Stanniectomy Increases Renal Magnesium and Calcium Excretion in Freshwater North American Eels (Anguilla Rostrata)

The corpuscles of Stannius are putative endocrine glands embedded in the ventral surfaces of the mesonephric kidneys of holostean (de Smet 1962; Youson and Butler 1976; Youson et al. 1976; Bhattacharyya et al. 1982) and teleostean (Stannius, 1839; Ristow and Piepho, 1963; Oguri, 1966; Fujita and Honma, 1967; Wendelaar Bonga and Greven, 1975; Bhattacharyya and Butler, 1978; Youson et al. 1989) fish.

There are two corpuscles of Stannius in eels. Fontaine (1964) showed that, when these glands were extirpated from freshwater Anguilla anguilla, there followed a gradual increase in plasma calcium and K⁺ and a reduction in plasma Na⁺ concentrations. Further research has attempted to show how stanniectomy leads to hypercalcaemia in freshwater teleost fishes (see review by Hirano, 1989). A possible hormonal link between the corpuscles of Stannius and branchial or renal magnesium transport has received less attention. In freshwater North American eels (Anguilla rostrata) there were no measurable changes in plasma or urinary magnesium concentrations 3 weeks after stanniectomy (Butler, 1969). There was, however, a significant increase in the relative magnesium clearance, which suggested that the corpuscles may directly or indirectly regulate the renal tubular transport of magnesium (Butler, 1969). The present work gives a more complete account of progressive changes in renal magnesium and calcium transport during a 14 day period following removal of the corpuscles of Stannius from freshwater North American eels.

Female freshwater eels (Anguilla rostrata LeSueur), weighing an average of 1003±42g (S.E.M., range 814–1473g) were collected from the St Lawrence River near Quebec City, Canada, in June 1992. They were held for at least 2 months in flowing dechlorinated tap water (Na⁺, 0.45; Cl⁻, 0.95; K⁺, 0.02; Ca²⁺, 0.98; Mg²⁺, 1.59mmol l⁻¹) at 11°C and fed earthworms until they were used for experiments. Selected eels were acclimated to 12°C for 1 week before and during the 14 day experimental period.

Experimental groups consisted of (a) nine sham-operated eels and (b) nine eels from which the corpuscles of Stannius had been removed. Experimental observations were made for 14 days while each eel was housed in a plastic holding tank supplied with aerated dechlorinated tap water at 13°C. The tank was covered with a sheet of black plastic so that the eel would not be disturbed and would remain quiescent. Venous and urinary catheters were pulled through ports in the holding tank so that blood and urine could be collected without disturbing the eel. Urine samples were collected over a 6-to 10-h period every day throughout the 14 day period of observation. Urine drained by gravity into a covered 45ml vial, where it collected under mineral oil. The urine catheter was checked regularly for blockages. Blood samples were collected from each eel before surgery, on the day after surgery and every 2 days thereafter.

Eels were anaesthetized in an aqueous solution of methane tricainesulphonate (1.0 gl⁻¹) and then wrapped in wet towelling during surgery. A heparin-filled blood catheter (Intramedic PE10) was inserted into the caudal vein, about 7.5cm from the end of the tail and tied in place with 5-0 silk sutures. A 7.5cm lateral incision through the body wall at the level of the posterior kidney allowed me to lift the bladder away from the ventral surface of the posterior kidney and remove the two corpuscles of Stannius. There was little, if any, bleeding. The same procedure was used for sham-operated eels, but the corpuscles of Stannius were left in place. A urine catheter made from a length of Tygon plastic tubing (o.d. 0.070 and i.d. 0.040; Norton Co., Performance Plastics) was inserted through the urinary papilla and tied in place with size 3-0 surgical silk. The incision through the body wall was closed tightly with size 3-0 stainless-steel wire and the eel was injected intramuscularly with 50mg of Ampicillin (Penbritin-500, Ayherst, in 1.0ml of 0.9% NaCl). Daily injections of antibiotic (50mg Ampicillin) were given for a further 4 days.

Caudal venous blood was collected over ammonium heparin and promptly centrifuged at 0°C. The plasma was stored at −30°C. Plasma and urine samples were later thawed and diluted with a 1% lanthanum solution. Phosphate is ‘swamped’ by lanthanum, thereby reducing interference during the measurement of plasma and urinary total calcium and magnesium by atomic absorption spectrometry (Instrumentation Laboratory, model 351, atomic absorption spectrometer).

Data are given as means ± S.E.M. A repeated-measures analysis of variance (SAS) was used for between-group comparisons. Sample means were compared using Tukey’s w-procedure. Linear regression analysis was based on the least squares method. The fiducial limit was P=0.05.

Table 1 shows that there was a 30% decrease in plasma magnesium concentration in both sham-operated and stanniectomized eels within 4 days after surgery. From then on, plasma magnesium concentration remained stable in the sham-operated group but it continued to decrease gradually in the stanniectomized group. The effect of stanniectomy becomes clearer in Fig. 1. In each sham-operated eel, plasma magnesium concentration was approximately constant in relation to plasma calcium concentration, whereas after stanniectomy, plasma magnesium concentration gradually declined as plasma calcium increased: there was a negative linear relationship between plasma magnesium and plasma calcium (y=–0.116x+1.54; r=–0.43; N=9 animals, 66 points; P<0.01, where x is magnesium concentration and y is calcium concentration).

Table 2 presents urine flow rates and urinary magnesium and calcium concentrations and excretion rates in sham-operated and stanniectomized eels. Urine flow rates did not change during the 2 week period following stanniectomy with the exception of day 2, when the flow rate dropped briefly. The clear increase in urinary magnesium concentrations following stanniectomy became evident on day 3. They remained significantly higher throughout the 14 day test period. A parallel change was seen for the rate of urinary magnesium excretion (Table 2). The gradual decrease in plasma magnesium levels together with the increase in urinary magnesium concentrations led to a marked increase in the magnesium urine/plasma ratio 6 days after stanniectomy and thereafter (Table 2).

Although there was a brief increase in urinary calcium concentration 2 days after stanniectomy, the sustained (and increasing) elevation did not occur until the ninth day after surgery. The rate of calcium excretion paralleled the increase in urinary calcium concentration (Table 2), the increase becoming statistically significant on day 9 and thereafter. The calcium urine/plasma ratio was unchanged following stanniectomy, indicating that the proportional increase in the renal excretion of calcium is a ‘passive’ process and is caused solely by the increase in plasma calcium concentration.

In both sham-operated (y=0.058x±0.46; r=0.33; N=9 animals, 109 points) and stanniectomized (y=0.073x±0.073; r=0.44; N=9 animals, 109 points, where x is the rate of calcium excretion and y is the rate of magnesium excretion) eels, there was a significant positive relationship (P<0.01) between the rates of urinary magnesium and calcium excretion (Fig. 2). No significant difference was detected between the regression lines. Although the rate of urinary calcium excretion increased following stanniectomy, there was a relatively greater increase in the rate of urinary magnesium excretion (Fig. 2).

Fig. 3 shows that, in sham-operated eels, the urinary molar ratio for magnesium/calcium began to decrease 3 days after surgery and then levelled off for the remainder of the 14 day observation period. After stanniectomy, the relationship changed. Even though urinary calcium concentration had increased, the increase in urinary magnesium concentration was relatively much greater, so the magnesium/calcium molar ratio increased.

Attempts have been made to show how the corpuscles of Stannius may regulate calcium transport through the gills (Fontaine, 1967; Fontaine et al. 1972; Fenwick and So, 1974; Fenwick, 1976, 1989) and kidneys (Rankin, 1967; Butler, 1969;Chan et al. 1969; Fenwick, 1974; reviewed by Hirano, 1989), but the link between the corpuscles of Stannius and the regulation of Na⁺, K⁺, Cl⁻and magnesium has often been overlooked. The present work examines the relationship between these glands and the renal handling of magnesium and calcium. Rankin (1967) was the first to study the renal response to stanniectomy in a teleost fish (Anguilla anguilla). After the corpuscles had been removed, there were no significant changes in urine flow rates [57.2±4.5mlkg⁻¹ day⁻¹ (N=5) compared with 47.0±5.1mlkg⁻¹ day⁻¹ (N=7)] or in urinary Na⁺ concentrations (13.1±3.5mmol l⁻¹ compared with 12.3±2.4mmol l⁻¹). In some stanniectomized eels, urine flow rates exceeded the glomerular filtration rate (GFR), showing that, under these experimental conditions, freshwater eel renal tubules secreted water, an observation made in North American eels by Butler (1969) and Schmidt-Nielsen and Renfro (1975). Chan et al. (1969) studied renal function in freshwater European silver eels during the first 7 days after stanniectomy. Again, urine flow rates did not change following stanniectomy but there were gradual decreases in urinary calcium and magnesium concentrations, causing a decreased rate of urinary excretion of both ions during the entire 7 day observation period; urinary Na⁺ and K⁺ concentrations were unchanged. These findings are the opposite of those made in the present experiments, in which the rates of renal excretion of magnesium and calcium increased in freshwater North American eels throughout the entire 14 day observation period (Table 2). However, they confirm the earlier observations of Butler (1969) and Fenwick (1974) on renal function in stanniectomized freshwater North American eels (3 weeks postoperative). There was no significant change in either GFR or urine flow rate in stanniectomized eels (29.9±3.6mlkg⁻¹ day⁻¹) compared with sham-operated controls (32.4±2.9mlkg⁻¹ day⁻¹). The temperature of acclimation was 10°C whereas in the present experiments it was higher. This accounts for the higher urine flow rates in both sham-operated (45.1±4.1mlkg⁻¹ day⁻¹) and stanniectomized (55.4±4.1mlkg⁻¹ day⁻¹) eels during the entire 14 day collection period (Table 2). Furthermore, stanniectomy had no effect on urine flow rates, with the exception of day 2 when they dropped sharply. However, this was a transient response and the flow rate returned to normal on the following day. One might speculate that this was an early response to the loss of a renin-like pressor substance (Chester Jones et al. 1966) or to angiotensin I (Takemoto et al. 1983; Hasegawa et al. 1982, 1984), which was then corrected by subsequent adjustments in renal and/or branchial blood circulation. Although the urine flow rates are higher (Table 2) than in my earlier observations at a lower temperature (Butler, 1969) they are in line with, for example, those of a sample group of sham-operated freshwater European eels (47.76±1.92mlkg⁻¹ day⁻¹) which were also acclimated to 12°C (Babiker and Rankin, 1978). Earlier experiments (Butler, 1969) showed that the urinary Na⁺ concentration fell from 19.8±2.8mmol l⁻¹ (N=10) to 11.9±1.4mmol l⁻¹ (N=18) and the urinary calcium concentration increased from 2.98±0.27mmoll⁻¹ (N=10) to 6.69±0.86mmol l⁻¹ (N=18) 3 weeks after stanniectomy. Also, urinary K⁺, Cl⁻and magnesium concentrations were not significantly different from those of sham-operated controls. The increased rate of urinary calcium excretion was caused by an early increase in plasma total calcium concentration and not by a change in the renal tubular transport of calcium because there was no change in the fraction of filtered calcium reabsorbed (Butler, 1969). Plasma magnesium levels did not change measurably following stanniectomy but the fraction of filtered magnesium reabsorbed decreased from 86.9±2.2% to 70.9±6.9%, which caused an increased rate of urinary magnesium excretion. Fenwick (1974) confirmed my observations when he reported that the GFR and urine flow rate in freshwater North American eels were unchanged 3 weeks after stanniectomy and that the filtered load of calcium and the relative calcium clearance (C_Ca/C_In) increased significantly. Thus, a clear pattern emerges: stanniectomy is followed by hypercalcaemia, and the increased filtered load of calcium spills over into the urine without concomitant changes in renal tubular transport of calcium.

Kenyon et al. (1980) studied the responses of freshwater European eels to stanniectomy. Plasma calcium concentration began to rise steeply on the third day of a 16 day observation period, whereas plasma magnesium began to gradually decline about 7 days after surgery. The latter finding is confirmed by the present observation that plasma magnesium concentration in individual eels decreased gradually following stanniectomy (Fig. 1), a change that is not evident when the data were pooled (Table 1). A rapid increase in plasma calcium concentration became statistically significant within 2 days after stanniectomy (Table 1), a trend which continued until plasma calcium levels reached 6.41±0.44mmol l⁻¹ on day 14 (compared with 2.79±0.15mmol l⁻¹ for sham-operated individuals). The relationship between decreasing plasma magnesium levels and increasing plasma calcium levels became apparent when the data for each sham-operated or stanniectomized eel were plotted separately (Fig. 1). In sham-operated eels the points are clustered and no such relationship is evident. The early increase in eel plasma calcium (Table 1) may have resulted from an increased net influx of calcium through the chloride cells of the gill laminar epithelium (Payan etal. 1981; Masoni etal. 1984) in response to a loss of the putative hypocalcaemic hormone, stanniocalcin (Wagner et al. 1986, 1988; Butkus et al. 1987; Lafeber et al. 1988). Tables 1 and 2 show that the increased rate of urinary calcium excretion by stanniectomized eels occurred after the preliminary rise in plasma calcium concentration. Because the urine/plasma calcium ratio was unchanged after stanniectomy, it is doubtful whether the increased rate of calcium excretion was caused by a change in renal tubular transport.

In the present experiments, the lowered plasma magnesium levels (Fig. 1) may reflect a decrease in net branchial influx of magnesium or simply an early and significant increase in the rate of urinary magnesium excretion following stanniectomy (Fig. 2; Table 2), a consequence of the increased fraction of filtered magnesium excreted by the kidney first reported by Butler (1969). Even though urinary calcium concentration increased following stanniectomy (Table 2), there was an even greater relative increase in the urinary magnesium concentration, with the result that the urinary magnesium/calcium ratio became clearly elevated throughout most of the 14 day observation period (Fig. 3). It is possible, therefore, that the corpuscles of Stannius secrete a hormone that normally increases the renal tubular influx of magnesium in freshwater eels. However, in view of the complexity of the possibly interdependent changes in Na⁺, Cl⁻, K⁺, calcium and magnesium that are often observed after stanniectomy, it is premature to think in terms of a single magnesium-or calcium-regulating hormone.

This investigation was supported by grant A2359 from the Natural Sciences and Engineering Research Council of Canada. The author acknowledges the excellent technical assistance of Ms M. Z. A. Cadinouche.