De Novo Serum Biochemistry and Electrophoretic Reference Intervals for Jamaican Fruit Bats (Artibeus jamaicensis)

Introduction

The Jamaican fruit bat (Artibeus jamaicensis, JFB) is becoming an increasingly popular species in biomedical research due to their unique immunologic and physiologic characteristics. Bats are flying mammals of the order Chiroptera, which derives from ancient Greek as “hand-wing” (χείρ: cheir “hand”; πτερόν: pteron, “wing”) due to evolutionary adaptations to the forelimbs allowing for flight.¹ Bats are the only mammals capable of sustained flight, allowing them to become one of the most widely distributed mammals in the world.² While flight is an efficient mode of locomotion, it creates a large energy demand requiring higher caloric intake and metabolic adaptations. Their metabolic rate is 2.5-3 times higher than the maximum observed during exercise in terrestrial mammals of comparable size.³ During rest periods, they enter a cyclic bradycardic state as low as 200 bpm, which reduces energy expenditure to counteract heart rates as high as 900 bpm during flight.⁴ Indeed, bats have developed strategies to combat sudden surges in activity required for flight.

Despite high metabolic rates and small body sizes, bats exhibit longer lifespans than comparable nonflying mammals. The JFB can live up to 10 years in captivity, whereas laboratory mice are considered aged at 18 months.^5,6 Their longer lifespans create ideal candidates for longevity, aging, and disease progression studies. In addition, a low incidence of spontaneous cancer and a limited degree of inflammation are some of the remarkable immunologic characteristics of bats.⁷ The ability of bats to avoid acute inflammatory reactions during infection likely contributes to this reduction in age-related illnesses and the ability to carry viral infections such as coronaviruses without developing disease themselves.⁸ JFB, in particular, are able to mount a robust, highly specific antibody response to viral exposure, which is uncommon in other bat species.⁹ These characteristics make the JFB an optimal species to study these viruses by avoiding morbidity and mortality in the animal model.

Notwithstanding all our knowledge of this species, its normal physiologic parameters remain largely undetermined. A previous study¹⁰ established reference intervals for CBC values, but serum chemistry values in this species still require investigation. Clinical biochemistry, like hematology, is often used in veterinary medicine to detect health abnormalities, which can aid in diagnosing and monitoring diseases. These values provide organ and tissue-specific information, such as that related to renal and hepatic function, which can reveal underlying health conditions with minimal discomfort to the patient. An understanding of normal serum biochemical parameters can make monitoring population health as well as individual health more feasible, as well as allowing for comparisons between species and study cohorts. Therefore, it is essential to determine reference intervals in this species to provide optimal health monitoring and to advance our understanding of JFBs as models of human disease.

The objectives of this study were to establish serum biochemical reference intervals for a clinically healthy population of JFBs and to compare the data by sex. The American Society of Veterinary Clinical Pathology (ASVCP) consensus guidelines were followed for the determination of de novo reference intervals in veterinary species, which mirror the 2008 Clinical Laboratory and Standards Institute recommendations.^11,12 We hypothesized that these values for JFBs would be similar to published intervals in other frugivorous bat species with increases in glucose and CK due to the stress of hand capture. The values obtained in our study can be used to expand on research studies and add to the general knowledge of normal physiologic parameters in this unique species of bat.

Materials and Methods

Statistical analysis and determination of reference intervals.

Reference intervals were established under the current ASVCP quality assurance guidelines.¹¹ Intervals were generated using Reference Value Advisor 2.1 for Excel (Microsoft).¹³ The program determines a Gaussian distribution using the Anderson-Darling goodness-of-fit test and identifies potential outliers using Dixon-Reed and Tukey tests. Nonparametric methods were used to establish upper (97.Fifth percentile) and lower (2.Fifth percentile) reference limits. When fewer than 120 samples were used, the 90% CI was determined using a bootstrap method. Data were provided as untransformed standard Gaussian or Box-Cox transformed data when statistically appropriate.

Any sample with icterus (icteric index >50), hemolysis (hemolytic index >500), or lipemia (lipemia index >100) was excluded from the study. Definite statistical outliers identified by the macro were confirmed using clinical judgement and consultation with an American College of Veterinary Pathologists (ACVP) board-certified clinical pathologist and excluded when appropriate. Other suspect outliers were identified based on visual examination of each parameter’s histogram within the cohort and removed based on clinical judgement and consultation with an ACVP board-certified veterinary clinical pathologist. Data analysis for potassium is provided as an example to demonstrate methods used to identify and exclude outliers (Figure 1). Few outliers were removed due to suspect sampling errors or suspected deterioration during sampling (ie, GGT of 0), prompting the exclusion of these animals from analysis. Outliers in any parameter involved in calculating anion gap or osmolality (Na⁺, K⁺, Cl⁻, and bicarbonate; and Na⁺, glucose, urea nitrogen, and creatinine, respectively) were removed from the subsequent calculation. Because sex was readily obvious, we chose to a priori partition biochemical and serum protein electrophoretic data by sex, and outliers were removed according to these methods. An unpaired t test with Welch correction was performed for each analyte to compare sex differences. Adjusted P values less than 0.05 were considered statistically significant and bolded in the corresponding tables (Tables 3 and 4). Statistical analysis was performed using Prism (version 10.0.2; GraphPad Software).

View Figure

• Download as Powerpoint
• Download Figure

Results

Blood was successfully collected from a total of 179 JFB (97 males and 82 females). Thirty-seven samples were excluded due to fibrin clots, inadequate blood volume, or laboratory error. All values from one female were excluded due to a lipemia index of 314 mg/dL, icteric mucous membranes, and hepatomegaly noted on necropsy. All values from 4 bats were also excluded due to skin or lip lesions, which may be attributed to prior trauma or trauma during catching. Three females were excluded from analysis due to pregnancy identified on necropsy. No specific findings were identified on gross necropsy of the rest of the animals. After removal of outliers and these exclusions, a total of 134 JFB (62 males and 72 females) were included in the data analysis.

The reference intervals for clinical chemistry analysis for the total population are reported in Table 1. For most analytes investigated, biochemistry values were similar to those reported in other frugivorous bat species. After outlier removal, 21 of the 23 analyzed variables followed a Gaussian distribution, either before or after transformation. Electrophoretic results and reference intervals are reported in Table 2. Chemistry analysis and serum protein electrophoresis reference intervals for males and females are reported in Tables 3 and 4, with extended data reported as Tables S2-S5. Significantly higher cholesterol concentrations were seen in male JFB, whereas significantly higher glucose and γ-globulin were seen in female JFB. Figure 2 shows the fractionation of serum proteins with clear subdivision into 6 fractions (albumin, α1-, α2-, β1-, β2-, and γ-globulins) in both males and females.

View Figure

• Download as Powerpoint
• Download Figure

Discussion

This study establishes reference intervals for serum biochemistry analytes and protein electrophoresis in Jamaican fruit bats (A. jamaicensis). Currently, the only biologic dataset for this species contains reference intervals for CBC values.¹⁰ Serum protein electrophoresis is performed to determine more specific fractions of albumin and globulins compared with serum biochemistry, allowing for a more accurate assessment of the health condition of the animal.¹⁴ In this study, we expand on our knowledge of this species to encompass serum biochemistry and electrophoretic data, providing a useful resource for investigators or veterinarians working with this species.

Our data show considerable variability in serum glucose concentrations and creatine kinase (CK). Elevated glucose levels are most likely attributed to stress hyperglycemia secondary to hand capture, which has been described in other bat species.¹⁵ Therefore, varying responses to stress may lead to the variability seen in glucose levels. Another likely contributing factor to an increased glucose concentration was the morning timing of collections. Glucose in other bat species is higher when measured at the end of their nightly feeding activity.¹⁶ Similarly, serum CK was variable and considerably high but comparable to ranges reported in Egyptian fruit bats.¹⁵ Increased CK activity is attributed to trauma to skeletal and cardiac muscle fibers and subsequent leakage of CK into the serum.¹⁷ Therefore, the higher individual CK values in our bat population are likely due to the common practice of hand capture, which stimulates flying activity in the room that can lead to microtrauma. Another less impactful variable contributing to elevated CK values is cardiocentesis, which may damage cardiac muscle fibers.

ALT, AST, and GGT levels were all considerably higher than in other species of bats. An elevated AST along with a concomitant elevation in CK is an indicator of capture myopathy.¹⁸ In contrast, ALT tends to be more specific to hepatic parenchymal cells, while GGT is indicative of biliary function. As this colony had no evidence of hepatobiliary disease and hepatic tissue appeared to be grossly normal, this likely indicates true differences in enzymatic activity or method differences.

The ALP activity and phosphorus levels were also skewed toward the higher end compared with other species of bats.^15,19 Although we focused on collecting from adult JFB in this study, young adults of this species can be difficult to differentiate from adults due to similarities in body size and a lack of knowledge of distinguishing characteristics. Therefore, the higher ALP activity can be attributed to the bone isoform of ALP in sexually immature bats. Similarly, phosphorus can be physiologically elevated in young animals to support bone development.²⁰

While most serum electrolyte ranges were similar to other species, serum potassium was consistently much higher compared with other species of bats, as well as other species where concentrations are maintained in a relatively narrow range (3.5-5.55 mEq/L).^19,21 Hyperkalemia can result from translocation of potassium from intracellular to extracellular fluid secondary to tissue breakdown from trauma.²¹ Therefore, hand capture may contribute to this process. The hyperkalemia seen in our JFB may also represent a true species difference, as seen in CD-1 mice.²²

Differences in serum chemistry and protein electrophoresis values between sexes were minimal. Cholesterol was higher in males, a difference that has been demonstrated as a natural phenomenon in rodents and primates due to sex differences in cholesterol synthesis.^23–25 Females in this study had higher glucose levels compared with males, and this may be related to sex differences in glucose homeostasis or response to stress.²⁶ Finally, γ-globulin was significantly higher in females, likely representing a true sex difference in this species.

While reference intervals can be useful in determining the impact of experimental manipulation and clinical health status, care should be taken while interpreting these results to a separate population. Many factors can influence results, including collection and sample processing methods, individual features such as signalment, genetic background, health status, and reproductive status, and environmental factors such as diet, housing, and handling. Furthermore, it is essential to establish baseline values for each bat species individually, as these can vary widely between species.²⁷ Therefore, these reference intervals are most transferable when the subject population is similar and when samples are collected and processed under similar conditions.

Freezing and thawing samples can affect certain analytic concentrations in serum, including but not limited to bilirubin, cholesterol, and certain liver enzymes. However, short-term storage at these temperatures and performing a single freeze-thaw cycle is not known to cause significant changes in the parameters we measured.^28–31 To confirm this, the samples that were run immediately were reevaluated and compared with the stored samples, revealing no notable differences. In addition, histology was not performed to confirm the presence or absence of subclinical disease that would possibly have impacted these serum chemistry results. However, our analysis included the removal of any samples reflecting a potentially unhealthy bat, and the robust outlier analysis would have likely eliminated any subclinical bats.

In summary, JFB are a unique species of bats used in biomedical research. We observed higher CK, glucose, ALT, AST, GGT, ALP, phosphorus, and potassium concentrations compared with other species of bats. While sex differences were minimal, cholesterol was higher in males, and glucose and γ-globulin were higher in females. These differences are important in interpreting biochemistry results. As the first known investigation into serum biochemical reference intervals in Jamaican fruit bats, we propose that this publication serves as a useful guide for researchers and veterinarians working with this species.

Supplementary Materials

Table S1. Analytic methods for biochemical measurands.

Table S2. Serum biochemical reference intervals (RIs) for 62 clinically normal, male, captive Jamaican fruit bats (Artibeus jamaicensis). RIs were calculated nonparametrically for all values. Data for descriptive statistics were untransformed. AG ratio, albumin-to-globulin ratio; CK, creatine kinase; G, Gaussian; LRL, lower reference limit; NG, non-Gaussian; TG, Gaussian after Box-Cox transformation; URL, upper reference limit.

Table S3. Serum electrophoresis reference intervals (RIs) for 62 clinically normal, male, captive Jamaican fruit bats (Artibeus jamaicensis). RIs were calculated nonparametrically for all values. Data for descriptive statistics were untransformed. G, Gaussian; LRL, lower reference limit; URL, upper reference limit.

Table S4. Serum biochemical reference intervals (RIs) for 72 clinically normal, female, captive Jamaican fruit bats (Artibeus jamaicensis). RIs were calculated nonparametrically for all values. Data for descriptive statistics were untransformed. AG, albumin globulin; CK, creatine kinase; G, Gaussian; LRL, lower reference limit; NG, non-Gaussian; TG, Gaussian after Box-Cox transformation; URL, upper reference limit.

Table S5. Serum electrophoresis reference intervals (RIs) for 72 clinically normal, female, captive Jamaican fruit bats (Artibeus jamaicensis). RIs were calculated nonparametrically for all values. Data for descriptive statistics were untransformed. G, Gaussian; LRL, lower reference limit; TG, Gaussian after Box-Cox transformation; URL, upper reference limit.