INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

HISTORICAL RECORDS dating back to 1785 recount observations made by travelers to South Africa concerning a peculiar behavior in cattle. These records describe cattle as eagerly searching for and gnawing bones after feeding on what the travelers called "harsh grass" (20). Because of the sickness (and subsequent economic losses) that usually developed in the cattle after the ingestion of decaying skeletal debris, initial investigations were initiated, and this behavior was linked to a mineral deficiency in the cattle.

A similar observation was made in 1925 when Henry Green reported what he termed to be a "perverted appetite" in cattle (16). Up to 90% of the cattle grazing on certain phosphate-deficient soils chewed on skeletal bones (a naturally occurring source of P_i). Green postulated that such behavior was due to a phosphate deficiency, but it was not until later that laboratory experiments performed by Denton et al. (9) confirmed that this osteophagic behavior in cattle was indeed brought on by a phosphate deficiency. By experimentally inducing a chronic (2 yr) P_i deficiency through a parotid fistula and a low-phosphate diet, Blair-West et al. (2) reported bone to be licked, picked up with the mouth, and chewed on much more avidly in the cattle deprived of phosphate compared with replete controls. Furthermore, in young heifers, P_i deficiency was also associated with growth retardation, and the appetitive behavior could be suppressed by raising the plasma phosphate concentration through a rapid intravenous infusion of a buffered P_i solution (6). These observations provide evidence suggesting that the P_i appetite in these cattle is linked to a phosphate deficiency. However, whether this phenomenon occurs in other mammals or is species dependent (rat vs. cattle) and related to an herbivorous or omnivorous behavior remains unknown.

The classic mineral appetite is that displayed for sodium (7, 26, 36). Given the important role sodium has in maintaining extracellular fluid volume, it is not surprising that a behavioral adaptation emerged to help maintain fluid homeostasis during times of sodium depletion (10, 12). Although other mineral deficiencies (e.g., potassium, iron, and calcium) result in sodium ingestion (28), P_i deficiency does not (35). This may suggest that the motivation to obtain P_i in the rat may be specific for P_i and in fact be an intrinsic response (2).

P_i is essential for growth and development in the immature animal as well as used in a myriad of metabolic processes in the adult animal. Physiological adaptations exist to ensure that an adequate supply of P_i is made available to the growing and adult animal. For example, urinary losses of P_i are limited in young rats, related in part to an enhanced ability to transport P_i. Indeed, the normal tubular capacity of the kidney to reabsorb P_i is enhanced in immature rats compared with adult rats (3, 18) and appears to be related to the increased demand for P_i during growth (32). Furthermore, during dietary P_i restriction, both the adult and juvenile animal significantly increase their maximal renal capacity to reclaim all available P_i, reducing urinary losses (23). In the adult animal, this adaptation occurs through an increase in P_i transport at both the proximal (1, 5, 22, 33) and sites distal to the proximal convoluted tubules (17). Thus there may be both behavioral and physiological adaptations initiated during P_i deprivation that would serve to increase the supply of P_i to the organism.

In light of the findings of increased appetite for bone in phosphate-deficient cattle and the overall importance of maintaining P_i homeostasis in young animals, the purposes of the present study were 1) to identify whether a P_i appetite can be induced in juvenile rats by feeding them a low-P_i diet and measuring their potassium phosphate water (P_iH₂O) intake; 2) to determine whether an appetite for P_i can be stimulated rapidly by 2 days of P_i restriction; 3) to examine whether this appetite is specific for P_i and not another mineral involved in the growth and development of the young animal (e.g., calcium); and 4) to examine the effects of P_i deficiency on growth, plasma, and CSF P_i levels and renal P_i reabsorption.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Animals, diets, and fluids. All experiments were performed using male, juvenile Wistar rats (Harlan, Indianapolis, IN) ranging from 23 to 25 days of age at the onset of testing. After arrival, the rats were initially fed standard Purina Rodent Laboratory Chow pellets (no. 5001) while they adapted to laboratory conditions (2 days). Each rat was individually housed in an 18 × 12 × 10 cm hard plastic cage with a 1-in. bedding of wood chips and fitted with a stainless steel cage top. Animals were housed in a 12:12-h light-dark cycle with lights on from 0700 to 1900. Unless otherwise mentioned, for each study animals were divided into an experimental group designed to receive a low-P_i diet (LPD, 0.02% P_i), consisting of a basic low-P_i rodent diet (Research Diets 93082402, New Brunswick, NJ) supplemented with 9.7 g/kg NaCl and 14.76 g/kg KCl, and a control group that was fed a normal P_i diet (NPD, 0.6% P_i), which was the basic LPD supplemented with 3.02 g/kg KH₂PO₄, 15.46 g/kg K₂HPO₄, 2.28 g/kg NaH₂PO₄, and 10.48 g/kg Na₂HPO₄ salt additives.

Concentration of P_i in plasma and cerebrospinal fluid. To determine the effects of P_i restriction on plasma P_i levels, juvenile Wistar rats were fed a LPD and were killed after days 0 (n = 3), 1 (n = 3), 2 (n = 3), and 7 (n = 3) of P_i restriction. Blood samples were obtained on these days and analyzed for plasma P_i. These values were compared with plasma P_i from animals fed LPD and P_iH₂O. To collect cerebrospinal fluid (CSF), 23-gauge stainless steel guide cannulas were implanted stereotaxically into the third ventricle, as previously reported (24), in rats fed either LPD (n = 4) or NPD (n = 4) for 3 days. Injectors were placed in the guide cannulas, and CSF was drawn into PE-10 tubing. The tubing was sealed at both ends by flame and stored at 4°C until microanalysis. Analysis was performed by using a flow-through microcolorimeter and a modified phosphomolybdate method of Chen et al. (4). CSF samples (100 nl) or P_i standards (50 nl) were transferred to separate tubing containing 2 µl of reagent (10% ascorbic acid, 10% 8 M H₂SO₄, and 10% 2.5% ammonium molybdate(VI) in distilled water). The samples were sealed, mixed, and incubated in a 37°C water bath for 90 min. After incubation, the samples were injected into the spectrophotometer port, and the absorbance was read as a change in voltage. All samples were run in duplicate in two separate assays, and P_i concentration was determined against the standard curve.

Effect of 7-day P_i restriction on P_i appetite. To assess whether a P_i appetite could be stimulated in juvenile rats, animals were pair fed either LPD (n = 11) or NPD (n = 11): the NPD rats were fed only the average amount of food consumed by the LPD rats the previous day. On day 7, both LPD and NPD rats were given free access to P_iH₂O for the remaining 8 days of the experiment. Body weight gain and food and water intakes were monitored daily throughout the experiment. Plasma P_i was determined at the end of the experiment as previously described.

Effect of 2-day P_i restriction on P_i appetite. To examine whether an appetite for P_i could be rapidly induced after just 2 days of P_i deprivation, juvenile rats were fed LPD (n = 6) and NPD (n = 6) for 9 days. After 2 days, rats fed both NPD and LPD received unlimited access to the P_iH₂O solution for the remaining 7 days. Body weight gain and food and water intakes were monitored daily. Plasma P_i was determined at the end of the experiment.

Effect of 2-day P_i deprivation on renal tubular P_i reabsorption. To monitor the renal effects of dietary P_i deprivation, a separate group of rats was placed on a NPD or LPD, and after 2 days, acute renal clearances experiments were performed (23) to assess the maximal capacity for tubular P_i reabsorption (T_mP_i) during changes in dietary phosphate intake. Briefly, animals were fed their respective diets (NPD, n = 3; LPD, n = 5). Another group of animals fed LPD were given access to P_iH₂O, and T_mP_i was assessed after 5 days (n = 5). On the day of the acute experiment, the rats were anesthetized with an intraperitoneal injection of Inactin (100 mg/kg, Promonta, Hamburg, Germany) and placed on a heated table. Body temperature was monitored via a rectal probe and maintained at 37 ± 0.5°C with a heated table and lamp. A tracheostomy was performed, and a tube was placed in the airway so the animals could breathe spontaneously. Catheters were placed in the jugular vein for infusions, carotid artery for blood pressure measurements and blood sampling, and the bladder for urine collection. Animals were then acutely thyroparathyroidectomized by heat cautery to remove the influence of endogenous parathyroid hormone. A 2% inulin solution was infused at 2% body weight/h throughout the experiment. After a 2-h recovery period, a 20-min baseline urine clearance was taken. To determine the T_mP_i, increasing concentrations of P_i (calculated to deliver 3-9 µmol/min) were added to the inulin solution and infused to increase the filtered load of P_i. Urine collections were made every 20 min for 2-3 h. Blood was sampled midway through each clearance period, and after the experiment, the animals were killed. Inulin concentrations in urine and plasma samples were determined by the anthrone method (15), and the glomerular filtration rate (GFR) was equated with the clearance of inulin. The T_mP_i was calculated as the mean of the highest individual values of reabsorbed P_i per milliliter GFR in each group.

Specificity of P_i appetite. Because both P_i and calcium are essential components of developing bone (as hydroxyapatite), a preliminary assessment of whether short-term (2-day) P_i deprivation in juvenile rats induces an appetite specific for P_i and not calcium was performed. Juvenile rats were fed LPD (n = 5) and NPD (n = 5) over 10 days. After 2 days, both LPD and NPD were given equal access to separate but equimolar (0.3 M) P_iH₂O and CaH₂O solutions as well as tap water for the remaining 8 days. Body weight gain and fluid and food intakes were monitored.

P_i appetite and plasma and CSF P_i concentrations after P_i replacement. The working hypothesis is that, during P_i deprivation, reductions in plasma and CSF P_i levels stimulate and maintain the behavioral response in the juvenile animal. Furthermore, it is proposed that if plasma and/or CSF P_i levels were restored to normal, the behavioral response to ingest additional P_i would be turned off. To test this notion, juvenile rats were fed NPD (n = 3) or LPD (n = 4) for 7 days. Body weights and food and water intake were monitored throughout the experimental period. After 2 days on the respective diets, animals were given free access to 0.3 M P_iH₂O for 5 additional days, and P_i appetite was monitored. On the 7th day of the experiment (after 5 days on P_iH₂O), all animals were placed on a high-P_i diet (HPD) for 3 days while P_iH₂O was still available. On the 10th day, animals were anesthetized, and cannulas were placed in the third ventricle of the brain to obtain a CSF sample. Blood was also obtained at the end of the experiment for determination of plasma P_i levels.

Statistical analysis. Comparisons within and between the groups were obtained by using paired and unpaired Student's t-tests, respectively. The rate of growth over the experimental periods was equated with the slope (determined by linear regression) of the individual growth curves. Means ± SE were determined for the slopes of each pre- and post-P_iH₂O group. Significance was designated as P < 0.05, and all results are expressed as means ± SE.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Effect of 7-day P_i restriction on P_i appetite. Rats fed LPD demonstrated a significantly reduced growth rate during the first 7 days of the experiment compared with animals fed NPD (Fig. 1). From days 0 to 7, the average daily gain in body weight in NPD rats was 4.5 ± 0.4 g/day compared with 1.5 ± 0.4 g/day in LPD-fed rats (P < 0.05). Once the P_iH₂O was provided to both NPD and LPD rats after day 7, the LPD rats showed an increase in daily body weight gain for the remaining 8 days that matched that observed in controls (6.0 ± 0.6 vs. 6.1 ± 0.5 g/day over 8 days in NPD, P < 0.05). Relative slopes of the growth curves were 4.23 ± 1.17 and 1.24 ± 1.27 for rats fed NPD and LPD, respectively, during days 0-7 (P < 0.01). The slope for days 8-15 (6.21 ± 0.92) reflects the increased growth rate of rats fed LPD plus P_iH₂O (P < 0.001 vs. LPD days 0-7). This elevated rate of growth was not different from that observed in rats fed NPD over days 8-15 (6.36 ± 0.67). Over days 8-15, rats fed LPD plus P_iH₂O also consumed more food (Fig. 2, top; 10 ± 0.2 over days 0-7 vs. 14 ± 0.3 g/day over days 8-15, P < 0.05). This may also explain the increase in daily body weight gain in the pair-fed rats fed NPD over days 8-15. Ingestion of P_iH₂O was significantly greater in the animals fed LPD compared with the NPD controls over days 8-15 (3.0 ± 0.2 vs. 1.8 ± 0.1 ml/day; Fig. 3). This P_iH₂O intake corresponds to ingestion of ~0.9 mmol P_i/day in the rats fed LPD. This was less than one-half of the normal amount of P_i ingested by the animals eating NPD (~2.1 mmol/day). Interestingly, rats fed NPD also exhibited a growth rate increase during the period that P_iH₂O was provided (days 0-7, 4.5 ± 0.4 g/day, compared with days 8-15, 6.1 ± 0.5 g/day), but this increase may be a result of the pair-fed design of the experiment and an increase in food ingestion in LPD-fed rats. Similarly, when P_iH₂O was provided at day 7, water consumption increased in the LPD-fed rats (Fig. 2, bottom).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Cumulative change in body weight (BW) in juvenile rats fed a low-phosphate diet (LPD, n = 11) and a normal phosphate diet (NPD, n = 11) before (days 0-7) and after access to a solution of 0.3 M potassium dihydrogen phosphate (PiH2O; days 8-15). m, Slope of growth curves from days 8-15.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Top: average daily food intakes in juvenile rats fed LPD (n = 11) and NPD (n = 11) before (days 0-7) and after (days 8-15) ingestion of a 0.3 M solution of PiH2O. Animals were pair fed throughout entire experiment. Bottom: average daily water intake in juvenile rats fed LPD and NPD before (days 0-7) and after (days 8-15) PiH2O ingestion. * P < 0.05 from days 0-7 values.

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Average daily ingestion of 0.3 M PiH2O in juvenile rats fed LPD (n = 11) and NPD (n = 11) over a 7-day period. * P < 0.001 vs. NPD control.

Effect of 2-day P_i restriction on P_i appetite. To determine whether the induction of a P_i appetite was a rapid process, a separate group of animals was given access to P_iH₂O after only 2 days of P_i deprivation. At this time point, the renal adaptation to P_i restriction was already evident, as seen by a significant T_mP_i elevation in rats fed LPD compared with those rats fed NPD (8.7 ± 0.6 vs. 5.6 ± 0.1 µmol/ml GFR in NPD controls, P < 0.001). These P_i-restricted animals also exhibited an increase in P_iH₂O consumption (3.2 ± 0.2 ml/day) compared with the controls (1.8 ± 0.2 ml/day), which was similar to that observed after 7 days of P_i restriction (Fig. 3). The reduction in daily body weight gain observed in rats fed LPD (0.1 ± 0.4 g/day; slope of 0.33 ± 0.89 vs. 6.30 ± 1.39 in rats fed NPD, P < 0.0001) was significantly increased during P_iH₂O ingestion (average daily gain in body weight during days 3-9 of 4.2 ± 0.5 g/day; slope of 3.71 ± 2.17 vs. 5.28 ± 0.34 in rats fed NPD, P = 0.14; P < 0.001 vs. 2-day LPD). Animals fed NPD during this period maintained a constant, rapid growth rate throughout the experimental period. Interestingly, the animals fed LPD continued to exhibit the P_i appetite during the entire period: the maintenance of the P_i appetite was also associated with continued low plasma P_i levels (1.9 ± 0.1 vs. 3.2 ± 0.1 mM in NPD controls, P < 0.01) as well as elevated T_mP_i (8.1 ± 0.3 µmol/ml GFR, P < 0.001 vs. NPD controls, but NS to LPD controls; Fig. 4). This indicated that both the behavioral and physiological adaptations were still stimulated, despite the increased supply of P_i, but suggests that the ingestion of P_iH₂O was not adequate to restore the system to normal.

View larger version (11K): [in this window] [in a new window]   Fig. 4.   Renal transport maximal for Pi (TmPi) in rats fed NPD (n = 3), LPD (n = 5), or LPD + 0.3 M PiH2O (n = 5) for 7 days. GFR, glomerular filtration rate. * P < 0.001 vs. NPD control.

Specificity of P_i appetite. Juvenile rats fed NPD and LPD for 2 days exhibited a preferential intake of P_iH₂O when given a choice between equimolar P_iH₂O and CaH₂O solutions. There was no significant difference in CaH₂O ingestion between animals fed NPD or LPD (0.2 ± 0.4 vs. 0.5 ± 0.5 ml, respectively). However, during the same period, rats fed LPD ingested an average 1.6 ± 0.3 ml/day of P_iH₂O, whereas rats fed NPD averaged 0.7 ± 0.1 ml/day (P < 0.002). In addition, the increased consumption of P_iH₂O in LPD rats was accompanied by an increased average growth rate of 3.4 ± 0.5 g/day during days 3-10 (P < 0.007 vs. 0.4 ± 0.6 g/day during initial 2 days of LPD). The amount of P_iH₂O ingested was consistent over the entire period, indicating that the P_i appetite was still in effect. Again, the maintenance of the P_i appetite was associated with significantly lower plasma P_i levels in LPD-fed rats receiving P_iH₂O (1.78 ± 0.15 mM) compared with those rats fed NPD also receiving P_iH₂O (3.19 ± 0.15 mM, P < 0.0001).

Plasma and CSF P_i concentrations during P_i restriction. Because the P_i appetite was evident early (after 2 days of P_i deprivation), we assessed the time course for changes in plasma P_i concentrations. Plasma P_i levels after 0, 1, 2, 3, and 7 days of P_i restriction are shown in Fig. 5. By day 2 of P_i restriction, the plasma P_i levels were significantly decreased from control levels (1.79 ± 0.03 vs. 2.79 ± 0.16 mM in controls, P < 0.003). This reduction in plasma P_i was not different after 7 days of P_i restriction. Furthermore, CSF P_i levels in rats fed LPD for 3 days were also significantly reduced compared with controls (0.20 ± 0.01 vs. 0.47 ± 0.01 mM in rats fed NPD, P < 0.05).

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Plasma Pi concentration in juvenile rats deprived of phosphate for 0, 1, 2, 3, and 7 days. Values are means of 3 animals at each day. * P  0.01 vs. day 0 control.

P_i appetite, plasma and CSF P_i concentrations after P_i replenishment. To assess whether plasma and/or CSF P_i levels are involved in the maintenance of the P_i appetite, a high concentration of P_i was replenished in the diets of P_i deficient juvenile rats. Figure 6 (left) summarizes data from the previous experiments. As previously observed, ingestion of P_iH₂O was stimulated in juvenile rats fed LPD after only 2 days of dietary P_i deprivation (1.6 ± 0.2 vs. 0.8 ± 0.1 ml P_iH₂O/day in NPD controls, P < 0.001; Fig. 6, bottom left). At this time, plasma and CSF P_i levels are still significantly reduced despite 5 days of ingestion of P_iH₂O (Fig. 6, top and middle left, respectively).

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Top: plasma Pi levels in juvenile rats fed NPD or LPD after 5 days of 0.3 M PiH2O ingestion (left) or after 3 days of PiH2O and further replenishment with high-Pi diet (HPD; right). Middle: cerebrospinal fluid (CSF) Pi levels in juvenile rats fed NPD or LPD after 5 days of PiH2O ingestion (left) or after 3 days of PiH2O and HPD (right). Bottom: PiH2O intake in juvenile rats fed NPD or LPD after 5 days of PiH2O ingestion (left) or after 3 days of PiH2O and HPD (right). * P < 0.05 vs. NPD.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The present study indicates that when juvenile rats are deprived of phosphate in their diet, plasma and CSF P_i levels decrease rapidly, the kidney increases P_i reabsorption, and the consumption of P_iH₂O is significantly increased. Interestingly, the ingestive response was not different between acute (2-day) and longer (7-day) P_i restriction. The P_i deficiency induced in this study resulted in the preferential ingestion of P_iH₂O rather than a solution of CaH₂O (2 minerals necessary for growth of young animals). Moreover, the reduction in plasma and CSF P_i levels may be part of the mechanisms by which the behavioral response is elicited, since restoration of plasma and CSF P_i levels was associated with quenching of the P_i appetite. Thus stimulation of an appetite for P_i may provide a behavioral mechanism, in addition to the intrinsic increase in renal P_i transport, by which the young animal can increase P_i supply during periods of phosphate depletion.

It is well known that P_i is crucial to the proper growth, metabolic processes, and general well being of animals. Reducing the supply of P_i in juvenile animals results in an immediate attenuation in somatic growth (21, 30). This phenomenon was readily observed in the present study, and consequently a positive change in the growth rate was used as the criterion for the efficacy of oral P_i ingestion. Indeed, after both 2 and 7 days of phosphate restriction, the stimulation of P_i ingestion resulted in significant increases in whole body weight gain. This positive correlation highlights the potential interrelationships between the behavioral and physiological responses to P_i deprivation and growth in the juvenile animal.

Because our studies were focused on the rapidly growing animal, which already has a high demand for P_i, it is unclear whether a comparable appetite for P_i can be stimulated in the adult rat. P_i is also of great importance to metabolic functions and bone remodeling in the adult animal (32), and we hypothesize that adult animals would exhibit a similar appetite to increase P_i supply, which could be exaggerated in times of high-P_i demand. The P_i-seeking behavior observed by previous investigators in cattle grazing on phosphate-poor vegetation support this notion (2, 6, 16). Typically, cows also have a high demand for P_i to maintain an adequate supply of the mineral for pregnancy and milk production. It is important to note, however, that herbivorous and omnivorous diets, as well as age and developmental status, may affect the degree of appetite induced for different minerals.

One of the effects of reducing dietary P_i is a decrease in plasma P_i concentration. In previous studies on cattle, the bone-gnawing behavior increased over months and was associated with a fall (over weeks) in plasma P_i, and in these studies, P_i concentrations in CSF were not shown to decrease (2). Denton et al. (9) and Blair-West et al. (2) have proposed that plasma P_i may regulate the P_i appetite in cattle. In the present study, there was a rapid fall in plasma P_i levels in juvenile rats, which was maximal after only 2 days and remained at the same low level through 7 days of P_i restriction. Furthermore, CSF P_i levels were also significantly reduced after only 3 days of P_i restriction, and the reduction most likely occurs earlier. This rapid reduction in peripheral and central phosphate concentrations, which coincided with induction of the P_i appetite, supports the concept that plasma and CSF phosphate levels, directly or indirectly, are part of the mechanism to stimulate a central behavioral response to increase P_i ingestion. Although the increase in P_iH₂O ingestion was significant enough to increase body weight gain in the animals, it evidently was not enough to restore plasma and CSF P_i concentrations and potentially turn off the appetite and renal adaptations. This notion is supported by the findings that a greater replenishment of P_i through HPD quenched the appetite and normalized plasma and CSF P_i levels. Further studies will help determine the temporal sequence of events contributing to the quenching of the appetite.

Although 7 days of P_i deprivation has been shown to elicit renal adaptations (18), the 1-wk time period may not have been sufficient to reduce plasma P_i concentrations enough to stimulate additional P_iH₂O consumption. Still, our findings argue against this notion, since although the P_iH₂O ingested was enough to significantly increase the growth rate, the plasma P_i level was not normalized, and both the renal and behavioral adaptations were still active. In light of these findings, we hypothesize that the P_iH₂O was aversive enough to restrict more avid ingestion (and hence animals fed LPD only received ~50% of phosphate ingested by animals fed NPD), the plasma and CSF P_i levels could not return to normal with the daily intake of P_iH₂O. In this scenario, the additional P_i consumption (via P_iH₂O) may have been available immediately for metabolic processes rather than remaining in the plasma pool. This could account for the lack of normalization of the plasma and CSF P_i levels and subsequent maintenance of the behavioral and physiological adaptations.

A potentially important finding was that the P_i appetite in the juvenile rat induced a preferential intake of P_iH₂O over CaH₂O in this preliminary study. Whether this highlights the relative importance of P_i in particular to the metabolic and growth requirements of the animal is unknown, and further study is required to fully assess the specificity of the behavioral response. Furthermore, we do not know to what extent the P_i ingestion is or is not similar to the well-characterized sodium appetite. First reported by Richter (26), the specific appetite for sodium that emerges when extracellular fluid and sodium are depleted is innate, linked to both a specific gustatory transduction mechanism involving cranial nerves (14a, 27, 31) and excitatory and inhibitory humoral signals that act in the central nervous system (8, 11, 13, 14, 29, 34). As with P_i deficiency, sodium restriction in the young severely attenuates somatic growth, highlighting its importance developmentally.

In conclusion, our initial results support the notion that there may be behavioral as well as the previously reported intrinsic renal mechanisms contributing to the regulation of P_i homeostasis in the rat. The appetite for P_i may be particularly conspicuous during development and, by analogy, during reproduction, when the demands for P_i are high. Moreover, it is conceivable that reductions in plasma and CSF P_i levels effect P_i transport in critical regions of the brain that regulate the P_i appetite and may provide the mechanism by which the behavioral response is mediated.

Perspectives