Molecular mechanisms of secondary hyperparathyroidism

The development of secondary hyperparathyroidism is a major risk factor for the development of renal osteodystrophy. In cases of chronic renal failure (CRF) where there is an inadequate increase in parathyroid hormone (PTH) secretion, adynamic bone disease may develop. There is, therefore, a pressing clinical need to be able to fine-tune the synthesis and secretion of PTH and the compensatory increase necessary in CRF. This requires an understanding of the physiological mechanisms involved in regulating PTH synthesis and secretion. PTH is tropic to the renal synthesis of 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) and there is a well-defined feedback loop whereby 1,25(OH)2D3 markedly decreases PTH transcription and subsequent secretion. The specific 1,25(OH)2D3 receptor (VDR) is present in the parathyroid at a similar concentration to that in the duodenum, which is the classical vitamin D target organ [1] (Figure 1). This underscores the physiological relevance of the action of 1,25(OH)2D3 on the parathyroid (Figure 2). The 1,25(OH)2D3 receptor (VDR) is localized to the parathyroid in a similar concentration to that found in the duodenum, indicating that the parathyroid is a physiological target organ for 1,25(OH)2D3. In situ hybridization with the VDR probe in rat parathyroid/thyroid and duodenum sections. (A) (1) Parathyroid/thyroid tissue from a control rat; (2) parathyroid/thyroid from a 1,25(OH)2D3-treated rat (100 pmol at 24 h); (3) duodenum from a 1,25(OH)2D3-treated rat. The white arrows point to the parathyroid glands. (B) A higher power view of (A2) showing the parathyroid gland (p) and thyroid follicles (t). The upper figures were photographed under bright-field illumination; the lower figures show dark-field illumination of the same sections. (Reproduced from reference 1 with permission of the American Society of Clinical Investigation.) Interrelationships between Ca2+ , phosphate, PTH and 1,25(OH)2D3 concentrations. There are sensitive physiological control mechanisms at the levels of secretion, gene expression and protein synthesis that maintain normal calcium and phosphate homeostasis. The feedback loops work at different time intervals, dependent upon the level of control. For instance, the effect of PTH to correct serum calcium is more rapid than its effect to increase 1,25(OH)2D3 synthesis. The increased levels of 1,25(OH)2D3 would take even longer to act on the parathyroid to decrease PTH gene transcription. In chronic renal failure, the retention of phosphate leads to an increase in serum PTH, which then acts to increase renal phosphate excretion, so the trade-off to maintain a normal serum phosphate concentration is hyperparathyoidism. Hyperparathyoidism involves increased PTH gene expression, parathyroid cell proliferation and, hence, total protein synthesis, leading to a large increase in PTH secretion. The effect of 1,25(OH)2D3 is a powerful one in all systems studied. In vivo in the rat, a single small dose of 1,25(OH)2D3 decreased PTH gene transcription by almost 100% [2]. This effect has an important application in the management of CRF patients with secondary hyperparathyroidism. Less VDR is expressed in secondary hyperparathyroidism, especially in nodular hyperplastic regions of the parathyroid [3]. A remarkable characteristic of the parathyroid is its sensitivity to small changes in serum calcium concentration, which is recognized by a seven-transmembrane-domain receptor with a large extracellular amino-terminal region, the calcium-sensing receptor [4]. A decrease in extracellular calcium concentration leads not only to an increase in PTH secretion but also to increases in PTH mRNA levels and parathyroid cell proliferation. Hypocalcaemia increases PTH secretion in the short term, PTH mRNA levels in hours and days and parathyroid cell number after a more prolonged stimulus [5-7]. We have studied the mechanism of the effect of calcium on PTH gene expression and have shown that in vivo it is post-transcriptional. Phosphate also regulates the parathyroid. The contribution of hyperphosphataemia to the pathogenesis of the secondary hyperparathyroidism of CRF has been documented for many years [8], but it was never possible to separate the effect of hyperphosphataemia from secondary decreases in serum calcium and 1,25(OH)2D3 [9]. This was first established by the work in our laboratory of Kilav et al. who succeeded in demonstrating that the effect of serum phosphate on PTH gene expression and serum PTH levels was independent of any changes in serum calcium or 1,25(OH)2D3 [10]. A visual example of the powerful effect of phosphate on the parathyroid is shown in Figure 3. To demonstrate the effect of phosphate on the parathyroid in vitro, it was imperative to maintain tissue architecture [11-13]. There was an effect in whole glands or tissue slices but not in isolated cells. In situ hybridization of parathyroid/thyroid sections with a PTH probe. (A) Parathyroid/thyroid tissue from a control rat with the parathyroid (P) staining darkly for the PTH mRNA. There is no staining of the thyroid (T) (× 125). (B) A higher power view of (A) showing the parathyroid cells (P) and thyroid follicles (T) (× 500). (C) Parathyroid/thyroid tissue from a rat that had been fed a low phosphate diet for 2 weeks, showing the much lighter staining for PTH mRNA in the parathyroid (P) and none present in the thyroid (T) (× 125). (D) A higher power view of C (× 500). (Reproduced from reference 10 by copyright permission of the American Society for Clinical Investigation.) The parathyroid responds to changes in serum phosphate concentration at the level of secretion, gene expression and cell proliferation, but by what mechanism? We can now provide some of the answers regarding the effect of phosphate and calcium on PTH gene expression. The clearest rat in vivo models for an effect of calcium and phosphate concentrations on PTH gene expression are hypocalcaemia with a large increase in PTH mRNA levels and hypophosphataemia with a large decrease in PTH mRNA levels. In both instances they were post-transcriptional as shown by nuclear transcript run-on experiments. Parathyroid cytosolic proteins were found to bind in vitro transcribed PTH mRNA with three bands at about 50, 60 and 110 kDa [6]. What was particularly interesting was that this binding was increased with parathyroid proteins from hypocalcaemic rats, where PTH mRNA levels are increased, and decreased with parathyroid proteins from hypophosphataemic rats, where PTH mRNA levels are decreased. Naveh-Many and colleagues used an in vitro degradation assay to study the effects of hypocalcaemic and hypophosphataemic parathyroid proteins on PTH mRNA stability [6] (Figure 4). In this assay, parathyroid cytosolic proteins from control rats led to the degradation of a radiolabelled PTH transcript after 40–60 min of incubation. Hypocalcaemic parathyroid proteins degraded the transcript only after 180 min, whilst hypophosphataemic parathyroid proteins had already degraded the transcript after 5 min. The rapid degradation of PTH mRNA by hypophosphataemic proteins was totally dependent upon an intact 3′-untranslated region (3′-UTR) and in particular the terminal 60 nucleotides. Proteins from other tissues in these rats were not regulated by calcium or phosphate concentrations. Therefore, calcium and phosphate act on the parathyroid cell to change the properties of cytosolic proteins which bind specifically to the PTH mRNA 3′-UTR and determine its stability (Figure 5). What are these proteins? Sela-Brown in our laboratory has now used affinity chromatography to isolate these RNA-binding proteins. She has recently presented preliminary information as to the identity of one of the PTH mRNA-binding proteins (a 50 kDa protein on a sodium dodecyl sulfate-polyacrylamide gel) and demonstrated its functionality [14]. In rats with experimental uraemia due to 5/6 nephrectomy, there is a modest increase in PTH mRNA levels due to a decrease in the activity of the degrading ribonucleases in the parathyroid with no change in the protective binding factors [15] (Figure 5). In vitro degradation of PTH mRNA by parathyroid cytosolic proteins. (A) Time response curves of intact full-length PTH mRNA after incubation with parathyroid cytosolic proteins. Each point represents the mean±SE of three or four different experiments, apart from the low-calcium rats at 240 and 300 min, which is the mean of two experiments. At some points the SE is smaller than the size of the graphic symbols. The PTH transcript was degraded very rapidly by proteins from low-phosphate rats, and remained intact for a longer with proteins from low-calcium rats. (B) Mapping a region in the PTH 3′-UTR that mediates degradation by proteins from low-phosphate rats. PTH mRNA probes used were: intact PTH mRNA (probe A), without the 3′-UTR (probe C) and without the 3′-terminal 60 nucleotides of the 3′-UTR (probe B). (Reproduced from reference 6 with permission of The American Society for Biochemistry and Molecular Biology.) Model of PTH mRNA including the 5′-untranslated region (5′-UTR), the coding region, the 3′-UTR and the parathyroid cytosolic proteins that interact with the 3′-UTR. The parathyroid proteins contain both protective factors (), measured by UV cross-linking, and degrading factors (endonucleases) (), measured by an in vitro degradation assay. In normal rats, the basal levels of PTH mRNA is determined by the balance between the protective and degrading factors in the cytoplasm. In hypocalcaemia, there is an increase in PTH mRNA associated with an increase in the binding of protective factors, which leads to a more stable transcript. In hypophosphataemia, there is a decrease in protective factors, which leads to a less stable transcript and a decrease in PTH mRNA levels. In rats with chronic renal failure due to 5/6 nephrectomy, there is no change in the protective factors together with a decrease in endonuclease activity, resulting in increased PTH mRNA levels. (Reproduced from reference 15 with permission from the American Society of Nephrology.) In CRF there is, with time, proliferation of the parathyroid cells (Figure 6). This is initially a polyclonal proliferation which may be complicated by a monoclonal proliferation characteristic of tertiary hyperparathyroidism [16,17]. In parathyroids removed surgically from patients with CRF, there are nodular areas where there is a decrease in the concentration of the vitamin D receptor protein and the calcium-sensing receptor [3,18]. These changes may be of pathophysiological significance. Hypocalcaemia induced by dietary means leads to a marked increase in parathyroid cell proliferation and diet-induced hypophosphataemia decreases parathyroid cell proliferation [7] (Figure 6). Another factor that decreases PTH cell proliferation is the administration of 1,25(OH)2D3 [19]. In experimental CRF there is also an increase in parathyroid cell proliferation, which can be markedly decreased by a low phosphate diet and increased by a high phosphate diet [7]. Factors that predispose to parathyroid cell proliferation in CRF are, therefore, hyperphosphataemia, hypocalcaemia and vitamin D deficiency. Experimental renal failure (5/6 nephrectomy) leads to an increase in parathyroid cell proliferation as measured by the number of parathyroid cells staining with PCNA. Parathyroid cell proliferation was increased by a high phosphate diet and markedly decreased by a diet with a low phosphate. (Reproduced from reference 7 by copyright permission of the American Society for Clinical Investigation.) The studies reported here shed some light on the mechanisms involved in the pathogenesis of secondary hyperparathryoidism and hopefully will lead to therapy designed to target the effects of phosphate on the parathyroid cell.