Slide 1

Ladies and Gentlemen, first I would like to thank the organisers of this meeting for giving me the opportunity to give this mini-lecture about the pathophysiology and consequences of renal phosphate leaks. So, the previous speaker presented data about high phosphatemic state, now I will talk about low phosphatemic conditions. I will give an overview of the various mechanisms responsible for renal phosphate leaks in human disorders.

Slide 2

Phosphate is free-filtered at the glomerulus, then it is almost exclusively reabsorbed in the proximal tubule. In subjects with normal or mildly altered glomerular filtration rate the amount of phosphate we absorb in the proximal tubules determines the levels of serum phosphate concentration. Renal phosphate leak is defined by the association of hypophosphatemia, low serum phosphate concentration and a decrease in the capacity of the kidney to reabsorb phosphate assessed by the TmP/GFR.

Slide 3

This slide presents a schematic view of the proximal tubular cells. 3 sodium phosphate transporters are expressed at the apical domain of these cells. Type I sodium phosphate transporter is a non- specific phosphate carrier. Type IIa, NaPT-IIa sodium/phosphate transporter is almost exclusively expressed in the kidney and in the kidney in the proximal tubule. This transporter generates a current that can be recorded in vitro to measure its activity. Type IIc transporter is an electroneutral transporter. The expression of this transporter is growth-dependent. The expression of its mRNA is higher in weaning than in adult rats. The activity of this transporter is controlled by hormones, PTH binds to type I PTH resulting in an increase in cyclic AMP synthases and decrease in PT2 expression. Recently new phosphateric factors have been identified, FGF23 and another factor, another peptide named klotho has also been involved in phosphate homeostasis. So, now I will present you disorders of these various proteins that have been identified in human pathology.

Slide 4

The first disorder in which mutations have been identified with renal phosphate leak is Jansen’s metaphyseal condrodysplasia. In these disorders the authors identified an activating mutation in type I PTH receptors. These mutations increase cyclic AMP synthases in cells in the absence of PTH as shown here, in the absence of PTH, basal cyclic AMP synthases is higher than in control. These patients present with mild hypophosphatemia probably due to a decrease in NaPT-IIa expression in the kidney. They also present hypercalcemia, hypercalciuria, serum phosphate concentration is low, there is no PTH-related peptide, concentration is undetectable in these patients but these patients present an increase in cyclic AMP excretion in urine. Jansen’s metaphyseal condrodysplasia is characterised by a short limb dwarfism, bowing of long bones, dysplastic growth plate and renal lithiasis.

Slide 5

The fibroblast growth factor 23 is a 34 kDalton circulating peptide. This peptide is synthesised in bone by osteoblasts and osteocytes. This peptide can be inactivated by an enzymatic cleavage between residues 176 and residue 179. Its expression is regulated by plasma phosphate concentration and phosphate intake as shown here in a healthy volunteer when a healthy volunteer was subjected to a low phosphate diet here serum phosphate concentration of FGF23 decreased.

Slide 6

When the content of phosphate in the diet is increased FGF 23 concentration increased. The expression of fibroblast growth factor is also regulated by serum calcitriol concentration. As shown here infusion of calcitriol in mice increases serum FGF23 concentration. This effect is due to the presence of a vitamin D response element in the promoter of the FGF23 gene. FGF23 decreases the expression of the mRNA of NaPT-IIa and NaPT-IIc and consequently decreases the expression of these two proteins and increases urinary phosphate excretion. But FGF23 also controls a level of calcitriol concentration in plasma. FGF23 inhibits the activity of 1 alpha hydroxylase that is the enzyme that synthesises calcitriol but it stimulates the 24 hydroxylase activity enzyme that catabolises calcitriol. So, consequently when FGF23 concentration is increased, calcitriol concentrations are decreased in plasma.

Slide 7

Increased concentration of FGF23 has been found in various disorders in humans. In tumour-induced osteomalacia, benign mesenchymal tumour or slow growing neoplast secrete FGF23. The patients complain of muscle weakness, pain or bone demineralisation. They present hypophosphatemia, a low TmP/GFR, serum calcium and calcium excretion in urine are normal or low. PTH is usually normal but can be slightly increased. The striking feature of this disorder is the decrease in the serum calcium concentration.

Slide 8

23 fibroblast growth factor gene has been identified in the human genome and only 4 genes encoding for fibroblast growth factor receptors. It is known also that FGF, the various FGFs including FGF23 can bind to the various FGF receptors. Furthermore, these fibroblast growth factor receptors are largely expressed in many tissues. It has been shown that the activation of FGF receptors causes the development of severe disorders, developmental disorders and cancers. So, it is difficult to understand why the effects of the over expression of FGF23 are restricted to disorders in phosphate homeostasis. This point has been studied by Kurosu. He showed in fact that FGF23 requires the presence of an FGF receptor but also of another peptide, klotho to be active. This is shown here in this insert. In CHO cells, CHO cells express fibroblast growth factor receptors. FGF23 in these cells is unable to induce a phosphorylation of the FGF receptor substrate, FRS-2 alpha and it is also unable to induce ERK phosphorylation. However, when these cells are transfected with the klotho protein here, FGF23 induces the phosphorylation of FRS-2 alpha and of ERK. The authors also showed that klotho can be coimmuno precipitated with FGF receptors, with different FGF receptors, type I, II and III. So this suggests that klotho can associate with the FGF receptors and this association is mandatory for FGF23 binding on these receptors and to activate the pathway, the downstream pathway.

Slide 9

In just one slide what do we know about klotho? Klotho is a single pathway membrane protein that is also detected in the plasma in a soluble form. Klotho deficient mice present a phenotype similar to FGF23 null mice. The over expression of klotho in mice, excellent life span however, we do not know if this over expression of klotho in these mice decreases serum phosphate concentration or increases urinary phosphate excretion. To my knowledge to date no activating mutation nor increase in the expression of klotho has been reported in disorders with renal phosphate leaks.
Klotho is expressed in the kidney but it is expressed in the distal tubule not in the proximal tubule, so it is difficult to understand, for the moment it isn’t clear how klotho can control FGF23 action in a specific manner in the proximal tubule.

Slide 10

I have shown you disorders that involve hormones or receptors to this hormone but mutations in the sodium phosphate transporters have also been identified in renal phosphate wasting disorders. We have identified 4 mutations in the human NaPT-IIa transporters. They are located here on the protein, NaPT-IIa protein presented on this slide. All these mutations changed aminoacids that are conserved throughout species. These mutations were identified in patients with nephrolithiasis, for these 3 mutations and in a patient with bone demineralisation for this mutation.
Recently, Lapointe also reported 2 mutations. The same as the ones we have identified in our patients but he reported a new mutation which is a deletion of 7 aminoacids in the end terminal part of the protein.

Slide 11

We studied the physiological consequences of these mutations on NaPT-IIa activity. We measured phosphate induced current in a xenopus oocyte injected with wild type or mutant NaPT-IIa RNA. As you can see on this scheme, phosphate-induced current increases with increasing amount of NaPT-IIa RNA injected in xenopus oocytes. But as you can see also, at all concentrations of RNA tested, phosphate-induced current was significantly lower in oocytes expressing the mutant NaPT-IIa RNA then that measured in oocytes expressing the wild type form of NaPT-IIa RNA.

Slide 12

Similar results were reported by Lapoint with the A133 free mutant and with the mutant with deletion of the 7 aminoacids. All the patients that we have identified with the mutations were heterozygous for these mutations. So we checked if these mutations, if a mutant protein could modify the activity of the wild type protein and we performed co-injection experiments to check this point. We measured phosphate induced current in xenopus oocytes injected with wild type NaPT-IIa RNA alone, 10 nanograms or with 10 nanograms of wild type plus 10 nanograms of the mutant RNA. We observed that phosphate-induced current was decreased by the co-expression of the mutant form of the protein suggesting that this mutant protein decreases the activity of the wild type form.

Slide 13

Other mutations of other sodium phosphate transporters in the kidney and PT2C have been recently reported by two groups at the beginning of this year. These groups identified 10 mutations localised throughout the protein. The patients were either homozygous, compound heterozygous or heterozygous for these mutations. They also described several mechanisms for these mutations, missense mutations, deletions leading to truncated proteins, intronic deletion or potential alteration of a splicing site but they also did not perform functional studies.

Slide 14

We do not understand all the mechanisms of phosphate homeostasis to date For example, Pfugers group identified a phosphateric factor from patients with tumour-induced osteomalacia. This factor is a secreted frizzle-related protein 4, sFRP4. The infusion of this peptide in mice significantly decreases sodium-dependent phosphate uptake in the animal, this is almost as potent as PTH and its concentration. This decrease in phosphate uptake is associated with a decrease in NaPT-IIa expression in the kidney of this animal. However, we do not know how sFRP4 induces the decrease in phosphate reabsorption. sFRP4 is a soluble factor that can bind Wnt. Wnt is a factor that can bind to a receptor that then activates and stabilises beta-catenin but we do not know why the lack of Wnt binding on its receptors, how it can induce a decrease in sodium phosphate excretion and how it can induce a decrease in NaPT-IIa expression. So something is missing. Many other phosphatanins have been identified. Phosphatanins are phosphateric factors. FGF7, MIB and dentin matrix protein, Dmp1 have also been identified but we do not know how this protein can modify phosphate reabsorption in the kidney.

Slide 15

Similarly, there are soluble factors but also proteins that are expressed in the proximal tubules that are involved, that are not transporters but that can control phosphate transport. For example, the disruption of the NERFH-1 gene in mice induces a phenotype that is similar to that observed in NaPT-IIa knockout mice as presented here. These mice present an increase in phosphate excretion and in the heterozygous and homozygous state and the homozygous mice are hypophosphatemic. This mouse presents renal phosphate leak, hypercalciuria, nephrolithiasis and various bone abnormalities. So, a decrease in phosphate uptake in these mice, in NHERF-1 null mice is due to a decrease in the expression of NaPT-IIa in the brush border membrane of renal cells.
To my knowledge to date no data in humans have been reported regarding NHERF-1 and mutation in NHERF-1.

Slide 16

What are the consequences of renal phosphate leaks? As I told you, renal phosphate leaks can be associated in humans with renal calculi or bone demineralisation. The analysis of the phenotype of our patients suggests that these differences in phenotype can depend on serum calcium concentration and urinary calcium excretion. For example, in most of our patients with NaPi-IIa mutation urinary phosphate excretion was increased of course, these patients presented an increase in serum calcium concentration and urinary calcium excretion and they have nephrolithiasis, except one patient. This patient presented only a mild increase in serum calcitriol concentration and urinary excretion and this patient did not have nephrolithiasis, she had bone demineralisation.

Slide 17

This point is supported by the study of the phenotype of double --- mice with a disruption in the NaPi-IIa and 1 alpha hydroxylase gene. These mice were obtained by methin NaPT-IIa knockout mice with 1 alpha hydroxylase mice. As you can see here, the level of phosphate excretion is increased and is similar in double knockout mice and in NaPi-IIa knockout mice. As expected, serum calcitriol concentration was undetectable in double knockout mice, it was increased in NaPT-IIa knockout mice and serum calcium excretion was normal in double knockout mice and was increased in NaPT-IIa knockout mice. Interestingly, NaPT-IIa knockout mice presented renal calcifications but these renal calcifications are markedly reduced in double knockout mice. So this supports the role of calcitriol and urinary calcium excretion to explain the phenotype of the --- of these mice.

Slide 18

How can we explain the formation of nephrolithiasis during renal phosphate leaks in patients with an increase in serum calcium concentration? When phosphate uptake is decreased in the proximal tubule, the concentration of phosphate is increased and is particularly increased in this part of the tubule, the thin part of Henle’s loop, the water is removed in the part of the tubule. If calcitriol concentration is increased, usually calcium absorption by its intestine is increased and the load of calcium filtered at the glomerulus is increased. So, here there is an increase in the calcium phosphate concentration, calcium phosphate product. Calcium phosphate crystals initiate in this part of the thick ascending limb. As shown by Even in a work published in the journal of Clinical Investigation, the calcium phosphate crystals can spread through the interstitium to the papilla. In the papilla the crystals grow by calcium cell deposits. This is the heterogeneous enucleation and we hope in our patients that calculi were made of calcium oxalate but the nucleus of the lithias were made of calcium phosphate. Thank you for your attention.