High rates of potassium intake are associated with protection from cardiovascular diseases in populations consuming primitive diets and in vegetarians living in industrialized cultures. In studies in humans and in animals, a strong inverse association between potassium intake and hypertension and stroke has been described. However, acceptance of the putative protective effect has been limited by inadequate understanding of 1) long-term potassium regulation, and 2) mechanisms by which small changes in plasma potassium concentration may affect development of cardiovascular diseases. In this review, we present results from analyses of long-term potassium regulation that indicated 1) changes in potassium intake may result in potassium concentrations from 3.1 to 4.6 mmol/l, and 2) when the initial rate is below normal, potassium concentration is very sensitive to changes in potassium intake rate. In addition, we present results that provide bases for possible mechanisms by which potassium may protect against cardiovascular diseases: 1) increases in potassium inhibit free radical formation from vascular endothelial cells and macrophages; 2) elevation of potassium inhibits proliferation of vascular smooth muscle cells; 3) platelet aggregation and arterial thrombosis are inhibited by elevation of potassium; and 4) renal vascular resistance is reduced and glomerular filtration rate is increased by elevation of plasma potassium. We propose that elevation of dietary potassium intake increases plasma potassium concentration, thereby inhibiting free radical formation, smooth muscle proliferation, and thrombus formation. As a result, the rate of atherosclerotic lesion formation and thrombosis will be diminished. In addition, we propose the increase in glomerular filtration rate will cause a shift in the relationship between arterial pressure and sodium excretion that will lead to a reduction in arterial blood pressure. By these actions, high levels of dietary intake of potassium could provide the observed protection against the cardiovascular diseases that have plagued humankind since we began eating a modern high-sodium, low-potassium diet.

We conducted this study to determine whether physiological changes in potassium concentration affect free radical formation by vascular cells. We assessed the effects of potassium on reactive oxygen species formed by cultured endothelial and monocyte/macrophage cells or freshly isolated human white blood cells by cytochrome c reduction or luminol chemiluminescence, respectively. Reducing potassium concentration of endothelial cell media (normally 5.1 to 6.1 mmol/L) to 3.0 mmol/L exponentially increased the rate of cytochrome c reduction, up to 8.4-fold at 2 hours; raising potassium concentration to 5.5 or 7.0 mmol/L at 1 hour reduced the maximal rate of cytochrome c reduction by 86% or 93%. Subsequent studies were done 30 to 75 minutes after media change. Potassium reduced the rate of cytochrome c reduction by 49% (endothelial cells) to 55% (monocytes/macrophages) between 3.0 and 7.0 mmol/L; the greatest decrement (20% to P otassium intake (60 to 80 1 mmol/d) and urinary potassium excretion 2 are inversely related to the incidence of stroke-associated mortality in adult women. Most studies also find a significant inverse relation between potassium intake (ranges: 23 to 54, 3 Moreover, the correlation coefficient of the relation between plasma potassium and diastolic blood pressure is three times greater in young (<36 years) than in old (>49 years) patients. 26 It has been proposed that potassium antagonizes the formation of early renovascular lesions 25 thought to initiate hypertension in youth. 27Support for this hypothesis has come from animal studies in which increased dietary potassium reduces the incidence of renovascular and aortic fatty-streak lesions in rats. -29There is a highly significant inverse correlation between plasma potassium (3.0 to 5.5 mmol/L) and blood pressure in hypertensive patients less than 36 years of age. 26 Varying potassium intake from 0 to 300 mmol/d for 5 days causes a corresponding 0.8 -mmol/L change in fasting plasma potassium.11 Lowering potassium intake from 96 to 16 mmol/d for 10 days reduces fasting plasma potassium from 4.2 to 3.4 mmol/L.12 Raising potassium intake from 80 to 200 mmol/d for 8 weeks raises fasting plasma potassium from 4.2 to 4.4 mmol/L.15 High potassium meals (100 mmol) further increase plasma potassium over 4 hours, peaking 2 hours after eating at 0.6 mmol/L above fasting levels. 30 Moderate potassium meals (25 mmol) do not alter postprandial plasma potassium. 30 These studies indicate that changes in dietary potassium can alter fasting plasma potassium by at least 1 mmol/L and peak postprandial plasma potassium by more than 1.5 mmol/L.Habitual changes in potassium intake may elicit greater changes in plasma potassium. The human body contains 2460 mmol of potassium, 31 35 times the average daily adult potassium intake, most of which is excreted. 32 Animal studies have shown that body potassium stores can change by 23% and that potassium balance is not achieved weeks after potassium intake is altered.

Stripped rabbit distal colonic mucosa was studied in vitro in Ussing chambers to investigate the effects of adrenergic stimuli on Na+, K+, and Cl- transport. The adrenergic stimuli epinephrine and norepinephrine decrease short-circuit current in a dose-dependent manner, with a half-maximal effect at 5 X 10(-7) M and a maximal effect between 10(-5) and 10(-4) M. The effects produced by norepinephrine and epinephrine can also be elicited by the beta 1-agonist dobutamine, but not by the beta 2-agonist terbutaline or the alpha-agonist phenylephrine. In addition, the effects of adrenergic stimulation can be inhibited by the beta-antagonist propranolol but not by the muscarinic antagonist atropine, the alpha 2-antagonist yohimbine, or tetrodotoxin. The decrease in short-circuit current elicited by adrenergic stimuli is accompanied by an increase in net K+ secretion with no change in net Cl- or Na+ transport. This increase in net K+ secretion elicited by beta-adrenergic stimulation can be inhibited by trifluoperazine but not by indomethacin. These studies suggest that K+ transport by the colon can be regulated by adrenergic agents acting via beta 1-receptors.

In vitro preparations of rabbit descending colon were studied under steady-state short-circuit conditions to determine 1) the K concentration dependence of unidirectional K fluxes; 2) the effects of the K channel blocker barium and the diuretic agent furosemide; and 3) the steady-state tissue specific activity of 42K when added to the luminal bathing solution. Results from these studies reveal that 1) labeling of cellular K from the mucosal solution is less than 25% of that from the serosal solution; 2) both unidirectional K fluxes are composed of saturable and nonsaturable components; 3) the serosal-to-mucosal saturable component is abolished by ouabain, and subsequent addition of 2,4-dinitrophenol abolishes the saturable component of the mucosal-to-serosal K flux; 4) luminal or serosal barium alters K transport in a manner consistent with the presence of barium-sensitive K conductances at both membranes; 5) luminal furosemide did not alter K transport; and 6) there is no shunt selectivity for K. We conclude that the majority of both unidirectional K fluxes follow a transcellular pathway and that both the apical and basolateral membranes possess active K uptake mechanisms and barium-sensitive K exit mechanisms.

Stripped rabbit colonic mucosa was studied in vitro in Ussing chambers to further investigate the role of Ca in regulating K and Cl secretion stimulated by the divalent cation ionophore A23187, prostaglandin E1 (PGE1), or 8-bromo-cAMP (8BrcAMP). To assess the effects of these secretagogues on the paracellular shunt permeability, we measured the Na concentration dependence of the serosal-to-mucosal Na flux in the absence or presence of these stimuli. Results from these studies reveal that changes in net K and Cl secretion produced by secretory stimuli cannot be accounted for by a change in shunt permeability. The possible involvement of Ca in the secretory response of the colon to these stimuli was investigated by measuring the changes in Cl and K transport elicited by A23187, PGE1, or 8BrcAMP in the absence or presence of trifluoperazine (10(-4) M) added to the serosal bathing solution. Trifluoperazine alone did not significantly alter basal Na or Cl fluxes or short-circuit current (Isc) but did decrease transepithelial conductance (Gt) and the serosal-to-mucosal K flux. Pretreatment of the tissues with trifluoperazine significantly reduced or abolished the changes in K fluxes elicited by A23187, 8BrcAMP, or PGE1 without altering the changes in Cl transport, Isc, and Gt. These results suggest that K secretion induced by these secretagogues involves an increase in intracellular Ca concentration and may be mediated by calmodulin.