When most omnivores and herbivores become sodium depleted they engage in the motivated behavior of sodium appetite (AKA salt appetite), or the seeking out and ingestion of salty substances. Sodium appetite is associated with psychological processes that serve to enhance the incentive and rewarding value of salty substances in order to attract animals to salty substances and reinforce the ingestion of them. The experience of sodium depletion also produces long-lasting changes in behavior; one of the most apparent changes being a seemingly life-long increase in hypertonic salt intake which indicates sodium appetite is sensitized. Two neural circuits have been implicated in the sensitization of sodium appetite: 1) a forebrain neural circuit that regulates body fluid homeostasis, and 2) the mesolimbic dopamine system which mediates motivated behaviors. This dissertation has three aims that serve the overall purpose of providing a better understanding of the neurobiological mechanisms that mediate the sensitization of sodium appetite. The first aim is to develop a model of sodium depletion that is amenable to pharmacological manipulation in order to determine whether the -blockade of N-methyl-d-aspartate receptors, which are critical for neural plasticity, will prevent the sensitization of sodium appetite. The second aim is to determine whether sensitization is associated with relatively long-term molecular changes in forebrain areas that regulate body fluid homeostasis. The third aim is to identify how forebrain areas involved in body fluid homeostasis may connect to and influence activity in the mesolimbic dopamine system.

Hypertension is the chronic elevation in blood pressure that is regulated in part through the retention and regulation of sodium retention and excretion in the kidneys. Hence the kidney has been considered the organ that regulates blood pressure. There are a cohort of patients that suffer with high blood pressure due to lack of 11β-hydroxysteroid dehydrogenase-type 2 (11β-HSD2) expression (which inactivates glucocorticoids (GCs), allowing selective activation of mineralocorticoid receptors (MR) by aldosterone) that results in hypertensive and increased salt appetite phenotypes - a condition known as syndrome of apparent mineralocorticoid excess (SAME). This disorder can be recapitulated in the mouse through the global deletion of 11β-HSD2, which results in over activation of the MR driving an elevation in blood pressure. However, the distinction between blood pressure elevation because of kidney dysfunction with loss of 11β-HSD2 or increased salt appetite due to loss of brain 11β-HSD2 expression is not clear from the global 11β-HSD2 knockout model. Salt appetite is regulated by regions of the brain out-with the blood-brain barrier, known as circumventricular organs. In the mouse, salt appetite is controlled by aldosterone-sensitive cells in the nucleus of the solitary tract (NTS) in the brain stem, where 11β-HSD2 is expressed to provide mineralocorticoid selectivity. However, in the fetal brain, 11β-HSD2 is widely expressed, protecting against adverse GC action that alters brain development and increases susceptibility to psychiatric disorders as adults. 11β-HSD2 deletion solely in the brain from embryonic day 12 resulting in GC fetal programming (HSD2BKO) causes effects on both behaviour and salt appetite. To determine the role of developmental versus adult expression of brain 11β- HSD2, mice with deletion of brain 11β-HSD2 from mid gestation (HSD2BKO) and mice with adult deletion of 11β-HSD2 in the NTS using lentivirus (HSD2.v- BKD) were compared. The phenotypes (salt appetite, blood pressure (BP), baroreceptor response (BRR) and cognition), can be categorised as either due to GC fetal programming (as indicated by HSD2BKO groups), or increased activation of MR in adult 11β-HSD2 expressing neurons (recapitulated in the HSD2.v-Cre groups). Salt appetite increased in both HSD2BKO and HSD2.v-BKD cohorts (mean percentage increase 65% n=8 and 46% n=6, compared to their respective controls), leading to an increased BP in both groups (+12% and +8%, respectively) as well as an impaired BRR, indicating all phenotypes are mediated by adult NTS neurons. However, spatial recognition memory (Object-in-Place task) is abolished in HSD2BKO mice, whereas, HSD2.v-BKD mice still retain short-term memory. Our data suggest that neural 11β-HSD2 protects against inappropriate activation of MR by corticosterone to regulate salt appetite and salt-induced rises in blood pressure. However, spatial recognition memory is not influenced by deletion of 11β-HSD2 in the adult brain, confirmation that this phenotype is underpinned by developmental programming by GCs, which is observed in the 11β-HSD2 brain KO. Salt appetite has been shown to be centrally regulated through the adult deletion of 11β-HSD2. From this, our data suggest that an increased salt appetite is due to adult loss of function of 11β-HSD2 rather than GC programming during development. Highlighting the NTS as a region for drug delivery to try and control salt appetite in salt sensitive individuals who struggle with administering a recommended change in diet. To develop this further, minimally invasive modes of delivery of viruses and drugs into the brain were investigated. In so doing, a non-invasive and reversible method to temporarily disrupt the blood brain barrier (BBB) was optimised. The technique required acoustic insonation of ultrasonic contrast agents (CAs) (gas microbubbles) adjacent to the BBB. These microbubbles (SonoVueTM, Bracco) were delivered via tail vein injection into the vasculature. To target the BBB, an ultrasonic transducer was suspended and focused through coupling gel onto the area of interest in the brain with skull the intact. The optimisation of this technique required determination of the focal position of the 3.5MHz transducer that was utilised, in addition to optimisation of the pulse length, pulse repetition frequency and power output of the ultrasound beam to enable the BBB to be disrupted. In addition, measurement of the attenuation of the ultrasound beam through ex vivo mouse skulls were measured. These results showed a 50% reduction in pressure amplitude from the baseline of 335.2mV (Baseline mean = 100% +/-SEM 0 n=3 (No skull), five regions across the skull averaged 47.79% +/-SEM 1.913 n=25 (using 5 different animals). In in vivo mice, after co-injection of the microbubbles with Evans Blue and insonation of the brain, disruption of the BBB was confirmed by the presence of Evans Blue dye in the brain, with no measurable damage occurring in the brain. This was confirmed by cell and nuclear morphology with no red blood cell extravasation into the surrounding tissue. The parameters used to open the BBB used a peak negative pressure of 2.1MPa (single pulse), transducer frequency 3.5MHz, 35,000 cycles over a 10ms burst at a pulse repetition frequency of 10Hz. The technique when applied in vivo in recovery animals is speculated to work by the focused ultrasound causing the microbubbles to oscillate within the vasculature adjacent to the BBB, resulting in high-shear stresses being generated on the tight junctions within the BBB. The resultant gaps in the BBB allow free circulating compounds (e.g. large dye molecules (Evans Blue - 960.8g/mol molecular weight) and adeno-associated-viruses (25nm with a packing capacity of 4.5kb) within the blood to pass into the brain, but there is no penetration of red blood cells (7μm). Longitudinal mouse experiments demonstrated that within 12-hours these gaps close with no long-term damage observed. Currently, utilising this technique, successful passage of an adeno-associated virus expressing GFP (as a marker) has been shown to pass into the brain (n=6 for each cohort including control) - indicating that the virus requires the ultrasound and microbubbles to facilitate its movement into the brain. Further technique optimisation is being explored looking at the role of CAs used in the opening and disruption of the BBB, comparing composition and size of the CAs. Microbubbles (2-3μm) and nanobubbles (200nm) were compared as well as lipid and non-ionic surfactant surface compositions, using volume of drug delivery and degree of disruption as outputs. Using this technique, the hydrophilic drug mimic calcein was delivered into the brain (n=5 non-ionic surfactant nanobubble, n=5 lipid nanobubble). Results have indicated that the delivery of calcein is most efficient when using non-ionic surfactant nanobubbles as opposed to lipid nanobubbles - with a greater volume of the drug being delivered into the cerebral tissue. Furthermore, the concentration and surface composition of the nanobubble have an effect as to the size and potential damage to the brain when opening the BBB. In conclusion, it has been shown that it is possible to non-invasively open the BBB and deliver viruses and dye into the brain. In addition, this thesis has investigated the use of nanobubbles as both facilitators to opening the BBB and delivery vectors for potentially therapeutic drugs. Finally, a non-invasive opening of the BBB has been achieved using focused ultrasound. Ultimately this non-invasive opening of the BBB can be used to achieve delivery of larger molecules (such as antibodies and viruses) into the brain to target treatments. Focused ultrasound brain targeting can be applied to the potential treatment of salt appetite regulation in the NTS. For the individuals who suffer from salt sensitive hypertension, the NTS can be targeted to reduce the drive to ingest high salt diets. Furthermore, the continuation of research into the central control of BP, salt appetite and baroreceptor reflex control can become better understood, using less invasive delivery techniques to the brain.