Effect of citrate on plasma aluminium concentration and aluminium excretion in the rat

Transcription

1 Clinical Science (1992) 83, (Printed in Great Britain) 43 I Effect of citrate on plasma aluminium concentration and aluminium excretion in the rat C. J. LOTE, H. C. SAUNDERS, 1. A. WOOD and A. J. SPENCER Department of Physiology, Medical School, University of Birmingham, Birmingham, U.K. (Received 16 April/8 June 1992; accepted 18 June 1992) 1. Plasma aluminium concentration and urinary aluminium excretion were monitored for 4.5 h in rats after the administration of 25pg or 8OOpg of aluminium as an intravenous bolus, either as aluminium chloride or as aluminium citrate (i.e. aluminium chloride together with sodium citrate). 2. Immediately after the bolus aluminium administration, the plasma aluminium concentration was higher in the groups given aluminium chloride than in those which received aluminium citrate, altho difference was significant (Pc0.05) only for the 25pg dose. This difference between aluminium chloride and citrate indicates that the citrate form has a higher volume of distribution (i.e. is able to leave the plasma). The calculated volume of distribution for the 25pg dose of aluminium chloride (17.5ml) was similar to the plasma volume of the rats used (15ml). 3. In experiments in vitro, the ultrafilterability of aqueous solutions of aluminium chloride and aluminium citrate were compared. Only 1.05 f 0.09% of the aluminium chloride solution was ultrafilterable (aluminium concentration f 730 ng/ml), whereas 97.3 f 2.4% of the aluminium citrate was ultrafilterable (aluminium concentration f 370 nglml). When the filterability of aluminium in plasma was examined, the aluminium chloride ultrafilterability was 1 with that in aqueous solution (1.06 f 0.13%, aluminium concentration & 2956 nglml), but the aluminium citrate was 79.8 f 7.1% ultrafilterable (aluminium concentration f 591 nglml). 4. The differences in ultrafilterability in vitro were reflected in the urinary aluminium excretion in viva Even though the plasma levels of aluminium were lower in the groups that received citrate, the aluminium administered as citrate was excreted much more rapidly than the aluminium chloride. With the 25pg dose of aluminium as chloride, 1339f0.87pg (i.e. 53%) was excreted in 4Sh, whereas with the citrate form f 3.05 pg (87%) was excreted in 4.5 h (P<O.Ol). For the 8OOpg dose of aluminium, the excretion was pg (chloride) and f36.76pg (citrate) (P<O.Ol). 5. It is clear that renal aluminium excretion is much more efficient if aluminium citrate is administered rather than aluminium chloride. The findings raise the possibility that intravenous citrate could have therapeutic applications in patients who have accumulated aluminium. INTRODUCTION Aluminium is present in the normal diet, as well as in a number of medications, and there is abundant evidence that the accumulation of aluminium in the tissues of the body is harmful [l-31. Although only a small proportion (O.l-l%) of dietary aluminium is absorbed [l, 23, tissue accumulation of aluminium would occur if this absorbed aluminium were not excreted. However, the renal excretion of aluminium is inefficient [4] for a number of reasons. These include protein-binding of aluminium, which limits its ultrafilterability, complexation of aluminium in lterable chemical forms, and the possibility of renal tubular reabsorption of aluminium (for a review, see [l]). It has been suggested by others that the protein-binding of aluminium in plasma increases with increased plasma aluminium concentration [S, 61. Our work, in which aluminium chloride was administered intravenously to conscious rats, has shown that while this is true in an absolute sense, at least for plasma aluminium levels up to about 200 ng/ml, nevertheless the fraction of aluminium which is protein-bound decreases with increasing plasma aluminium concentration [4]. At very high plasma aluminium levels (e.g ng/ ml), the ultrafilterability of aluminium is very low (about l%), not primarily as a consequence of protein binding, but as a consequence of the insolubility (i.e. colloid formation) of aluminium at these high concentrations. Indeed, the complexity of aluminium speciation in body fluids is a factor which complicates all data on renal aluminium excretion [4, 71, and has received little attention. Citrate is extremely effective in enhancing the gastrointestinal absorption of aluminium [S, 91, although the mechanism of this effect is still not entirely clear. It is likely that, in the gut, a soluble complex of citrate and aluminium is formed [S, 103, which could be absorbed, In addition, citrate is Key words: aluminium, citrate, kidneys. Correspondence: Dr C. J. Lote, Department of Physiology, Medical School, University of Birmingham, Birmingham B15 2ll, U.K.

2 432 C. J. Lote et al. known to open epithelial tight junctions in cultured cells [ 113 and in intestinal loops in uitro [9]. Since there are similarities between many of the epithelial transport processes in the gut and in the kidney, the effect of citrate on the renal handling of aluminium is of interest. Although there have been previous studies which have shown that when citrate and aluminium are administered orally, aluminium excretion increases (due to increased gastrointestinal absorption of aluminium), there have to date been no investigations in which the effects of citrate on the renal handling of aluminium per se have been studied. This was the question investigated in the present paper, in which we have used intravenous administration of both citrate and aluminium in order to avoid the complications of altered gastrointestinal absorption. METHODS Animals and protocol Experiments were performed on male Sprague- Dawley rats (weight range g), which had been previously maintained on a rat cake diet (containing 9.59 & 0.44 pg of aluminium/g), with free access to water. On the day of the experiment, rats were anaesthetized with ether, and a tail artery and tail vein were cannulated. Seven groups of animals were used. (1) Control group (n = 6). This group received 1 ml of 0.7xNaCl via the venous cannula, while still under ether anaesthesia. (2) Low dose aluminium chloride group (n=7). This group received lml of 0.7XNaCl containing 25 pg of aluminium (as A1C13.6 H,O). (3) High dose aluminium chloride group (n=6). This group received 1 ml of 0.7XNaCl containing 8OOpg of aluminium (as AlC13.6H,0). (4) Low dose citrate control group (n=5). This group received lml of a solution of 0.09% sodium citrate/0.7% NaCl. (5) Low dose aluminium citrate group (n=5). This group received lml of the solution as in group 4, but with 25pg of aluminium (as AlC13.6H,0). (6) High dose citrate control group (n=6). This group received lml of a solution of 3% sodium citrate/0.7% NaCl. (7) High dose aluminium citrate group (n=6). This group received lml of the solution as in group 6, but with 8OOpg of aluminium (as AlC13.6H,0). After the above bolus administrations, the animals were placed in individual Perspex restraining cages, and after recovery from the anaesthetic, each rat received a 4.5 h infusion (6ml/h) of 0.7xNaCl solution via the venous cannula. Urine was collected by spontaneous voiding, and after appropriate dilution was analysed for aluminium by graphite furnace atomic absorption. Details of the diluents and of the furnace conditions are given in [4]. Precautions were taken at all stages of the experi- ments to prevent aluminium contamination of infusion fluids or samples. Aluminium-free water was prepared by passing deionized water through an Elga UHQ2 water purification unit (Elga, High Wycombe, Bucks, U.K.). All containers were washed in this aluminium-free water, and all infusates were prepared using it. For the determination of plasma aluminium concentration blood (0.3 ml) was taken from the arterial line at the beginning of the experiment (before administration of the bolus), then immediately after administration of the bolus, and subsequently at hourly intervals. The blood withdrawn was replaced by an equal volume of heparinized (50 units/ml) saline (1 50 mmol/l NaC1). The blood was immediately centrifuged and the plasma was separated. At the end of the experiment (4.5h) a 2ml terminal blood sample was taken. Ultrafilterability of aluminium citrate for plasma and aqueous solutions An in uitro ultrafiltration technique (Amicon Micropartition System MPS 1; Amicon, Danvers, MA, U.S.A.) was used to determine the ultrafilterability of aluminium chloride with sodium citrate, in the way we have already described for aluminium chloride alone [4]. Determinations were made in aqueous solution (19 mmol/l NaH, P0,.2 H20; 8lmmol/l Na,HPO,, ph7.4) and also in plasma obtained from rats immediately after the administration of the aluminium chloride/sodium citrate bolus. Statistics Differences (in corresponding time periods) between experimental groups were assessed using the Wilcoxon rank sum test (for unpaired samples). RESULTS Animal experiments The low dose of aluminium administered in this study (in groups 2 and 5), 25pg, is 0.94pmol. For group 5, this dose of aluminium was administered together with 3.1pmol of sodium citrate. The measured osmolality of that 1ml bolus of aluminium chloride and sodium citrate was 236 mosmol/kg H,O. The high dose of aluminium used in groups 3 and 7, 8OOpg, is 29.6pmol. For group 7, this dose was administered with 102pmol of sodium citrate (measured osmolality of administered bolus, 572 mosmol/kg H,O). Thus for both groups (5 and 7) that received aluminium and citrate, there was at least a threefold molar excess of citrate over administered aluminium. Fig. 1 shows the aluminium excretion in the seven groups over the 4.5h infusion period, and Table 1 shows the total aluminium excretion for the groups over the 4.5h period. There were no consistent

3 Effect of citrate on aluminium excretion 433 (4.I - \ ooo-,M s g i :$ g IOOO I i Time (hl Fig. I. Urinary aluminium excretion over the duration of the experiment. Values are means +SEW Note the different scales in (a) and (b). (a) m, Group I (control); 0, group 2 (low dose aluminium chloride); A, group 4 (low dose citrate control); +, group 5 (tow dose atuminium citrate); 17, group 6 (high dose citrate control). P values are for differences between groups 2 and 5. (b) m, Group 3 (high dose aluminium chloride); 0, group 7 (high dose aluminium citrate), P values are for differences between the two groups. Table I. Total urinary aluminium excretion (0-4.5h) in the seven groups. Values are means +SEM. For both doses of aluminium, the aluminium excretion when citrate was administered was significantly higher (P <0.01) than when aluminium chloride only was administered. Urinary aluminium excretion (pg) Control Zspgof 8M)pgof (vehicle only) aluminium aluminium 0.7%NaCI k k % sodium citrate in 0.68 k k %naci 3.0%sodium citrate in 1.80k f O.7%NaCI differences between any of the control (i.e. nonaluminium) groups. A higher fraction of the administered aluminium was excreted when the aluminium was given as citrate rather than chloride; the Pre 0 I bolus Time (h) (4 Pre 0 I bolus Time (h) Fig. 2. Plasma aluminium concentrations over the duration of the experiment. The zero time sample was taken immediately after (i.e. within I min) bolus administration. Values are meansksw. Note different scales in (a) and (b). (a) W. Group I (control); 0, group 2 (low dose aluminium chloride); A, group 4 (low dose citrate control); +, group 5 (low dose aluminium citrate); 0, group 6 (high dose citrate control). P values are for difference between groups 2 and 5. (b) m, Group 3 (high dose aluminium chloride); 0, group 7 (high dose values are for differences between the two groups. significant. difference was significant for both the low dose and the high dose of aluminium (kk0.01). Fig. 2 shows the plasma aluminium concentrations during the time course of the experiment. It is obvious that aluminium administered as the chloride (AlC1,. 6 H20) produced higher plasma concentrations than did the same dose of aluminium administered as AlC1,.6 H20 plus sodium citrate. Nevertheless, in the citrate groups the urinary excretion from the lower plasma level was greater than when AlC1,.6 H20 alone was administered. This indicates greater ultrafilterability of the plasma aluminium in the presence of citrate, since we have already shown that glomerular filtration rate is not affected by the presence of aluminium in the plasma in the concentrations achieved in the present study c41.

4 434 C. j. Lote et al. Table 1. Ultrafilterability of aluminium in aqueous buffered solution at ph 7.4 (19mmol/l NaH,PO, - 1 H,O 81 mmol/l NqHPO,) and in rat plasma after intravenous administntion of 25pg of aluminium (low dose) or Wpg of aluminium (high dose) as the chloride or with sodium citnte. The data for aluminium chloride are from [4]. Values are means ~SEM. Sample Aluminium concn. Amount (nelml) filterable (%) Aluminium chloride Plasma (low dose) l54f f7.8 Plasma (high dose) I.06 f 0. I3 Aqueous f 730 I.05 f 0.09 Aluminium citrate Plasma (low dose) 1077* f 8.1 Plasma (high dose) * f 7. i Aqueous f2.4 Ultrafilterability of aluminium The ultrafilterability of aluminium chloride with sodium citrate was determined in an aqueous buffered solution, and in rat plasma after the administration of 800pg or 25 pg of aluminium as a 1 ml bolus in 0.7xNaCl with either 3% sodium citrate (for the 800,ug dose) or 0.09% sodium citrate (for the 25pg dose). The results are shown in Table 2, together with previously published data for aluminium chloride alone. DISCUSSION The findings of the present study indicate that aluminium administered with citrate is more efficiently excreted than aluminium given as the chloride. It is clear both from the present work and from our previous observations [4] that there are several reasons for this. The initial concentration of aluminium in plasma after aluminium chloride (25 pg) administration (1427 k 233 ng/ml) indicates a volume of distribution of 17.5m1, i.e. the aluminium is confined to the vascular space. (The calculated plasma volume for a 350g rat is 4.7% of body weight, i.e. 15ml). When the 25 pg of aluminium is administered together with citrate, the significantly lower (P c 0.05) initial concentration (594 k 56.7 ng/ml) indicates a volume of distribution of 42.1 ml; clearly, the citrate form of aluminium is able to leave the plasma and enter other body fluid compartments. The differences in volume of distribution are consistent with differences in ultrafilterability. We have previously shown [4] that 41.9f7.8% of the 25pg dose of aluminium as the chloride is ultrafilterable from the plasma. The non-ultrafilterability of the remaining approximately 60% is due to binding; this is likely to be mainly to the iron- binding protein transferrin [l 11, but combination with plasma phosphate as AlPO,. 2 H20 [which could also be written as A1(OH2)H,PO4] to form insoluble variscite could also be involved [ 1 I]. In contrast, aluminium (25 pg) administered together with citrate is 84.2 & 8.4% ultrafilterable from plasma, indicating much less binding to substances which render the aluminium non-ultrafilterable than is the case with the administration of aluminium chloride alone. With the 800pg dose of aluminium, only 1.06 f0.13% was ultrafilterable from the plasma if the aluminium was administered as chloride [4], whereas 79.8 k 7.1% was ultrafilterable when the aluminium chloride was administered with sodium citrate. Clearly, the citrate is maintaining the aluminium in an ultrafilterable form. The plasma aluminium concentration immediately after the high dose aluminium bolus was not significantly different between the group given chloride and that given citrate (Fig. 2). This may be due to insufficient time for such a large dose of aluminium citrate to diffuse from the plasma into other body fluid compartments. In all subsequent periods, however, the plasma level of aluminium in the group given chloride was higher than that in the group that received citrate, and this mainly reflects the higher rate of aluminium excretion when the citrate form is administered. Do the findings imply that intravenous administration of citrate would have beneficial affects for patients with high plasma aluminium concentrations? It is at present difficult to give an unequivocal answer to this question. Certainly renal aluminium excretion is more efficient if citrate is administered, but this is because the aluminiumcitrate complexes formed are more ultrafilterable than the forms in which aluminium is bound in the absence of citrate. However, because the increased ultrafilterability of aluminium-citrate complexes is reflected in a larger volume of distribution, this could make the aluminium more accessible to tissues where it could be deposited. Nevertheless, after low dose aluminium administration, almost twice as much was excreted in 4.5 h in the presence of excess citrate than in its absence, and after high dose aluminium administration, the enhancement of excretion in the presence of excess citrate was almost threefold. Clearly, more information is still required about the longer term excretion and longer term tissue deposition of aluminium in the presence of excess citrate, but it nevertheless does appear that intravenous administration of citrate could have therapeutic applications in patients who have been exposed to aluminium. ACKNOWLEDGMENTS This work was supported by a Wellcome Trust

5 Effect of citrate on aluminium excretion 435 grant to C.J.L. H.C.S. is in receipt of an MRC studentship. REFERENCES I. Lote, C.J. & Saunders, H. Aluminium: gastrointestinal absorption and renal excretion. Clin. Sci. 1991; 81, Alfrey, A.C. Aluminium metabolism. Kidney Int. 1986, 29 (Suppl. 18), SCI I. 3. Wilhelm, M., lager, D.E. & Ohnesorge, F.K. Aluminium toxicokinetics. Pharmacol. Toxicol. 1990; W, Lote, C.J. Wood, J.A. & Saunders, H.C. Renal filtration, reabsorption and excretion of aluminium in the rat. Clin. Sci. 1992; 82, Burnatowrka-Hledin, M.A., Mayor, G.H. & Lau, K. Renal handling of aluminium in the rat: clearance and micropuncture studies. Am. 1. Physiol. I 985; US, F Rahman, H., Skillen, A.W., Channon, S.M., Ward, M.K. & Kerr, D.N.S. Methods for studying the binding of aluminium by serum protein. Clin. Chem. 1985; 31, Yokel, R.A. & McNamara, P.J. Influence of renal impairment, chemical form and serum protein binding on intravenous and oral aluminium kinetics in the rabbit. Toxicol. Appl. Pharmacol. 1988, 95, Molitoris, B.A., Froment, D.H., Mackenzie, T.A., Huffer, W.H. & Alfrey, A.C. Citrate: a major factor in the of orally administered aluminium compounds. Kidney Int. 1989; Froment, D.H., Molitoris, B.A., Buddington, B., Miller, N. & Alfrey, A.C. Site and mechanism of the gastrointestinal absorption of aluminium citrate in the rat. Kidney Int. 1989; 36, Martin, R.B. The chemistry of aluminium as related to biology and medicine. Clin. Chem. 1986; 32, I I. Martinez-Paloma, A., Meza, I. Beaty, G. & Cereijido, M. Experimental modulation of occluding junctions in a cultured transporting epithelium. J. Cell Biol. 1980, 87,