Magnesium metabolism studies

Abstract

Five experiments, including in vivo and in vitro studies, were carried out to provide quantitative data for modelling of how potassium (K), sodium (Na), potential difference (p.d.) and volatile fatty acids (VFA) influence magnesium (Mg) metabolism in sheep.

In the first experiment, four levels of K were infused intra-ruminally to raise total K intake to a range (15, 25, 35 and 45 g K/d) which animals experience grazing New Zealand pastures. A technique for measuring p.d. across the ruminal wall was established and was used to monitor the p.d. continuously during 48 h. Diurnal variation in p.d. was relatively small, but significant treatment differences were associated with changes in ruminal K concentration. The p.d. increased linearly from 45.8 to 52.6 mV when K concentration in ruminal fluid increased from 58.7 to 80.6 mmol/l as a result of K intake from 15 to 25 g/d and increased slightly further from 52.6 to 53.8 mV with a further increase in ruminal K concentration from 80.6 to 117.2 mmol/l. Change in urinary Mg excretion, which was used to observe apparent Mg absorption, followed a similar pattern, decreasing from 0.43 to 0.37 g/d when K intake was increased from 15 to 25 g/d. Further decreases were small from 0.37 to 0.34 g/d as K intake was increased from 25 to 45 g/d. Unexpectedly, there was a linear relationship between plasma Mg concentration and K intake extending beyond the range (25 g K/d) at which K affected Mg absorption.

These results confirm the important effect of p.d. and K intake on Mg absorption in ruminants, but suggest a second effect of K on Mg homeostasis may exist.

Analysis of data from experiment 1 and data from the literature have suggested that increase in water consumption often occurs as a result of dietary addition of Na or K salts. The second experiment, therefore, was assigned to attempt to separate the effects of increasing Na intake from that of increasing water consumption on Mg metabolism. Increasing Na intake by ten-fold (13 g Na/d), but not water loading (infusion of approx. 3 l/d into the rumen), resulted in increases in the Na:K ratio in ruminal fluid from 0.7:1 to 1:1 and in urinary Mg excretion from 0.27 to 0.38 g/d, but apparent absorption of Mg was not altered. A comparable increase in water loading in animals consuming a low Na diet (1.3 g Na/d) did not change either absorption or urinary excretion of Mg.

Volatile fatty acids, a major end-product of fermentation in the large intestine have been suggested in the literature to increase Mg absorption in rat or rabbit caecum, but this is not well documented in ruminants. The effects of intra-ileal infusion of VFA at rates of 0, 50, 100 and 150 mmol/d on urinary Mg excretion was studied in sheep. Flow of digesta and Mg was determined using ⁵¹Cr and ¹⁴¹Ce and p.d across ruminal, duodenal and colonic walls was monitored continuously. A preliminary measurement of p.d. across the duodenal and the colonic walls showed no significant differences among VFA treatments and those values averaged 8.85 and 20.3 mV, respectively. These figures were much lower than p.d. across the ruminal wall (54.2 mV). Increasing VFA infusion from 0 to 150 mmol/d tended to decrease colonic pH from 8.29 to 8.09 and increase fractional solubility of Mg in colonic digesta from 0.28 to 0.37. Data from the flow of Mg showed that increasing infusion of VFA into the terminal ileum did not alter net Mg absorption. Across all treatments, 43% of total Mg was absorbed from the forestomachs and 57% from the large intestine.

In a later experiment, magnesium chloride (MgCl₂) solutions were infused intra-ileally at rates of 0, 0.2, 0.6, 1.0 and 1.6 g Mg/24 h in five sheep offered a lucerne hay diet in the first stage and a ryegrass diet in a subsequent stage. On the hay diet urinary Mg excretion tended to increase from 7.7 to 11.7 mg/h when infusion of Mg was increased from 0 to 0.2 g/24h but no further change occurred with increase in rate of Mg infusion. This was not observed on the ryegrass diet. These results on their own may suggest that a transport mechanism for Mg occurs in the hind-gut which is saturated at relatively low Mg concentration.

Finally, in vivo and in vitro experiments were carried out to obtain more quantitative data on differences between diets in ability to supply Mg, the examples chosen being concentrate, lucerne hay and ryegrass. First, an in vivo experiment with nine ewes was conducted to determine absorption of Mg and the p.d. across ruminal wall and colonic wall. The p.d. across the ruminal wall (42.2 mV) and the colonic wall (17.6 mV) of sheep offered the concentrate diet was apparently lower than that of those offered the hay (49.4 and 37.0 mV) and the ryegrass (55.7 and 32.0 mV) diets. The observed relationship between p.d. and concentration of K in digesta in the rumen was also observed in the colon but with a lower elevation. As expected, Mg absorption as judged by urinary Mg excretion in sheep offered the concentrate (0.38 g Mg/d) diet was higher than in those offered the hay (0.28 g Mg/d) and the ryegrass (0.21 g Mg/d) diets. These data taken together suggest that conditions created by diet, viz. digesta pH, p.d., and Mg concentration may vary between sites and suggest the possibility that absorption of Mg may occur at different sites on different diets.

The pH of digesta from the rumen and the caecum of animals offered these diets was manipulated in vitro in range of 2-12 using inorganic acids or alkali. Solubility of Mg was pH-dependent: differences between diets were small for ruminal digesta, but were large in caecal digesta. At pH 8, the estimated fractional solubility of Mg in caecal digesta was 0.36, 0.29 and 0.19 for the hay, the concentrate and the ryegrass diets, respectively. On all diets and at both sites, most Mg was associated with the water-soluble and the alkali-insoluble fractions. The soluble fraction for ruminal contents on the hay diet (0.18) was low compared to that on the ryegrass (0.45) and the concentrate (0.36) diets, but much higher for caecal digesta on the hay diet (0.61 vs 0.39 and 0.43) than on the ryegrass diet and the concentrate diet, respectively, despite high caecal pH.

In conclusion, the results in this thesis confirm in, in vivo studies, work from in vitro studies demonstrating the importance of p.d. and ruminal K concentration for Mg absorption in ruminants. They also confirm more recent studies which suggest that the hind-gut can be equally important as the fore stomachs in Mg absorption. Some evidence is presented to suggest a saturable transport process of Mg absorption in the hind-gut. A much lower p.d. observed between the contents in the colon than in the rumen would be favourable for Mg absorption, but the studies suggest that large variation may exist between diets in p.d. generated in the colon, and which needs to be further characterised. A major and potentially important finding was of differences among diets in proportion of Mg associated with the water-soluble fraction in both the rumen and the hind-gut which suggest that sites of absorption of Mg may differ between diets.