Gene Targeting and Generation of trpv4-Null Mice. The trpv4 gene was inactivated by a deletion of exon 12 that encodes the pore-loop domain and adjacent transmembrane domains 5 and 6. Mouse trpv4 genomic DNA clones were isolated from a 129SvJ l -phage genomic library (Stratagene). The 5¢ homology arm harbored exons 10 and 11 (2.3 kb). It was Pfu-amplified from a 21-kb genomic l clone. The 3¢ homology arm harbored the final three exons 13-15 (5.5 kb). It was a HindIII fragment from the 21-kb genomic clone. These arms were inserted into a targeting vector, pKO-NTKV (Stratagene). The neomycine resistance marker in this vector had been replaced with a cassette loxP-exon12-loxP-pGKneo-loxP. This vector harbored a TK cassette for negative selection with ganciclovir. The pGKneo was devoid of a pA⁺ sequence for trapping of the endogenous trpv4 pA⁺. This targeting vector was electroporated into E14.1 embryonic stem cells, and 16 of 196 neomycin-gancyclovir-resistant colonies were correctly targeted as judged by Southern blotting with a 3¢ external probe (see Fig. 5). Correct targeting at the 5¢ end was confirmed with PCR and subsequent digestion of the PCR products with PacI (see Fig. 1A in the main text). Six of the former 16 clones were correctly targeted at the 5¢ end. Two correctly targeted clones were injected into C57Bl6/J-derived blastocysts. Resulting chimeric mice were mated to C57Bl6/J females and were transmitted the induced mutation. Males heterozygous for the mutated allele were mated to C57Bl6/J females transgenic for the cre-recombinase driven by an adenoviral promoter [EIIa-cre⁺ mice, (1)]. The resulting heterozygous mice were tested for the successful excision of exon 12 and the neo marker by PCR and Southern blot. Heterozygous offspring harboring the correct excision were mated with one another and produced ^-/-, ± , and ^+/+ offspring in a Mendelian ratio. In the nulls, the absence of the trpv4 mRNA was verified by RT-PCR (see Fig. 6).

Genotyping of mice was performed by Southern blotting and PCR on tail genomic DNA. Mice were separated by sex, housed in groups of five, and were maintained under standard husbandry conditions by using a 12-h light:dark cycle with food and water available ad libitum. All experiments were conducted according to National Institutes of Health and institutional guidelines and with approval of the institutional animal review board.

Immunohistochemistry. The absence of the TRPV4 protein was verified by immunocytochemistry with an anti-TRPV4 rabbit antibody (Immunogen CDGHQQGYPRKWRTDDAPL peptide, peptide, and antibody generated by Washington Biotechnology, Bethesda, MD; in a manner similar to ref. 2). TRPV4 protein was present in the distal tubule of the kidney, and in the choroid plexus in trpv4^+/+ mice, and it was absent in trpv4^-/- mice (data not shown). In addition, sensory ganglia and circumventricular organs were examined by TRPV4 immunohistochemistry on fresh-frozen and formalin-fixed, paraffin-embedded sections (10-20 and 4 m m). Tissue was fixated with acetone and/or formalin. Because the above antibody did not lead to a strong signal in Western blotting, another anti-TRPV4 rabbit antibody was raised against the first 445 amino acids of the rat TRPV4 protein (a generous gift from Dr. Stefan Heller, Harvard University, Boston). This antibody recognized heterologously expressed TRPV4 and failed to recognize TRPV1 and TRPV2. Secondary antibodies were fluorescently labeled (Molecular Probes) or coupled to biotin (Jackson ImmunoResearch, West Grove, PA) and used with an avidin-based detection system (Vector Laboratories), according to the suggestions of the manufacturer.

Mouse Physiology and Behavior. Osmotic regulation and drinking behavior. All experiments with mice were performed in accordance with institutional and National Institutes of Health policies and guidelines. For investigation of natural drinking behavior, mice were housed individually and without food. Fluid volume consumed over time was determined by weighing drinkometers (modified 50-ml syringes) before and after the test period with a precision scale. Recorded drinking volumes were adjusted per 20 g of body weight.

The osmotic equilibrium of the mice was challenged by i.p. injection of 0.5 M NaCl solution at a dose of 0.4 ml per 10 g of body weight. Mice were handled on previous days to habituate them to the stress of handling. In response to the injection, we recorded drinking behavior and amounts. Amounts were recorded as described above at 45- and 120-min time points. Initiation of drinking behavior was recorded for each individual mouse by determining the time interval between i.p. injection and the first licking/water intake behavior. Blood osmolality was determined in plasma of i.p.-injected mice after 90 min. Animals were injected as pairs of a null with a wild-type littermate, and blood was collected after 90 min by decapitation. Osmolality was determined with a vapor pressure osmometer (Wescor, Logan, UT). To determine c-FOS immunoreactivity in the CNS, brains of i.p.-stimulated mice were sampled by fixation in 10% neutral buffered formalin by pressure perfusion of the left cardiac ventricle and subsequent overnight immersion. Vibratome sections (50-m m) of the lamina terminalis were immunostained for the immediate early response gene, c-FOS, with a rabbit c-FOS-specific antibody (Oncogene, Cambridge, MA), and fluorescently labeled secondary antibody (Molecular Probes). Micrographs were recorded with a Zeiss Axioplan upright microscope, Zeiss digital camera, and Zeiss AXIOVISION software program. Labeled cells were counted in a blinded fashion. For measurement of systemic ADH, EDTA-plasma was collected 30 min after the i.p. injection of 0.5 M NaCl, treated with trasylol (Sigma), and kept on ice at all times. Plasma was extracted with reverse-phase columns (Alpco, Windham, NH), and ADH was determined by an RIA, according to the instructions of the manufacturer (Alpco). The osmotic equilibrium of the mice was also challenged by 48-h water deprivation. Mice were individually housed without food and water at a regular 12-h light:dark cycle. After 48 h, the animals were killed by decapitation and blood osmolality was determined as described above.

To test the effect of chronic ADH infusion, the synthetic analogue, dDAVP (desmopressin acetate; Sigma), was used. One microgram per 20 g of body weight per mouse was infused over a 3-day period. Before receiving dDAVP, mice were implanted with an s.c. osmotic minipump (Alzet/Durect, Cupertino, CA) that was loaded with isotonic saline (100 m l). Drinking amounts and blood osmolality of single-housed mice that did not have access to food were recorded. After dDAVP infusion, urine was also sampled for osmolality.

Somatosensory evaluation. All experiments with mice were performed in accordance with institutional and National Institutes of Health policies and guidelines. Mechanical and thermal stimuli were applied to the hindpaws. Mice were tested by an investigator who was unaware of their genotype.

In one assay, mice were subjected to squeezing of their paw by a slowly increasing force applied by a paw pressure analgesy meter for the rodent paw according to the Randall-Selitto test (Ugo Basile, Varese, Italy). Mice were repeatedly handled before testing to habituate them to the stress of being handled. Force increased gradually (30 g per cm per second; maximum distance 25 cm) and led to withdrawal. The force until mice showed first signs of discomfort (hindpaw licking, hindleg shaking) was recorded. The experiment was repeated three times with at least 5 min between trials. The mean force (in percent of maximum force) leading to withdrawal was calculated. Mice that did not show any distress or withdrawal were recorded as maximum force. In another assay ( the von-Frey test for rodents), mice were subjected to prodding of their paw from underneath by a flexible metal rod. A dynamic plantar aesthesiometer for mice (Ugo Basile) was used for this experiment. Mice were habituated for at least 30 min to the test environment. The stimulus was applied to either hindpaw. It had to be firmly at rest and the mouse had to rest its weight on the paw. Stimulation was not applied when the mouse was grooming. Withdrawal thresholds of the mechanically stimulated hindpaw were recorded by the instrument in a series of five trials per mouse. Mice were not restrained during this experiment. Handling of the mice was not involved when performing this test. The assay excluded some of the subjectivity inherent in the above pressure analgesy assay and reduced the bias introduced by the stress of handling the mice. For the determination of c-FOS immunoreactivity in the spinal cord dorsal horn, the hindpaw was squeezed for 90 s with 80% maximum force (lever at 20 cm of a maximum of 25 cm). Ninety minutes later, mice were perfused transcardially with 10% neutral-buffered formalin and processed for c-FOS immunostaining. The spinal cord was sectioned at levels L3-S2 for frozen sections.

To assess the threshold of mice for noxious heat, they were subjected to thermal stimulation of their hindpaws with a plantar test apparatus (Ugo Basile), also known as a Hargreave’s test. As is the case with the plantar Von-Frey test apparatus, this device stimulates the unrestrained mouse, and the withdrawal latency in response to the stimulus is determined automatically. There is no handling involved in this testing. The stimulus consists of an infrared beam that is focused onto the plantar surface of the paw from underneath. Mice were allowed to habituate for at least 30 min before testing. Four latencies until withdrawal per mouse were determined, and the latency was averaged.

1. Lakso, M., Pichel, J. G., Gorman, J. R., Sauer, B., Okamoto, Y., Lee, E., Alt, F. W. & Westphal, H. (1996) Proc. Natl. Acad. Sci. USA 93, 5860–5865.

2. Delany, N. S., Hurle, M., Facer, P., Alnadaf, T., Plumpton, C., Kinghorn, I., See, C. G., Costigan, M., Anand, P., Woolf, C. J., et al. (2001) Physiol. Genomics 4, 165–174.