Original article

Peculiarities of vitamin D and of the calciumand phosphate homeostatic system in horses

Alexander Breidenbach. Christina Schlumbohm,Johein Harmeyer

Physiologisches Institut, Tierärztliche Hochschule Hannover, Bischofsholerdamm 15,30173 Hannover, Germany

(Received 28 october 1997; accepted 5 December 1997)

Abstract - The aim of the present study was to investigate the importance of putative regulatoryfactors of the calcium (Ca) and inorganic phosphate (Pj) homeostatic system in the horse. The con-centrations of Ca, Pi, vitamin D metabolites, parathyroid hormone (PTH), the activity of thealkaline phosphatase (AP) and the concentration and binding properties of vitamin D binding pro-tein (DBP) were measured in the plasma. In addition, the ability of the renal cortex to hydroxy-late calcidiol into 24,25(OH)!D-j and 1,25(OH)ZD3 was evaluated in vitro. The plasma concen-tration of Ca (3.2 ± 0.15 mmol ! L!! , N= = 1 00) showed no significant differences between differenthorse breeds and was not influenced by Ca intake, exercise or by indoor maintenance. The con-centration of plasma Pi which ranged from 0.58 to 1.99 mmo1 . L-! was negatively correlated withage and positively correlated with the P content of the feed. AP activities in plasma rangingfrom 131 to 852 U - L-1 were also negatively correlated with age and tended to be higher inhorses than in other domestic animals. Plasma concentrations of calcidiol and 24,25(OH)2Dwere much lower than in most other mammals and birds. The concentration and binding propertiesof DBP to calcidiol were not markedly different from those of other mammals. The mean plasmaconcentration of calcitriol (55 ± 24 pmol . L-1, N 19) was much lower than in other mammals.The plasma concentration of PTH was 218 8 181 ng ’ L-1, In renal cortex homogenates, only25-hydroxycholecalciferol-24-hydroxylase activity could be detected (V.a,:0.42 ± 0.11 I pmol . min-] . mg-I protein; Km: 373 ± 263 nmol . L 1). In conclusion, this study pro-vided evidence that in contrast to other species, vitamin D does not appear to play a key role inregulating Ca and P! homeostasis in horses. © Inra/Elsevier, Paris

horse / vitamin D / calcium homeostasis / phosphate homeostasis / renal 25-hydroxychole-calciferol-hydroxylases

* Correspondence and reprintsTel.: (49) 51 856 7626; fax: (49) 511 8567687; e-mail: jharmey@physiology.tiho-hannover.de

Résumé - Particularités du système homéostatique de la vitamine D, du calcium et duphosphate chez le cheval. Le but de cette étude était de déterminer l’importance des facteurs poten-tiels intervenant dans la régulation du système homéostatique du calcium (Ca) et du phosphate inor-

ganique (Pi) chez le cheval. Nous avons mesuré, dans le plasma, les concentrations en Ca, P,, enmétabolites de la vitamine D, en hormone parathyroïdienne (PTH), ainsi que l’activité de laphosphatase alcaline (AP) et la concentration et les propriétés ligantes de la protéine liant lavitamine D (vitamin D binding protein, DBP). De plus, nous avons déterminé in vitro, la propriétédu cortex rénal à hydroxyler le calcidiol en 24,25(OH)2D et en 1,25(OH)zD. La concentration plas-mique en Ca (3,2 ± 0,15 mmo1 . L-!, n = 100) n’était pas significativement différente entre les dif-férentes races équines et n’était pas modifiée par la prise orale de Ca, l’exercice ou le maintienen écurie. La concentration plasmique en Pi variait entre 0,58 et 1,99 mmol - L-1 et était corrélée

négativement avec l’âge mais positivement avec la concentration en P dans la nourriture. L’acti-vité de l’AP dans le plasma variait entre 131 et 852 U - L-!, était corrélée négativement avec l’âgeet apparaissait plus importante chez les chevaux que chez les autres animaux domestiques. Laconcentration plasmique en calcidiol et en 24,25(OH)ZD est beaucoup plus faible que chez la majo-rité des autres mammifères et oiseaux. En revanche, la concentration en DBP et les propriétésligantes de la DBP au calcidiol n’étaient pas différentes de celles des autres mammifères. Laconcentration plasmique en 1,25(OH)ZD (55 ± 24 pmol - L-!, n = 19) était plus faible que chezles autres mammifères. La concentration plasmique en PTH était de 218 ± 181 ng ’ L-!. Dans deshomogénats de cortex rénal, est seule mesurable l’activité 25-hydroxycholecalciferol-24-hydroxy-lase. Cette étude apporte donc de nouvelles preuves que, chez le cheval, et contrairement aux autresespèces, la vitamine D ne semble pas jouer un rôle primordial dans la régulation de l’homéosta-sie du calcium et du phosphate. @ Inra/Elsevier, Paris

cheval / vitamine D / homéostasie du calcium / homéostasie du phosphate / 25-hydroxy-cholecalciferol-hydroxylases rénales

1. INTRODUCTION

The equine calcium (Ca) and inorganicphosphate (Pi) homeostatic system differsin many aspects from that of most mam-malian species. Although the Ca concen-tration in horse blood is regulated as accu-rately as in the blood of other speciesincluding non-reproducing birds [1, 21 itsconcentration is remarkably high, rang-ing from 2.75 to 3.25 mmol - L-!. Thisrange must be regarded as hypercalcemicwhen compared with other species inwhich the concentration ranges from 2.3 to2.5 mmol - L-1. Such high Ca concentra-tions are found in blood of different

breeds, e.g. thoroughbreds [22], warm-bloods [14] and ponies [12]. The high Caconcentration is paralleled by a remark-ably high concentration of ionized cal-cium (Ca2+) [32]. This high plasma Caconcentration, however, is not the onlypeculiar feature of the equine Ca and Pihomeostatic system. The high Ca level iscomplemented by an unusually low con-centration of Pi in plasma which ranges

from 0.7 to 1.7 mmol - L-’ [6, 22, 32]. Inaddition, hormonal control of intestinalabsorption of Ca appears to be poor. Incontrast with other species, elevation ofCa in feed leads to an almost linearincrease in intestinal absorption of Ca [27].

If one looks at the three hormones, cal-citriol, parathyroid hormone (PTH) andcalcitonin, which are responsible for reg-ulating Ca and Pi metabolism, it is inter-esting to note that the circulating concen-tration of calcidiol in horse plasma is alsoremarkably low (- 10 nmot. L-1) !26, 301when compared with other mammalianspecies, including humans (50 to

I 15 nmot - L 1) [4, 16]. This is worth men-tioning since the calcidiol concentrationin plasma usually functions as a sensitiveand reliable indicator of the vitamin D sta-tus [25]. The low concentrations of cal-cidiol, usually present in horses would beconsidered rachitogenic if present in othermammals. Despite this fact, spontaneousdevelopment of rickets or osteomalacia israre in horses, if not completely absent.El Shorafa et al. [I I] have demonstrated

that it was very difficult to induce rickets

experimentally in horses by either vita-min D or sunlight deprivation or by bothmeans.

In view of these peculiarities, it wasthe aim of the present study to furtherinvestigate the regulatory factors of theequine Ca and Pi homeostatic system withparticular emphasis to the physiologicalrole of vitamin D (lack of an identifyingsubscript indicates both vitamins D2 and

D3) in regulating the Ca and Pi metabolic

system.

2. MATERIALS AND METHODS

2.1. Animals

2.1.1. Field studies

Plasma samples were collected from 100horses, which came from 21 different farms.The age of the animals ranged from I to 30years. Daily intakes of Ca, phosphate and vita-min D were estimated from their concentra-tions in feeds. These values were obtained fromcommon feed tables or from the listings whichwere attached to the feed products. Mean daily yintake of vitamin D in the horses of this studywas 0.65 ± 0.29 pmol (N = 85) but reached avalue of 5 and 8.7 pmol in some cases (N = 4).Samples of 40 mL of heparinized blood werecollected from each horse by venipuncture priorto the morning feed. The samples were imme-diately cooled on ice, then centrifuged and theplasma was stored at-20 °C for further analy-ses.

2.1.2. Vitamin D metabolites

Volumes of 20 mL of plasma were collectedfrom 13 horses and I I ponies, which were ran-domly selected with respect to breed, age (0.8to 20 years) and sex (1 1 females, 10 geldingsand 3 stallions). The animals were fed purehay and had free access to water and paddocks.The heparinized blood was cooled on ice, cen-trifuged and the plasma was stored at-20 °C.

2.1.3. 25-Hydroxycholecalciferol-hydroxylases

Six ponies (3 females and 3 geldings) from6 to 18 years of age and 147 to 310 kg bodyweight were used for the experiments. Two ofthe ponies were fed pure hay and four ponieswere given a mixture of 10 % hay and 90 %concentrate. Daily intakes of Ca and phosphatewere 10 and 5 g per 100 kg BW (pure haygroup) and 11 and 10 g (hay-concentrate group),respectively. The animals were killed by stun-ning and bleeding 5 h after the morning meal.The left kidneys were removed from the abdom-inal cavity 3 min after the end of bleeding.

2.1.4. Vitamin D binding protein

Plasma was collected from three horses.The animals had been brought to the clinic forhorses at the School of Veterinary Medicinein Hannover because of orthopedic disorders.

2.1.5. Parathyroid hormone

Plasma samples were collected from threehorses on six occasions throughout the year.In all samples, the Caz+ concentration was mea-sured (ISE-Analyzer 987S, AVL, Bad Hom-burg, Germany).

2.2. Analytical procedures

2.2.1. Calcium, inorganic phosphateand activity of alkalinephosphatase (AP) in plasma

Ca, Pi and activity of AP were photome-trically determined by use of test kits

(Boehringer, Mannheim, Germany). Determi-nation of AP activity was included in this studybecause of the unusually low concentrationsof calcidiol and calcitriol in the plasma. Alldeterminations were carried out in duplicate. Ifthe two values differed by more than 6 %, themeasurement was repeated again in duplicate.

2.2.2. 250HD and 24,25(OH)2Din plasma

Aliquots of stored frozen plasma (I to

4 mL) were used for the assays. The procedure

consisted of two purification steps, followedby radiometric quantification of individual vita-min D metabolites. First, a liquid/liquid parti-tion (extraction) according to a modified pro-cedure outlined by Hollis and Frank [15] wascarried out, followed by extraction of the super-natants. These were further purified in a twostep solid phase extraction on porous Bond-Elut@ Cis and Bond-Elut° silica cartridges(ICT-ASS Chem GmbH, Bad Homburg, Ger-many). The extraction procedure yielded threeeluate fractions containing 250HD,24R,25(OH),D and Ia,25(OH),D. 250HDand 24R,25(OH)ZD were quantified byradioimmunoassay using an antibody raised insheep against calcitriol-semicarbazide. Thisantibody cross-reacted with both metabolites.For details see Clemens et al. [7]. Concentra-tions of the metabolites were calculated by useof standard curves which were set up withmetabolites kindly provided by Hoffmann-LaRoche (Basel, Switzerland).

To determine the concentration ofI a,25(OH),D in the plasma, a receptor bindingassay based on the calf thymus vitamin Dreceptor (VDR) was used. Preparation of thereceptor and the assay followed essentially themethod outlined by Horst et al. 1 7 The con-centrations of calcitriol in plasma were calcu-lated by use of standard curves.

2.3. Renal 25-hydroxycholecalciferol-hydroxylase activities

Measurement of calcidiol hydroxylation invitro followed the procedure described by Win-kler et al. [33J. In brief, supernatants of renalcortex homogenate were incubated withincreasing amounts of calcidiol (final concen-trations between 29 and 2445 nmol ! L-1) and3.7 kBq of 25-hydroxy-[26,27-methyl-3H]-cholecalciferol (Amersham Buchler, Braun-schweig, Germany) for 10 min in a shakingwater bath (37 °C). The reaction was stoppedby adding of 6 mL of methanol. Samples wereextracted using the method of Bligh and Dyer[SJ. The efficiency of metabolite extractionwas controlled by adding of 0.8 kBq of labeledcalcidiol and calcitriol to extra aliquots ofsupernatant in which hydroxylation of calci-diol was prevented by adding methanol at thestart of the experiment. Next, 50 ng each, ofunlabeled 24,25(OH)!. 25,26(OH)ZD3 andi,25(OH)!D., were added to the labeled

hydroxylation products of calcidiol to markthe position of labeled products on the UV-chromatogram. This mixture was separated byHPLC (Waters Millipore, Eschborn, Germany)under straight phase conditions on a Zorbax-Silo column with 5 pm particles (Knauer, BadHomburg, Germany). Eluate fractions werecounted by liquid scintillation (Tri-Carb 460C,Canberra Packard, Frankfurt/Main, Germany). ).

In two samples from two horses, radioac-tivity and UV-absorption of the eluate werecontinuously recorded by simultaneous UV-and radioactive flow through detection (LB507A, Berthold, Wildbad, Germany) with thetwo detectors mounted in series (figure 4).

The quantity of newly hydroxylated prod-ucts was calculated from the amount of radioac-

tivity and its specific radioactivity in the eluate.It was expressed as pmol of product formedper min and per mg of protein. Recovery ofthe vitamin D metabolites after extraction wastaken into account. Michaelis-Menten con-stants (Km) and maximal turnover rates (vmaX>of calcidiol hydroxylations were calculatedfrom plots according to methods outlined byLineweaver and Burk [24].

2.4. Vitamin D binding protein

The binding capacity and dissociation con-stant of DBP were determined according toKaune et al. [21 L In brief, a 250 pL amountof diluted plasma was combined with increas-ing amounts of 250HD3 (from 1.56 to 801 1nmol - L-1) and with 66.7 Bq of labeled250HDj. After mixing and equilibrating inpolypropylene tubes for 18 to 24 h at 4 °C, thetracer bound to DBP was separated by adsorp-tion of the unbound fraction to activated char-coal. After centrifugation, the radioactivity wascounted in 250 p L aliquots of supernatant byliquid scintillation. Total radioactivity (T), non-specific binding (BN) and zero binding (B!)were measured in quadruplicate. Specific bind-ing of 250HD3 to DBP (Bs, pmol - L-1) wascalculated by subtracting BN from the radioac-tivity measured in each supernatant. The max-imal number of binding sites (Bmax) and thedissociation constant (Ka) for 250HD3 were

computed after plotting B, versus free 250HDj 3

(F). F was calculated by subtracting B. fromtotal unlabeled 250HD3 concentration. Thedata were fitted to a double rectangular hyper-bola (GRAPH PAD, San Diego, CA, USA).

2.5. Parathyroid hormone

A two-sited immuno-radiometric assaybased on two different polyclonal antibodiesfrom goats raised against human intact PTH(Nichols Institute Diagnostics, San Juan Capis-trano, CA, USA) was used to determine theconcentration of intact PTH in horse plasma.Concentrations of intact PTH were calculated

by the use of standard curves which were set upconcurrently with the samples. The resultsobtained are relative values due to the use of a

heterologous PTH for the standard curve.

2.6. Chemicals

All chemicals and reagents, unless indicatedotherwise, were purchased from Merck (Darm-stadt, Germany). Activated charcoal was pur-chased from Sigma Chemical Co. (St. Louis,MO, USA). Labeled calcidiol and

24,25(OH)-,D! were from Amersham-Buchler(Braunschweig, Germany).

2.7. Statistics

Data are expressed as arithmetic means (x)and standard deviations (SD) unless indicatedotherwise. Linear regression analysis was per-formed to i) test the intluence of phosphate infeed or calcitriol concentration in plasma onP! concentration in plasma, ii) test the rela-tionship of Ca content in feed with plasma Caconcentration and iii) examine the influenceof PTH concentration in plasma on ionizedplasma Ca. Nonlinear regression analysis wasused to test the influence of age on Pi concen-

tration and on AP activity in plasma, respec-tively.

3. RESULTS

3.1. Plasma concentration of calcium

(Ca), phosphate (P!) and activityof alkaline phosphatase (AP)

The Ca concentration in plasma was3.2 ± 0.15 5 mmol . L-1. Despite great vari-

ability Ca content in feed (from 0.39 to1.1 % in dry matter (DM)) and in vitaminD uptake, no relationship was observedbetween the Ca concentration in plasmaand the Ca content in feed or vitamin D

uptake. The age of the animals also hadno significant influence on plasma Ca con-centration.

The concentration of P, in the plasmaranged from 0.58 to 1.99 mmol - L- anddecreased significantly with age (figure l ).There was a positive correlation

(y = 1.02x + 0.71 mmol - L r = 0.23,P < 0.05) between its concentration in theplasma and its content in the feed (from0.16 to 0.39 % in DM). Plasma Ca con-centration was unaffected by the P con-tent in feed. Activity of AP in plasmashowed great variability (from 13 to852 U - L-1) but was found to be nega-tively related to age figure 2). Horse breedhad no influence on Ca, Pi or activity ofAP in plasma (table n.

3.2. Calcidiol (250HD), 24,25(OH)2D,calcitriol (1,25(OH)!D) andvitamin D binding protein (DBP)

Plasma concentrations of calcidiol were

very low and often (N = 15) below thelower detection limit of the method (4.7to 22 nmol - L-1), which varied accordingto the plasma volume used for the assay(table II). Examination of 1,25(OH),Dconcentrations in 19 horses by the recep-tor assay yielded plasma concentrationsfrom 26 to 106 with a mean of 55 ± 24

pmol . L-1 (table 1!. Vitamin D metaboliteconcentrations were not different betweenhorses and ponies and showed no corre-lation with plasma Ca, Pi (figure 3), breed,sex or age of the animals (data not shown).

In 13 of the 22 plasma samples,24,25(OH),D concentration was near thelower end of the calibration curve andcould not be quantified accurately. In the

remaining samples the concentrationranged from 2 to 34 nmol . L-1 (table 11).

3.3. Renal 25-hydroxycholecalciferol-la-hydroxylase and25-hydroxycholecalciferol-24-hydroxylase activities

Renal cortex homogenates from sixponies showed no I a-hydroxylase activitywhen incubated with labeled calcidiol as

precursor for 10 min. No radioactivelylabeled 1,25(OH)zD! could be detected inthe lipid extracts prepared from the tissue

homogenates after separation on HPLC.This was true for both sexes.

Crude renal tissue converted, however,labeled 250HD3 into 24,25(OH)ZD3 indi-cating the presence of an active 24-hydrox-ylase. Depending on the substrate con-centration, 1.5 to 19 % of the labeled

precursor was converted into

24,25(OH):P3’V max and the Km of the renal 24-

hydroxylase were 0.42 ± 0.11 pmol - min- I. i-ng-i protein and 373 ± 263 nmol ! L-1,respectively. Type of diet (hay or con-centrate) had no effect on 24-hydroxylaseactivity and there was also no influenceexerted by the sex of the animal.

The HPL-radio- and -UV-chro-

matograms, simultaneously recorded afterseparation on a Zorbax-Si]O column,showed a peak at the position of

24,25(OH)2D3 but no signal at position of25,26(OH)!D! or 1,25(OH)ZD3 (figure 4).

3.4. Vitamin D binding protein

The binding capacities of DBP for cal-cidiol in plasma (B max) of the three horseswere 3.5, 5.5 and 10.1 umoi L-1, and the

corresponding dissociation constants (Kd)of the binding protein were 0.12, 0.15 and0.18 pmol. L-1.

3.5. Parathyroid hormone

In 18 plasma samples, the mean PTHconcentration was 218 ± 181 ng ’ L-!. Thecorresponding mean Ca2+ concentrationwas I .59 ± 0.07 mmol - L-1 (adjusted topH 7.4). Figure 5 illustrates a negativecorrelation between PTH and Ca2+.

4. DISCUSSION

4.1. Ca, Pi and AP in plasma

The plasma Ca concentration of 3.2mmol - L-1 found in this study underlinesthe view that the concentration of extra-cellular Ca in horses and ponies is higherthan in all other domestic animals. Little,if any, influence was exerted on plasmaCa by breed, by intake of Ca, vitamin D orby maintenance conditions, such as exer-cise or exposure to sunlight. This agreedwith findings of others. No effect onplasma Ca concentration (x -2.88 mmol - L-1, N = 4) was found ingrowing ponies fed diets containing 1.5,

0.8 and 0.15 % of Ca [27]. Deprivation ofdietary vitamin D and/or sunlight did notresult in unphysiological concentrationsof plasma Ca (x = 3.09 mmol - L-1) in 12 2young Shetland ponies [11]. Thus, it islikely that the lack of correlation in ourstudy between plasma Ca concentrationand Ca intake was probably not primar-ily due to the uncertainties associated withthe estimation of Ca concentrations infeed. It is still unknown which factors ormechanisms are responsible for the controlof the plasma Ca in horses. Though quiteunresponsive to many influencing factorsit shows much variation and does not

appear to be as precisely controlled as inother species. The negative correlation

between P uptake and plasma Ca concen-tration which was reported earlier [28, 29]was not seen in this study. This might bedue to less marked variation of P intakein our study as compared with the exper-iment of Schryver et al. [28, 29].

Plasma Pi concentrations behaved dif-

ferently from plasma Ca. They respondedto P intake [28, 29] and declined with age(figure 1 ). The decline appears to be char-acteristic for horses. The Pi levels in

plasma, however, showed no relationshipto vitamin D intake (figure 3). The con-

centrations were remarkably low whencompared to other species [3, 9, 23]. Thismight have to do with the high Ca con-centration present in plasma of these ani-mals, since an inverse relationship hasbeen shown to exist between plasma Piand Ca, e.g. in pigs (own observations).

As with P!, the activity of AP in plasma(ranging from 131 to 852 U - L-1) wasalso negatively correlated with age (P <

0.0001,figure 2). It showed, however, nosignificant relationship with plasma Ca or

Pi concentrations. It is not clear at presentwhether the comparably high activity ofAP in horses when compared with otherspecies [8-10] is either related to the highCa concentration in plasma or to the lowconcentrations of vitamin D metabolites

(see below).

4.2. Vitamin D metabolites in plasma

The remarkably low concentrations of250HD which were found in 24 horsesand ponies demonstrated that the physio-logical significance of vitamin D in theseanimals differed from that of other species.

Clearly, the equine circulating level of cal-cidiol cannot serve as an indicator of vita-min D status when the standards derivedfrom other species are applied. In otherspecies, e.g. in humans, pigs, chicken, etc.,such low concentrations of calcidiol wouldindicate a condition of serious vitamin D

deficiency.

Calcidiol is the metabolic precursor forthe biologically active form of vitamin D,i.e. calcitriol. Therefore, it was not too

surprising that the concentration of cal-citriol in horse plasma was also found tobe unusually low. The calcitriol concen-trations of 55 ± 24 pmol - L-1 were even

lower than those usually seen in a pigstrain with an inherited defect of renal 1 a-

hydroxylase activity. These pigs, whichsuffer from pseudo vitamin D deficiency,type I rickets, have calcitriol concentra-tions in plasma of 72 ± 26 pmol . L-1 [20].This level is too low to protect the ani-mals from developing florid rickets. Buthorses despite having even lower concen-trations of calcitriol do not develop anysymptoms of rickets. Nephrectomizedhumans have circulating calcitriol con-centrations of up to 48 pmol ! L-1 [19].Horses, however, are capable of main-taining supraphysiological Ca concentra-tions in their plasma irrespective of suchlow circulating concentrations of vitaminD hormone.

Since none of the horses suffered from

any vitamin D related metabolic bone dis-

orders, the question arises as to what thephysiological significance of calcidiol andcalcitriol in horses is.

4.3. Renal calcidiol hydroxylases

Renal 1 a-hydroxylase activity couldnot be detected in vitro, even in the pres-ence of substrate concentrations about 70times higher than those present in plasma.Production of 24,25(OH)ZD3, however,occurred always. It remains unansweredwhether the lack of renal I a-hydroxylaseactivity resulted from a down regulation ofthe enzyme due to the high circulating lev-els of plasma Ca or whether renalla-hydroxylase is virtually absent inequidae.

The small amounts of calcitriol whichcould be detected in the horse plasmacould be explained by the presence ofextrarenal hydroxylases. This has beendeduced from studies with nephrectomizedhumans [ 19] and pigs [ 16] which showedsimilar concentrations of calcitriol in

plasma as seen in our horses. Theextrarenal synthesis of calcitriol, however,

is not sufficient in humans and pigs to pre-vent them from developing rickets orosteomalacia.

4.4. DBP in plasma

Since the possibility of a reduced affin-ity of DBP to calcidiol could compensatefor a low total calcidiol concentration in

plasma it appeared important to examinethe concentrations of its free and biologi-cally diffusable forms. Our results for theBmdX and Kd of DBP showed, however,that the free form of calcidiol must be aslow as its total concentration. Both, Bmaxand Kd of DBP did not differ significantlyfrom those of other species, such as piglets[2 1 orruminants [31].

4.5. Parathyroid hormone

PTH concentrations appeared to be lowin horses under basal conditions when

compared with sheep [ 13] or cows [ 18]and exhibited a negative relationship withplasma Ca2+ (figure 5). The statistical sig-nificance of this relationship between PTHand Caz+ concentration in plasma, shouldnot be comfounded by the method. Thefact that only relative values of PTH wereobtained by our method could perhapsaffect the slope of the relationship but notits statistical significance. Thus, the find-ing suggests that the renal 1 a-hydroxy-lase activity is probably down regulated,and indicates that, in horses, PTH is per-haps more closely involved in Ca and P!homeostasis than it is the case in other ani-mal species.

4.4. General conclusions

The present findings showed that theregulatory mechanisms in horses for keep-ing the plasma Ca and Pi concentrationsconstant differ markedly from most other

mammals, including all domestic animalspecies and birds. Evidence was providedthat due to their low circulating concen-trations, calcidiol and calcitriol are prob-ably not required for maintaining the Caand Pi concentrations in plasma at theirphysiological levels. Although many fea-tures of the vitamin D- and Ca-system dif-fer between horses and other mammals,horses have also things in common. Thisincludes the concentration and bindingproperties of DBP in plasma and the activ-ity of renal 25-hydroxycholecalciferol-24-hydroxylase. It also includes the occur-rence of soft tissue calcifications whichcan develop in horses following theadministration of pharmacological dosesof vitamin D. Further studies are requiredfor better understanding of the regulatoryfactors which are responsible for control-ling the Ca and Pi homeostasic system inhorses.

ACKNOWLEDGMENTS

The authors gratefully acknowledge thehelp of Dr M. Neunlist and Dr M. Venner intranslating the abstract.

REFERENCES

[ 11 J Basudde C.D., Humphreys D.J., The effectof the active principle of Solanum mala-coxylon on rabbits and the inhibition of itsaction by actinomycin D, Calcif. Tissue Res.18 (1975) 133-139.

[21 Basudde C.D., Humphreys D.J., The effectof the administration of Solanum malacoxy-lon on the chick, Res. Vet. Sci. 18 (1975)330-331. 1 .

[3] Bell N.H., Epstein S., Shary J., Greene V.,Oexmann M.J., Shaw S., Evidence of a prob-able role for 25-hydroxyvitamin D in the reg-ulation of human calcium metabolism, J.Bone Miner. Res. 3 (1988) 489-495.

[4] Bikle D.D., Gee E., Halloran B., KowalskiM.A., Ryzen E., Haddad J.G., Assessmentof the free fraction of 25-hydroxyvitamin D inserum and its regulation by albumin and thevitamin D-binding protein, J. Clin.Endocrinol. Metab. 63 (1986) 954-959.

[51 Bligh E.G., Dyer W.J., A rapid method oftotal lipid extraction and purification, Can.J. Biochem. 37 (1959) 911-917.

[6] Boyd J.W., The interpretation of serum bio-chemistry test results in domestic animals,Vet. Clin. Pathol. 13 (1984) 7-14.

[71 Clemens T.L., Hendy G.N., Papapoulos S.E.,Fraher L.J., Care A.D., O’Riordan J.L.H.,Measurement of 1,25-dihydroxycholecalcif-erol in man by radioimmunoassay, Clin.Endocrinol. (Oxf.) 11 (1979) 225-234.

[81 Czerwek H., Bleuel H., Normal values ofalkaline phosphatase, glutamic oxaloacetictransaminasc and glutamic pyruvic transam-inase in the serum of experimental animalsusing optimised methods, and the effects ofhaemolysis on these values, Exp. Pathol. 19(1981) 161-163.

[9] Doornenbal H., Tong A.K., Murray N.L.,Reference values of blood parameters in beefcattle of different ages and stages of lacta-tion, Can. J. Vet. Res. 52 (1988) 99-105.

[ 10] Eastman J.R., Bixler D., Serum alkaline phos-phatase: normal values by sex and age, Clin.Chem. 23 ( 1977) 1769-1770.

111] 1 ! El Shorafa W.M., Feaster J.P., Ott E.A.,Asquith R.L., Plasma calcium, phosphorus,magnesium, alkaline phosphatase and 25-OHvitamin D levels in vitamin D-deprived youngponies, J. Anim. Sci. 49 Suppl I (1979)239-240.

[12J Elfers R.S., Bayly W.M., Brobst D.F., ReedS.M., Liggitt H.D., Hawker C.D., BaylinkD.J., Alterations in calcium, phosphorus andC-terminal parathyroid hormone levels inequine acute renal disease, Cornell. Vet. 76(1986) 317-329.

[ 13J J Elias E., Shainkin Kestenbaum R., Hypocal-caemia and serum levels of inorganic phos-phorus, magnesium parathyroid and calci-tonin hormones in the last month of

pregnancy in Awassi fat-tail ewes, Reprod.Nutr. Dev. 30 (1990) 693-699.

[ 14j Geiser D.R., Faulk D., Ionized calcium in thehorse, Equine Practice 11 (1989) 25-28.

[ 15J Hollis B.W., Frank N.E., Solid phase extrac-tion system for vitamin D and its majormetabolites in human plasma, J. Chromatogr.343 (1985) 43-49.

[ 16J Horst R.L., Littledike E.T., Riley J.L., NapoliJ.L., Quantilation of vitamin D and itsmetabolites and their plasma concentrationsin five species of animals, Anal. Biochem.116(1981)189-203.

[17] Horst R.L., Reinhardt T.A., Hollis B.W.,Improved methodology for the analysis ofplasma vitamin D metabolites, Kidney Int.Suppl. 29 (1990) S28-35.

[ 18! Hove K., Cyclic changes in plasma calciumand the calcium homeostatic endocrine sys-

tem of the postparturient dairy cow, J. Dairy.Sci. 69 (1986) 2072-2082.

[19] Jongen M.J.M., van der Vijgh W.J., Lips P.,Netelenbos J.C., 1,25-Dihydroxyvitamin Dmeasurements in anephric subjects, in: Nor-man A.W., Schaefer A., Herrath V.D.,Grigoleit H.-G. (Eds.), Vitamin D, Chemi-cal, Biochemical and Clinical Endocrinologyof Calcium Metabolism, Walter de Gruyter,Berlin, New York, 1982, pp. 797-799.

[20] Kaune R., Harmeyer J., Vitamin D3metabolism in a pig strain with pseudo vita-min D-deficiency rickets, type I, ActaEndocrinol. (Copenh.) 115 (1987) 345-352.

[21] ] Kaune R., Schr6der B., Harmeyer J., Bind-ing properties of plasma vitamin D-bindingprotein and intestinal 1,25-dihydroxyvitaminD3 receptor in piglets with pseudo-vitaminD-deficiency rickets, type I: treatment effectswith pharmacological doses of vitamin D3,Arch. Biochem. Biophys (1990) 282,326-332.

[22] Keenan D.M., Changes of blood metabolitesin horses after racing, with particular refer-ence to uric acid, Aust. Vet. J. 55 (1979)54-57.

!23] ] Lachenmaier-Currle U., Harmeyer J., Pla-cental transport of calcium and phosphorusin pigs, J. Perinat. Med. 17 (1989) 127-136.

[241 Lineweaver H., Burk D. The determinationof enzyme dissociation constants, J. Chem.

Soc. 56 (1934) 658-666.

[25] Mawer E.B., Hann J.T., Berry J.L., DaviesM., Vitamin D metabolism in patients intox-

icated with ergocalciferol, Clin. Sci. (Lond.)68(1985)135-141.

[26] M5enp!! P.H., Koskinen T., Koskinen E.,Serum profiles of vitamins A, E and D inmares and foals during different seasons, J.Anim. Sci. 66 (1988) 1418-1423.

[27] Schryver H.F., Craig P.H., Hintz H.F., Cal-cium metabolism in ponies fed varying levelsof calcium, J. Nutr. 100 (1970) 955-964.

[28] Schryver H.F., Hintz H.F., Craig P.H., Phos-phorus metabolism in ponies fed varying lev-els of phosphorus, J. Nutr. 101 (1971)1257-1263.

129J ] Schryver H.F., Hintz H.F., Craig P.H., Cal-cium metabolism in ponies fed a high phos-phorus diet, J. Nutr. 101 ( 1971 ) 259-264.

[30] Smith B.S., Wright H., 25-Hydroxyvitamin Dconcentrations in equine serum, Vet. Rec.I 15 (1984) 579.

[31 1 ] Vieth R., Kessler M.J., Pritzker K.P., Speciesdifferences in the binding kinetics of 25-hydroxyvitamin D3 to vitamin D binding pro-tein, Can. J. Physiol. Pharmacol. 68 (1990)1368-1371. 1 .

[32] Weisweiler B., Twehues R., Kuczka A.,Meyer H., Harmeyer J., Einfluss hoher Vita-min-D-Dosierungen auf Calcium-, Magne-sium- und Phosphatstoffwechsel von Pfer-den, klinische und pathomorphologischeBefunde, Pferdeheilkunde 9 ( 1993) 343-352.

[33] Winkler I., Schreiner F., Harmeyer J.,Absence of renal 25-hydroxycholecalciferol-1-hydroxylase activity in a pig strain withvitamin D-dependent rickets, Calcif. TissueInt. 38 (1986) 87-94.