Abstract
Vasopressin is a neuropeptide that is essential for body fluid balance and blood pressure regulation. Vasopressin maintains body fluid homeostasis by promoting renal water retention in response to increased plasma osmolality. Vasopressin is synthesised by hypothalamic neurons and is secreted into the circulation from the posterior pituitary gland in response to action potential firing. Vasopressin neurons respond directly to changes in osmolality through expression of an N-terminal truncated variant of the transient receptor potential vanilloid (TRPV)-1 channel, ΔN-TRPV1, which is mechanically activated by osmotically-induced membrane shrinkage, depolarising the neurons towards action potential threshold. While osmosensitivity is conferred by ΔN-TRPV1, vasopressin neurons also express TRPV4 but the role of TRPV4 in vasopressin neuron function is unknown. Full-length TRPV1 can form heteromers with TRPV4, which modifies channel function. However, it is unknown whether ΔN-TRPV1 can form heteromers with TRPV4. Therefore, the aim of this project was to test the hypothesis that ΔN-TRPV1 forms heteromeric channels with TRPV4 to modulate mechanosensitivity of ΔN-TRPV1. Patch-clamp electrophysiology was used to determine biophysical properties and mechanosensitivity in ΔN-TRPV1-transfected, TRPV4-transfected and ΔN-TRPV1+TRPV4-transfected cells.
Initial studies were completed using two-electrode voltage-clamp electrophysiology to study TRPV function in Xenopus oocytes. However, the results showed that TRPVs do not form functional channels in Xenopus oocytes because TRPV-transfected oocytes did not respond to non-specific TRPV agonists or antagonists (Chapter III). Therefore, the project was continued using single-channel patch-clamp electrophysiology on transfected HEK293 cells.
Cell-attached single-channel patch-clamp electrophysiology on transfected HEK293 cells verified that homomeric rat ΔN-TRPV1 is mechanically activated by positive pressure, which resulted in a higher open probability (p = 0.0004; paired t-test) and maximum current amplitude (p = 0.007; paired t-test) at +60 holding potential. Furthermore, I validated the use of rats as an animal model for human comparison because human ΔN-TRPV1-transfected HEK293 cells were also mechanically activated by positive pressure (NPo p = 0.003; maximum current amplitude p = 0.004; paired t-tests). Neither human nor rat ΔN-TRPV1 were mechanically activated by negative pressure (Chapter IV).
Before testing heteromeric TRPV channel function, I measured homomeric human TRPV4 single-channel properties and found that TRPV4 appears to have a larger conductance (~50 pS ± 5) and longer channel open time (11.7 ms ± 1.9) than human ΔN-TRPV1 (14 pS ± 1.4; 0.7 ms ± 0.01) at positive holding potentials. Furthermore, TRPV4 was not mechanically activated by positive or negative pressure using cell-attached single-channel configuration (Chapter V).
HEK293 cells were transfected with equal quantities of human ΔN-TRPV1 DNA and human TRPV4 DNA. All human ΔN-TRPV1+TRPV4 expressing HEK293 cells had different single-channel properties than the homomers, suggesting that heteromers formed. The single-channel conductance of putative ΔN-TRPV1+TRPV4 was larger at positive holding potentials (~50 pS ± 3.4) than at negative holding potentials (~10 pS ± 1.4). Putative ΔN-TRPV1+TRPV4 was mechanically activated by positive pressure, causing an increase in open probability (p = 0.006; paired t-test) and maximum current amplitude (p = 0.04; paired t-test) at +60 mV holding potential. Some putative heteromers were also activated by negative pressure, a trait neither homomer expressed (maximum current amplitude p = 0.01 paired t-test). Finally, putative ΔN-TRPV1+TRPV4 activation by positive pressure was reduced when microtubules were destabilised by nocodazole treatment. However, nocodazole did not appear to completely prevent mechanical activation in some ΔN-TRPV1+TRPV4-transfected cells (Chapter V). Therefore ΔN-TRPV1+TRPV4 might have a greater connection to the cytoskeleton via actin.
Finally, a single point mutation of human TRPV4 (Ser319Leu) has been identified but not yet characterised. Unlike TRPV4, Ser319Leu-TRPV4-transfected HEK293 cells had a high open probability, independent of holding potential (mean NPo ~0.5). Furthermore, Ser319Leu-TRPV4 did not appear to form heteromeric channels with ΔN-TRPV1 as readily as the common TRPV4 variant did with ΔN-TRPV1. Finally, Ser319Leu-TRPV4-containing putative heteromers appeared less conductive (32.1 pS ± 3.6), less likely to open (NPo = 0.3) and less strongly mechanically activated by positive pressure than TRPV4-containing putative heteromers at +60 mV holding potential (Chapter VI).
Taken together, the results in this thesis suggest that ΔN-TRPV1 and TRPV4 form functional heteromeric channels that have broader mechanosensitivity than homomeric ΔN-TRPV1. Ser319LeuTRPV4 also formed putative heteromers with ΔN-TRPV1, but these channels appeared less functional than putative ΔN-TRPV1+TRPV4 heteromers.
Hence, the ΔN-TRPV1 and TRPV4 could form heteromers in vasopressin neurons, providing a mechanism that might allow plasticity in the osmosensitivity of vasopressin neurons evident in different (patho) physiological states such as pregnancy and hypertension.