Cyclooxygenase type 2 (COX-2) expression is enhanced in cortical loops of Henle (cTAL) during postnatal rat kidney development. The cTAL proliferates significantly in the postnatal period and COX-2 is crucial for normal development of the kidneys. We asked whether COX-2 expression and kidney proliferation was regulated by salt balance and/or by mineralocorticoids in the postnatal period. During brief anesthesia (ketamin 0.25 mg/10 g rat and xylasin 0.05 mg/10 g rat) at postnatal day 10, rats were either adrenalectomized (ADX) or sham operated. All rats were supplied with cholesterol pellets containing only cholesterol (sham and ADX), corticosterone (ADX-C) and corticosterone and DOCA (ADX-CD). ADX rats were injected with isotonic NaCl to sustain survival. At postnatal day 20 (P20), the rats were humanely killed. For immunohistochemistry rats were anaesthetized with mebumal (0.5 mg/10 g rat), then perfused with 4% paraformaldehyde through the left ventricle for 5 min. ADX prevented the peak in postnatal corticosterone seen between P13-P21 (18.5 ± 1.9 vs 137.3 ± 30 ng/ml plasma, n = 7-11, P<0.05), reduced body growth rate and led to a 5-fold increase in urinary Na/K ratio (0.18 ± 0.05 vs 0.99 ± 0.16, n = 6-7, P < 0.002). The kidney/body weight ratio was slightly elevated in the ADX-CD group compared to the other groups (10.46 ± 0.18 vs 9.83 ± 0.10 mg/g body weight, n = 8-11, P = 0.005). Growth rate was reduced in ADX and ADX-C compared to sham and ADX-CD (body weight increase P10-P20; 13.5 ± 0.9 g; 13.4 ± 1.2 g; 20.9 ± 0.7 g, respectively, n = 7-10, P<0.001). COX-2 expression was significantly elevated in ADX and ADX-C rats despite NaCl supplementation (mRNA 4-fold (1.99×10^-4 ± 2.27×10^-5 and 6.23×10^-4 ± 1.51×10^-4 vs 8.71×10^-4 ± 2.51×10^-4 cpm, n = 7-8, P<0.02, respectively); protein 2-fold (2.7 ± 0.7 and 6.6 ± 1.9 vs 6.1 ± 1.2 OD u × mm2, n = 4, P<0.05, respectively), whereas COX-2 expression was not changed in ADX-CD rats (1.65×10^-4 ± 3.99×10^-5 cpm and 1.2 ± 0.6 OD u × mm2). Immunohistochemical labelling for COX-2 revealed a wide distribution along the full length of cTAL in ADX and ADX-C while in sham and ADX-CD rats COX-2 was hardly detectable. Urine concentrating ability was decreased in ADX and ADX-C animals, measured as lower papillary osmolality (311.1 ± 21 and 392.4 ± 29 vs 552.5 ± 62 mosm/kg, n = 7-9, P<0.002, respectively) and Na concentration (44.7 ± 4 and 48.0 ± 4 vs 85.5 ± 10 mmol/l, n = 7-9, P<0.002, respectively) after 22 h dehydration. We did not detect any change in TAL transporter mRNAs; NKCC-2, ROMK and Na-K-ATPase-α1. In control kidneys at various stages, COX-2 and the proliferation marker PCNA were always mutually exclusive along TAL and there were no obvious differences in PCNA labelling distribution at postnatal day 20 between sham and ADX animals. Thus we conclude that NaCl substitution in the absence of adrenal steroids does not normalize cortical COX-2 expression; that mineralocorticoid exerts a stronger negative influence on COX-2 compared to corticosterone; that COX-2 is not localized to proliferating TAL cells; and that mineralocorticoids and NaCl supply are both required to sustain normal growth and postnatal rat kidney development.