The Journal of Nutrition Vol. 129 No. 1 January 1999, pp. 165-169

Calcium and Oxalic Acid Kinetics Differ in Rats^1,2

Denise A. Hanes, Connie M. Weaver³, and Meryl E. Wastney^*

Purdue University, West Lafayette, IN 47907-1264 and ^* Georgetown University Medical Center, Washington DC 20007-2197

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Small molecular weight calcium salts, if absorbed intact, could provide a nutritional source of calcium in subjects with impaired absorption of calcium by the saturable pathway. An understanding of the mechanism of absorption of calcium oxalate (as a representative salt) may be important nutritionally and therapeutically. The aim of the present study was to develop models to study absorption, distribution and retention of calcium and oxalate in rats as a basis for studying calcium oxalate absorption. Labeled compounds (⁴⁵Ca and [¹⁴C]-oxalic acid) were administered to separate groups of rats orally (n = 8-11) or intravenously (n = 3-5) and blood was sampled for up to 240 min. Data were analyzed using SAAM/CONSAM. Calcium kinetics were fitted by a model with three compartments in the body and one absorption pathway from the intestine. By contrast, oxalic acid kinetics were fitted by two pools in the body and two absorption pathways from the intestine. Calcium and oxalic acid, therefore, demonstrate different absorption and distribution kinetics in rats.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Calcium is essential to human nutrition, and its absorption occurs by two routes: active and passive. Active transport occurs transcellularly with saturable kinetics and involves binding of calcium ions by a vitamin D-dependent, calcium-binding protein in the intestinal mucosa (Kretsinger et al. 1982, Pansu et al. 1983). By contrast, passive transport occurs paracellularly with nonsaturable kinetics, and a constant fraction of calcium is absorbed at high loads (Heaney et al. 1975). Absorption of calcium by either route is thought to occur only after dissociation of calcium from complexes in foods (Allen 1982), although there is a lack of experimental evidence. The models described in this paper for calcium and oxalic acid were used to compare the kinetics of double-labeled calcium oxalate (⁴⁵Ca-[¹⁴C]-oxalate). The results provide evidence that calcium oxalate is absorbed intact (Hanes et al. 1999).

Previous experiments studying kinetics of calcium absorption in rats have used soluble calcium salts and in situ, ligated gut loop procedures. Pansu et al. (1983) applied the in situ, ligated loop procedure to dissect transmural calcium transport into two components in the rat intestine, saturable and nonsaturable processes. The saturable process existed only in the proximal small intestine, was vitamin D-dependent and predominated only at lower calcium loads. By contrast, the nonsaturable component was similar in intensity throughout the small intestine and predominated at higher calcium loads. Bronner et al. (1986) performed kinetic analysis of transmural calcium transport in rats for the calcium load range 0.001-0.2 mmol, also using in situ techniques. Below 0.01 mmol calcium, absorption was nearly complete in 10-20 min and showed saturable kinetics, characteristic of transcellular active transport. When calcium load increased to 0.2 mmol, only about half the available calcium was absorbed in 2.5 h and the rate was nearly constant, characteristic of passive diffusion. Bronner et al. proposed that calcium moved by the nonsaturable process travels largely via the paracellular route at a rate of 16%/h, regardless of load. To determine the relationship between load and absorption in the whole animal, absorption was first studied as a function of calcium load.

Because of the relationship of dietary oxalate and genesis of urinary calcium oxalate stones, oxalic acid absorption kinetics have been studied to a limited extent (Prenen et al. 1984) Mild hyperoxaluria is more important in the etiology of calcium-containing renal stones than hypercalciuria (Marshall et al. 1972). In this paper, we contrast metabolism of calcium and oxalic acid as a basis against which calcium oxalate metabolism can be compared.

	MATERIALS AND METHODS

Abstract Introduction Methods Results Discussion References

Animals and protocols.  The animals and protocols are described in detail elsewhere (Hanes et al. 1999). Animal care and use procedures during all phases of the study were approved by the Purdue University Animal Care and Use Committee. Briefly, young adult male Sprague-Dawley rats weighing approximately 250 g purchased from Harlan Industries (Indianapolis, IN) underwent surgery to catheterize the jugular vein for repeated blood draws. Rats were studied following recovery to presurgery weight and <6 h food deprivation.

Absorption of soluble calcium.  Absorption of soluble calcium and oxalic acid were studied at equimolar loads for ultimate comparison with the calcium oxalate intact salt. An appropriate load was considered to be one at which passive absorption of soluble calcium was dominant, to minimize the impact of any dissociated calcium on absorption of the intact salt. To study the time course of soluble calcium absorption, rats weighing approximately 250 g (range 240-260 g) were gavaged with ⁴⁵Ca ascorbate following 6 h of food deprivation (n = 10). Unlabeled calcium ascorbate was dried overnight in a vacuum drying oven at room temperature then transferred to a dessicator prior to weighing. Ca ascorbate was mixed with ⁴⁵CaCl₂ in deionized H₂O to provide 0.025-0.375 (0.025, 0.098, 0.150, 0225, 0277, and 0.375) mmol Ca and 1.11 mBq ⁴⁵Ca (Amersham, Arlington Heights, IL) in a volume of 400 µL per rat.

Following gavage of the labeled salt, 300 µL of blood was drawn at each of 12 staggered time points (up to 12 per rat) between 0-240 min after dosing. Serum (100 µL) was obtained by centrifugation at 15,000 × g, then bleached by adding 25 µL of 3 mol KOH/L followed by 100 µL of 8.8 mol H₂O₂/L, letting the mixture stand for 30 min, then neutralizing the sample with 25 µL of 3 mol HCl/L. Fifteen mL scintillation cocktail was then mixed with each aliquot prior to counting in a Beckman LS 1800 scintillation counter (Fullerton, CA). Calcium appearance in serum was expressed as fraction of ⁴⁵Ca administered per L serum and was plotted for each rat vs. time in minutes following gavage.

Absorption of oxalic acid.  Knowledge of the time course of absorption of soluble oxalic acid along with that of soluble calcium at an equimolar load to allow comparison with the absorption profile of the insoluble calcium oxalate salt is necessary . The above experiment was repeated with [¹⁴C]-oxalic acid (n = 11) given a gavage load of 0.375 mmol oxalic acid and 1.48 mBq ¹⁴C. Blood (600 µL) was drawn at each of 6 time points up to 240 min. Because oxalic acid is not substantially metabolized by rats (Williams and Wandizilak et al. 1989), the appearance of the ¹⁴C radiolabel in the blood reflected oxalate and not a metabolite.

Distribution of calcium and oxalic acid.  To determine distribution of calcium and oxalic acid, 0.025 mmol calcium ascorbate containing 0.56-0.74 mBq ⁴⁵Ca or 0.025 mmol oxalic acid containing 0.56 mBq ¹⁴C were given intravenously (IV) through the catheter (n = 3-5). Following administration of the tracer, 300 µL of blood was drawn at each of 12 staggered time points for up to 240 min.

Kinetic modeling.  Tracer in serum (expressed as fraction of the dose/L serum) measured following oral or intravenous ⁴⁵Ca and ¹⁴C-oxalic acid were analyzed by compartmental modeling using the SAAM (Simulation Analysis and Modeling) program (Berman et al. 1983, Berman and Weiss 1978).

Notation.  The compartments represented pools in the body that turn over at distinct rates. Transfer coefficients [L(I,J), fraction/min] represented the fraction of compartment J moving into compartment I per unit time. Transport rates [R(I,J), mmol/min] represented the mass transferred per unit time and was calculated as the product of fractional transfer [L(I,J)] and compartment mass [M(J), mmol]:R(I,J) = L(I,J) × M(J).

Data fitting.  Intravenous tracer data from each rat were fitted initially to determine the distribution of calcium and oxalic acid within the body. Tracer (⁴⁵Ca and ¹⁴C-oxalic acid) measured in serum following intravenous administration were fitted initially to determine the number of compartments within the body. The initial volume distribution was determined by the model as the zero intercept of the fraction of dose/L serum.

Calculations.  Absorption was calculated using a modified compartmental model described for humans (Neer et al. 1967, Wastney et al. 1996). Calcium absorption was calculated as the ratio of tracer that moved from the gut into serum divided by the sum of material moving out of the gut compartment.

Statistics.  ANOVA was performed using SPSS (SPSS, Chicago, IL). A P value of <0.05 was considered significant.

	RESULTS

Abstract Introduction Methods Results Discussion References

Calcium kinetics. 

Intravenous administration.  The serum profile of ⁴⁵Ca (Fig. 1) following IV administration of 0.025 mmol ⁴⁵Ca ascorbate (Fig. 1A) was fitted by a three compartment model (Fig. 1C). Compartment 1 includes the serum pool where the dose was introduced, and Compartments 2 and 3 are exchange pools in the body. The initial volume of distribution (plasma plus interstial fluids of 20% of body weight) for calcium is 50 ml (At time = 0, fraction of dose/L is 20, therefore distribution of the dose is L/20 or 50 mL, Fig. 1A). Compartment 1 therefore contains serum plus some extravascular fluid. Serum calcium had a rapid turnover time ~10 min); Compartment 2, 12 min and Compartment 3, 66 min. The mass of calcium in each compartment was calculated by assuming a serum calcium concentration of 2.5 mmol/L (Corning Clinical Laboratory, 1988) in adult male Sprague-Dawley rats.

View larger version (21K): [in this window] [in a new window]   View larger version (14K): [in this window] [in a new window]   Fig 1. Calcium kinetics in rats. (A). Serum profiles of three rats distinguished by different symbols following intravenous administration of 0.025 mmol 45Ca ascorbate. (B). 45Ca serum profiles following gavage of 0.025-0.375 mmol 45Ca ascorbate for two rats at each load. Different symbols indidcate different individual rats. (C). Model for serum appearance 45Ca following intravenous administration of 0.025 mmol 45Ca ascorbate, indicated by *(IV), or gavage administration of 0.375 mmol 45Ca ascorbate, indicated by *(G). Compartment 1 represents serum, Compartments 2 and 3 calcium exchange pools in the body, and Compartment 4 the gastrointestinal tract. Values next to arrows are transfer coefficients (fraction/min). (D). Simulated average 45Ca serum profiles following gavage of 0.025-0.375 mmol 45Ca ascorbate. Each curve was obtained by averaging parameters of individual fits to 5-8 rats. (E). Predicted values (lines) and observed data (symbols) of 45Ca in serum after oral administration with three calcium loads. The predicted values were calculated using a compartmental model (Fig. 1C) modified to include two pathways for absorption from the intestine: one was saturable with Michaelis-Menten kinetics and the other nonsaturable (or constant).

Oral administration.  Serum data obtained following oral tracer administration were fitted by adding an intestinal compartment (Fig 1B). Serum profiles following oral tracer administration of soluble ⁴⁵Ca ascorbate at three calcium loads are shown: 0.025, 0.150 and 0.375 mmol. Data were fitted by allowing fractional transfer of ⁴⁵Ca from the gastrointestinal tract to serum, L(1,4), and the turnover of Compartment 1, L(2,1), to vary at each load (Table 1). The differences in shape of the serum curves could be explained by faster turnover of the serum or intestinal compartments at lower calcium loads.

The simulated average of ⁴⁵Ca plasma profiles following gavage of soluble ⁴⁵Ca ascorbate at the following calcium loads 0.025, 0.095, 0.150, 0.225, 0.277 and 0.375 mmol is shown in Figure 1D. These curves were obtained using SAAM by averaging parameters of individual modeled serum profiles of 5-8 rats for each calcium load using the three compartment model of Figure 1C. The fraction absorbed was inversely related to load until 0.277 mmol (Table 1). At higher loads fractional absorption was constant, consistent with passive absorption (Pansu et al. 1983).

Bronner et al. (1986) proposed two pathways for calcium absorption, a saturable, transcellular pathway (Vm 22 µmol Cag intestine¹h¹, Km 3.85 mM) and a passive paracellular pathway (0.16/h) based on in vitro studies using isolated intestinal loops from rats. As an alternative to the linear model described above, this hypothesis was tested against the in vivo data from the current study, by using a dynamic model with two pathways for absorption. Two models (one representing tracer and the other total calcium) were used with two calcium pools in the intestine: absorption from the first pool was represented by a saturable mechanism, whereas absorption from the second intestinal pool was first order. To fit the saturable pathway, it was necessary to convert the amount of calcium administered to a concentration by assuming a volume for the intestine: a volume of 10 ml was used. When concentration varied , this model predicted the in vivo data (Fig. 1E) and the values calculated for Vm was 0.134 mM/min and the Km was 4.6 mM. Therefore, this is consistent with the hypothesis of two absorption pathways of calcium in rats, although the value for Vm, 8 µmol/mL pd h¹, is lower than the value calculated by Bronner et al. (1986), 22 µmol/g pd h¹. The values for fractional calcium absorption were similar between the linear and dynamic models. Studies longer than 80 min would be required to more fully define the saturable versus nonsaturable pathway at each calcium load.

Oxalic acid kinetics. 

Intravenous administration.  Serum [¹⁴C]-oxalic acid profiles (Fig. 2) were fitted by a two compartment model (Fig. 2C). The volume of distribution is 1 L (from the intercept at time zero in Fig. 2A) in contrast to the 50 mL for calcium. Compartment 11 includes serum and has a turnover time of about 18 min and Compartment 12 an exchange pool in the body with turnover time of 140 min. The serum profiles for two rats (data not shown) had a short delay (4-16 min) before exchanging with Compartment 12. These rats were not fasted as long as the other three and some species from the diet may have bound incoming free oxalate ion in serum.

View larger version (16K): [in this window] [in a new window]   Fig 2. 14C-oxalic acid kinetics in rats. (A). Serum 14C profiles for individual rats distinguished by different symbols following intravenous administration of 0.025 mmol [14C]-oxalic acid in rats. (B). Appearance of oxalic acid in serum following gavage of 0.375 mmol [14C]-oxalic acid. (C). Model for serum appearance of [14C]-oxalic acid following intravenous administration of 0.025 mmol [14C]-oxalic acid, indicated by *(IV), or gavage of 0.375 mmol (14C)-oxalic acid, indicated by *(G). Compartment 11 represents plasma, Compartment 12 an [14C]-oxalate exchange pool in the body, and Compartments 14, 17 and 16 compartments in the gastrointestinal tract. Values next to the arrows or transfer coefficients (fraction/min). Means ± SD.

The mass of Compartment 11 was 1.1 µmol and Compartment 12 5 µmol oxalate, assuming a steady state oxalic acid concentration in serum of 1.22 µmol/L, similar to human serum (Williams and Wandzilak 1989).

Oral administration.  To fit serum data for rats gavaged with 0.375 mmol of ¹⁴C-oxalic acid (Fig. 2B) three intestinal compartments were added to the model (Compartments 14, 17and 16 Fig. 2C), with Compartment 14 likely the stomach and/or small intestine, 16 the large intestine and 17 an intestinal compartment where no [¹⁴C]-oxalic acid absorption occurred, but which accounted for a gut transit time of approximately 125 min. Absorption occurred from Compartments 14 and Compartment 16 (6 and 16%, respectively).

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Passive calcium transport is characterized by a constant fraction of absorption regardless of load (Bronner et al. 1986). The constant fraction of absorption above 0.277 mmol calcium load is consistent with absorption that was predominantly passive. The fractional transfer of calcium from the intestinal compartment to serum in rats administered high loads, 0.003/min or 0.18/h, was comparable to the value of 16%/h for passive calcium transport reported by Bronner et al. following in situ experiments on rat intestinal segments. They reported a maximal rate of saturable transport of 22 µmol/h pd g¹ intestine. Intestinal loops are a closed system, and therefore it is difficult to compare values directly to those in vivo, where material passes through the intesine in an open system.

With a load of 0.375 mmol calcium, serum ⁴⁵Ca showed a steady rise over time to an average peak within 50 min, followed by a slight decline due to exchange with other body pools. By contrast, the serum profile of [¹⁴C]-oxalic acid was characterized by an initial sharp rise to a peak, within 5 min following gavage, followed by a rapid decline to near baseline within 50 min continuing over the period of study. In addition, the average peak fraction of dose/L serum of [¹⁴C]-oxalic acid (0.061/L) was 5% of ⁴⁵Ca (1.32/L).

The models fitting the serum profiles following gavage of equimolar amounts of oxalic acid and calcium ascorbate were also different. Unlike soluble calcium, the oxalate model distinctly showed absorption at the two sites, possibly upper and lower intestine, although the fraction absorbed in the lower intestine was larger than in the upper gastrointestinal tract. As in rats, oxalate absorption in humans ranges from 2-30% and occurs in the small and upper intestine (Diamond et al. 1988). Another contrast in calcium and oxalate kinetics was that serum oxalic acid exchanged with only one other pool in the body and was lost by one pathway, probably to urine, whereas calcium exchanged with two other body pools and had one loss representing bone deposition, urinary and fecal excretion. The high value of volume of distribution for oxalate in Compartment 11 could have arisen from binding of the oxalate ion with positively charged species in the plasma and interstitial fluid, giving the pool a much greater capacity for oxalate ions or the equivalence of a much larger volume.

The serum profiles of both ionized calcium and calcium oxalate were different in shape and markedly different from that of soluble oxalic acid (Hanes et al. 1999), which underwent rapid clearance following entry into the plasma pool. Oxalic acid is not metabolized in the body to any appreciable extent (Williams and Wandizilak 1989). Therefore, the appearance of the ¹⁴C radiolabel in the blood reflected the behavior of oxalate and not a metabolite.

In summary, modeling soluble ⁴⁵Ca and soluble [¹⁴C]-oxalic acid serum profiles showed differences in the distribution kinetics. This makes feasible a study of calcium oxalate metabolism to determine if the salt is absorbed intact. Characterization of multiple sites of oxalate absorption is important in understanding methods, such as pharmacological agents, for reducing urinary oxalate and calcium oxalate renal stone disease.

	FOOTNOTES

Manuscript received 13 January 1998. Initial reviews completed 3 April 1998. Revision accepted 9 September 1998.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Allen H. Calcium bioavailability and absorption: A review. Am. J. Clin. Nutr. 1982; 35:783-808[Free Full Text]
• Berman, M., Beltz, W. F., Greif, P. C., Chabay R. & Boston, R. C. (1983) CONSAM User's Guide. U.S. Govt. Printing Office 421-132:3279, Washington, DC.
• Berman, M & Weiss M. F. (1978) SAAM Manual. U.S. Printing Office, [DHEW Publications No. (NIH) 78-180], Washington, DC.
• Bronner F., Pansu D., Stein W. D. An analysis of intestinal calcium transport across the rat intestine. Am. J. Physiol. 1986; 250:G561-G569[Abstract/Free Full Text]
• Corning Clinical Laboratory. (1988) Serum chemistry/complete blood counts, male Sprague-Dawley rats, 12 weeks old. Mid Atlantic Regional Laboratory, Rockville, MD.
• Diamond K. L., Fox C. C., Borch D. H. Role of cecal pH in intestinal oxalate absorption in the rat. J. Lab. Clin. Med. 1988; 112:352-356[Medline]
• Hanes D. A., Weaver C. M., Heaney R. P., Wastney M. E. Absorption of calcium oxalate does not require dissociation in rats. J. Nutr. 1999; 129:170-173[Abstract/Free Full Text]
• Heaney R. P., Saville P. D., Recker R. R. Calcium absorption as a function of calcium intake. J. Lab. Clin. Med. 1975; 85:881-890[Medline]
• Kretsinger, R. H., Mann J. E. & Simmons, J. G. (1982) Model of facilitated diffusion of calcium by the intestinal calcium binding protein. In: Vitamin D Chemical Biochemical and Clinical Endocrinology of Calcium Metabolism (Norman, A. W., Schaefer, K., Herrath, D. V. and Gringoleit, H.G., eds). pp. 233-246. DeGruyter, Berlin and New York.
• Marshall R. W., Cochran M., Hodgkinson A. Relationships between calcium and oxalic acid intake in the diet and their excretion in the urine of normal and renal-stone-forming subjects. Clin. Sci. 1972; 43:91-99[Medline]
• Neer R., Berman M., Fisher L., Roseberg L. E. Multicompartmental analysis of calcium kinetics in normal adult males. J. Clin. Invest. 1967; 46:1364-1379[Medline]
• Pansu D., Bellaton C., Roche C., Bronner F. Duodenal and ileal calcium absorption in the rat and effects of vitamin D. Am. J. Physiol. 1983; 244:G695-G700[Abstract/Free Full Text]
• Prenen J.A.C., Boer P., Mees E.J.D. Absorption kinetics of oxalate from oxalate-rich food in man. Am. J. Clin. Nutr. 1984; 40:1007-1010[Abstract/Free Full Text]
• Wastney M. E., Ng J., Smith D., Martin B. R., Peacock M., Weaver C. M. Differences in calcium kinetics between adolescent girls and young women. Am. J. Physiol. 1996; 271:R208-216[Abstract/Free Full Text]
• Williams H. E., Wandizilak T. R. Oxalate synthesis, transport, and hyperoxaluria syndromes. J. Urol. 1989; 141:742-749[Medline]

0022-3166/99 $3.00 ©1999 American Society for Nutritional Sciences