Department of Biochemistry and Molecular Biology
University of Connecticut, Storrs, CT, Ph.D., 1975
Worcester Foundation for Experimental Biology, Shrewsbury, MA (American Cancer Society Postdoctoral Fellow and Senior Research Associate)
Publications: Click here
Polycystic kidney disease (PKD) is characterized by the growth of numerous, large epithelial-lined cysts from the nephrons and collecting ducts of affected kidneys. It is one of the most common lethal genetic diseases in the U.S. and worldwide. Cyst formation is initiated early in fetal kidneys, suggesting that the disease is caused by an abnormality in the differentiation of renal tubules. Most cases of human PKD are known to be caused by mutations in two unlinked genes, PKD1 and PKD2, with mutations in the PKD1 gene being associated with ~85% of PKD cases. The PKD1 protein product, polycystin-1, is a very large, membrane-associated glycoprotein. The N-terminal, extracellular region is ~3,000 amino acids in length, the membrane-spanning region (including 11 transmembrane domains) is ~1,000 amino acids in length, and the C-terminal, cytosolic domain is ~200 amino acids in length.
A number of protein motifs have been identified in the N-terminal, extracellular region and in the C-terminal, cytosolic region that suggest that polycystin-1 may function as a plasma membrane receptor, possibly to regulate cell differentiation as a ciliary mechanosenor. The PKD2 protein product, polycystin-2, is a calcium channel that directly interacts with polycystin-1. Thus, polycystin-1 may regulate polycystin-2 channel activity. The proposed role of polycystin-1 as a membrane receptor is supported by evidence from other labs and by our own work showing that expression of polycystin-1 fusion proteins in transiently transfected cells activates a number of cellular signaling pathways, including activation of the AP-1 and NFAT transcription factors (Parnell et al., 2002; Puri et al. 2004). Much of our current work is based on our evidence that polycystin-1 functions as a heterotrimeric G protein coupled receptor (Parnell et al., 1998; 2002).
To test this, we carried out in vitro protein binding assays using bacterially expressed polycystin-1 fusion constructs, which showed that the C-terminal cytosolic domain can stably bind G proteins. The binding domain was found to be within an evolutionarily conserved region of polycystin-1, and to contain a 20-amino acid peptide sequence that can activate purified G proteins in vitro. These results were the first to demonstrate that polycystin-1 may function by initiating heterotrimeric G protein coupled signal transduction.
Polycystin-1 mediated calcium and cAMP signaling
Several key observations made over the past several years have suggested that cAMP is central to the pathogenesis of PKD. First, cAMP has been shown to be the driving force in the activation of the Ras/MAPK pathway and in the resulting cell proliferation that is required for cyst formation and cyst growth. Second, cAMP activation of CFTR has been shown to be the driving force for the fluid secretion required for cyst enlargement after cysts pinch off from the nephron. The central role of cAMP has recently been further confirmed by a series of experiments in which animal models of PKD were successfully treated with a vasopressin V2 receptor antagonist, presumably because the antagonist lowered renal cAMP levels.
However, while these experiments support the view that cAMP is important, it is not known how a defect in the polycystins can cause misregulation of cAMP-mediated mechanisms associated with increased cell proliferation and fluid secretion, both integral to cyst growth and enlargement. The polycystins are thought to regulate intracellular calcium mobilization in response to some yet-to-be-determined ligand-mediated or mechanosensory stimulus. As stated above, polycystin-2 has been shown to be a calcium-regulated non-specific cation channel, and polycystin-1 is thought to regulate polycystin-2 activity. Recently, we showed that polycystin-1 alone is capable of elevating intracellular calcium through a heterotrimeric G protein-coupled mechanism. Both polycystin-1 and polycystin-2 have been shown to be required for ciliary mechanosensory increases in intracellular calcium; however, it is not known how these calcium fluxes are linked to cAMP signaling or to cell proliferation.
Recently, we showed, using three cell culture model systems, that PKD-like cell proliferation is dependent on cAMP-mediated activation of the Ras/B-Raf/ERK pathway (Sutters et al., 2001; Yamaguchi et al., 2004). This was shown in primary cultures of ADPKD cyst epithelial cells, in M-1 cells stably transfected with a dominant-negative polycystin-1 construct, and in M-1 cells treated with calcium channel blockers. All three systems displayed the same phenotypic switch from cAMP-inhibited to cAMP-stimulated cell proliferation. Yet, despite these common observations, it is not clear whether and how the polycystins may be involved.
As such, we are determining if the PKD-like cAMP-stimulated phenotype is a characteristic of Pkd1 deficient mouse cells. For these studies, we are using the Pkd1(m1bei) mouse, which has a single amino acid missense mutation that causes a PKD phenotype, and we are generating two new mouse models that will carry, single amino acid (aa) mutations in the heterotrimeric G protein binding region of polycystin-1 (to mimic known human mutations), and a 52 aa frameshift mutation in the heterotrimeric G protein binding region of polycystin-1. We are also utilizing metanephric kidney cultures from these mice to determine whether the PKD-like phenotypic switch occurs in embryonic kidney cells at a time when cysts begin to grow in human PKD (Magenheimer et al., in preparation). Finally, we are testing polycystin-1 mediated signaling activity under conditions of fluid flow mediated ciliary bending to determine if polycystin-1 can generate a calcium signal that can lead to altered calcium-regulated gene expression.
Regulation of PKD1 gene expression
Another project in the lab is looking into the regulation of PKD1 gene expression by analyzing 5' proximal promoter elements that function to up- or down-regulate PKD1 mRNA levels. We were the first to isolate and analyze the PKD1 promoter, showing that it is a target of the beta-catenin pathway (Rodova et al., 2002). More recent work has shown regulation of PKD1 promoter activity by Ets factors (Puri et al. in press) and by retinoic acid (Islam et al., in preparation). We have also determined that the last intron of the PKD1 gene (intron 45) has exceptionally high sequence conservation (Rodova et al., 2003). Pairwise comparisons for intron 45 showed 91% identity (human vs. dog) to 100% identity (mouse vs. rat) for an average for all four species of 94% identity. In contrast, introns 43 and 44 of the PKD1 gene had average pairwise identities of 57% and 54%, and exons 43, 44, and 45 and the coding region of exon 46 had average pairwise identities of only 80-84%.
Intron 45 is 90 to 95 bp in length, with the major region of sequence divergence being in a central 4-bp to 9-bp variable region. RNA secondary structure analysis of intron 45 predicts a branching stem-loop structure in which the central variable region lies in one loop and the putative branch point sequence lies in another loop, suggesting that the intron adopts a specific stem-loop structure that may be important for its removal. Although intron 45 appears to conform to the class of small, G-triplet–containing introns that are spliced by a mechanism utilizing intron definition, its high sequence conservation may be a reflection of constraints imposed by a unique mechanism that coordinates splicing of this last PKD1 intron with polyadenylation.
James P. Calvet, Ph.D.