from anesthetized snakes on day 0. A 2-cm skin and muscle incision was then made 5 cm distal to the biopsy sites with a CO2 laser to induce inflammation. On day 7, biopsy specimens of skin and muscle tissues were collected from the incision sites. Inflamed and noninflamed tissue specimens were evaluated for production of COX-1, COX-2, phosphorylated protein kinase B (AKT), total AKT, nuclear factor κ-light-chain-enhancer of activated B cells, phosphorylated extracellular receptor kinases (ERKs) 1 and 2, and total ERK proteins by western blot analysis. Histologic evaluation was performed on H&E-stained tissue sections.

RESULTS All biopsy specimens of inflamed skin and muscle tissues had higher histologic inflammation scores than did specimens of noninflamed tissue. Inflamed skin specimens had significantly greater production of COX-1 and phosphorylated ERK than did noninflamed skin specimens. Inflamed muscle specimens had significantly greater production of phosphorylated ERK and phosphorylated AKT, significantly lower production of COX-1, and no difference in production of COX-2, compared with production in noninflamed muscle specimens.

CONCLUSIONS AND CLINICAL RELEVANCE Production of COX-1, but not COX-2, was significantly greater in inflamed versus noninflamed skin specimens from ball pythons. Additional research into the reptilian COX signaling pathway is warranted.

Abstract

OBJECTIVE To determine degrees of production of cyclooxygenase (COX)-1 and -2 and other mediators of inflammation in noninflamed and inflamed skin and muscle tissues in ball pythons (Python regius).

ANIMALS 6 healthy adult male ball pythons.

PROCEDURES Biopsy specimens of noninflamed skin and muscle tissue were collected from anesthetized snakes on day 0. A 2-cm skin and muscle incision was then made 5 cm distal to the biopsy sites with a CO2 laser to induce inflammation. On day 7, biopsy specimens of skin and muscle tissues were collected from the incision sites. Inflamed and noninflamed tissue specimens were evaluated for production of COX-1, COX-2, phosphorylated protein kinase B (AKT), total AKT, nuclear factor κ-light-chain-enhancer of activated B cells, phosphorylated extracellular receptor kinases (ERKs) 1 and 2, and total ERK proteins by western blot analysis. Histologic evaluation was performed on H&E-stained tissue sections.

RESULTS All biopsy specimens of inflamed skin and muscle tissues had higher histologic inflammation scores than did specimens of noninflamed tissue. Inflamed skin specimens had significantly greater production of COX-1 and phosphorylated ERK than did noninflamed skin specimens. Inflamed muscle specimens had significantly greater production of phosphorylated ERK and phosphorylated AKT, significantly lower production of COX-1, and no difference in production of COX-2, compared with production in noninflamed muscle specimens.

CONCLUSIONS AND CLINICAL RELEVANCE Production of COX-1, but not COX-2, was significantly greater in inflamed versus noninflamed skin specimens from ball pythons. Additional research into the reptilian COX signaling pathway is warranted.

Reptile analgesia has advanced in the past decade; however, challenges remain, including species differences in response to analgesics and lack of a thorough understanding of reptilian pain and inflammation. A survey of members of the Association of Reptile and Amphibian Veterinarians revealed that veterinarians may not provide appropriate or effective analgesia for reptile patients undergoing noxious procedures.1 However, an investigation2 of the pharmacodynamic properties of NSAIDs, such as meloxicam, revealed that although commonly used in reptilian species, meloxicam might not be as effective as previously reported.2

Nonsteroidal anti-inflammatory drugs are routinely used in humans and domestic animal species to inhibit COX enzymes from catalyzing the conversion of arachidonic acid into various functional prostaglandins that mediate inflammation; this inhibition thereby decreases the degree of inflammation and associated pain. The COX-1 isoform is constitutively produced in most healthy tissues, and production of the COX-2 isoform is typically induced in inflamed and neoplastic tissues in mammals.3–5

The inflammatory cascade is a complex system, wherein the COX signaling pathway interacts with others, such as the MAPK signaling pathway. The MAPK signaling pathway involves cellular response to inflammation as well as cell proliferation and differentiation. The AKT signaling pathway is also linked to inflammatory processes in the skin.6 In mammalian species with experimentally induced inflammation, the COX, NF-κΒ, and MAPK signaling pathways play key roles in induction of inflammation in targeted tissues.7,8 However, an unwanted adverse effect of some NSAIDs is gastrointestinal tract irritation. Cyclooxygenase-2 selective inhibitors were developed with the aim of blocking the predominantly inducible production of COX-2, while leaving the constitutive production of COX-1 unaffected so that the adverse gastrointestinal effects associated with NSAID administration might be avoided.9 Exotic animal species, including reptiles, are commonly given COX-2–selective inhibitors, such as meloxicam.10,11

Although little information has been published regarding inflammation in reptilian species, previous research12 has shown the possibility of differences between mammalian and nonmammalian species in distributions, functions, and amounts of COX isoforms. Prostaglandins, the end product of COX signaling pathway, are associated with various processes, including thermoregulation, nociception, and homeostatic activities in some reptilian species.13 In injured box turtles (Terrapene carolina carolina), COX-2 has been identified in liver, kidney, and muscle tissue and a significant degree of COX-1 and a nonsignificant degree of COX-2 production occurs in traumatized muscle tissue.14 In common garter snakes (Thamnophis sirtalis), the cellular response to cutaneous injury is characterized by a mixed heterophilic and histiocytic response, with heterophils predominating longer in the tissues than macrophages, suggesting different inflammation patterns than those in mammals and birds.15

The purpose of the study reported here was to investigate the molecular mechanisms of inflammation in ball pythons (Python regius) by evaluating the role of COX, NF-κΒ, ERK, and AKT signaling pathways in noninflamed and laser-induced inflamed skin and muscle tissues. A second objective was to use a histologic grading scheme to confirm the presence of inflammation in selected tissues.

Materials and Methods

Animals

Six healthy adult male ball pythons were acquired from a US commercial breeder of snakes and allowed to acclimate to the study environment for 2 weeks before the study began. Snakes were housed individually in plastic containers (approx 63.5 × 38.1 × 15 cm), with a hiding box and paper substrate in accordance with published animal care guidelines.16 Water was provided ad libitum. The room in which snakes were housed was maintained at a mean ± SD daytime temperature of 29.4 ± 2.2°C and nighttime temperature of 27.7 ± 1.2°C, with ambient humidity > 30% for the duration of the study. Automated lighting provided a 12-hour photoperiod. Food was withheld from snakes for 1 week prior to the study. At the beginning of the study, snakes received a complete physical examination, including recording of body weight and snout-to-vent length, and a blood sample (0.5 mL) was collected for measurement of hematologic and plasma biochemical variables. The study protocol was approved by the University of Tennessee Institutional Animal Care and Use Committee.

Biopsy of noninflamed tissues on day 0

All biopsy procedures were performed in an aseptic operating room with an ambient temperature of 29.4°C. Anesthesia was induced in each snake with ketamine (10 mg/kg) and morphine (1 mg/kg) administered IM. Thirty minutes after drug administration, snakes were positioned on top of a conductive warming blanketa (43°C). Absence of the righting reflex, response to noxious stimulus (tail pinch), and heart and respiratory rates were monitored and recorded throughout anesthesia. Planned skin and muscle incision sites were aseptically prepared with alternating washes with 4% chlorhexidine scrub and sterile saline (0.9% NaCl) solution. A 2 × 1-cm rectangular skin incision was made with a No. 15 scalpel blade between the scales on the lateral aspect of the body, over the epaxial musculature at the cranial third of each snake. A biopsy specimen of skin tissue (as large as 2 × 1 × 1 cm) was collected from each snake with DeBakey thumb forceps and Metzenbaum scissors via sharp dissection to separate the skin from the underlying musculature. Afterward, a similarly sized specimen of muscle tissue was obtained from the epaxial musculature directly deep to the skin biopsy site.

Skin and muscle biopsy specimens were cut in half. One half of each specimen was placed into a small tube with a preservative solutionb and stored in a freezer at −80°C for western blot protein analysis. The other half was placed in a container with neutral buffered 10% formalin solution for histologic evaluation. Each biopsy site was closed by apposing or slightly everting the scales with sterile surgical skin staples. To provide postoperative analgesia for snakes, 0.1 mL of a solution of containing 0.3 mL of 2% lidocaine and 0.15 mL of sodium bicarbonate solution was injected around each biopsy site.

Laser-induced inflammation on day 0

Inflammation of skin and muscle tissue was induced by laser incision on day 0 as previously described.17 Briefly, at a position 5 cm distal to the initial biopsy site, a 2-cm skin and muscle incision (inflammation site) was made via CO2 laserc with a 0.4-mm gold tip at 9-W output on a continuous setting with a focused beam. Skin incisions were closed by apposing or slightly everting the scales with sterile surgical skin staples. Each snake was continuously monitored until it responded to touch, its righting reflex had returned, and it was moving on its own. Postoperative analgesia was provided by IM administration of morphine (1 mg/kg, q 12 h as needed). Each snake was returned to its enclosure and monitored 3 times/d for 7 days for signs of discomfort and pain.

Biopsy of inflamed tissues on day 7

Seven days after the initial biopsy procedure, each snake was anesthetized and biopsy specimens were collected from the skin and muscle sites at which the CO2 laser had been applied with the same protocol as described for day 0. A 7-day inflammation period was chosen on the basis of peak inflammation scores achieved in another study17 in which healing of laser incisions was investigated in ball pythons. Skin and muscle biopsy specimens were divided as stored until analysis as described for day 0. Each biopsy site was closed with 3–0 synthetic sutured in a horizontal mattress pattern, and 0.1 mL of a solution of containing 0.3 mL of 2% lidocaine and 0.15 mL sodium bicarbonate solution was injected around the site. Each snake was continuously monitored until it responded to touch, its righting reflex had returned, and it was moving on its own. Postoperative analgesia was provided by IM administration of morphine (1 mg/kg, q 12 h as needed). Each snake was returned to its enclosure and monitored 3 times/d for 7 days for signs of discomfort and pain.

Histologic evaluation

Formalin-fixed tissue specimens were routinely processed. Once embedded in paraffin, specimens were sectioned at 5 µm and stained with H&E and Masson trichrome stains. One pathologist (KMN), who was blinded to specimen identity, examined each tissue section via light microscopy to identify the extent of inflammation (presence of granulocytes, macrophages, and lymphocytes), healing, and necrosis and presence of granulomas.

Degree of inflammation was subjectively scored on the basis of severity of granulocytic (heterophil) infiltrates into the wound bed and severity of perivascular infiltrates of lymphocytes and macrophages within examined sections (0 = no infiltrates, 1 = mild infiltrates, 2 = moderate infiltrates, and 3 = dense infiltrates). Degree of healing was subjectively scored on the basis of degree of epithelial migration from the wound margin (0 = healing absent, 1 = 1% to 49% healed, 2 = 50% to 99% healed, and 3 = 100% healed); incisions with a score of 3 were considered completely healed. Granulomas were subjectively scored on the basis of the previous criteria (0 = no granuloma present and 1 = granuloma present).

Necrosis was defined as coagulated, hypereosinophilic (necrotic) dermal collagen or loss of differential staining with retention of cellular architecture (coagulation necrosis) of the epithelial cells in the epidermis. A representative width of necrotic tissue on 1 side of the wound bed was measured. Measurements used to assess the width of necrotic tissue were performed in the most representative section for each site. A histologic wound score was assigned to each specimen to represent degree of severity by width of the affected tissues (0 = no necrosis, 1 = 1 to 15 μm, 2 = 16 to 99 μm, and 3 = ≥ 100 μm). These scores were combined with the total inflammation score to calculate a histologic response score for each biopsy specimen.

Western blot analysis

Primary antibodies used for western blot analysis included COX-1,e COX-2,e a blocking peptide specific for COX-2 N-20 terminus (sc-1746P),e total ERK,e phospho-ERK (Tyr204),e actin,e total AKT,f phospho-AKT (Ser437/Thr308),f and NF-κB (p65).g

Skin and muscle biopsy specimens were sonicated on ice for 10 seconds twice in ice-cold bufferh (1% nonyl phenoxypolyethoxylethanol, 0.5% sodium deoxycholate, and 0.1% SDS) supplemented with protease and phosphatase inhibitors (aprotinin [1 μg/mL], leupeptin [1 μg/mL], 1mM phenylmethylsulfonyl fluoride, 0.1mM Na3VO4, and 10mM NaF) as described elsewhere18,19 and kept at −80°C until western blot analysis. Protein concentrations were assessed by bicinchoninic acid assayi in accordance with the manufacturer's protocol. Equal amounts of proteins were assayed via SDS PAGE and transferred onto a nitrocellulose membrane.j Lysates from human squamous cell carcinoma cellsk and a canine transitional cell line20,l with high production of COX-2 and COX-1 were used as positive control samples to test the ability of the antibodies used to detect snake COX-1 and COX-2 by western blot analysis. After blocking with 5% nonfat milk in a mixture of tris-buffered saline solution and Tween 20i for 1 hour at room temperature (21°C), membranes were incubated with primary antibodies (pERK, ERK, pAKT, AKT, COX-1, COX-2, NF-κB, and actin) overnight at 4°C. Afterward, membranes were incubated with horseradish peroxidase–conjugated secondary antibodiesf for 1 hour at room temperature.

The blocking peptide for COX-2 (N-20) antibody was used to evaluate the specific reactivity of COX-2 antibody against snake COX-2 (70–72 kDa) by western blot analysis. The COX-2 antibody was neutralized by addition of blocking peptide at a ratio of 1:5 and mixing on a shaker for 2 hours at room temperature before incubating on a membrane overnight at 4°C. Immunoreactive bands were viewed with an enhanced chemiluminescence system.m Each western blot analysis was run in 2 independent replicates. Quantitative analysis of the bands was performed by use of imaging software,n which yielded densitometric values.

Statistical analysis

Histologic scores were compared between noninflamed (day 0) and inflamed (day 7) tissue specimens from the 6 snakes by use of the Wilcoxon signed rank test. Densitometric values for the various tested proteins (COX-1, COX-2, pERK, pAKT, ERK, NF-κB, and AKT) and for the control protein actin were tested for normality by use of the Shapiro-Wilk test. Values are summarized as mean ± SD. Densitometric values for the proteins of interest were normalized to those of actin, and resulting values were compared between noninflamed and inflamed tissue specimens. The Student paired t test was used to identify significant differences in densitometric values between noninflamed and inflamed tissues specimens for each protein by use of statistical software.n Values of P ≤ 0.05 were considered significant.

Results

Animals

All snakes were judged as healthy on the basis of the results of complete physical examination. Hematologic and plasma biochemical values were within the reference limits established for ball pythons at the investigators’ institution. All snakes recovered from all surgical procedures without complication and behaved typically throughout the study period. No additional analgesia was required for any snake enrolled in the study. By 30 days after collection of biopsy specimens of inflamed skin and muscle tissue, all biopsy sites appeared to have healed, with minimal scarring as assessed by 2 investigators (RAS and JPS). All snakes were able to shed their skin as usual, and no retained skin was identified at the biopsy sites.

Histologic comparison of noninflamed and inflamed tissue specimens

Histologic findings for noninflamed skin and muscle specimens collected from snakes on day 0 were unremarkable, with no evidence of underlying dermatologic disease. Median scores were significantly higher for inflamed versus noninflamed skin specimens with respect to granulocytes, lymphocytes and macrophages, necrosis width, and total combined scores (Table 1). Median scores were significantly higher for inflamed versus noninflamed muscle specimens with respect to granulocytes, lymphocytes and macrophages, granuloma formation, necrosis width, and total combined scores. Both sets of findings confirmed that the laser incision had resulted in inflammation of affected tissues.

Table 1—

Median (95% CI) histologic inflammation scores for H&E-stained noninflamed (day 0) and inflamed (day 7) skin and muscle tissue sections from ball pythons (Python regius).

VariableDay 0Day 7P value
Skin tissue
 Granulocytes03.0 (2.2–3.0)0.03
 Lymphocytes and monocytes02.0 (2.0–2.8)0.03
 Granuloma00.5 (0.0–1.0)NA
 HealingNA1.5 (1.0–2.0)NA
 Necrosis02.0 (2.0–3.0)0.03
 Overall score010.0 (7.4–10.0)0.03
Muscle tissue
 Granulocytes01.5 (1.0–2.8)0.03
 Lymphocytes and monocytes02.5 (1.2–3.0)0.03
 Granuloma01.0 (0.0–1.0)NA
 Healing001.0
 Necrosis02.5 (1.2–3.0)0.03
 Overall score06.5 (5.2–8.8)0.03

Inflammation was subjectively scored on the basis of severity of granulocytic (heterophil) infiltrates into the wound bed and severity of perivascular infiltrates of lymphocytes and macrophages within examined sections (0 = no infiltrates, 1 = mild infiltrates, 2 = moderate infiltrates, and 3 = dense infiltrates). Healing was subjectively scored on the basis of degree of epithelial migration from the wound margin (0 = healing absent, 1 = 1% to 49% healed, 2 = 50% to 99% healed, and 3 = 100% healed). Granulomas were scored as present (1) or not present (0). For necrosis, a representative width of necrotic tissue on 1 side of the wound bed was measured and a score was assigned to represent degree of severity by width of the affected tissues (0 = no necrosis, 1 = 1 to 15 μm, 2 = 16 to 99 μm, and 3 = ≥ 100 μm). Values of P ≤ 0.05 were considered significant.

NA = Not applicable.

Western blot comparison of noninflamed and inflamed skin specimens

Western blot analysis revealed significantly greater COX-1 (P = 0.03) and pERK (P = 0.01) production in inflamed versus noninflamed skin specimens (Figure 1). Greater production of pAKT was also detected in inflamed versus noninflamed skin specimens; however, because of large SEs, that difference was not significant (P = 0.13). Only weak production of COX-2 was detected in skin specimens when C-terminus or N-terminus-specific antibody against COX-2 was used, and only weak signals were detected via densitometry. Production of total AKT and total ERK was detected, although no significant difference in values was identified between noninflamed and inflamed skin specimens. Western blot analysis with antibody that recognizes human p65 failed to detect production of NF-κB in any snake skin specimen.

Western blot comparison of noninflamed and inflamed muscle specimens

Significantly greater production of pERK (P = 0. 001) and pAKT (P = 0.001) was identified in inflamed versus noninflamed muscle specimens (Figure 2). Moderately greater production of total AKT and ERK was detected in inflamed versus noninflamed muscle specimens; however, these increases were not significant (P = 0.74 and P = 0.48, respectively). Interestingly, significantly (P = 0.043) lower COX-1 production was detected in inflamed versus noninflamed muscle specimens. In contrast to COX-1, slightly increased production of COX-2 was identified in inflamed versus noninflamed muscle specimens, although this increase was not significant (P = 0.73). Similarly to findings for skin specimens, western blot analysis with antibody that recognizes human p65 failed to detect production of NF-κB in any snake muscle specimen.

Densitometric comparison of noninflamed and inflamed tissue specimens

Although evidence of low COX-2 production was visually detected in skin specimens via western blot analysis, production appeared more pronounced when densitometry was used (Figure 1). To rule out the possibility that nonspecific binding of COX-2 (N-20) antibody had occurred in snake skin and muscle specimens, blocking peptide was used to prevent binding of COX-2 specific antibody (70 to 72 kDa), confirming the specificity of COX-2 (N-20) antibody for the snake COX-2 in the specimens (Figure 3).

Figure 1—
Figure 1—

Photographs of results of western blot analysis (A) and mean values of the densitometric analysis (arbitrary units) of western blot data (B) representing degrees of expression of various proteins involved in COX signaling pathway in noninflamed (day 0; D0) and inflamed (day 7; D7) skin tissue specimens from 6 ball pythons (Python regius). A—Numbers 1 through 6 represent individual snakes. B—Densitometric readings were normalized to those of actin. Error bars represent 95% confidence intervals. *Value differs significantly (P ≤ 0.05) between inflamed and noninflamed tissues by paired t test. Weak expression of COX-2 was detected with both types of antibodies (N-terminal [N-20] and C-terminal [C-20]) in skin tissue, and no NF-κB expression was detected in any tissue specimen. Six specimens (1/snake) were assessed for all proteins except COX-2, for which only 5 specimens were available because of artifact smearing on the western blot membrane for 1 specimen.

Citation: American Journal of Veterinary Research 77, 5; 10.2460/ajvr.77.5.487

Figure 2—
Figure 2—

Photographs of results of western blot analysis (A) and mean values of densitometric analysis (arbitrary units) of western blot data (B) representing degrees of expression of various proteins involved in the COX signaling pathway in noninflamed (day 0; D0) and inflamed (day 7; D7) muscle tissue specimens from 6 ball pythons. Actin was used as a control protein. Weak expression of COX-2 was detected when C-terminal COX-2 antibody was used, although better expression was detected when N-terminal COX-2 antibody was used. Again, no NF-κB expression was detected in any tissue specimen. See Figure 1 for remainder of key.

Citation: American Journal of Veterinary Research 77, 5; 10.2460/ajvr.77.5.487

Figure 3—
Figure 3—

Photographs of results of western blot analysis of skin (A) and muscle (B) specimens from the snakes in Figures 1 and 2 after application of blocking peptide (BP) specifically designed to prevent binding of N-terminal COX-2 antibody to COX-2 protein. Cell lysates from human squamous cell carcinoma cells (hSCC-25) and canine transitional cell carcinoma cells (K9TCCLilly) were used as positive control samples for COX-2 expression.

Citation: American Journal of Veterinary Research 77, 5; 10.2460/ajvr.77.5.487

Discussion

Although investigations2,21,22 have been conducted into pharmacokinetics and pharmacodynamics of various NSAIDs in particular reptilian species, the clinical effects of these drugs remain largely unknown, as does information regarding inflammatory processes in reptiles. Similar to findings of a study14 regarding COX production in chelonians (eastern box turtles) with naturally inflamed muscle tissues, findings of the present study involving snakes with experimentally induced inflammation indicated that COX-1 production was higher in inflamed versus noninflamed skin specimens and that COX-2 production did not differ significantly between skin tissue types. However, contrary to the findings in turtles,14 we did not identify a significantly greater degree of COX-1 production in inflamed versus noninflamed muscle specimens from snakes.

The study14 involving turtles also revealed an increase in production of both COX enzymes in many traumatized tissues; however, the degree of production was not normalized by comparison with that of a housekeeping protein, such as actin, to verify equal loading of proteins for each specimen during western blot analysis. The presence of COX-2 in noninflamed tissues other than those of renal origin suggests possible widespread inflammation, and therefore, suitable control techniques might have been needed to verify the results in that study.14 The inflammation in snake tissues 7 days after laser induction might have resulted in different degrees of COX enzyme production than in the other study, given that 7 days is reportedly the peak time for histological inflammation assessment in ball pythons17 and the peak period for turtles is unknown.14

Snake tissues in the present study were tested for production of several proteins by western blot analysis involving antibodies that were originally designed for detection of human proteins. Some of those antibodies cross-react in other mammalian species, but none of them had been tested for reactivity in reptiles specifically. Results of the study reported here also suggested that human-specific antibodies (N-terminus) could be used to detect COX enzymes in tissues of ball pythons. Significant increases in COX-1 but not COX-2 production were detected in inflamed versus noninflamed skin specimens. Although enzymes from other inflammatory signaling pathways were significantly increased in inflamed versus noninflamed tissue specimens (ie, pERK in skin specimens and pERK and pAKT in muscle specimens), only COX-1 production was significantly greater in inflamed versus noninflamed skin specimens and significantly less in inflamed versus noninflamed muscle specimens.

Sonication and homogenization of the skin specimens were technical challenges for isolation of proteins for western blot analysis. Some of the tested proteins were not clearly separated, resulting in poor quality of some visualized bands. Detection of COX-2 bands at the molecular weight of 70 to 72 kDa for both control and tissue specimens indicated that COX-2 was present, but the lack of a significant increase in COX-2 production with inflammation, despite other evidence of inflammation, could have indicated that COX-2 production had not been substantially induced at the point when inflamed skin and muscle specimens were obtained or that there was an insufficient number of cells producing this enzyme to allow detection by western blot analysis. In particular, skin specimens consisted largely of tough scale and fibrous tissues, compared with muscle specimens, which resulted in lower amounts of proteins available for extraction. This circumstance could have explained the poor detectability of bands for COX-2 in western blot analysis of skin specimens.

As has been reported for common garter snakes,15 a predominantly heterophilic response to dermal injury in ball pythons in the present study may have resulted in less production of COX isoforms than would be achieved with a predominantly histiocytic response. Histologic evaluation of tissue specimens revealed that the degrees of granulocytic and histiocytic cellular responses were similar in the inflamed tissue specimens examined, compared with responses in control tissues; therefore, the degree of inflammatory response should have elicited COX production. On the other hand, NF-κB had no detectable signal in either skin or muscle specimens, likely because of failure to extract detectable amounts of protein, insufficient expression in cells, or incompatibility of tested NF-kB antibody used in the western blot analysis with snake NF-κB. Several slides of tissue sections from noninflamed and inflamed tissue biopsy specimens from select snakes were evaluated by use of immunohistochemical techniques and COX-2– and COX-1– specific antibodies (data not shown), and the positive signal detected on these slides supported specific binding of COX-2 and COX-1 antibodies in snake tissues. The combined mixed histiocytic and granulocytic histologic findings and immunohistochemical signal detection suggested that a sufficient inflammatory response had been induced to allow accurate measurement of COX response.

Surprisingly, we identified a significant decrease in COX-1 production in inflamed versus noninflamed muscle specimens. This downregulation of COX-1 production in inflamed muscle specimens, with concurrent significant upregulation of pERK and pAKT production, appeared contradictory. Low COX-1 production in inflamed tissue specimens might have been attributable to the presence of excessive necrotic debris, lack of migration of appropriate inflammatory cells capable of COX production to the target tissue, poor protein extraction from tissue specimens, or failure to collect specimens at the point of greatest protein production by COX signaling pathways. Common garter snakes develop a fibroelastic response to integumentary injury that occurs in the laterally adjacent dermis rather than in the deeper subcutaneous tissues as in mammals,15 and if the situation were similar in ball pythons, such a response could have resulted in excessive fibrous tissue formation in inflamed tissue specimens from the snakes in the present study.

In the study reported here, COX-1 production in inflamed skin specimens from ball pythons was significantly greater than in noninflamed skin specimens, but the same was not true for muscle specimens. However, COX-2 production was not significantly greater in inflamed versus noninflamed skin and muscle specimens, despite histologic evidence of inflammation and concurrent significant increases in production of other inflammatory signaling pathways, such as pERK in skin specimens and pERK and pAKT in muscle specimens. The lack of changes in COX-2 production in response to inflammation in snakes suggested that additional investigation should be pursued to identify specific roles of inflammatory mediators in COX signaling pathways of reptiles to further elucidate the cellular response to injury in various reptilian species so that appropriate analgesic and anti-inflammatory treatments can be identified for reptiles.

Acknowledgments

Supported by the University of Tennessee College of Veterinary Medicine Companion Animal Fund, the University of Tennessee Center of Excellence in Livestock Diseases and Human Health, and the University of Tennessee Department of Small Animal Clinical Sciences.

Presented in abstract form at the 46th Annual American Association of Zoo Veterinarians Conference, Orlando, Fla, October 2014.

The authors thank Steve Tinkel, Janet Pezzi, Lillian Gerhardt, and Rick Adams for technical assistance.

ABBREVIATIONS

AKT

Protein kinase B

COX

Cyclooxygenase

ERK

Extracellular receptor kinase

MAPK

Mitogen-activated protein kinase

NF-κB

Nuclear factor κ-light-chain-enhancer of activated B cells

pAKT

Phosphorylated protein kinase B

pERK

Phosphorylated extracellular receptor kinase

Footnotes

a.

HotDog, Augustine Biomedical Design, Eden Prairie, Minn.

b.

RNAlater solution, Sigma-Aldrich, St Louis, Mo.

c.

Nova-Pulse CO2 laser LX-20SP, Luxar Co, Bothell, Wash.

d.

PDS II, Ethicon Inc, Somerville, NJ.

e.

Santa Cruz Biotechnology, Santa Cruz, Calif.

f.

Cell Signaling Corporation, Beverly, Mass.

g.

Cayman Chemical Corp, Ann Arbor, Mich.

h.

RIPA buffer, Fisher Scientific, Pittsburgh, Pa.

i.

Fisher Scientific, Pittsburgh, Pa.

j.

Osmonics, Minnetonka, Minn.

k.

SCC-25, American Type Culture Collection, Manassas, Va.

l.

K9TCC-Lilly, University of Tennessee College of Veterinary Medicine, Knoxville, Tenn.

m.

Chemiluminescent camera, Pierce Biotechnology, Rockford, Ill.

n.

Image J, National Institutes of Health, Bethesda, Md.

o.

MedCalc Software, version 1511, Ostend, Belgium.

References

  • 1. Read MR. Evaluation of the use of anesthesia and analgesia in reptiles. J Am Vet Med Assoc 2004; 224: 547552.

  • 2. Olesen MG, Bertelsen MF, Perry SF, et al. Effects of preoperative administration of butorphanol or meloxicam on physiologic responses to surgery in ball pythons. J Am Vet Med Assoc 2008; 233: 18831888.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3. Vane JR, Botting RM. Mechanism of action of nonsteroidal antiinflammatory drugs. Am J Med 1998; 104: 2S8S.

  • 4. Eberhart CE, Coffey RJ, Radhika A, et al. Up-regulation of cyclooxygenase 2 gene expression in human colorectal adenomas and adenocarcinomas. Gastroenterology 1994; 107: 11831188.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 5. Dannenberg AJ, Lippman SM, Mann JR, et al. Cyclooxygenase-2 and epidermal growth factor receptor: pharmacologic targets for chemoprevention. J Clin Oncol 2005; 23: 254266.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6. Datta SR, Brunet A, Greenberg ME. Cellular survival: a play in three AKTs. Genes Dev 1999; 13: 29052927.

  • 7. Doddareddy MR, Rawling T, Ammit AJ. Targeting mitogen-activated protein kinase phosphatase-1 (MKP-1): structure-based design of MKP-1 inhibitors and upregulators. Curr Med Chem 2012; 19: 163173.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8. Laveti D, Kumar M, Hemalatha R, et al. Anti-inflammatory treatments for chronic diseases: a review. Inflamm Allergy Drug Targets 2013; 12: 349361.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9. Marnett LJ. The COXIB experience: a look in the rearview mirror. Annu Rev Pharmacol Toxicol 2009; 49: 265290.

  • 10. Duncan A. Reptile and amphibian analgesia. In: Miller RE, Fowler M, eds. Fowler's zoo and wild animal medicine. St Louis: WB Saunders, 2012: 247253.

    • Search Google Scholar
    • Export Citation
  • 11. Sladky KK, Mans C. Clinical analgesia in reptiles. J Exot Pet Med 2012; 21: 158167.

  • 12. Reed DW, Bradshaw WS, Xie W, et al. In vivo and in vitro expression of a non-mammalian cyclooxygenase-1. Prostaglandins 1996; 52: 269284.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 13. Seebacher F, Franklin CE. Prostaglandins are important in thermoregulation of a reptile ( Pogona vitticeps). Proc R Soc Lond B Biol Sci 2003; 270: S50S53.

    • Search Google Scholar
    • Export Citation
  • 14. Royal LW, Lascelles BDX, Lewbart GA, et al. Evaluation of cyclooxygenase protein expression in traumatized versus normal tissues from eastern box turtles ( Terrapene carolina carolina). J Zoo Wildl Med 2012; 43: 289295.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15. Smith DA, Barker IK. Healing of cutaneous wounds in the common garter snake ( Thamnophis sirtalis). Can J Vet Res 1988; 52: 111119.

    • Search Google Scholar
    • Export Citation
  • 16. National Research Council. Guide for the care and use of laboratory animals. Washington, DC: National Academies Press, 2011;4149.

  • 17. Hodshon RT, Sura PA, Schumacher JP, et al. Comparison of firstintention healing of carbon dioxide laser, 4.0-MHz radiosurgery, and scalpel incisions in ball pythons (Python regius). Am J Vet Res 2013; 74: 499508.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18. Cekanova M, Uddin MJ, Bartges JW, et al. Molecular imaging of cyclooxygenase-2 in canine transitional cell carcinomas in vitro and in vivo. Cancer Prev Res (Phila Pa) 2013; 6: 466476.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 19. Cekanova M, Uddin MJ, Legendre AM, et al. Single-dose safety and pharmacokinetic evaluation of fluorocoxib A: pilot study of novel cyclooxygenase-2-targeted optical imaging agent in a canine model. J Biomed Opt 2012; 17: 116002.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20. Rathore K, Cekanova M. Animal model of naturally occurring bladder cancer: characterization of four new canine transitional cell carcinoma cell lines. BMC Cancer 2014; 14: 465.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21. Tuttle AD, Papich M, Lewbart GA, et al. Pharmacokinetics of ketoprofen in the green iguana (Iguana iguana) following single intravenous and intramuscular injections. J Zoo Wildl Med 2006; 37: 567570.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22. Divers SJ, Papich M, McBride M, et al. Pharmacokinetics of meloxicam following intravenous and oral administration in green iguanas ( Iguana iguana). Am J Vet Res 2010; 71: 12771283.

    • Crossref
    • Search Google Scholar
    • Export Citation

Contributor Notes

Address correspondence to Dr. Cekanova (mcekanov@utk.edu).