Microtubules (MT) are involved in the delivery of secretory vesicles, including those carrying AQP2 to the apical surface of renal collecting duct principal cells. MT are polar structures, which typically have their minus end anchored in an organising centre (MTOC), while their plus ends grow away into the cell. Different motor proteins are involved in vesicle transport towards the plus and minus ends of MT, so which motors are required for movement towards a particular cell membrane will depend on MT organisation within the cell. In fibroblastic and neuronal cells, the MTOC is perinuclear, with the plus ends of microtubules being near the plasma membrane. In contrast, there is some evidence that in epithelial cells MT are nucleated from a diffuse apical network, and project towards the basal pole of the cell. To investigate this, we have immunolabelled microtubules and AQP2 in the collecting duct cell line mpkCCDc14 (grown on glass coverslips and filters), before, during, and after treatment with 33 μM nocodazole and/or 1 nM dDAVP. Cells were photographed on a Zeiss LSM 510 confocal microscope, and images analysed using ImageJ software. In control cells, there was an abundant apical network of MT, and a sparser basal network. In between, there appeared to be bundles of MT running around the nucleus. Nocodazole treatment (30 min) decreased MT labelling by 95± 2% in the apical two thirds of the cells (p<0.05, n= 10), although a few wavy microtubules could still be seen. After 40 minutes washout, microtubules had started to reappear, predominantly in the form of an aster in the apical pole of the cells. dDAVP treatment was able to increase the fraction of AQP2 in the apical third of the cell from 30% to 50% of the total (p<0.05, n=10): this redistribution was prevented by nocodazole treatment. These results are consistent with the hypothesis that MT in these cells are organised with their “organising centres” (and thus their minus ends) in the apical part of the cell, and their plus ends projecting down into the cell. This is consistent with our previous evidence that dynein (a minus-end directed motor) associates with AQP2-containing vesicles, and is likely to be important in AQP2 shuttling in response to vasopressin (Marples et al, 1998).