Increased numbers of caveolae in retinal endothelium and pericytes in hypertensive diabetic rats.

PURPOSE: Long-term clinical studies have now shown that tight control of blood pressure in type 2 diabetes reduces the risk of diabetes-related death and common diabetic complications, including diabetic retinopathy. However, the mechanisms by which hypertension enhances diabetic microvascular disease, especially diabetic retinopathy, are poorly understood. We developed an experimental model of hypertension in diabetic rats and studied the early ultrastructural changes in retinal capillaries under these conditions. METHODS: Hypertension was induced in diabetic BioBreeding (BB) rats by unilateral nephrectomy, weekly subcutaneous mineralocorticoid and 0.9% oral saline. Serial blood pressures and ultrastructural features of retinal capillaries were recorded in four groups: normotensive Wistar rats, normotensive diabetic rats, hypertensive Wistar rats and hypertensive diabetic rats. RESULTS: A significant and sustained increase in systolic blood pressure occurred in both groups of nephrectomised rats. There was a significant increase in the number of caveolae (i) in both pericytes and endothelial cells in animals with hypertension and diabetes together compared with all other groups and (ii) in pericytes in animals with diabetes alone. The number of direct contacts between pericytes and endothelial cells was reduced in diabetic and hypertensive diabetic animals. Hypertension and diabetes had an interactive effect in producing retinal capillary basement membrane thickening. CONCLUSIONS: In the BB rat hypertension and diabetes have an interactive effect in increasing the number of caveolae in both endothelial cells and pericytes. We speculate that this may be a reflection of changes in calcium and nitric oxide metabolism in these animals.

Cyclic GMP-linked pathway for renin secretion.

The role of cGMP as a second messenger for renin secretion is contentious. This was investigated using a superfused collagenase-dispersed rat kidney cortex cell preparation devoid of indirect influences on renin secretion. Nitroprusside, atriopeptin II and 8-Br-cGMP all increased renin release but the dose-response relationships were biphasic. At low dose ranges there was a positive correlation between increasing drug concentration and renin secretion, but at high drug concentrations, a negative correlation was apparent. Methylene blue, a guanylate cyclase inhibitor, also suppressed baseline renin release at 10(-5) and 10(-6) M, but stimulated release at 10(-3) M. Using mid-range drug concentrations, the cGMP specific phosphodiesterase inhibitor MB22948 potentiated renin release in response to nitroprusside and 8-Br-cGMP. Inhibition of guanylate cyclase with either methylene blue or LY83583 attenuated renin release in response to nitroprusside, but, as expected, had no effect on 8-Br-cGMP induced release. We conclude that, under physiological conditions, cGMP is a stimulatory second messenger for renin release. This activity is mimicked at low dose ranges by 8-Br-cGMP, nitroprusside and atriopeptin II. In response to high doses of these drugs an unknown inhibitory pathway is activated and this opposes, in a dose-related manner, the stimulatory actions of cGMP for renin release.

Plasma active and acid-activatable renin during the development of one-kidney, one-clip and two-kidney, one-clip hypertension in the rabbit.

Systolic blood pressure in the central ear artery of eight rabbits increased by 21 mmHg (1 mmHg = 133.32 Pa) over 40 days following renal artery clipping and contralateral nephrectomy (one-kidney, one-clip). Plasma active and acid-activatable (pH 2.8) renin did not change significantly. Similar data were obtained from a group of 12 rabbits following renal artery clipping alone (two-kidney, one-clip) except that blood pressure in this group increased for 26 days but then declined until 40 days. Two animals with one-kidney, one-clip hypertension and three rabbits with two-kidney, one-clip hypertension had large increases in plasma active and inactive renin levels, which followed a more exaggerated rise in blood pressure than in the previous two groups. Forty days after unilateral renal artery clipping, the unclipped kidney was removed in 10 animals with two-kidney, one-clip hypertension. A further increase in blood pressure (+29%) occurred in seven of the animals but no change in plasma active or inactive renin. Results were compared with two groups of control animals, a unilateral nephrectomy group and a laparotomy group. None of the surgical procedures used produced a consistent pattern of change in the relative amounts of active and inactive renin in plasma. No marked changes in sodium, potassium, or water balance occurred in any group of animals.

Beta-adrenoceptor mediated secretion of active and inactive renin in conscious sheep.

1. Beta-adrenoceptor linked regulation of plasma active and inactive renin was investigated in sheep with indwelling artery, vein and bladder catheters. Concurrent changes in renal function were also monitored. 2. The beta-adrenoceptor agonist isoprenaline (bolus intravenous (i.v.) dose 0.5 mg/kg, 0.05 microgram/kg per min) increased plasma active renin concentration within 15 min and reached a plateau increase of 306%. Initially plasma inactive renin remained unchanged but after 105 min it decreased. 3. The non-selective beta-adrenoceptor antagonist propranolol (bolus i.v. dose 4 mg/kg, 40 micrograms/kg per min) did not significantly alter basal plasma active renin but selectively reduced inactive renin. The absence of any change in active renin may reflect the relatively low level of stress-induced sympathetic activation in these conscious animals compared to equivalent studies carried out on human subjects. 4. Using the same dose schedules, propranolol completely blocked the isoprenaline-induced increases in plasma active renin secretion even though it had no effect on the basal active renin levels. 5. The selective beta 1-adrenoceptor antagonist atenolol (bolus i.v. dose 5 mg/kg) decreased basal plasma active renin by 24% (+ 135 min) but reduced plasma inactive renin by 75% compared to control. 6. Therefore, both a beta-adrenoceptor agonist (isoprenaline) and two beta-adrenoceptor antagonists (propranolol and atenolol) decreased plasma inactive renin concentration in conscious sheep. 7. These results are discussed in relation to the accompanying changes in arterial blood pressure and aspects of renal function.

Inhibitory role of Ca2+ in the control of renin secretion: a study using superfused dispersed rat renal cortical cells.

1. The role of Ca2+ in the control of renin release was investigated using a collagenase-dispersed rat kidney cortex cell preparation. 2. Superfusion with a series of low [Ca2+] buffers in either ascending or descending order of concentration increased renin release. Exposure to 0.06 mmol/l Ca2+ increased release by 120% (P less than 0.001) when presented as the first buffer in ascending order of concentration and by 79% (P less than 0.001) when presented as the fourth and last in a series of descending order. 3. The Ca2+ entry blocking drug diltiazem in a range of concentrations increased renin release and at 10(-5) mol/l diltiazem the mean stimulation was 35% (P less than 0.01). 4. 8-(N.N-Diethylamino)octyl-3,4,5-trimethoxybenzoate (TMB-8) reduces the release of Ca2+ from intracellular stores and, studied over a range of concentrations, this compound increased renin release. At 10(-5) mol/l TMB-8 the mean increase was 44% (P less than 0.001). 5. None of these experimental manipulations, low [Ca2+], diltiazem or TMB-8, had any effect on the release of adenosine 3':5'-cyclic monophosphate into the cell superfusate, indicating that a decrease in intracellular [Ca2+] increases renin release by a mechanism which is independent of changes in adenosine 3':5'-cyclic monophosphate production. 6. Effects of low [Ca2+], diltiazem and TMB-8 on renin secretion were all shown to be reversible when superfusion with control buffer was resumed.

Secretion control for active and inactive renin: effect of ouabain on release from rabbit kidney cortex slices.

Release of active and inactive renin by rabbit kidney cortex slices was investigated. Inactive renin was estimated as the increase in renin activity after acidification (pH 2.8) of slice supernatant solutions. For kidney slices incubated in complete Krebs bicarbonate buffer, the Na-K-ATPase inhibitor ouabain (100 microM) reduced the secretion of both active (-19.2%) and inactive (-78.9%) forms of renin. In low Na buffers ([Na+] = 23 mM) active renin release was increased and inactive renin was suppressed. Both of these changes were abolished by addition of ouabain (100 microM). The reduction in inactive renin secretion produced by ouabain in complete Krebs buffer did not occur in low [Na+] buffers. In zero Ca2+ buffers containing EGTA (5 mM), secretion of both active and inactive renin was increased but these changes were abolished by addition of ouabain (100 microM). Incubating kidney slices in low Na+, zero Ca2+ media revealed differences between the secretion control mechanisms for the two forms of renin. The separate stimulatory effects of low Na+ and low Ca2+ were not additive for the release of active renin and inclusion of ouabain resulted in similar secretion rates to those under control conditions. For inactive renin secretion, in the absence of Ca2+ release mechanisms still respond to reduction in Na+ with decreased secretion. Conversely, in low Na+ buffers, removal of Ca2+ still promotes inactive renin secretion. These changes were abolished by the addition of ouabain (100 microM). Slices did not change in weight during incubation in media which did not contain ouabain. Addition of this inhibitor to control buffers and low Na buffers did result in an increase in weight. This correlated with the presence of Ca2+ in the buffer and did not appear to be related to [Na+]. These studies again show that the mechanisms regulating the secretion of active and inactive renin are not identical and support the hypothesis that Na+ have differing roles to play in the regulation of these two forms of renin.

Verapamil-induced secretion of active and inactive renin in conscious sheep.

Regulation of plasma active and inactive renin was investigated using conscious sheep with indwelling artery, vein and bladder catheters. Control and experimental studies were carried out in the same animals on different days. The calcium antagonist drug verapamil was given as an initial bolus injection (0.5 mg/kg) followed by a continuous infusion (0.1 mg/kg per h) over a 2.5 h period. Plasma active and inactive renin changed in parallel. Both were significantly increased within 15 min of the initial drug dose and both attained a peak increase after 45 min. Thereafter, the two forms of renin returned to basal levels despite continued infusion of the drug. Effective renal plasma flow (CPAH) was also transiently increased by verapamil and followed a similar time course to changes in plasma active and inactive renin concentration. Arterial blood pressure, however, remained suppressed by verapamil for the duration of the study. Verapamil did not alter urine flow or sodium and potassium excretion rates. These results are discussed in relation to the possible link between intrarenal haemodynamics and renin secretion in conscious and in anaesthetized animals and also in relation to the concept that variation in the relative amounts of active and inactive renin secreted in differing physiological situations represents a mechanism for regulating the renin-angiotensin system.

Haemorrhage-induced secretion of active and inactive renin in conscious and pentobarbitone-anaesthetized sheep.

The response of plasma levels of active and inactive renin to haemorrhage was investigated in sheep with indwelling artery and vein catheters. In conscious animals, loss of 10% of estimated blood volume over a 5 min period increased plasma active renin by a mean of 59%, a surprisingly small change. Plasma inactive renin also increased, but only by 86%. Maximum increases in both forms of renin occurred within 1 h of the haemorrhage. The effects of an equivalent blood loss were investigated in pentobarbitone-anaesthetized sheep maintained in an upright posture using padded slings. Anaesthesia per se had no effect on plasma active or inactive renin. In anaesthetized sheep, 3 h after haemorrhage, plasma active renin had increased by 403% and inactive renin by 299% above control values, but a plateau (maximum) response was not reached during this time. In both conscious and anaesthetized animals the haemorrhage-induced increases in active and inactive renin occurred in parallel. It appears that haemorrhage of this intensity is a comparatively mild stimulus to increase plasma renin concentration in conscious sheep but is much more effective in anaesthetized animals. This may be linked to anaesthetic-induced increases in prostaglandin synthesis within the kidney.

Fluid balance and secretion of antidiuretic hormone following transsphenoidal pituitary surgery. A preliminary series.

Hyponatremia developing some days after transsphenoidal pituitary adenectomy is a treacherous complication of uncertain cause. Of 19 patients monitored in a pilot study at the Wessex Neurological Centre, plasma sodium fell below 125 mmol/liter in three patients at times ranging from 6 to 9 days postoperatively. One patient had evidence of inappropriate secretion of arginine vasopressin (AVP), and the other two probably had steroid insufficiency despite apparently adequate steroid cover. In a more detailed study, the fluid and sodium balance of a further 16 patients was monitored for 7 to 11 days following transsphenoidal surgery together with plasma cortisol, renin, and AVP concentrations. No patient became severely hyponatremic. Three developed partial diabetes insipidus. Two patients with Cushing's disease had evidence of postoperative corticosteroid insufficiency despite normal steroid protection. An inappropriately low plasma cortisol concentration was recorded in both. Plasma AVP concentrations did not show a delayed surge postoperatively. Delayed hyponatremia appears to occur most often in patients with hypoadrenalism, as glucocorticoid cover is decreased. It results from water retention combined with natriuresis, and is reversed by glucocorticoid treatment.

Secretion of active and inactive renin by rabbit kidney cortex slices: effects of verapamil, flunarizine and A23187.

Secretion control for active and inactive (acid-activatable) renin was investigated using a rabbit kidney cortex slice preparation. In complete Krebs-Ringer bicarbonate buffer (pH 7.4) the calcium-channel-blocking drugs verapamil (50 microM) and flunarizine (50 microM) had no effect on active renin release but increased inactive renin secretion. In Ca-free or Ca-depleted (+EGTA) buffers secretion of both active and inactive renin was increased. Neither verapamil nor flunarizine under these conditions had any additional effect on active renin secretion but inactive renin release was stimulated by verapamil and inhibited by flunarizine. Although having no effect in the presence of Ca++ ions, the ionophore A23187 (17 microM) selectively abolished the secretion of inactive renin in the absence of calcium. The mechanisms regulating the secretion of the two forms of renin are clearly not identical. Differential release of active and inactive renin by the kidney may contribute to the overall control of the renin-angiotensin system.

Secretion control for active and inactive renin: effects of calcium and potassium on rabbit kidney cortex slices.

Release of active and inactive renin by rabbit kidney cortex slices was investigated. Inactive renin was estimated as the increase in renin activity after acidification (pH 2.8) of slice supernatant solutions. In Ca2+-free media, release of both active and inactive renin was increased but the changes in inactive renin were more marked. The percentage of total renin released which was inactive ranged from 8.3% ([Ca2+] = 9.2 mM) to 34.5% (zero [Ca2+]) with a linear relationship (r = -0.96) over the range of [Ca2+] studied. Depolarizing media ([K+] = 20 mM) suppressed release of inactive renin more than release of the active form. This effect, for both forms of renin, was lost when Ca2+ ions were omitted from the incubation media. This suggests that influx of Ca2+ ions was responsible for the reduced renin secretion following depolarization of the juxtaglomerular cells. Reducing the [K+] of the incubation buffer from 5.7 mM (control) to 1.0 mM did not alter active renin but increased release of inactive renin. Low [K+] media abolished the stimulatory effect of low [Ca2+] on release of both forms of renin. In incubation media with low [Ca2+] or with low [K+] the mixture of renins released by the kidney cortex slices correlated with that found in extracts of non-incubated kidney: that is, about 35% of the total renin was in the inactive form. Mechanisms controlling the secretion of active and inactive renins by the kidney are at least partially independent. Differential secretion of the two forms, perhaps linked to regulation of the activation of inactive renin before release, appears to be the basis of a control step in the over-all expression of the renin-angiotensin system. This may be coupled in some way to juxtaglomerular cell Na+ ion flux.

Furosemide fails to alter plasma active or inactive renin in conscious sheep but does so in anaesthetized animals.

Regulation of plasma levels of active and inactive renin was investigated using sheep with indwelling artery, vein and bladder catheters. Control and experimental studies were carried out in the same animals on different days. Volume depletion during any single experiment was limited to a maximum of 50 ml. Despite large changes in sodium and water excretion, the diuretic furosemide at two dose levels, 1 and 10 mg/kg, failed to alter plasma levels of either active or inactive renin in conscious sheep. Induction of pentobarbitone anaesthesia in the sheep did not, per se, alter either plasma active or inactive renin. Furosemide (10 mg/kg) in anaesthetized animals produced a similar diuresis and natriuresis response to conscious sheep, but plasma active renin increased by 270% and inactive renin decreased to zero. In conscious sheep given an infusion of papaverine, furosemide also produced an increase in plasma active renin and a concurrent decrease in the inactive form. In both anaesthetized animals and in conscious sheep infused with papaverine, furosemide-induced intrarenal vasodilation, as evidenced by changes in clearance of p-aminohippuric acid, was much reduced in comparison to the conscious animals. This may be significant in relation to the control of renin secretion. It appears that the macula densa sodium receptor, which is considered to regulate renin release, will only function after it has been primed by other intra- or extrarenal factors. This is discussed, particularly in relation to the possible role played by the prostaglandin system.

Active and inactive renin release from rabbit kidney cortex slices: effect of sodium concentration and of furosemide.

1. Active and inactive renin release by rabbit kidney cortex slices was investigated. Inactive renin was estimated as the increase in renin activity after acidification (pH 2 . 8) of slice supernatant solutions. 2. Active renin release was increased when incubation medium [Na+] was reduced. This relationship was linear (r2 = 0 . 96) over the range [Na+] = 23-133 mM. 3. For the same range of [Na+] inactive renin secretion decreased when medium [Na+] was reduced (r2 = 0 . 92). Therefore, the proportion of total renin which was in the inactive form decreased linearly as [Na+] was reduced (r2 = 0 . 97). 4. Chloride ions did not appear to be important in altering the secretion of either active or inactive renin. 5. Adding furosemide to the incubation medium in concentrations up to 40 micrograms/ml. did not change secretion of either form of renin. The action of furosemide on secretion of active and inactive renin in vivo is therefore secondary to altered renal function. 6. Regulation of the relative amount of active and inactive renin in plasma could be entirely an intrarenal event. It is not essential to invoke a plasma activating enzyme for inactive renin in order to explain changes in plasma levels of the two forms of renin. 7. This paper supports the hypothesis that release of inactive renin by the kidney is controlled by a sodium-sensitive mechanism.

Secretion control for active and inactive renin: interactions with calcium and sodium ions.

Excitation-secretion coupling for active and inactive renin was investigated by manipulating the ionic composition of media in which rabbit kidney cortex slices were incubated. Alterations in [Ca++] and [K+] produced qualitatively similar, but quantitatively distinct, changes in the release of the two forms of renin. Under control conditions, the slices secreted 15-20% of total renin in the inactive form and exposure to Ca++-free or low [K+] media increased this to 30-40%, a similar proportion to that stored within the kidney. Decreasing [Na+] stimulated release of active renin but reduced inactive renin. There appeared to be antagonism between the effects of Na+ and Ca++ ions in respect of active renin but synergism for release of the inactive form. It is suggested that the characteristics of a single renin release mechanism in the rat kidney resemble those for secretion of inactive, rather than active, renin in other species.

Isoprenaline-induced secretion of active and inactive renin in anaesthetized rabbits and by kidney cortex slices.

1. The effect of the beta-adrenoceptor agonist isoprenaline on the secretion of active and inactive renin was investigated in two preparations. 2. In ten urethane-anaesthetized rabbits isoprenaline, given as a renal artery infusion, had relatively minor effects on renal sodium excretion (increased) and systemic arterial blood pressure (decreased). Urine volume, potassium excretion, creatinine clearance and serum electrolytes were all unchanged. Plasma active and inactive renin both increased immediately and returned to basal values after ceasing the isoprenaline infusion. 3. No significant changes in either plasma renin activity or renal function were observed in a group of ten control animals. 4. The magnitude of the isoprenaline-induced changes in plasma active renin was similar to that in a previous study of frusemide diuresis, but the time course was quite different. Inactive renin disappeared from plasma during frusemide diuresis. 5. Renin release by rabbit kidney cortex slices was also studied. Isoprenaline, added to the incubation medium, caused a dose-related increase in active renin secretion, but inactive renin release remained unchanged. This is in marked contrast to a previous study where reducing [Na+] increased active renin and inhibited inactive renin output. 6. These data support our previous suggestion that activation of inactive renin is regulated by a sodium-sensitive intrarenal mechanism.

Inactive renin in rabbit plasma: effect of frusemide.

1. An inactive form of renin exists in rabbit plasma. This can be activated, and therefore measured, after acidification (pH 2.8). 2. The effect of frusemide diuresis, with replacement of volume losses, on plasma levels of active and inactive renin was studied over a 3 1/2 h time course. Plasma active renin increased during frusemide diuresis but inactive renin disappeared from the circulation. The time courses for the changes in the two forms of renin were similar. 3. The peak of the frusemide-induced changes in renal function (urine flow, sodium and potassium excretion and creatinine clearance) preceded the maximum changes in the two forms of renin by 90 min. 4. The response of plasma levels of inactive renin to physiological stimuli depends on the nature of the stimulus, as well as its duration. Some form of sodium-sensitive mechanism may control the activation of inactive renin.

Blockade of pressor responses to angiotensins I and II and noradrenaline using phentolamine, propranolol and hexamethonium in conscious rabbits.

Angiotensin I is known to have direct agonist activity at some target tissues, independent of its conversion to angiotensin II. Aspects of its possible direct role as a pressor agent were investigated in conscious rabbits. Phentolamine (3 mg/kg i.v.), a dose which did not affect baseline blood pressure, reduced the pressor response to angiotensin II by 17% (P less than 0.05) but did not alter the sensitivity to angiotensin I. Noradrenaline activity was reduced by 56%. Propranolol (2 mg/kg i.v.), a dose which did not affect baseline blood pressure but which substantially reduced the depressor response to isoprenaline, had no effect on the pressor activity of either angiotensin I or II. Noradrenaline potency was reduced by 32%. Hexamethonium (20 mg/kg i.v.), caused a potentiation of the response to angiotensin I and II and noradrenaline. This was probably a non-specific action associated with the decrease in baseline blood pressure. No differential effect of ganglion blockade on the two angiotensins was noted. The dissociation of the effects of phentolamine on angiotensins I and II provides further evidence that pressor actions of these two compounds act through partially different mechanisms.