Angiotensin converting enzymes from human urine of mild hypertensive untreated patients resemble the N-terminal fragment of human angiotensin I-converting enzyme.

Angiotensin I-converting enzyme (ACE) activity was analyzed in human urine collected from mild hypertensive untreated patients. DEAE-cellulose chromatography using linear gradient elution revealed two forms of angiotensin I-converting enzyme, eluted in the conductivity of 0.75 and 1.25 mS. The fractions of each conductivity were pooled and submitted to direct gel filtration in an AcA-34 column, and the apparent molecular weights of urinary ACEs were estimated as 90 kDa (for ACE eluted in 0.75 mS) and 65 kDa (for ACE eluted in 1.25 mS). Both enzymes have a K(i) of the order of 10(-7) M for the specific inhibitors studied, and are able to hydrolyze luteinizing hormone-releasing hormone and N-acetyl-Ser-Asp-Lys-Pro as described for N-domain ACE. By Western blot analysis, both peaks were recognized by ACE-specific antibody Y4, confirming the molecular weight already described. A plate precipitation assay using monoclonal antibodies to the N-domain of ACE showed that both forms of ACE binds with all monoclonal antibodies to the active N-domain ACE, suggesting that these forms of human urine ACEs resemble the N-fragment of ACE. The HP2 ACE (65 kDa) is similar to low molecular weight (LMW) ACE from normal subjects, and the HP2 ACE (90 kDa) is different from high molecular weight (190 kDa) and LMW (65 kDa) normal ACEs. The 90 kDa ACE could have an important role in development of hypertension. It will be fundamental to elucidate the molecular mechanism responsible for the genesis of this isoform.

Hypokalemia, glucose intolerance, and hyperinsulinemia during diuretic therapy.

Hypokalemia and glucose intolerance may result from diuretic therapy. Increases in plasma insulin and glucose levels have been observed in thiazide-treated hypertensive patients and have been attributed to a diminished insulin sensitivity induced by diuretic therapy. To investigate the effects of hypokalemia on glucose tolerance and insulin secretion, we studied 21 essential and nine diabetic hypertensive patients after 4 weeks of placebo and after 4 weeks of chlorthalidone therapy (25 mg/day). Plasma glucose and insulin levels were measured for a 3-hour period after a 75-g glucose oral dose. Hypokalemia developed in seven of the essential hypertensive patients (HK group), whereas only one diabetic patient had decreased plasma potassium levels to below 3.5 meq/l. The results obtained in the HK group after chlorthalidone showed that plasma glucose and insulin values increased after the oral glucose load to levels significantly higher than those observed after placebo. In contrast, the patient who remained normokalemic after chlorthalidone did not show any change in plasma insulin and glucose levels during glucose tolerance testing. These results show that diuretic therapy may induce hyperglycemia and hyperinsulinemia and suggest that potassium depletion is involved in the increase in insulin resistance that has been demonstrated during thiazide therapy.