Plasma leptin concentration in donkeys

Abstract

Background: Donkeys appear to be more predisposed than large breed horses to suffer from hyperlipemia. The reason for that predisposition is unknown but anorexia is a consistent feature of the disease. Leptin, a protein synthesized in fat tissue, is one of the major inhibitors of appetite in mammals.

Objective: We hypothesized that donkeys could have elevated plasma leptin concentrations compared to horses.

Animals and Methods: Blood samples were obtained from 50 donkeys for measurement of leptin, triglycerides (TGs), glucose, and insulin. Glucose/insulin ratio, modified insulin to glucose ratio, and reciprocal of the square root of insulin were calculated. Based on their body condition score (BCS), donkeys were classified as lean (n = 18), normal (n = 16), or overweight (n = 16). The results were compared with reference values from our laboratory and with a group of horses (n = 25) used as an internal control.

Results: Values of both leptin and TGs in donkeys were above the horse reference range and also significantly higher than those of the control horses: leptin (11.2 ± 1.7 versus 5.8 ± 0.5 µg/L, p < 0.05) and TGs (0.93 ± 0.1 versus 0.54 ± 0.1 mmol/L, p < 0.01). Overweight donkeys had leptin (19.3 ± 2.9 µg/L) and TG (1.3 ± 0.2 mmol/L) concentrations that were significantly (p < 0.01) higher than normal (9.4 ± 3.3 µg/L and 0.85 ± 0.1 mmol/L, respectively) and lean (5.5 ± 1.0 µg/L and 0.66 ± 0.1 mmol/L, respectively) donkeys. A significant positive correlation (p < 0.01) was found between BCS and leptin (r = 0.43), TGs (r = 0.46), glucose (r = 0.41), and insulin (r = 0.40).

Conclusion: Donkeys have higher plasma leptin concentrations than horses and leptin is correlated with BCS.

Keywords:

• leptin
• donkey
• hyperlipemia
• insulin resistance

1. Introduction

Hyperlipemia is a serious condition that develops secondary to anorexia and is highly prevalent in donkeys (Moore et al. 1994; McKenzie 2011). The reason for donkey predisposition is unknown. It is often argued that donkeys are thriftier and, when overfed, develop overweight and become predisposed to hyperlipemia. However, hyperlipemia may appear in donkeys that are not obese. A prominent feature of hyperlipemia is fasting, which is known to decrease sensitivity to insulin (Forhead and Dobson 1997). A positive correlation between insulin and triglycerides (TGs) has been reported in hyperlipemic donkeys (Forhead et al. 1994). Thus, insulin resistance could be involved in the pathophysiology of hyperlipemia in donkeys. Donkeys have been described to be less insulin-resistant than ponies or horses (June et al. 1992). However, a recent study by Mclean (McLean et al. 2009) has reported that, in donkeys, parameters related to insulin resistance are within reference range of horses. Therefore, additional mechanisms, different from insulin resistance, may be involved in the predisposition to hyperlipemia exhibited by donkeys.

Since anorexia is a consistent feature in the development of hyperlipemia, factors that regulate appetite may play a role in the pathophysiology of this disease. Leptin, a protein synthesized in fat tissue, is one of the major inhibitors of appetite in mammals (Radin et al. 2009). Leptin is known to be positively correlated with body condition score (BCS) and negatively correlated with insulin sensitivity in horses (Frank et al. 2006; Kearns et al. 2006).

We hypothesize that donkeys have elevated plasma leptin concentrations compared to horses. Thus, the objective of this study is to investigate plasma leptin concentrations in donkeys and to compare them with horses.

2. Materials and methods

2.1. Animals

Blood samples from 50 adult healthy donkeys were collected in three farms located in southern Spain between March and May 2009. The sampled population included a variety of local donkey breeds. Mean age was 7.8 ± 5.3 years (range 1–20 years). The study group included 41 females and nine males. All donkeys lived in small paddocks and were fed grass supplemented with hay and/or concentrate.

Based on their BCS donkeys were classified as lean (n = 18), normal (n = 16), or overweight (n = 16). In a scale from 1 to 9 (Pearson and Ouassat 1996), lean donkeys had BCS <5 (3.9 ± 0.2), normal donkeys BCS = 5, and overweight donkeys BCS >5 (6.2 ± 0.5).

Blood samples were also obtained from 25 healthy horses with a BCS ranging from 5 to 7 (6.1 ± 0.7). Horse plasma samples were used as an internal control and were run in parallel with the donkey samples in all biochemical assays. All horses, 19 males and six females, were Andalusians with an age of 4.6 ± 3.3 years (range 2–15 years), lived in the same farm, and received a diet similar to the donkeys.

Animals were determined to be healthy based on known history, physical exam, and hematology. Donkeys and horses were required to have normal (<144 pmol/L) insulin concentrations to be included in the study. All experimental procedures were approved by the Ethics Committee of the University of Cordoba.

2.2. Sampling

Blood samples were obtained from all animals before the morning meal. Samples were collected from the jugular vein into 10 mL heparin, 2 mL fluoride, and 2 mL EDTA tubes. Samples were maintained on ice until arrival to the laboratory. A complete blood count was performed upon arrival using the EDTA samples. Plasma was separated from the heparin and fluoride tubes and immediately frozen at −20°C.

2.3. Biochemistry measurements

Glucose (from the fluoride tubes) and TGs (in heparinized plasma) were measured by spectrophotometry (Vettest 8008, Idexx laboratories, Inc.). Leptin and insulin were measured in heparinized plasma by radioimmunoassays (Linco Research RIA kits) validated for horses (Kearns et al. 2006; Gordon et al. 2007) and previously used in donkeys (Salimei et al. 2007). Glucose/insulin ratio (G/I), modified insulin to glucose ratio (MIRG), and reciprocal of the square root of insulin (RISQI) were calculated as follows:

For comparison, non SI units were used for glucose and insulin in these proxies.

2.4. Statistics

Statistical analysis was performed by using the computer package SPSS for Windows 11.0.1 (SPSS Inc., Chicago, USA). Donkeys were segregated in three groups based on their BCS. A Kolmogorov–Smirnov test was carried out to test for normality. All variables under study, except leptin and insulin, were normally distributed. One-way analysis of variance followed by a Fisher LSD test was used to assess the inter-group differences of normally distributed variables. The Mann–Whitney test was used to compare leptin and insulin values in the three groups in a one to one basis. Comparisons between donkeys and horses were conducted using t-tests. The Pearson test was used for the study of correlations. A p value < 0.05 was considered significant. Values are presented as the mean ± standard error.

3. Results

Plasma concentrations of glucose, insulin, and the glucose–insulin proxies in donkeys were within the reference ranges for horses. In contrast, values of both leptin and TGs were above the horse reference range. When compared with the horses used as an internal control, donkeys also had significantly higher plasma concentrations of leptin (11.2 ± 1.7 versus 5.8 ± 0.5 µg/L, p < 0.05), TGs (0.93 ± 0.1 versus 0.54 ± 0.1 mmol/L, p < 0.01), glucose (4.73 ± 0.1 versus 3.97 ± 0.1 mmol/L, p < 0.01), and insulin (49.5 ± 3.6 versus 25.1 ± 5.0 pmol/L p < 0.01) (Table 1).

Table 1. Plasma concentrations of leptin, TGs, and parameters related to glucose metabolism (see Section 2 for details) in the donkeys under study and in the horses used as an internal control.

	Leptin (µg/L)	TG (mmol/L)	Glucose (mmol/L)	Insulin (pmol/L)	G/I	MIRG	RISQI
Donkeys n = 50	11.2 ± 1.7^a	0.93 ± 0.1^a	4.73 ± 0.1^a	49.5 ± 3.6^a	15.0 ± 0.8^a	4.3 ± 0.2	0.42 ± 0.1^a
Horses n = 25	5.8 ± 0.5	0.54 ± 0.1	3.97 ± 0.1	25.1 ± 5.0	37.2 ± 4.6	3.9 ± 0.7	0.69 ± 0.5
Horse reference range	<7	0.05–0.57	4.4–6.7	<144	>10	<5.6	>0.32
Notes: ^a p < 0.05 versus horses. Horse reference range represents the normal range of the parameters under study in our laboratory based on n = 100 control horses.

Values of the parameters under study in donkeys with different BCS are presented in Table 2. Although plasma leptin, TGs, glucose, and insulin were consistently lower in lean donkeys no significant differences were found between lean and normal donkeys. By contrast, overweight donkeys had leptin (19.3 ± 2.9 µg/L) and TG (1.3 ± 0.2 mmol/L) concentrations significantly (p < 0.01) higher than normal (9.4 ± 3.3 µg/L and 0.85 ± 0.1 mmol/L, respectively), and lean (5.5 ± 1.0 µg/L and 0.66 ± 0.1 mmol/L, respectively) donkeys. A significant positive correlation (p < 0.01) was found between BCS and leptin (r = 0.43), TG (r = 0.46), glucose (r = 0.41), and insulin (r = 0.40). BCS was also negatively (p < 0.01) correlated to RISQI (r = −0.43).

Table 2. Plasma concentrations of leptin, TGs, and parameters related to glucose metabolism in the donkeys under study.

	Leptin (µg/L)	TG (mmol/L)	Glucose (mmol/L)	Insulin (pmol/L)	G/I	MIRG	RISQI
Lean donkeys n = 18	5.5 ± 1.0^a	0.66 ± 0.1^a	4.48 ± 0.1^a	35.2 ± 2.2^a	17.2 ± 1.0^a	3.9 ± 0.2^a	0.46 ± 0.01^a
Normal donkeys n = 16	9.4 ± 3.3^a	0.85 ± 0.1^a	4.68 ± 0.1^a	46.6 ± 7.2^a	15.8 ± 1.3^a	4.1 ± 0.3	0.43 ± 0.02^a
Overweight donkeys n = 16	19.3 ± 2.9	1.3 ± 0.2	5.07 ± 0.1	66.7 ± 7.9	11.8 ± 1.3	4.9 ± 0.4	0.35 ± 0.02
Notes: Donkeys were allocated into three groups (lean, normal and overweight) based on their BCS (see Section 2 for details).
^a p < 0.05 versus overweight donkeys.

4. Discussion

This study was designed to test the hypothesis that leptin, an adipokine that regulates appetite, has higher plasma concentrations in donkeys (a species predisposed to anorexia-induced hyperlipemia) than in horses. Our results demonstrate elevated plasma leptin concentrations in donkeys and a significant correlation between BCS and leptin.

Since insulin resistance and hyperinsulinemia could influence leptin concentrations (Cartmill et al. 2003; Caltabilota et al. 2010), only donkeys with baseline insulin lower than 144 pmol/L were included in the study. Donkey leptin values and other parameters under study were compared with our laboratory reference ranges in horses. In addition, a group of horses was also studied as “internal control” to account for inter-assay variation (i.e., to confirm that inter-assay variation was not responsible for excessively high or low values in any analyte). This population of horses was intentionally selected to be in the normal-high range of BCS to avoid any artifact derived from a lower adiposity in the control group.

Plasma leptin concentrations were higher in donkeys than in horses. As previously reported in horses (Buff et al. 2002; Carter et al. 2009), BCS was positively correlated with plasma leptin concentrations. Only lean donkeys had plasma leptin concentrations within horse reference range, while many “normal” and most “overweight” donkeys exhibited very high-leptin concentrations. It is important to note that BCS of the donkeys (5.0 ± 0.1) was lower (p < 0.01) than BCS of the horses (6.1 ± 0.1). Thus, even though a clear relationship between BCS and leptin concentration has been found, species differences and not obesity itself seems to be the main factor in the hyperleptinemia found in our study.

Leptin has been previously measured in donkeys (Magistrelli and Rosi 2006; Salimei et al. 2007). These studies reported values of 5.6 ± 0.4 µg/L in pregnant lactating dairy asses with a BCS = 3.5/5 (Salimei et al. 2007) and 6–8 µg/L in lactating non-pregnant dairy asses with BCS = 3.2/5 (Magistrelli and Rosi 2006). Our leptin values are slightly higher, however comparison between studies is difficult since previous reports included animals in which reproductive status could have influenced their results. Both lactation and pregnancy have been reported to reduce leptinemia in equids (Heidler et al. 2003).

Although our study only included donkeys with normal insulin values, there was some relationship between leptin and insulin resistance as evidenced by the high correlation found between plasma leptin and insulin (r = 0.68, p < 0.01) and the higher leptin concentrations found in donkeys with lower G/I and RISQI and higher MIRG values. Since the hyperleptinemic donkeys also had the higher BCS, our results reinforce the concept that adiposity, hyperleptinemia and insulin resistance are closely related in this species. This is similar to horses as hyperleptinemic horses have been found to be less insulin sensitive than horses with normal leptin concentrations of the same body condition (Cartmill et al 2003; Caltabilota et al. 2010).

There is some controversy about the normal range of insulin concentrations in donkeys and how to define insulin resistance in this species. Mclean et al. (2009) obtained values related with insulin resistance in donkeys that were within reference range for horses. However, June et al. (1992) reported a higher insulin response to glucose tolerance test in donkeys than in horses or ponies. Donkey insulin concentrations in this study were within reference range for horses and higher than those reported in Dugat et al. (2010). Basal concentrations of insulin can be altered by numerous factors such as: stress, type of food and time since last meal (Firshman and Valberg 2007) which may explain the differences between studies.

TGs in the donkeys of this study were significantly higher than in control horses and reference values for horses. This is in agreement with a previous report by Dugat et al. (2010). The elevated TG values found in donkeys are not a consequence of obesity since, as mentioned above, the BCS of the donkeys was lower than the BCS of the control horses. However, as previously reported (Watson et al. 1990), within our donkey population TGs were also positively correlated with BCS. It is interesting to note that the correlation between TGs and BCS is not clear in horses (Kedzierski and Kapica 2008; Kedzierski et al. 2010).

In conclusion, donkeys appear to have higher plasma leptin concentrations than horses and hyperleptinemia is correlated with BCS in donkeys. Future studies should be aimed to elucidate the role of these elevated leptin concentrations in the pathophysiology of metabolic problems like hyperlipemia.