Effect of vitamin E and selenium supplementation on haematology , blood chemistry and thyroid hormones in male buffalo ( Bubalus bubalis ) calves *

An experiment was conducted on 20 male buffalo calves (average body weight 75.30±2.20 kg) to elucidate the effect of supplemental vitamin E and/or selenium (Se) on their haematology, blood metabolic profi le, serum enzymes and thyroid hormones status. Group I served as control (without any supplementation), groups II, III and IV were supplemented with 0.3 mg Se, 300 IU of DL-α-tocopheryl acetate, and both 300 IU DL-α-tocopheryl acetate and 0.3 mg Se, respectively. Animals were fed on wheat straw and concentrate mixture to meet their nutrient requirements. Blood was collected at zero day and subsequently at 45, 90, 135 and 180 days and analysed for haematological parameters, blood chemistry and level of thyroid hormones. Results revealed no signifi cant difference (P>0.05) in haematological parameters (haemoglobin, packed cell volume, red blood cell, white blood cell) due to the supplementation of vitamin E ,Se or both vitamin E + Se in buffalo calves. Similarly, the concentration of various blood biochemical components (glucose, total protein, albumin, globulin, urea, creatinine, total cholesterol, and triglycerides) except high density lipid (HDL) cholesterol was statistically similar in 4 groups. Results revealed signifi cantly (P<0.01) higher concentration (mg/dl) of HDL cholesterol in buffalo calves given Se and Se + vitamin E as compared to control and vitamin E supplemented groups. There was no signifi cant difference in serum enzymes (serum glutamate oxaloacetate transaminase, serum glutamate pyruvate transaminase, alkaline phosphatase, creatine kinase and lactate dehydrogenase) activities due to different treatments. Level of triiodothyronine was signifi cantly (P<0.01) higher in buffalo calves supplemented with selenium, and vitamin E + Se, as compared to non-supplemented control group; whereas concentration of thyroxine (nmol/l) was statistically similar in 4 groups, without showing * This research work was carried out under AP Cess Fund Ad Hoc scheme supported by Indian Council of Agricultural Research, Krishi Bhawan, New Delhi (India) 1 Corresponding author: e-mail: rsd@ivri.up.nic.in Journal of Animal and Feed Sciences, 18, 2009, 241–256 242 VITAMIN E AND SE SUPPLEMENT THYROID HORMONES IN CALVES any effect of different treatments. It may be concluded that supplementation of Se and vitamin E in the diet of buffalo calves had no benefi cial effect on blood haematology and blood chemistry except an increase in the serum levels of high density lipid cholesterol and triiodothyronine as compared to non-supplemented control buffalo calves.


INTRODUCTION
The nutritional essentiality of selenium arose from the work of Patterson et al. (1957) in chickens.Subsequently, it was found that vitamin E and selenium are essential for proper health, immunity and reproductive functions of animals, and their inter-relationship became an active and rewarding fi eld of research, as vitamin E and Se are the micronutrients that share a common biological function.Selenium is a component of glutathione peroxidase enzyme, which destroys free radicals in the cytoplasm, whereas, vitamin E is a non-enzyme scavenger of free radicals that functions as a specifi c lipid soluble antioxidant in cell membranes.Subsequently, it was confi rmed that Se and vitamin E act synergistically and protect the tissues against oxidative damage.Both Se and vitamin E have been shown to improve immune responses (Shinde et al., 2007).Selenium is also a component of enzyme type I deiodinase that is required for the conversion of thyroxine (T 4 ) into more active tri-iodothyronine.NRC (1989) has recommended a dietary level of 0.3 ppm Se and 15-40 mg/kg DM of vitamin E for growing cattle.However, research suggests that relatively higher levels of vitamin E supplementation may improve calf performance (Reddy et al., 1987), which may be due to enhanced immunity (Rajeesh et al., 2008).Selenium is required for the development and expression of non-specifi c, humoral and cell mediated immune responses.However, the mechanism by which Se affects the immune system is speculative.Probably, the effects of Se through glutathione peroxidase and on the cellular levels of reduced glutathione, as well as the ability of selenium to interact with cell membranes, represent the immune-enhancing role of Se.As such, there appears to be no standard report on the requirement of Se and vitamin E for buffaloes and often values recommended for cattle are used for buffaloes.Keeping in view that very little information is available on the requirement and role of vitamin E and Se on blood haematology, blood chemistry and thyroid hormones in buffaloes, the present research work was planned to evaluate the effect of vitamin E and Se supplementation on these aspects in male buffalo calves.

Animals, feeding and management
To carry out this experiment, twenty male buffalo (Bubalus bubalis) calves were procured from the local market.These animals were maintained on the experimental diet comprising of concentrate mixture and wheat straw for a period of two months during which they were treated against ecto and endo parasites before the start of the experiment, and subsequently at regular intervals.All the calves were vaccinated against foot and mouth disease.The buffalo calves were distributed in four different groups of fi ve animals in each on the basis of their body weights following complete randomized design.During the experimental period all the calves were kept in a well ventilated shed with individual feeding and watering arrangements.Calves in all the four groups were fed on wheat straw and concentrate mixture to meet their nutrient requirements for a daily weight gain of 500 g (Kearl, 1982).The concentrate mixture consisted of, %: crushed maize grain 25, soyabean meal 30, wheat bran 42, mineral mixture 2 and common salt 1. Treatments were: group I (control), without any supplementation, group II supplemented with 0.3 mg Se/kg DM from sodium selenite, group III supplemented with 300 IU DL-α-tocopheryl acetate (Impextraco, Belgium) and group IV supplemented with both 0.3 mg Se/kg DM and 300 IU DL-α-tocopheryl acetate, mixed well in the concentrate mixture.Available green fodder (2 kg) was given once a week to all the calves to meet their vitamin A requirement.Clean and fresh drinking water was provided twice a day to all the animals.The chemical composition of wheat straw and concentrate mixture is presented in Table 1.The crude protein content of the concentrate mixture and wheat straw was 20.4 and 3.6%, respectively, whereas the basal α-tocopherol and Se concentration were 13.15, 2.00 and 0.17 and 0.13 mg kg -1 , respectively.

Collection of blood
About 12 ml blood was collected from each calf through jugular venipuncture in the morning (before watering and feeding) at zero day and subsequently at 45 days interval.Out of 12 ml, 10 ml blood was collected into clean and dry test tube and kept in slanting position for 45 min for the separation of serum for blood chemistry and thyroid hormones.Remaining 2 ml was taken in another clean and dry ependroph tube (2 ml) containing anticoagulant (heparin) for the haematological studies.The blood samples were centrifuged at 3000 rpm for 10 min at 4 o C and serum was separated.The serum was collected in plastic vials and kept at -40 o C until further analysis.

Estimation of α-tocopherol in feeds
The concentration of α-tocopherol in concentrates mixture and wheat straw offered to the experimental calves was estimated by using HPLC.The α-tocopherol standard was procured from M/s Sigma (USA), and was diluted using ethanol (Merck, Germany).Methanol (HPLC grade) was used as a mobile phase to maintain a fl ow rate of 2 ml/min and α-tocopherol was detected at 292 nm.

Estimation of selenium in feeds
Selenium in feed samples was estimated by Atomic Absorbance Spectrophotometer (Model 4141, Electronic Corporation of India Ltd., Hyderabad, India) using a nitrous oxide-acetylene fl ame, nitrogen as inert gas and sodium borohydrate (0.6% w/v in 0.5% NaOH) as a reducing agent.Samples of feed were digested using double acid (HNO 3 , HClO 4 ; 4:1) mixture and volume was made with double distilled water.

Estimation of proximate principles and fi bre fractions in feed
Feed samples were analysed for proximate principles and fi bre fractions using standard methods.

Estimation of haematological parameters
Blood kept for haemoglobin estimation was treated with Drabkin's solution.Ferricyanide from Drabkin's solution converted haemoglobin to methaemoglobin, which was further converted to cyanmethaemoglobin by the action of cyanide.The cyanmethaemoglobin has an absorbance which is proportional to the haemoglobin concentration (g/100 ml blood).To estimate packed cell volume (PCV) Wintrobe tubes were fi lled with blood up to the ten mark and centrifuged at 3000 rpm for 30 min.PCV was expressed as percentage.Haemocytometer was used to determine the total leukocyte count (TLC) and total erythrocyte count (TEC) in blood.Blood was drawn up to the 0.5 mark of the WBC pipette followed by WBC diluting fl uid up to 11 marks.The pipette was then rotated between the fi ngers for a few seconds in order to facilitate proper mixing of the contents.After few minutes, which allowed haemolysis of RBCs, the counting chamber was charged after discarding the fi rst few drops of the diluted sample.Once the cells were settled down, WBCs were counted in the four large squares.This number multiplied by 50 gave the TLC (10 3 cells/ µl).Anticoagulated blood was drawn up to the 0.5 mark of the RBC pipette followed by RBC diluting fl uid up to 101 marks.The pipette was then rotated between the fi ngers for a few seconds in order to facilitate proper mixing of the contents.After few minutes, the counting chamber was charged after discarding the fi rst few drops of the diluted sample.Once the cells were settled down, RBCs were counted in the fi ve squares.This number multiplied by 10 6 gave the TEC (10 6 cells/µl).

Estimation of serum biochemical components
Diagnostic kits manufactured by Span Diagnostic Limited, Surat (India), were used for the analysis of all the blood biochemical components.Serum total protein (TP) and albumin were estimated by biuret and BCG dye binding method (Dumas et al., 1971), whereas serum globulin was calculated by subtracting serum albumin from TP. Urea in serum was determined by diacetylmonooxime (DAM) method and serum creatinine was determined by the alkaline picrate method.High density lipoproteins were obtained in the supernatant after the centrifugation of the serum.Cholesterol was estimated according to Wybenga et al. (1970).The amount of triglycerides present in serum samples was estimated by following the enzymatic method as given by McGowan (1983).

Estimation of serum enzymes
Serum alkaline phosphatase (ALP) activity was estimated by the method of Kind and King (1954).Serum glutamate oxaloacetate transaminase (SGOT) and serum glutamate pyruvate transaminase (SGPT) activities in blood serum were determined (Reitman and Frankel, 1957).Serum lactate dehydrogenase (LDH) activity was estimated by the method of Wroblewski and Duean (1955), creatine kinase (CK) activity in blood serum by modifi ed standard method according to the recommendations of European Committee for Clinical Laboratory Standards and of International Federation of Clinical Chemistry (Szasz et al., 1976).

Estimation of serum thyroid hormones
Estimation of triiodothyronine (T 3 ) and thyroxin (T 4 ), in blood serum samples was done by radioimmunoassay technique using an automatic gamma counter (Packard, USA, Model Cobra II).RIA kits for T 3 and T 4 were procured from M/s Immunotech Radiova (Czech Republic).

Statistical analysis
Data generated was analysed using SPSS (1996) by one way analysis of variance.Treatment means are presented along with standard errors of the mean (SEM).Individual animal was the experimental unit for analysis of all the parameters.When signifi cant differences were noticed, means were separated by using Duncans test.The P-values compared the differences between different groups.

RESULTS
Results of haematological parameters studied in this experiment are presented in Table 2.The mean haemoglobin (Hb) values were in normal range and were found to be comparable (P>0.05) in all the four groups.It was found that the Hb values were signifi cantly (P<0.001)lower at zero day as compared to other periods, followed by a signifi cant (P<0.001) increase on 45 d; and on 90, 135 and 180 days Hb values were found to be comparable.The cumulative group mean PCV (%) values were found to be statistically (P>0.05)similar in all the four groups.The mean treatment and period RBC values (at 0, 45, 90, 135 and 180 d) were also statistically (P>0.05)similar in 4 groups.Results revealed no signifi cant effect of treatment and time intervals on the total erythrocyte count and total leukocyte counts in blood of buffalo calves.
The data concerning blood glucose, total protein, albumin, globulin, albumin: globulin ratio, total cholesterol, HDL cholesterol, triglyceride, urea and creatinine are presented in Table 3. Results revealed no signifi cant effect of treatments on blood glucose level.Serum glucose level was found to be signifi cantly (P<0.001) higher at 45 and 90 days of experimental feeding as compared to 0, 135 and 180 d.The mean total protein (TP) values in groups I, II, III and IV were found to be statistically (P>0.05)comparable.Mean albumin concentrations were statistically (P>0.05)similar in 4 groups and the overall period means of serum albumin were found to be signifi cantly (P<0.001)lower at zero day of experimental feeding at 2.47 mg/dl, followed by a signifi cant (P<0.001) increase on 45 d, whereas it was highest (3.24 mg/dl) on 180 d of experimental feeding.The mean serum globulin values were statistically (P>0.05)alike in all the 4 groups of calves.The mean albumin: globulin ratio in different treatments were also found to be statistically (P>0.05)alike.The mean serum total cholesterol concentration (mg/dl) were numerically higher in Se, and vitamin E + Se supplemented groups as compared to control and vitamin E supplemented groups, but did not reveal any statistical (P>0.05)differences.The mean HDL cholesterol values were found to be signifi cantly (P<0.01)higher in group II (supplemented with 0.3 mg Se/kg DM) and group IV(supplemented with both Se 0.3 mg/kg DM intake and vitamin E 300 IU/head/day) as compared to group III (supplemented with vitamin E 300 IU/head/day) and group I (control).HDL cholesterol values at zero day of experimental feeding were signifi cantly (P<0.001)lower followed by a signifi cant increase on 45 and 135 d and signifi cantly higher values were recorded on 180 days of experimental feeding.There was a signifi cant (P<0.02)treatment x period interaction regarding HDL cholesterol values.At 90 days of experimental feeding the HDL values in group II were found to be signifi cantly (P<0.05)higher as compared to group I and III, but were found to be comparable to that of group IV.Further, it was observed that at 180 d the HDL cholesterol values were signifi cantly (P<0.01)higher in groups II and IV than control, but HDL cholesterol values of group II and IV were found to be comparable to that of group III.Although numerically higher triglyceride (mg/dl) values were noted in all the supplemented animals as compared to control, but on statistical comparison they were found to be similar.The overall mean values for serum urea and creatinine (mg/dl) did not differ (P>0.05)among different treatments and at different time intervals.
The results of the blood enzyme viz., serum glutamate oxaloacetate transaminase (SGOT), serum glutamate pyruvate transaminase (SGPT), lactate dehydrogenase (LDH), creatine kinase (CK), and alkaline phosphatase (ALP) are presented in Table 4.The mean SGOT and SGPT, activities (unit/ml) did not reveal any statistical signifi cant (P>0.05)differences due to treatments.The mean LDH and CK activities (IU/L) were also comparable (P>0.05)among the different groups.The ALP (KA units) values in groups I to IV did not reveal any statistically signifi cant differences.The levels of triiodothyronin (T 3 ) and thyroxine (T 4 ) in different groups as well as at different periods has been presented in Table 5.The serum T 3 levels (nmol/l) at zero day were found to signifi cantly (P<0.001)lower at 1.15 nmol/l as compared to other periods showing a trend of a progressive increase towards the end of the experiment.The mean serum T 3 levels were found to be 1.50, 1.84, 1.66 and 1.92 nmol/l, respectively in groups I, II, III and IV.The statistical analysis revealed that the serum T 3 levels were signifi cantly (P<0.01)higher in group II and IV as compared to control, whereas the T 3 levels between group I and III were found to be comparable.The overall mean values of T 4 (nmol/l) were 39.69, 42.86, 40.16 and 44.14, in groups I, II, III and IV, respectively.The statistical analysis revealed a non-signifi cant difference (P>0.05) when different groups were compared.When different periods were compared the mean T 4 value was found to be signifi cantly (P<0.001)lower on zero day as compared to other periods, while the highest values were observed on 180 d.The T 4 : T 3 ratio were found to be comparable (P>0.05)among the different treatments as well as in different periods.

DISCUSSION
Results indicated that the supplementation of Se, vitamin E or both had no effect on the haematological parameters of buffalo calves.Similar results were reported earlier by Bednarek et al. (1996), who observed that the erythrocytes count, haemoglobin and PCV values of the vitamin E and Se supplemented calves were similar to that of control group.Contrary to the above fi ndings, Qureshi et al. (2001) reported signifi cantly (P<0.05)higher haemoglobin concentration, RBC count and PCV values in buffaloes supplemented with vitamin E and selenium.
The overall mean values of serum glucose (mg/dl) were found to be similar (P>0.05) in the four groups and were within the normal range, indicating that the supplementation of Se, vitamin E, or both had no effect on the serum glucose level.Similar results were reported by Cipriano et al. (1982), who did not fi nd any signifi cant (P>0.05)effect of supplemental vitamin E (500 mg dl-alpha tocopheryl acetate per day) on blood glucose level in calves.However, Nayyar et al. (2003) reported signifi cantly (P<0.01)higher level of blood glucose in anoestrous buffalo heifers supplemented with vitamin E + Se as compared to control group.Contrary to above, Singh et al. (2002) observed low blood glucose concentration in buffalo calves fed wheat straw containing high Se (8.54 ppm).
The mean total protein (TP) values in groups I, II, III and IV were found to be statistically (P>0.05)comparable.Similar to our fi ndings, Cipriano et al. (1982) and Reddy et al. (1987) did not observe any signifi cant effect of supplemental vitamin E on the serum total protein values in calves.Similar fi nding were reported by Arthur et al. (1988), where in they did not fi nd any effect of Se supplementation on plasma albumin levels in Friesian steers.Similarly, Reddy et al. (1987) also did not observe any effect of supplemental vitamin E on the serum albumin indicating that supplemental vitamin E, Se or both had no effect on the serum total protein levels of buffalo calves.The mean serum globulin and albumin: globulin ratio in different treatments were also found to be statistically (P>0.05)alike in four groups without showing any effect of Se, vitamin E or Se + vitamin E supplementation.Similarly, Reddy et al. (1987) also did not fi nd any effect of graded levels of vitamin E supplementation on serum globulin and A: G ratio.Similarly, Arthur et al. (1988) in Friesian steers and Singh et al. (2002) in buffalo calves reported that Se supplementation (0.1 and 8.54 ppm, respectively), had no effect on the serum globulin and A: G ratio.
The mean values of total cholesterol, although numerically higher in Se and vitamin E + Se supplemented groups as compared to control and vitamin E supplemented groups, but did not reveal any statistical (P>0.05)differences.Similarly, Njeru et al. (1995) reported that graded levels of vitamin E supplementation had no effect on the cholesterol levels of yearling cattle.Similarly, Arthur et al. (1988) in Friesian steers and Singh et al. (2002) in buffalo calves reported that Se supplementation at a level of 0.1 and 8.54 ppm, respectively, had no effect on the blood cholesterol levels.Contrary to our fi ndings, Nayyar et al. (2003) reported that supplementation of vitamin E (500 IU) and vitamin E + Se (500 IU vitamin E + 2 mg Se) per head per day to anoestrous buffalo heifers caused a signifi cant (P<0.01)increase in the levels of cholesterol, as compared to unsuppleemnted anoestrous buffalo heifers.The mean HDL cholesterol values were found to be signifi cantly (P<0.01)higher in group II (supplemented with 0.3 mg Se/kg DM) and group IV (supplemented with both Se 0.3 mg/kg DM intake and vitamin E 300 IU/head/day) as compared to group III (supplemented with vitamin E 300 IU/head/day) and group I (control).These results showed that Se and vitamin E supplementation increased the serum HDL cholesterol levels and the increase was more pronounced due to Se supplementation as compared to vitamin E supplementation.Similarly, Mudgal et al. (2008) also reported that Se supplementation in buffalo calves signifi cantly (P<0.05)increased their HDL cholesterol levels.The reason for an increase in the serum HDL cholesterol values due to Se and vitamin E supplementation may be due to the fact that Se is essential to maintain the integrity of pancreas and hence effi cient digestion and absorption of fats, whereas regarding vitamin E, it is transported though circulation as components of lipoproteins, and cattle maintain majority of their serum vitamin E in the HDL fraction.
The overall mean triglyceride values (mg/dl) though numerically higher in all the supplemented animals as compared to control, but on statistical comparison they were found to be similar.Similar to our fi ndings, Njeru et al. (1995) reported that supplementation of graded levels of vitamin E to yearling cattle had no effect on the serum triglyceride concentration.Contrarily, Wojcicki et al. (1991) reported that supplementation of vitamin E partially reduced the serum triglyceride levels of adult rabbits which were fed a high fat diet.
Non-signifi cant (P>0.05)variation in serum urea and creatinine in 4 groups of buffalo calves was similar to the fi ndings of Reddy et al. (1987) who did not notice any effect on serum urea and creatinine values of calves, supplemented with graded levels of vitamin E. Cipriano et al. (1982) also did not observe any signifi cant differences (P>0.05) in the serum creatinine levels of calves that were fed a control, vitamin E supplemented (500 mg/day) or vitamin E defi cient diet.Mudgal et al. (2008) also reported that supplementation of Se to buffalo calves had no effect on their serum urea and creatinine levels.Since serum urea and creatinine are indicators of the normal physiological status and N metabolism of the animals, hence it can be concluded that supplementation of Se, vitamin E or both had no effect on the nitrogen metabolism of buffalo calves.
Non-signifi cant variation in the activities of SGOT and SGPT were similar to earlier fi ndings of Samanta and Dass (2007) and Mudgal et al. (2008), who did not fi nd any effect of supplemental vitamin E and Se on the activity of these enzymes in crossbred and buffalo calves, respectively.Contrary to present fi ndings, Kursa and Kroupova (1976) reported signifi cantly higher activity of SGOT and SGPT in serum of calves given extra vitamin E (15-30 mg/kg) in milk replacer.Regarding Se supplementation, Singh et al. (2002) observed that buffalo calves fed high Se (8.54 ppm) had higher activity of plasma SGOT and SGPT.The levels of vitamin E and Se that were present in the basal diet used in this experiment may be suffi cient to maintain the normal enzyme activities of SGOT and SGPT and hence supplemental vitamin E and Se did not show any effect on the SGOT and SGPT activities in buffalo calves.
LDH isoenzyme profi les were the fi rst to be used in veterinary medicine to detect organ injury, whereas CK have also been used as a sensitive indicator to detect muscle damage in clinical cases.Elevated levels of LDH and CK are suggestive of a tentative diagnosis of vitamin E and/or Se defi ciency.Supplementation of vitamin E and Se were found to decrease the activity of these enzymes in buffalo calves (Amer et al., 1986); but, in the present experiment, it was found that supplemental vitamin E and/or Se had no effect on the serum activities of LDH and CK, which might be due to the fact that the levels of these micronutrients in the basal diet was suffi cient enough to maintain the normal activities of these enzymes.The ALP (KA units) values observed in this study were not affected by treatments, and Hatfi eld et al. (2002) also did not fi nd any effect of supplemental vitamin E (330 IU) in ewes on their serum ALP activity.Contrary to this, Kursa and Kroupova (1979) reported higher ALP activity in calves supplemented with vitamin E as compared with control calves.With regards to Se supplementation, Mudgal et al. (2008) did not fi nd any effect of supplemental Se (0.3 ppm) on the ALP activity of buffalo calves.Contrary to these, Singh et al. (2002) reported higher ALP activity in buffalo calves fed wheat straw containing 8.54 ppm Se, where as Arthur et al. (1988) reported a decrease in ALP activity in Friesian steers fed a Se defi cient diet (0.015 ppm).
Results indicated that buffalo calves that were supplemented with Se and vitamin E+Se, had signifi cantly (P<0.01)higher levels of T 3 as compared to control.Similarly, Nayyar et al. (2003) also reported that buffalo heifers that were supplemented with vitamin E and vitamin E + Se had signifi cantly (P<0.01)higher levels of T 3 as compared to unsuppleemnted control animals and the increase in the levels of T 3 was reported to be due to supplementation of Se.Arthur et al. (1988) reported that the peripheral concentration of T 3 was reduced and that of T 4 was increased in calves that were fed a synthetic diet that was defi cient in Se as compared to the calves that were fed the same diet supplemented with Se.Nonsignifi cant variation in the concentration of serum T 4 and T 3 : T 4 ratio observed in the present study was similar to the fi ndings of Donald et al. (1994) in ewes supplemented with different levels of Se (0-8 mg) as an aqueous solution of sodium selenite.This may be due to the fact that type I iodothyronine-5´-deiodinase is a Se dependent enzyme, which is responsible for the deiodination of T 4 to T 3 .

CONCLUSIONS
It may be concluded that supplementation of selenium and vitamin E in the diet of buffalo calves had no benefi cial effect on blood haematology and blood chemistry except an increase in the serum levels of high density lipid cholesterol and triiodothyronine as compared to non-supplemented control buffalo calves.

Table 1 .
Chemical composition of feeds, % DM basis

Table 2 .
Haematological parameters of buffalo calves in different groups

Table 3 .
Blood biochemical profi le of male buffalo calves in different groups

Table 4 .
Serum enzyme profi le of buffalo calves in different groups

Table 5 .
Thyroid hormone profi le of buffalo calves in different groups