2004 Global Researcher Conference Proceeding
April 09 - 11, 2004
| Conference: | 2004 Global Researcher Conference |
|---|---|
| Title: | Lithium-induced Nephrogenic Diabetes Insipidus: A Cell Culture Model |
| Authors: | Shaw, Stephen; Kamsteeg, Erik-Jan; Vandewalle, Alain; Deen, Peter M.T. |
| Institutions: | Nijmegen Centre for Molecular Life Sciences, University of Nijmegen, INSERM U773 |
Lithium, a widely used treatment for bipolar affective disorders, often causes nephrogenic diabetes insipidus. In vivo studies have demonstrated that chronic lithium treatment-induced NDI is associated with a reduction in the expression of the vasopressin (VP) sensitive water channel aquaporin-2 (AQP2). However, the mechanism by which lithium causes NDI is unknown. In this study we have used a mouse cortical collecting duct cell line, mpkCCDc14, which expresses AQP2 endogenously when stimulated with VP, to establish and analyse a cell model of lithium-induced NDI.
Incubating confluent monolayers of filter-seeded mpkCCDc14 cells with the VP analogue dDAVP resulted in a time-dependent increase in the expression of AQP2, with maximal levels of AQP2 expression being reached between 72 and 96 hours of treatment. In all subsequent experiments cells were stimulated with dDAVP for a total period of 96 hrs, to induce maximal expression of AQP2, whilst being simultaneously incubated with lithium for the times indicated. Treating cells with lithium (1 mM for 48 hours, or 10 mM for 24 or 48 hours) resulted in a marked reduction in AQP2 protein levels compared to cells treated with dDAVP alone. Analysis of AQP2 mRNA levels by Northern blotting revealed that lithium, when given under the same conditions as above, also caused a decrease in AQP2 mRNA.
Immunocytochemical detection of AQP2 in cells treated with dDAVP for 96 hours showed intense labelling for AQP2 at both apical and basolateral membranes, as well as some intracellular labelling. Lithium, at 1 mM for 24 hrs, resulted in reduced plasma membrane localisation of AQP2, with a reduction in AQP2 expression evident after 48 hrs. AQP2 expression was virtually abolished after 10 mM lithium treatment (for either 24 or 48 hours), with very little labelling of the plasma membrane. Additionally, the plasma membrane expression of AQP2 in cells treated with 10 mM lithium was restored by briefly adding the adenylate cyclase activator forskolin; suggesting that forskolin is able to activate adenylate cyclase, even in the presence of lithium. As inhibition of adenylate cyclase is thought to contribute to lithium-induced NDI, we tested whether this coincided with AQP2 down-regulation using cAMP assays, and found that lithium had no effect on dDAVP-stimulated cAMP generation.
In conclusion, we have generated a cell model of lithium-induced NDI using concentrations of lithium that are typically found in the blood plasma (1 mM) and urine (10 mM) of patients receiving lithium therapy. In this system, the reduction of dDAVP-induced AQP2 protein and mRNA expression by lithium was also associated with decreased plasma membrane targeting of AQP2. Strikingly, the onset of AQP2 down-regulation by lithium was not associated with reduced adenylate cyclase activity. Thus implying that the mechanism by which lithium actually causes NDI may not be through adenylate cyclase inhibition, but via effects on other proteins and/or intracellular pathways.
One possible side effect of taking the drug, lithium, is NDI. Researchers know that lithium use is associated with the reduction of the number of AQP2 in cells, however they do not know how lithium actually causes NDI. Shaw, et al., created a laboratory cell culture that allowed them to begin to understand the dynamics of lithium-induced NDI. They added concentrations of lithium similar to the concentrations found in the blood and urine of people taking lithium. Treating cells with lithium resulted in a reduction in both the number of AQP2s in the cell membrane, and later a reduction of AQP2 and AQP2 mRNA.
When the researchers added the adenylate cyclase activator, forskolin, expression of AQP2 in the cell membrane was restored. This indicates that forskolin is able to activate adenylate cyclase even when there is lithium in the cells. Additionally, tests determining the degree of adenylate cyclase activity demonstrated that vasopressin could maximally stimulate this enzyme in the presence of lithium. This is an important finding because until now, an inhibition of adenylate cyclase by lithium was thought to contribute to lithium NDI. This research indicates that lithium must cause NDI by some other means.