Parathyroid Hormone (PTH) Homeostasis in Bartter Syndrome

The parathyroid glands play a pivotal physiological function by maintaining blood calcium levels, specifically blood ionized calcium concentrations, within a very narrow range. They do so by modulating the minute-to-minute release of parathyroid hormone (PTH) into the circulation. Such changes have almost immediate effects on calcium excretion in the urine and on calcium efflux from bone and, if sustained for hours or days, affect renal vitamin D metabolism and ultimately the efficiency of intestinal calcium absorption. The capacity of chief cells of the parathyroid to detect small changes in blood ionized calcium levels, modify PTH release accordingly, and initiate these adaptive responses is mediated by a calcium-sensing receptor (CaSR) located at the cell surface.

The importance of the CaSR in parathyroid tissue extends beyond its traditional role as a modifier of calcium-regulated PTH secretion to involve other key components of parathyroid gland function that are frequently abnormal in clinical disorders characterized by excess parathyroid gland activity such as hyperparathyroidism. These include disturbances in the control of PTH gene transcription and hormone synthesis and the development of parathyroid gland enlargement due to tissue hyperplasia.

PTH acts mainly on renal proximal tubule phosphorus (Pi) reabsorption and bone osteoclast calcium and Pi resorption. CASR is also expressed in nephron thick ascending limb (TAL), where it interacts with luminal potassium ROMK channel. Mutations in several TAL channels and proteins (including ROMK, NKCC2, ClCKb, Barttin and CASR) cause Bartter syndrome (BS), a normotensive hypokalemic tubulopathy. Whereas the effects of CASR mutations on ROMK function in the kidney have been described , it is unknown whether ROMK mutations exert any effects on CASR function or PTH regulation. We describe here a group of children and adolescents with type II BS (due to mutations in ROMK) with abnormal PTH homeostasis.

Preliminary Data:

We compared laboratory data of 12 children with type II BS (4M, 8F) and 17 children (7M, 10F) with type IV BS (d/t mutations in the Barttin gene, a beta subunit of the ClCKb basolateral chloride channel in the TAL ), followed in our center over the past 10 years. A total of 86 and 105 datasets of blood and urine analyses (average datasets/pt: 7.3±4.1 and 6.9±2.9), for type II and IV BS respectively, were analyzed.Potassium levels were normal in all BS-II children without additional salt supplementation, whereas BS-IV children were usually mildly hypokalemic. Estimated GFR remains normal in all children. There was no hypomagnesemia. Average PTH values were significantly higher in BS-II (102±39 Vs 46±24 pg/ml in BS-IV, p<0.001) and were above upper normal limit in 93% of cases Vs 13% in BS-IV (p<0.001). Levels of 25(OH) vitamin D were not different. Total serum calcium was mildly decreased (within the normal range) and serum Pi increased in BS-II, both in absolute values or when normalized for age (PiSDS). The threshold for phosphate excretion (TpGFR) was slightly higher in BS-II. There was no difference in the degree of hypercalciuria between groups. Based on these preliminary data we concluded that the elevated PTH levels only in type II BS are not related to a decrease in GFR or vitamin D levels or decreased serum calcium or hypercalciuria. The elevated Pi levels are associated with a decrease in phosphate excretion, but are not correlated with PTH levels. The possibility that a mutation in ROMK may cause a dysregulation of PTH secretion via possible interaction with CASR should now be investigated.

Methods:

The investigative protocol has been submitted to the local Committee for Human Experimentation. Informed consent will be obtained from affected children (or young adults) and their parents.

We expect to recruit 5 patients from each BS subgroup. In addition we will recruit 5 normal volunteers to serve as an additional control group.

Subjects will be evaluated during 2-day admissions to the General Pediatric Ward as previously described . On the first day of study, 2-h I.V. infusions of sodium citrate will be done to gradually lower blood ionized calcium concentrations to a level of at least 0.2 mmol/L below preinfusion values; the dose of sodium citrate ranges usually from 28-118 mg/kg/h. Blood samples for measurements of ionized calcium and PTH will be obtained 30, 15, and 0 min before and every 10 min during sodium citrate infusions. The following day, 2-h IV infusions of 10% calcium gluconate will be done to gradually raise blood ionized calcium concentrations to a level at least 0.2 mmol/L above preinfusion values. The dose of calcium gluconate usually ranges from 2-8 mg/kg/h. Blood samples for measurements of ionized calcium and PTH will be obtained as described previously for infusions of sodium citrate. The average of measurements obtained 30, 15, and 0 min before starting each infusion will be used to determine basal values for blood ionized calcium and serum PTH for each day of study.

Blood ionized calcium levels will be monitored during calcium infusions using a calcium-specific electrode (Radiometer ICA-II, Copenhagen, Denmark); blood samples will be collected anaerobically, and measurements will be obtained immediately thereafter. Serum samples for PTH determinations will be separated by centrifugation immediately after collection, snap frozen on solid CO2, and stored at -70 oC until assay. Ionized calcium levels will be monitored after stopping calcium infusions until values returned to baseline levels.

The sigmoidal curve that describes the relationship between blood ionized calcium and serum PTH levels will be determined for each study subject using the combined results obtained during sodium citrate and calcium gluconate infusions. According to the four parameter model, the set point for calcium-regulated PTH release represents the ionized calcium concentration at which serum PTH levels are midway between the maximum value achieved during hypocalcemia and the minimum value attained during hypercalcemia, as reported previously .

Results obtained during calcium gluconate infusions will be separately analyzed to assess the inhibitory effect of increasing blood ionized calcium concentrations on PTH release. To improve the linear fit of the data, serum PTH levels, expressed as the natural logarithm (ln) of percent preinfusion values, will be plotted against the corresponding blood ionized calcium concentration at each 10-min interval as previously described .

Statistical Analysis:

Linear regression analysis will be done using the method of least squares, and slope and y-intercept values will be compared using the t statistic. A mono-exponential curve fitting algorithm of the form y =A e-2kt + B will be also used to examine the curvilinear relationship between blood ionized calcium and serum PTH levels during I.V. calcium infusions; these results will be presented as mean values with 95% confidence intervals.