Autosomal dominant polycystic kidney disease (ADPKD) affects more than 1 in 1000 live births and is the most common monogenic cause of kidney failure in humans. Two genes, PKD1 and PKD2, were discovered using positional cloning approaches. Mutations in either of these proteins cause virtually indistinguishable clinical presentations.
The discovery of the genes for ADPKD was based on genetics and not function, and the respective protein products, polycystin-1 and -2, were novel although they did share structural similarity with each other. Polycystin-1 is a ~4300 amino acid integral membrane glycoprotein predicted to have a ~3000 amino acid extracellular domain with a multitude of putative interaction domains, 11 membrane spans and a small cytoplasmic tail. Polycystin-2 is predicted to have six membrane spans and is the prototypical member of the polycystin-2 subfamily of the TRP superfamily of calcium permeable channels. Polycystin-2 also shares sequence similarity with the last six membrane spanning domains of polycystin-1, raising the possibility that polycystin-1 may participate in channel function as well. These proteins are highly conserved throughout metazoan evolution, but are absent from unicellular organisms.
Location of polycystin-2 in kidney
To address the issue of the location of functional polycystin-2, we fractionated whole kidney tissue from wild type mice on iodixanol-based linear gradients and compared the distribution of polycystin-2 with the distribution of proteins from the plasma membrane, Golgi apparatus and ER. The distribution of polycystin-2 across the gradient closely paralleled the distribution of the ER membrane protein calnexin, and differed significantly from a pair of integral proteins of the basolateral and apical plasma membranes, epithelial growth factor receptor (EGFR) and sodium proton exchanger 3 (NHE3), respectively. The distribution of polycystin-2 also differed from that of the Golgi 58 kDa protein (panel a). We next treated all fractions individually with Endo H to determine if polycystin-2 in all fractions retained sensitivity to Endo H as an indication of expression in the ER compartment. Polycystin-2 in every fraction was sensitive to Endo H (panel b). Taken with additional biochemical assays described in our paper, these data, support the hypothesis that at least 99.5% of the cells polycystin-2 is restricted to the ER in native kidney tissue.
Location of polycystin-2 in primary cilia
A second important location of functional polycystin-2 is the primary cilia of cells. Primary cilia are found in virtually all dividing cells, but their function is unclear. However, the loss of polycystin-2 in mutated mouse cells will allow investigations into the functions of both polycytin-2 and primary cilia.
Polycystin-2 forms a calcium permeable channel
Divalent cation currents were observed when ER vesicle preparations over-expressing polycystin-2 were fused to planar lipid bilayers (second trace, after perfusion). No such currents were observed when ER vesicle preparations from LLC-PK1 cells expressing the empty vector alone were fused to bilayers (bottom trace, after perfusion). The top and third traces show that potassium and chloride channels from both types of cells were fused with the bilayer. When ER vesicles from cells expressing the naturally occurring missense mutation D511V were fused to bilayers, no divalent cation currents were observed (not shown).
Calcium transients are enhanced by expression of polycystin-2
To test the hypothesis that activation of polycystin-2 would lead to enhanced calcium transients in intact cells, LLC-PK1 cells stably expressing wild type and mutant polycystin-2 were loaded with the calcium-sensitive fluorescent dye fluo-3 and stimulated with vasopressin to induce intracellular calcium release via the InsP3R. LLC-PK1 cells expressing only endogenous polycystin-2 responded to stimulation by vasopressin with a single transient increase in cytosolic calcium. The calcium transient declined to baseline levels within 60 seconds after addition of vasopressin. Similar results were obtained with LLC-PK1 cells stably transfected with the empty expression vector (vector), the L703X truncation mutant (PKD2-L703X) and the D511V missense variant (PKD2-D511V). By contrast, LLC-PK1 cells over-expressing full length polycystin-2 showed an approximately two-fold larger increase in cytosolic Ca2+ and an approximately ten-fold longer duration of the Ca2+ transient (PKD2-WT). Identical results were obtained in the absence of extracellular calcium, consistent with the hypothesis that the increase in cytosolic calcium results from release from intracellular stores.
Measuring the pore size of polycystin-2
The conductance of a series of organic cations of increasing diameter was used to examine the pore size of the PC2 channel. We found that PC2 forms a functional pore with a sieve diameter larger than 11 Å, which means that multiple subunits are required to form a functional pore. From the size of the pore it is possible to calculate the number of transmembrane helices needed to make a large pore. With the tight packing of the transmembrane domains as shown for the K+ channels (Doyle et. al., 1998), at least eight transmembrane helices are needed to form such a large pore. Assuming that two transmembrane domains (S5 and S6) are contributed to form the pore (as in RyR/InsP3R/TRP), then our present experiments suggests a tetrameric assembly of PC2 protein.
Models for polycystin interaction and function
A number of models for polycystin function have been proposed; two are presented here. In all cases polycystin-1 is localized to the basolateral membrane and complexed with polycystin-2. Polycystin-2 may be either in the plasma membrane or the endoplasmic reticulum (ER) in close apposition to the plasma membrane. Signals from external stimuli are transduced by polycystin-1 in association with effector molecules and other potential participants. This activation results in increased local calcium. If polycystin-2 is on the cell surface, this increase in intracellular calcium results primarily from movement of extracellular calcium into the cell but may be augmented by local release of calcium from intracellular stores. If polycystin-2 is in the ER, then cation translocation across the plasma membrane, perhaps through polycystin-1 and associated proteins, triggers specific augmentation by calcium release from cytoplasmic stores through opening of the polycystin-2 channel. The role of the inositol 1,4,5-trisphosphate receptor in intracellular Ca2+ release in this pathway is uncertain. In either configuration, store operated calcium channels (SOC) may be activated in response to intracellular store depletion.
Regulation of the function of the polycystin proteins in isolated systems and in intact cells will be the focus of future studies. It is hoped that an understanding of the how the function of these proteins goes awry in disease will yield answers about normal biology and eventually will suggest therapeutic strategies for the disease.