Influence of ET(B) receptor antagonism on pregnancy outcome in rats.

OBJECTIVE: To determine the effect of endothelin-B (ET(B))-selective receptor antagonism on pregnancy outcome in normal rats. METHODS: ET(B) receptor antagonist (A-192621; 5.0, 10.0, and 15.0 mg/kg per day) or vehicle was infused subcutaneously for 7 days by osmotic pump. Infusion was begun on day 14 of a 22-day gestation. Nonpregnant animals were treated similarly, and blood pressure (BP) responses and plasma antagonist levels were compared to those in pregnant animals. Mean arterial pressure (MAP) was measured on days 1, 4, and 7 of the infusion. Plasma ET(B) antagonist levels were measured on day 7 of infusion. On gestational day 21, fetal and placental weights and viability were evaluated at hysterotomy. Data were analyzed by analysis of variance and are presented as mean +/- standard error of the mean. RESULTS: Fetal and placental weights were significantly lower at doses of 10 and 15 mg/kg per day of the ET(B) antagonist compared with vehicle-treated controls (P <.001); these effects were less severe at 15 than at 10 mg/kg per day despite a fourfold higher plasma level of antagonist. Mean arterial pressure was significantly higher at 10 and 15 mg/kg per day compared with controls, but only on infusion day 1 (P <.05). In contrast, MAPs for nonpregnant rats were elevated throughout the infusion at all doses of the ET(B) antagonist (P <.05). CONCLUSIONS: ET(B) receptor antagonism inhibited fetal growth and increased maternal MAP in a dose-dependent manner, although the effect on BP was not sustained in pregnant animals. ET(B) receptor antagonism is detrimental to pregnancy outcome in the rat.

Risk and patterns of bacteraemia after splenectomy: a population-based study.

During a period in which vaccination of splenectomized patients has been recommended, we analysed the patterns of severe post-splenectomy infections (i.e. bacteraemia or meningitis) in a defined population-based cohort. A total of 561 patients undergoing splenectomy were identified during 1984-93 in a Danish county, and the 538 eligible patients were followed for 1731 person-years. After splenectomy, 38 patients contracted a bacteraemia, of which 45% occurred within 30 d (i.e. during the postoperative period). No cases of meningitis were found. Among splenectomized patients the incidence rate of bacteraemia was 2.3 per 100 person-years at risk. Beyond the postoperative period we found an 8-fold increased risk of bacteraemia. Enterobacteria were the predominant cause (45%), and only 1 case due to Streptococcus pneumoniae was recorded. 89 (17%) died during the postoperative period, and the overall mortality rate was 18.4 per 100 person-years at risk. In all, 60% of the patients had been given a pneumococcal vaccination, and a Cox proportional hazard regression model showed that vaccination significantly reduced the risk of bacteraemia of any cause beyond the postoperative period. We conclude that splenectomy increases the risk of severe infections, and that vaccinated patients carry a lower risk of infection than non-vaccinated ones.

Fetal growth and blood pressure in a Danish population aged 31-51 years.

During the past decade, studies have shown an inverse association between birth weight and blood pressure and risk of coronary heart disease in adult life. From old public archives we were able to trace the birth records of 545 out of 905 persons (60.2%) aged 31-51 years who participated in the Ebeltoft Health Promotion Project in Denmark. We examined the associations between birth weight, length at birth, Ponderal Index and systolic and diastolic blood pressure. No associations were found for women. For men, the mean systolic blood pressure fell from 131.1 mmHg with a birth weight of less than 3300 g to 129.6 mmHg with a birth weight of more than 4000 g, and for diastolic blood pressure 81.6 mmHg to 80.3 mmHg, respectively. For men, the mean systolic blood pressure fell from 135.7 mm Hg with a birth length of 30-51 cm to 131.6 with a birth length of 55-62 cm, and for diastolic blood pressure 83.0 mmHg to 78.8 mmHg, respectively. The associations may reflect organ programming in fetal life.

Secular trends in incidence and mortality of bacteraemia in a Danish county 1981-1994.

We estimated the incidence and mortality of bacteraemia in the County of North Jutland and examined factors that could explain the changes observed. A population-based survey of bacteraemia was conducted in the Danish County of North Jutland during 1981-1994. Data were retrieved from a regional bacteraemia register. The mortality was determined through linkage to the Danish Civil Registration System. A total of 7198 bacteraemias were detected, and the annual incidence increased from 76 per 100,000 person-years in 1981 to 153 in 1994. One major determining factor was a change in blood culture system with a higher volume of blood per sample, but annual numbers of blood cultures also increased. The 30-day mortality rate increased from 17 to 40 per 100,000 person-years during the study period, whereas the case-fatality rate remained constant (23.6%; 95% confidence intervals 22.6%-24.6%). The number of bacteraemias increased significantly. This observation could be explained only partly by changes in demography, in blood culture system, and in diagnostic activity. The case fatality rate remained constant despite the fact that more people were diagnosed with bacteraemia; this indicates that, with recent blood culture practice, more clinically significant bacteraemias are diagnosed.

Developmental expression of aquaporin 1 in the rat renal vasculature.

Aquaporin 1 (AQP-1) is a water channel protein that is constitutively expressed in renal proximal tubule and descending thin limb cells as well as in endothelial cells of the descending vasa recta. Studies in the developing rat kidney have demonstrated that AQP-1 is expressed in renal tubules before birth. However, nothing is known about the expression of AQP-1 in the renal vasculature during kidney development. The purpose of this study was to establish the distribution of AQP-1 in the renal vasculature of the developing rat kidney and follow the differentiation of the vascular system during kidney development. Kidneys from 16-, 17-, 18-, and 20-day-old fetuses and 1-, 4-, 7-, 14-, 21-, and 28-day-old pups were preserved and processed for immunohistochemical studies using a preembedding immunoperoxidase procedure. AQP-1 immunoreactivity was detected using affinity-purified rabbit polyclonal antibodies to AQP-1. AQP-1 was expressed throughout the arterial portion of the renal vasculature of the fetal and neonatal kidney from gestational age 17 days to 1 wk after birth. AQP-1 immunoreactivity gradually disappeared from the renal vasculature between 1 and 2 wk of age and remained only in the descending vasa recta. In contrast, AQP-1 immunoreactivity was not observed in lymphatic vessels until 3 wk of age and persisted in the adult kidney. AQP-1 was also expressed in a population of interstitial cells in the terminal part of the renal papilla at 3 wk of age as well as in the adult kidney. The transient expression of AQP-1 in the arterial portion of the renal vasculature in the developing rat kidney suggests that AQP-1 is important for fluid equilibrium and/or drainage in the developing kidney or, alternatively, plays a role in the regulation of growth and/or branching of the vascular tree during kidney development.

Angiotensin II-stimulated nitric oxide release from porcine pulmonary endothelium is mediated by angiotensin IV.

In this study, a nitric oxide (NO) sensor was used to examine the ability of angiotensin II (AngII), AngIV, and bradykinin (Bk) to stimulate NO release from porcine pulmonary artery (PPAE) and porcine aortic endothelial (PAE) cells and to explore the mechanism of the AngII-stimulated NO release. Physiologic concentrations of AngII, but not Bk, caused release of NO from PPAE cells. In contrast, Bk, but not AngII, stimulated NO release from PAE cells. AngIII-stimulated NO release from PPAE cells required extracellular L-arginine and was inhibited by L-nitro-arginine methyl ester. AT1 and AT2 receptor inhibition had no affect on AngII-mediated NO release or activation of NO synthase (NOS). AngIV, an AngII metabolite with binding sites that are pharmacologically distinct from the classic AngII receptors, stimulated considerably greater NO release and greater endothelial-type constitutive NOS activity than the same amount of AngII. The AngIV receptor antagonist, divalinal AngIV, blocked both AngII- and AngIV-mediated NO release as well as NOS activation. The results demonstrate that AngIV and the AngIV receptor are responsible, at least in part, for AngII-stimulated NO release and the associated endothelium-dependent vasorelaxation. Furthermore, these results suggest that differences exist in both AngII- and Bk-mediated NO release between PPAE and PAE cells, which may reflect important differences in response to these hormones between vascular beds.

Intercalated cell subtypes in connecting tubule and cortical collecting duct of rat and mouse.

At least two populations of intercalated cells, type A and type B, exist in the connecting tubule (CNT), initial collecting tubule (ICT), and cortical collecting duct (CCD). Type A intercalated cells secrete protons via an apical H+-ATPase and reabsorb bicarbonate by a band 3-like Cl-/HCO3-exchanger, AE1, located in the basolateral plasma membrane. Type B intercalated cells secrete bicarbonate by an apical Cl-/HCO3- exchanger that is distinct from AE1 and remains to be identified. They express H+-ATPase in the basolateral plasma membrane and in vesicles throughout the cytoplasm. A third type of intercalated cell with apical H+-ATPase, but no AE1, has been described in the CNT and CCD of both rat and mouse. The prevalence of the third cell type is not known. The aim of this study was to characterize and quantify intercalated cell subtypes, including the newly described third non A-non B cell, in the CNT, ICT, and CCD of the rat and mouse. A triple immunolabeling procedure was developed in which antibodies to H+-ATPase and band 3 protein were used to identify subpopulations of intercalated cells, and segment-specific antibodies were used to identify distal tubule and collecting duct segments. In both rat and mouse, intercalated cells constituted approximately 40% of the cells in the CNT, ICT, and CCD. Type A, type B, and non A-non B intercalated cells were observed in all of the three segments, with type A cells being the most prevalent in both species. In the mouse, however, non A-non B cells constituted more than half of the intercalated cells in the CNT, 39% in the ICT, and 22% in the CCD, compared with 14, 7, and 5%, respectively, in the rat. In contrast, type B intercalated cells accounted for only 8 to 16% of the intercalated cells in the three segments in the mouse compared with 26 to 39% in the rat. It is concluded that striking differences exist in the prevalence and distribution of the different types of intercalated cells in the CNT, ICT, and CCD of rat and mouse. In the rat, the non A-non B cells are fairly rare, whereas in the mouse, they constitute a major fraction of the intercalated cells, primarily at the expense of the type B intercalated cells.

[Use of antiretroviral chemoprophylaxis after occupational HIV exposure]. / Anvendelse af antiretroviral kemoprofylakse efter erhvervsbetinget eksposition for HIV.

The study describes the use of postexposure antiretroviral prophylaxis (PEP) after occupational exposure to HIV in Denmark in the period 1/1 1997-31/12 1998. Forty-seven accidentally exposed persons received PEP in this period, of whom 14 were nurses, 13 physicians and 11 were not employed in the health care system. In 23 cases side effects to PEP were described. In 18 cases the index person was i.v. drug abuser and in 11 cases homosexual/bisexual men. In 18 cases the HIV status of the index person was unknown at the time of exposure. Ten of these index persons were subsequently tested and all found HIV-negative. It is concluded, that the frequency of occupational exposure to HIV is unacceptably high. Further it is emphasised that in cases where the HIV status of the index person is unknown, PEP should only be instituted if the index person is at risk of being HIV infected and has signs of HIV infection.

Trend in incidence and case fatality of meningococcal disease over 16 years in Northern Denmark.

The incidence and case fatality rates of meningococcal disease were assessed in the county of Northern Jutland, Denmark, during the 16-year period from 1980 to 1995. A total of 320 patients were identified from the Meningococcal Research Database, which comprises information from the following sources: (i) the Department of Public Health, to whom notification of meningococcal disease is obligatory; (ii) the Regional Hospital Discharge Registry; and (iii) the register of the regional department of clinical microbiology. In order to assess prognostic indicators assessable at admission, information was collected for each patient from hospital records regarding contacts, symptoms and signs on arrival, laboratory data, and course of disease. The mean incidence was 4.3 cases per 100000 persons per year (range, 2.7-7.7). The incidence increased slightly during the period studied. Overall, the case fatality rate was 9.7%, with a significant rise occurring during the period (P=0.016) and a peak occurring in 1992. Advanced age (> or = 50 years), seizures, impaired consciousness, and skin bleeding on arrival at hospital were predictors of death.

The risk of bacteremia in patients with monoclonal gammopathy of undetermined significance.

We estimated the risk of bacteremia in patients with monoclonal gammopathy of undetermined significance (MGUS) compared with the general population; 1237 cases of MGUS were identified by linking information on detected monoclonal components in the North Jutland County with the Danish Cancer Registry. We evaluated the risk of bacteremia in the MGUS cohort during the 13-yr period from 1981 to 1993 by linkage to the Bacteremia Registry in the County. Follow-up for the occurrence of bacteremia started 30 d after detection of the M-component and continued until malignant transformation, death or until 31 December 1993, whichever came first. The expected numbers of bacteremia was based upon county, age, sex and period-specific incidence rates. The median follow-up period was 3.8 yr. Forty episodes of bacteremia occurred during 5500 person-years versus 18 expected. The crude standardized incidence ratio of bacteremia was 2.2 (95% confidence interval, 1.6-3.0). There was no distinct pattern of bacterial etiology in the MGUS cohort. Although we found an association between MGUS and risk of bacteremia, the overall risk is small and this finding hardly affects the clinical handling of MGUS patients.

Can hospital discharge diagnosis be used for surveillance of bacteremia? A data quality study of a Danish hospital discharge registry.

OBJECTIVE: To assess the data quality of septicemia and sepsis registration in a hospital discharge registry in the County of Northern Jutland, Denmark. DESIGN: Comparison of data from the discharge registry of an 880-bed, public, urban hospital in the County of Northern Jutland with data from a computerized bacteremia database at the regional department of clinical microbiology. SETTING: Urban hospital with approximately 45,000 admissions per year. PATIENTS: The study included 406 episodes of bacteremia in the bacteremia database and 83 discharges with the diagnosis of septicemia registered in the hospital discharge registry between January 1, 1994, and December 31, 1994. INTERVENTIONS: None. RESULTS: Eighteen episodes were registered in both the hospital discharge registry and the bacteremia database. Using the bacteremia database as reference standard, the sensitivity for the diagnosis of septicemia in the hospital discharge registry was 4.4% (18/406; 95% confidence intervals [CI95, 2.4%-6.4%]). By review of hospital records, we estimated the positive predictive value of septicemia registration in the hospital discharge registry as 21.7% (18/83; CI95, 12.8%-30.5%). No blood culture had been obtained in 44.4% (36/81; CI95, 33.6%-55.3%) of the cases with a discharge diagnosis of septicemia. In 33.3% (27/81; CI95, 23.1%-43.6%), the discharge diagnosis of septicemia was given, although blood cultures were negative. CONCLUSIONS: The hospital discharge registry revealed numerous misclassifications, and the system was found not suited for surveillance of, or research in, bacteremia at present.

Continued increase in prevalence of obesity in Danish young men.

BACKGROUND: Previous studies of Danish young men showed a steep increase of obesity since the birth years of the early 1940s with a levelling off in the 1950s. OBJECTIVE: To study the current prevalence of obesity and its recent changes in Danish young men. DESIGN: Cross-sectional study based on weight and height measured at the military board for the birth cohorts 1955, 1965 and 1973-75. SUBJECTS: We obtained information from the old paper files comprising the two cohorts of men born in 1955 (6549 adults) and 1965 (6404 adults) and prospectively examined 4300 from the birth cohorts 1973-75 in a conscription district in Denmark during the period 1 August 1993 to 31 July 1994. RESULTS: The prevalence of obesity (defined as body mass index (BMI) > or = 30 kg/m2) was 1.5% in 1955, 2.1% in 1965 and 4.6% in 1973-75 birth cohorts. The median BMI has increased during the period from 21.7-22.8 kg/m2. CONCLUSION: Obesity is relatively common and has increased among Danish young men.

Ultrastructural localization of osteopontin in the kidney: induction by lipopolysaccharide.

Osteopontin is a secreted phosphoprotein that is expressed in the normal kidney and induced during various pathologic conditions associated with tubulointerstitial injury. However, the exact cellular location of osteopontin in the kidney has been a matter of controversy, and little is known about the role of osteopontin in the kidney. The purpose of this study was to establish the cellular and intracellular distribution of osteopontin in the rat kidney under normal conditions and after injection of a bacterial endotoxin lipopolysaccharide (LPS). Animals received injections of LPS or vehicle at different time intervals from 4 to 20 h before sacrifice. Kidneys were preserved by in vivo perfusion with paraformaldehyde-lysine-periodate (PLP) and processed for light and electron microscope immunocytochemistry using monoclonal antibodies to rat osteopontin. By light microscopy, immunostaining was observed in the descending thin limb and the papillary surface epithelium of both control and LPS-treated animals. After injection of LPS, osteopontin immunostaining was observed throughout the distal nephron and was also present in segments of the proximal tubule, where it was distributed in a punctate pattern. Staining was already present 4 h after injection of LPS and was maximal 6 h after injection. Electron microscopy revealed that osteopontin immunoreactivity in the descending thin limb and distal tubule cells was located in the Golgi apparatus and in small cytoplasmic vesicles, whereas in the proximal tubule labeling was observed in the vacuolar-lysosomal system. Western blot analysis demonstrated a band at approximately 70 kD and confirmed the increase in osteopontin expression after administration of LPS. These results demonstrate that osteopontin is constitutively expressed in cells of the descending thin limb and papillary surface epithelium and is induced throughout the distal tubule after administration of LPS.

Axial distribution of band 3-positive intercalated cells in the collecting duct of control and ammonium chloride-loaded rabbits.

Recent studies have suggested that less than 10% of intercalated cells in the rabbit outer cortical collecting duct (CCD) [1, 2] and less than 3% in the connecting segment (CNT) [3] are identifiable by functional criteria as acid-secreting (type A or alpha) intercalated cells. Other studies, using peanut lectin-binding and the absence of apical endocytic activity to identify bicarbonate-secreting (type B or beta) intercalated cells, have suggested that acid-loading increases the percentage of alpha intercalated cells in the CCD. Because our preliminary observations of band 3 immunoreactivity suggest that the percentages of alpha intercalated cells in the rabbit outer CCD and the CNT are underestimated by physiologic studies and are not altered by chronic acid-loading, we quantified the percentage of alpha intercalated cells in various segments of the collecting duct using light microscopic immunohistochemistry in kidneys of rabbits receiving tap water (control) or 75 mM NH4Cl for 12 days plus 8 daily gavages of 2 to 6 mEq NH4Cl/kg body wt. Mean urine pH values were 5.96 in acid-loaded animals versus 8.47 in controls. Kidneys were preserved by in vivo perfusion with periodatelysine-paraformaldehyde fixation and processed for immunohistochemical colocalization using sequential labeling with monoclonal antibodies and peanut lectin, followed by immunoperoxidase detection. Colocalization of band 3 and carbonic anhydrase II immunoreactivity revealed the following percentages of band 3-positive intercalated cells in control versus NH4Cl rabbits: CNT, 49.0 versus 52.8; initial collecting tubule (ICT), 27.2 versus 34.5; outer CCD, 33.5 versus 30.3; inner CCD, 38.2 versus 41.8; corticomedullary CD, 67.9 versus 58.8. There were no differences between the groups for all comparisons. Similar results were obtained using band 3 protein immunoreactivity and peanut lectin-binding to identify intercalated cell subtypes. However, in NH4Cl-loaded rabbits, peanut lectin-binding was observed in band 3 positive intercalated cells in the outer medullary CD. We conclude that: (1) the percentage of alpha intercalated cells in rabbit CCD subsegments are approximately 50% in the CNT, 30% in the ICT and the outer CCD, 40% in the inner CCD, and 60% in the corticomedullary CD; (2) the percentage of alpha intercalated cells is not altered by chronic NH4Cl-loading; (3) peanut lectin is not a specific marker of beta intercalated cells.

SSCP analysis of paraffin wax embedded tissues in a family with an atypical form of Fabry disease.

To investigate the distribution of a single base pair mutation within a family with one known case of Fabry disease, DNA from paraffin wax embedded necropsy material was studied using single-strand conformation polymorphism (SSCP) analysis. The proband, who presented with an atypical form of Fabry disease, had a G to A transition in exon 6 of the alpha-galactosidase A gene. This patient had mainly cardiac symptoms and late onset disease. Further cases of coronary disorders occurred in this family, including the proband's brother who died at 42 years of age of a cardiac disorder. Formalin fixed, paraffin wax embedded material from the brother and two more distant relatives was available for analysis. SSCP analysis showed that the proband's brother also carried the G to A transition. Thus, the atypical form of Fabry disease and unrelated cardiac diseases with similar clinical symptoms occurred within a single family. The variant form is rare but may account for a few of the numerous cases of cardiac disease in men and should be considered when clusters of cases of cardiac disease occur within a single family.

Role of apoptosis in development of the ascending thin limb of the loop of Henle in rat kidney.

At birth, the rat renal papilla has the structural composition of the mature inner stripe of the outer medulla. All loops of Henle have the configuration of short loops, and there are no ascending thin limbs. This study examines the role of apoptosis in the differentiation of the loop of Henle and the development of the ascending thin limb in the rat kidney. Kidneys of 20-day-old fetuses and 1-, 3-, 5-, 7-, 14-, and 21-day-old pups were preserved for immunohistochemistry and electron microscopy. Using a preembedding immunoperoxidase method, we identified thick ascending limbs by labeling with antibodies to the serotonin receptor, 5-HT1A, and descending thin limbs were identified by labeling with antibodies to aquaporin-1. Three methods were used to identify apoptotic cells as follows: 1) in situ nick end labeling using the ApopTag kit, 2) toluidine blue staining on plastic sections followed by etching, and 3) transmission electron microscopy. At birth, tubules with 5-HT1A immunoreactivity were present throughout the renal papilla, and there were no ascending thin limbs. From 1 to 14 days of age, staining for apoptosis was observed in numerous cells in the 5-HT1A-positive epithelium, beginning at the papillary tip and ascending to the border between outer and inner medulla. This was associated with transformation from a cuboidal to a squamous epithelium and subsequent disappearance of 5-HT1A immunostaining from the transformed cells. Electron microscopy confirmed the presence of apoptotic cells and phagocytosed apoptotic bodies in the thick ascending limb in the renal papilla. We conclude that the ascending thin limb is derived from the 5-HT1A-positive thick ascending limb by apoptotic deletion of thick ascending limb cells and transformation of the remaining tubule cells into the 5-HT1A-negative ascending thin limb.

Effects of chronic hypokalemia on renal expression of the "gastric" H(+)-K(+)-ATPase alpha-subunit gene.

Chronic potassium restriction leads to active potassium reabsorption in the late distal nephron and collecting duct, segments known to express "gastric" H(+)-K(+)-adenosinetriphosphatase (H(+)-K(+)-ATPase) alpha-subunit mRNA. In this study, the cellular distribution and relative abundance of mRNA encoding this isoform were examined in kidneys of normal and potassium-deprived (2 wk) rats. In situ hybridization with isoform-specific cRNA probes demonstrated prominent expression of this gene in the connecting segment (CNT), entire collecting duct, and renal papillary surface epithelium in a comparable distribution in both groups of rats. Hypertrophy of the outer medullary collecting ducts in the inner stripe of potassium-restricted rats was observed. Competitive polymerase chain reaction analysis revealed twofold greater levels of gastric H(+)-K(+)-ATPase alpha-subunit mRNA (normalized to the level of beta-actin mRNA) in the cortex, but roughly comparable levels in the outer and inner medulla, of potassium-restricted rats compared with controls. These data suggest that chronic potassium restriction results in modestly enhanced renal cortical expression of the gastric H(+)-K(+)-ATPase alpha-subunit gene and that this isoform may participate in potassium conversation by the CNT and cortical collecting duct during potassium deprivation.

Role of apoptotic and nonapoptotic cell death in removal of intercalated cells from developing rat kidney.

In the developing rat kidney, both type A and type B intercalated cells are present throughout the medullary collecting duct (MCD), as well as the papillary surface epithelium. After birth, intercalated cells gradually disappear from the papillary surface epithelium and the terminal MCD, and type B cells disappear from the entire MCD. The purpose of this study was to establish the mechanism(s) by which intercalated cells are deleted from the MCD during development. Kidneys from 14-, 16-, 18-, and 20-day-old fetuses and 1-, 3-, 7-, and 14-day-old pups were preserved for light microscopic immunohistochemistry and electron microscopy. Intercalated cells were identified by immunostaining for H(+)-adenosinetriphosphatase (H(+)-ATPase) and band 3 protein. Apoptosis was identified by nick end labeling of DNA fragments, staining with the vital dye toluidine blue, and transmission electron microscopy. Two distinct mechanisms of elimination of intercalated cells were detected. Cells with apical labeling for H(+)-ATPase and basolateral labeling for band 3 protein protruded into the lumen of the MCD as if they were being extruded from the epithelium, and many had lost contact with the basement membrane. Extrusion of the cells with basolateral H(+)-ATPase or with no labeling for H(+)-ATPase was never observed. Apoptosis was observed in the MCD from shortly before birth to 7 days after birth, gradually progressing from the papillary tip toward the outer medulla. Staining for apoptosis was present in H(+)-ATPase-positive apoptotic bodies, located in cells that were negative for H(+)-ATPase. Staining was also occasionally observed in apoptotic cells with basolateral H(+)-ATPase but never in cells with apical H(+)-ATPase. Electron microscopy confirmed the presence of apoptotic intercalated cells in the MCD and demonstrated that apoptotic bodies were located in inner medullary collecting duct (IMCD) cells and principal cells. These results demonstrate that intercalated cells are deleted from the MCD by two distinct mechanisms, one involving apoptosis and subsequent phagocytosis by neighboring principal cells or IMCD cells. Elimination by extrusion affects only type A intercalated cells, whereas deletion by apoptosis appears to occur only in type B intercalated cells.