Blood Cell Morphology and Plasma Biochemistry of Captive Mauremys caspica ( Gmelin , 1774 ) and Mauremys rivulata ( Valenciennes , 1833 )

Metin K., Y. Başımoğlu Koca, F. Kargın Kıral, S. Koca, O. Türkozan: Blood Cell Morphology and Plasma Biochemistry of Captive Mauremys caspica (Gmelin, 1774) and Mauremys rivulata (Valenciennes, 1833). Acta Vet. Brno 2008, 77: 163-174 Morphological characteristics of peripheral blood cells, micronucleated erythrocyte counts and plasma biochemistry profile were examined in fourteen healthy captive Mauremys caspica and in twenty-three Mauremys rivulata. The size of erythrocyte cells were 19.07 × 11.68 μm and 19.76 × 11.44 μm for M. caspica and M. rivulata, respectively. Nucleus sizes were 6.50 × 5.30 μm for M. caspica and 6.79 × 5.45 μm for M. rivulata. The micronucleated erythrocyte (MNE) values were 0.0008 and 0.0037 for the males and females of M. caspica, respectively. The MNE values were 0.0002 for male and female M. rivulata. We found sex-dependent differences only in the Ca value in the blood biochemistry profile for healthy M. caspica. Sex-dependent differences were found only in albumin and P values in the blood biochemistry profile for healthy M. rivulata. No significant differences were found between males of both species in question with respect to plasma biochemistry values. However, only plasma total protein and Ca content levels differed significantly between the females of the two species. Reptilia, haematology, micronucleus, sex, plasma indicators, Ca, P, enzymes, albumin, urea Turtles inhabit all parts of the world with a temperate to warm climate and are especially abundant in the tropics and subtropics. Water turtles are found in a wide variety of habitats, including ponds, swamps, small pools thick with vegetation, lakes of all sizes, large streams and rivers. Blood biochemistry profiles and haematology are often used to assess the physiological status of lower vertebrate patients, such as fish, amphibians, reptiles, and birds. However, there is a general lack of controlled studies designed to clarify the meaning of changes in the blood chemistry of these animals compared to those of domestic mammals. Therefore, the clinical chemistry of lower vertebrates has not achieved the same degree of critical evaluation as demonstrated in domestic mammalian medicine. Many papers characterize the blood of land tortoises (Duguy 1970; Lawrence and Hawkey 1986; Garner et al.1996; Muro et al.1998; Christopher et al. 1999; Knotková et al. 2002), however, freshwater turtles are little known as to their haematology and blood biochemistry (Pagés et al. 1992; Kölle et al. 1999; Uğurtaş et al. 2003; Metin et al. 2006). Comparative studies of clinically healthy and diseased turtles may provide useful information for their management and conservation (Bolten and Bjorndal 1992; Hasbun et al. 1998). Understanding the blood composition of turtles is very important for preventing and treating many illnesses as well. Many animal species can be used as bioindicators, either for testing the effects of some chemicals in laboratory strains, or for assaying natural populations to investigate the ACTA VET. BRNO 2008, 77: 163-174; doi:10.2754/avb200877020163 Address for correspondence: K. Metin Adnan Menderes University Faculty of Science and Arts Department of Biology 09010 Aydın, Turkey Phone: +90 256 212 84 98 / 2218 Fax: +90 256 213 53 79 E-mail: kmetin@adu.edu.tr http://www.vfu.cz/acta-vet/actavet.htm presence of pollution in a territory (Cristaldi et al. 2004).The micronucleus test detects the effect of mutagenetic agents on chromosomes by the identification of acentric fragments or lagging chromosomes, those remaining separate from the nucleus (Zuniga-Gonzalez et al. 2000). The scoring of MN is simpler, requires shorter training and is less timeconsuming. Although many studies have been made on the haematology of reptiles, blood cell morphology and plasma biochemistry and micronucleus for many species are not available. Various authors have described different circulating blood cells of different reptile species (Duguy 1970; Mateo et al. 1984; Canfield and Shea 1988; Knotková et al. 2002; Azevedo and Lunardi 2003; Başımoğlu Koca et al. 2006; Metin et al. 2006). Identification of morphologic characteristics of different peripheral blood cells and plasma biochemistry profile of M. caspica and M. rivulata kept in captivity was the purpose of this study. Materials and Methods Animals Fourteen adult Caspian turtles (7 ♂♂ and 7 ♀♀) and twenty-three adult Balkan terrapins (13 ♂♂ and 10 ♀♀) were studied on a private captive breeding farm in the month of July. The specimens were kept in vessels (200 × 200 × 60 cm) at the farm and fed commercial trout food, dog food, sardines, anchovies, chopped chicken liver and sheep stomach. The mean straight carapace lengths of both species were measured. The females were checked by manual examination through the cloaca for eggs in oviducts. All of them were determined as non-pregnant. Blood sampling Plasma biochemistry analyses and blood cell morphology were analyzed in 14 adult M. caspica (7 ♂♂, 7 ♀♀) and 23 adult M. rivulata (13 ♂♂, 10 ♀♀). Blood specimens of two turtle species were taken by venepuncture from caudal vein. Blood (1-2 ml) was collected using 21 gauge needles and 5 ml syringes. Blood specimens were transferred into vacutainer tubes containing lithium heparin and placed on ice until processing in the laboratory 4-6 h after capture. Plasma was separated by centrifugation at 800 g for 10 min (Nüve NF 800 R, Turkey) and split in two or more vials. Blood cell morphology Blood samples were taken from the caudal vein. Blood smears were prepared immediately and air-dried. Wright stained blood smears were used for the measurement and assessment of blood cells. Four to five blood smears were prepared per individual. On each slide lengths (EL) and widths (EW) of randomly selected 100 mature erythrocytes and their nuclei (NL and NW), 50 thrombocytes, heterophils, eosinophils, basophils, lymphocytes and monocytes were measured by an Olympus ocular micrometer at a magnification of × 600 (Olympus BX51, Japan). Erythrocyte and nuclear sizes (ES and NS) were calculated according to formulas [(EL × EW × π) / 4] and [(NL × NW × π) / 4], respectively. In addition, micronucleated erythrocytes were counted among 1000 erythrocytes on each blood smear by the same micrometer at a magnification of × 1000. Plasma biochemistry Biochemical indices of plasma were measured spectrophotometrically (Microlab, Merck 200, Deutschland) by means of commercial kits (Biomedical Biosystems/Spain). Na, K and Cl were measured by means of an ion-selective device (Ion selective, Easy lite, England). Samples that appeared haemolysed were discarded. The following plasma was measured: aspartate aminotransferase (EC 2.6.1.1; AST), alanine aminotransferase (EC 2.6.1.2; ALT), gamma glutamyl transferase (EC 2.3.2.2; GGT), amylase (EC 3.2.1.1), total protein, albumin, globulin, glucose, creatinine, urea, triglycerides, cholesterol, Ca, P, Na, K and Cl. In addition, calculated values included globulin and albumin/globulin. Blood chemical values are expressed in SI units. Statistical analyses Haematological and biochemical indicators were summarized as mean, standard deviation (SD), standard error of the mean (SEM) and range. We used analysis of variance (ANOVA) and t-test for comparison between sexes within species, and between species within sex. Results were considered significant at p < 0.05. Statistical analyses were carried out by using STATISTICA version 6.0.

Morphological characteristics of peripheral blood cells, micronucleated erythrocyte counts and plasma biochemistry profile were examined in fourteen healthy captive Mauremys caspica and in twenty-three Mauremys rivulata.The size of erythrocyte cells were 19.07 × 11.68 μm and 19.76 × 11.44 μm for M. caspica and M. rivulata, respectively.Nucleus sizes were 6.50 × 5.30 μm for M. caspica and 6.79 × 5.45 μm for M. rivulata.The micronucleated erythrocyte (MNE) values were 0.0008 and 0.0037 for the males and females of M. caspica, respectively.The MNE values were 0.0002 for male and female M. rivulata.
We found sex-dependent differences only in the Ca value in the blood biochemistry profile for healthy M. caspica.Sex-dependent differences were found only in albumin and P values in the blood biochemistry profile for healthy M. rivulata.No significant differences were found between males of both species in question with respect to plasma biochemistry values.However, only plasma total protein and Ca content levels differed significantly between the females of the two species.Reptilia, haematology, micronucleus, sex, plasma indicators, Ca, P, enzymes, albumin, urea Turtles inhabit all parts of the world with a temperate to warm climate and are especially abundant in the tropics and subtropics.Water turtles are found in a wide variety of habitats, including ponds, swamps, small pools thick with vegetation, lakes of all sizes, large streams and rivers.
Blood biochemistry profiles and haematology are often used to assess the physiological status of lower vertebrate patients, such as fish, amphibians, reptiles, and birds.However, there is a general lack of controlled studies designed to clarify the meaning of changes in the blood chemistry of these animals compared to those of domestic mammals.Therefore, the clinical chemistry of lower vertebrates has not achieved the same degree of critical evaluation as demonstrated in domestic mammalian medicine.Many papers characterize the blood of land tortoises (Duguy 1970;Lawrence and Hawkey 1986;Garner et al.1996;Muro et al.1998;Christopher et al. 1999;Knotková et al. 2002), however, freshwater turtles are little known as to their haematology and blood biochemistry (Pagés et al. 1992;Kölle et al. 1999;Uğurtaş et al. 2003;Metin et al. 2006).Comparative studies of clinically healthy and diseased turtles may provide useful information for their management and conservation (Bolten and Bjorndal 1992;Hasbun et al. 1998).
Understanding the blood composition of turtles is very important for preventing and treating many illnesses as well.
Many animal species can be used as bioindicators, either for testing the effects of some chemicals in laboratory strains, or for assaying natural populations to investigate the presence of pollution in a territory (Cristaldi et al. 2004).The micronucleus test detects the effect of mutagenetic agents on chromosomes by the identification of acentric fragments or lagging chromosomes, those remaining separate from the nucleus (Zuniga-Gonzalez et al. 2000).The scoring of MN is simpler, requires shorter training and is less timeconsuming.
Although many studies have been made on the haematology of reptiles, blood cell morphology and plasma biochemistry and micronucleus for many species are not available.Various authors have described different circulating blood cells of different reptile species (Duguy 1970;Mateo et al. 1984;Canfield and Shea 1988;Knotková et al. 2002;Azevedo and Lunardi 2003;Başımoğlu Koca et al. 2006;Metin et al. 2006).Identification of morphologic characteristics of different peripheral blood cells and plasma biochemistry profile of M. caspica and M. rivulata kept in captivity was the purpose of this study.

Animals
Fourteen adult Caspian turtles (7 ♂♂ and 7 ♀♀) and twenty-three adult Balkan terrapins (13 ♂♂ and 10 ♀♀) were studied on a private captive breeding farm in the month of July.The specimens were kept in vessels (200 × 200 × 60 cm) at the farm and fed commercial trout food, dog food, sardines, anchovies, chopped chicken liver and sheep stomach.The mean straight carapace lengths of both species were measured.The females were checked by manual examination through the cloaca for eggs in oviducts.All of them were determined as non-pregnant.

Blood sampling
Plasma biochemistry analyses and blood cell morphology were analyzed in 14 adult M. caspica (7 ♂♂, 7 ♀♀) and 23 adult M. rivulata (13 ♂♂, 10 ♀♀).Blood specimens of two turtle species were taken by venepuncture from caudal vein.Blood (1-2 ml) was collected using 21 gauge needles and 5 ml syringes.Blood specimens were transferred into vacutainer tubes containing lithium heparin and placed on ice until processing in the laboratory 4-6 h after capture.Plasma was separated by centrifugation at 800 g for 10 min (Nüve NF 800 R, Turkey) and split in two or more vials.

Blood cell morphology
Blood samples were taken from the caudal vein.Blood smears were prepared immediately and air-dried.Wright stained blood smears were used for the measurement and assessment of blood cells.Four to five blood smears were prepared per individual.On each slide lengths (EL) and widths (EW) of randomly selected 100 mature erythrocytes and their nuclei (NL and NW), 50 thrombocytes, heterophils, eosinophils, basophils, lymphocytes and monocytes were measured by an Olympus ocular micrometer at a magnification of × 600 (Olympus BX51, Japan).Erythrocyte and nuclear sizes (ES and NS) were calculated according to formulas [(EL × EW × π) / 4] and [(NL × NW × π) / 4], respectively.In addition, micronucleated erythrocytes were counted among 1000 erythrocytes on each blood smear by the same micrometer at a magnification of × 1000.

Statistical analyses
Haematological and biochemical indicators were summarized as mean, standard deviation (SD), standard error of the mean (SEM) and range.We used analysis of variance (ANOVA) and t-test for comparison between sexes within species, and between species within sex.Results were considered significant at p < 0.05.Statistical analyses were carried out by using STATISTICA version 6.0.

Body size
The mean straight carapace lengths (SCL) of M. caspica and M. rivulata were measured 14.53 ± 3.74 cm and 15.02 ± 2.86 cm for males, 18.20 ± 3.17 cm and 15.60 ± 1.87 cm for females, respectively.Body size distributions between genders in both species were not Therefore, the effect of size was independent of the effect of sex on the blood chemistry values.
Interspecific differences in 166  3 and 4. Heterophils were easily identified by the presence of numerous elongated pinkred cytoplasmic granules.These granules were hardly packed in the cytoplasm.The nucleus was round, oval or mostly bilobed and eccentric (Fig. 1A).The nucleus of eosinophils was sometimes bilobed and eccentrically placed.The cytoplasm was filled with deep eosinophilic round granules (Fig. 1B).Basophils were smaller than heterophils and eosinophils.Basophils were characterized by the presence of round basophilic (dark blue) granules of various sizes.The round nucleus was in the centre of the cell (Fig. 1C).Lymphocytes had a compact, large and dark, centrally positioned nucleus.Light-blue thin cytoplasm covered a narrow area around the nucleus (Fig. 1D).The monocytes mostly had a kidney-shaped nucleus, which was less intense chromatin than in lymphocytes.The cytoplasm of monocyte was blue-grey and covered a larger area (Fig. 1E).Thrombocytes often clump together in blood smears.The nucleus of thrombocytes was 167 Almost the same cases were recorded in females having smaller eosinophils (F = 63.74 p < 0.0001), heterophils (F = 57.82,p < 0.0001) and basophils (F = 31.65,p < 0.0001) in M. rivulata.The sizes of leukocytes for these species were summarized in Tables 3 and 4.

Plasma biochemistry
We found significant differences only in the Ca value between genders in the blood biochemistry profile of healthy M. caspica.The mean Ca value was significantly higher in females (2.60 ± 0.42 mmol/l) than males (2.12 ± 0.28 mmol/l) (t = -2.49;df = 12; p < 0.05).Results of plasma biochemistry analyses are summarized in Table 5.
Sex-dependent differences were found only in albumin and P values in the blood biochemistry profile of healthy M. rivulata.The mean albumin value was significantly higher in females (17.73 ± 5.19 g/l) than in males (13.56 ± 2.24 g/l) (t = -2.61;df = 21; p < 0.05).Nevertheless, the mean P value was significantly higher in males (1.67 ± 0.29 mmol/l) than in females (1.36 ± 0.26 mmol/l) (t = -2.67;df = 21; p < 0.05).Results of plasma biochemistry analyses are summarized in Table 6.No significant differences were found between males of both species in question with respect to plasma biochemistry values.However, only plasma total protein and Ca content levels differed significantly between females of the two species.The mean total protein value was significantly higher in M. rivulata (30.19 ± 7.19) than in M. caspica (23.70 ± 4.65 g/l) (F (1,12) = 4.90, p < 0.05).However, the mean Ca value was significantly higher in M. caspica (2.60 ± 0.42 mmol/l) than in M. rivulata (2.20 ± 0.26 mmol/l) (F (1,12) = 5.89, p < 0.05).

Micronucleated erythrocytes (MN)
The micronucleated erythrocyte (MNE) values of M. caspica were 0.0008 for the males and 0.0037 for the females; the (MNE) values of M. rivulata were 0.0002 and 0.0002 for the males and females, respectively.The micronucleated erythrocyte (MNE) values of M. caspica and M. rivulata are given in Table 7.

Discussion
One of the most important functions of erythrocytes is to carry oxygen and carbon dioxide, and its surface area to size ratio is also a determining factor in the tissues.Thus, a small erythrocyte offers the possibility of a higher rate of exchange than a larger one (Hartman et al. 1964;Sevinç et al. 2000).According to Wintrobe (1933) the erythrocyte size reflects the position of a species on the evolutionary scale: in lower vertebrates and those with a not-so successful evolutionary past, i.e. in cyclostomes, elasmobranches and urodeles, the erythrocytes are large, but in higher vertebrates (mammals) the same cells are smaller and do not contain nuclei.
Nucleus in mature erythrocyte is round in green turtles (Chelonia mydas) (Samour et al. 1998).The only blood cell values that were comparable to those of some natural terrapins from Turkey were EL, EW, ES, EL/EW and NS/ES as taken from Uğurtaş et al. (2003).When the means of our blood cell values were tested with one sample t-test against the natural M. caspica and M. rivulata population means, the EW, ES EL/EW and NS/ES The classification of reptilian leukocytes poses many problems since these cells show morphological variation within the class and several different nomenclatures have been used to describe them (Knotková et al. 2002).For example, Taylor and Kaplan (1961) divided leukocytes into neutrophils, basophils, eosinophils, lymphocytes and monocytes on the basis of light microscopy in turtles.Saint Girons (1970) reported the presence of eosinophils, azurophils, neutrophils and plasma cells in reptiles, Sypek and Borysenko (1988) described eosinophils and heterophils in reptilian blood.Cannon et al. (1996) divided granulocytes into basophils and eosinophils on phase-contrast microscopy.Wood and Ebanks (1984) described eosinophils and neutrophils.Widely accepted opinion is that reptilian (Montali 1988) and avian heterophils (Brooks et al. 1996) have functions similar to mammal neutrophils.
According to Canfield (1998) the mammalian neutrophil is equivalent to the nonmammalian heterophil.The heterophil, excluding mammals, has coarse, red to brown, speculated to irregular granules of variable size and either a bilobed (birds and some lizards) or unlobed nucleus (most reptiles and amphibians).Azevedo and Lunardi's (2003) observations show that 2 types of eosinophilic granulocytes are present in blood of Chrysemys dorbigni.Eosinophils and neutrophils are granulocytic leukocytes present in the blood of most vertebrates.The existence of these two cell types in reptiles is a matter of controversy.To avoid confusion, some researchers suggest that the term neutrophils be restricted to mammals and the term heterophil to non-mammals (Zapata et al. 1981;Canfield 1985).In fact Azevedo and Lunardi (2003) determined 2 types of eosinophilic granulocytes which are called type I and type II for Chrysemys dorbigni.After analyzing morphological characteristics of type I, it is seen that they are similar to those of eosinophils of some birds, and in comparison of cytochemical characteristics they are similar to those of eosinophils of both birds and mammals.Type II cells are morphologically similar to those of bird heterophils, and cytochemical characteristics are similar to those of neutrophils of birds and mammals.
In this study, it appears that on the basis of light microscopic findings there are three main types of granulocytes (heterophils, eosinophils, basophils) and two main types of agranulocytes (lymphocytes and monocytes) in M. caspica and M. rivulata (Figs. 1, 2).We identified heterophils as having an eccentrically placed nucleus and being round-oval or mostly bilobed in shape.The cytoplasm was filled with numerous elongated granules.The eosinophils in the present study had a blue round or oval nucleus.The nucleus sometimes consisted of two lobes and was eccentrically placed.The cytoplasm was pink-red and filled with deep eosinophilic round granules.Basophils were round and provided with a centrally positioned nucleus.The cytoplasm was filled with large rounded granules, the colour of which varied from dark blue to black.These basophilic granules partially masked the nucleus, as previously stated by Canfield (1998).
Lymphocytes may be small, medium or large.Canfield (1998) stated that cytoplasm may contain small vacuoles and azurophilic granules.In the present study, the nucleus of lymphocytes almost filled the cytoplasm of the cell.The amount of cytoplasm was lower and of light blue colour.Monocytes are large cells with unlobed or lobed nuclei and a large amount of lightly basophilic cytoplasm.Monocytes in captive M. caspica and M. rivulata had a mostly kidney-shaped nucleus, which was less intense and contained pale chromatin.The cytoplasm was grey blue and expanded over more area.Both species contained a higher amount of lightly basophilic cytoplasm in comparison to lymphocytes.
The similarity of thrombocytes and lymphocytes in reptiles is known (Saint Girons 1970;Frye 1991).Canfield and Shea (1988) reported that thrombocyte morphology was influenced by the degree of aggregation . Saint Girons (1970) reported that thrombocytes were small, oval cells characterized by elongate, centrally located, highly chromophilic nuclei.Knotková et al. (2002) identified two types of thrombocytes in Russian tortoises, Agrionemys horsfieldi: oval with a good visible membrane, a faintly stained cytoplasm; and rectangular with small projections of lightly basophilic cytoplasm.They attributed this variability to ageing, function and artifact.The similarity of thrombocytes and lymphocytes in reptiles is known (Frye 1991).The present study reports that thrombocytes are formed in cell groups, with centrally located dark-stained nuclei and their cytoplasm is difficult to see at the light microscopic level.
There was an only plasma Ca difference detected in the plasma biochemistries between genders in M. caspica.Calcium concentration was significantly higher in females than in males.However, plasma P value between genders was not significantly different; it was found to be higher in females than in males.The elevated Ca and P concentrations in female turtles were not unexpected.Female turtles routinely mobilize Ca and P during their reproductive cycle for egg production and vitellogenesis.Female reptiles exhibit features of Ca metabolism similar to those of birds during egg production.During egg development, female reptiles exhibit hypercalcaemia in response to oestrogen and reproductive activity.This situation shows that vitellogenesis and ovulation continued during the sampling period in females and they were very active metabolically.
M. rivulata showed sex-dependent difference as to albumin and P. The Ca concentration was higher in males than in females (non-significant difference).This might be a sign of finishing ovulation and vitellogenesis in females.The higher albumin level could be the explanation of an active period during summer in females.These results coincide with previous studies.The seasonal fluctuations of the albumin concentration followed that of the total protein.It was the lowest in the spring, increased in the summer and peaking in the autumn.Active reptile species with a high metabolic rate have higher albumin concentrations (Masat and Dessauer 1968;Dessauer 1970).
The normal plasma concentration of Ca, P and albumin for turtles and tortoises in the data found in literature ranges between 0.70 -6.33 mmol/l, 0.67 -4.45 mmol/l and 2.89 -18.4 g/l, respectively (Hutton and Goodnight 1957;Jackson et al. 1974;Mosquera et al. 1976;Taylor and Jacobson 1982;Jacobson et al. 1991;Bolten and Bjorndal 1992;Pagés et al. 1992;Kölle et al. 1999;Knotková et al. 2002;Metin et al. 2006).Plasma Ca, P and albumin levels of both species in the present study were within the range reported for other turtles, although the normal plasma Ca, P and albumin concentrations vary with the species and physiological status of the reptiles and most likely other ectotherms.
No significant differences were found between males of both species in question with respect to plasma biochemistry values.However, only plasma total protein and Ca content levels differed significantly between the females of the two species.The mean total protein value was significantly higher in M. rivulata than in M. caspica.However, the mean Ca value was significantly higher in M. caspica than in M. rivulata.The physiological plasma total protein concentration of lower vertebrates is generally lower than that of mammals.Female reptiles demonstrate a marked increase in the plasma total protein concentration during active folliculogenesis.The plasma total protein concentration returns to original values following ovulation.According to data found in literature, physiological plasma concentrations of total protein in turtles and tortoises range between 22 -75 g/l (Masat and Musacchia 1965;Taylor and Jacobson 1982;Jacobson et al. 1991;Bolten and Bjorndal 1992;Kölle et al. 1999;Knotková et al. 2002;Metin et al. 2006).
The micronucleus count is an indicator of a genetic damage in mature animals.An increased number of micronucleated cells indicate poor health.However, Zuniga-González et al. (2000) suggested that in the case of new-born animals the presence of MNE could be increased, since the reticuloendothelial system might be immature in the young of some species.They also noted that the reticuloendothelial system matures with age.In some reptile species such as Crocodylus acutus, Pituophis depei, Macroclemys temminckii (Zuniga-González et al. 2000), Emys orbicularis (Metin et al. 2006), Neurergus crocatus (Basimoglu Koca et al. 2006) very low or no MNE count was found.

Table 3 .
Differential leukocyte size in peripheral blood of captive Caspian turtle, Mauremys caspica Table 4. Differential leukocyte size in peripheral blood of captive Balkan terrapin, Mauremys rivulata MALE

Table 5 .
Blood biochemistry values obtained from healthy captive Caspian turtle, Mauremys caspica MALE

Table 6 .
Blood biochemistry values obtained from healthy captive Balkan terrapin Mauremys rivulata