 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
REVIEW |
Mount Sinai School of Medicine, New York, New York.
Correspondence to Dr. Patricia D. Wilson, Department of Medicine/Division of Nephrology, Mount Sinai School of Medicine, 1425 Madison Avenue, New York, NY 10029. Phone: 212-659-9383; Fax: 212-849-2434; E-mail: pat.wilson{at}mssm.edu
 |
Abstract |
|---|
 Top Abstract IntroductionStructure of Polycystin-1Function of Polycystin-1Regulation of Polycystin-1Structure of Polycystin-2Function of Polycystin-2ConclusionsReferences |
|---|
200 amino acid intracellular C-terminal
portion with several phosphorylation signaling sites. Polycystin-1 is highly
expressed in the basal membranes of ureteric bud epithelia during early
development of the metanephric kidney, and disruption of the PKD1
gene in mice leads to cystic kidneys and embryonic or perinatal death. It is
proposed that polycystin-1 functions as a matrix receptor to link the
extracellular matrix to the actin cytoskeleton via focal adhesion proteins.
Co-localization, co-sedimentation, and co-immunoprecipitation studies show
that polycystin-1 forms multiprotein complexes with 
2ß1-integrin,
talin, vinculin, paxillin, p130cas, focal adhesion kinase, and c-src in normal
human fetal collecting tubules and sub-confluent epithelial cultures. In
normal adult kidneys and confluent epithelial cultures, polycystin-1 is
downregulated and forms complexes with the cell-cell adherens junction
proteins E-cadherin and ß-, 
-, and -catenin. Polycystin-1
activation at the cell membrane leads to intracellular signaling via
phosphorylation through the c-Jun terminal kinase and wnt pathways leading to
activation of AP-1 and TCF/LEF-dependent genes, respectively. The C-terminal
of polcystin-1 has been shown to be phosphorylated by c-src at Y4237, by
protein kinase A at S4252, and by focal adhesion kinase and protein kinase X
at yet-to-be identified residues. Inhibition of tyrosine phosphorylation or
increased cellular calcium increases polycystin-1 focal adhesion complexes
versus polycystin-1 adherens junction complexes, whereas disruption
of the actin cytoskeleton dissociates all polycystin-1 complexes. Genetic
evidence suggests that PKD1, PKD2, NPHP1, and tensin are in the same
pathway. |
Introduction |
|---|
|
TopAbstractIntroductionStructure of Polycystin-1Function of Polycystin-1Regulation of Polycystin-1Structure of Polycystin-2Function of Polycystin-2ConclusionsReferences |
|---|
The major ADPKD disease gene, PKD1, on human chromosome 16p13.3 was the first to be identified and shown to have 46 exons and to encode a 14.5-kb transcript (7). Cloning of the full-length cDNA predicted a large, 4303 amino acid protein, termed polycystin-1, which was unique in terms of its modular structure and hydropathy characteristics (8,9). Since that time, it has become clear that polycystin-1 is a member of an ancient and conserved family of proteins that includes the PKD2-encoded protein polycystin-2, as well as homologues in the mouse, the Fugu fish, the sea urchin, and the worm Caenorhabditis elegans (10,11,12,13). Recently obtained molecular, cellular, and biochemical information has led to important insights into the function of the polycystins. Similar studies are beginning to shed light on how polycystin function can be regulated, an important prerequisite for therapy to restore polycystin function to normal in ADPKD patients.
|
Structure of Polycystin-1 |
|---|
|
TopAbstractIntroductionStructure of Polycystin-1Function of Polycystin-1Regulation of Polycystin-1Structure of Polycystin-2Function of Polycystin-2ConclusionsReferences |
|---|
462 kD
with its N-terminal extracellular and C-terminal intracellular
(Figure 1). The presence of a
signal sequence suggested that the protein would be synthesized and processed
through the secretory pathway of the cell, and hydropathy plot analysis
suggested 9 to 11 transmembrane spanning regions
(8,14).
These predictions of membrane insertion of polycystin-1 have since been
confirmed by many investigators using a variety of anti-poly-cystin-1
antibodies or epitope-tagged transfected cDNA and labeled proteins
(15,16,17).
|
Extracellular Domain
After clipping of a characteristic signal sequence at the N-terminal of the
protein, the first >3000 amino acids form a long extracellular portion that
contains several putative structural domains. The first of these is two
cysteine-flanked, leucine-rich repeats that, when present in other proteins,
are known to be associated with protein-protein interactions involved in
adhesion. Typically, however, leucine-rich repeats occur in much larger
multiples than the two found in polycystin-1. For instance, there are 21 or
more in the Drosophila Toll protein, decorin, and biglycan
(18). Cysteine-rich flanking
is typical of extracellular proteins. The next putative domain is homologous
to a C-type lectin motif that predicts protein-carbohydrate interactions,
because they are found in proteins such as the selectins
(19). An interesting property
of C-type lectins is their calcium sensitivity. The presence of a 40 amino
acid cysteine-rich low-density lipoprotein (LDL-A) homologous domain in the
N-terminal of polycystin-1 is difficult to interpret, although this type of
hydrophobic extracellular domain has been implicated in ligand binding.
A distinctive and unique feature of polycystin-1 is the presence of 16 copies of the so-called "Ig-like" PKD domains. The first PKD repeat is located between the C-lectin and LDL-A domains; the rest are located after the LDL-A domain and are linked by 5 to 7 amino acid spacers (see Figure 1). Although the overall homology to immunoglobulins and to one another is low, there is a conserved portion of sequence (WDFGDGS) in all human PKD domains. Nuclear magnetic resonance structural analysis of PKD repeat number 10, which is conserved from human to Fugu fish, revealed an Ig-like ß-sandwich fold, again implicating a role in ligand binding for the extracellular portion of the polycystin-1 protein (20). Overall, these structural sequence analysis studies suggest that the long N-terminal portion of polycystin-1 might be capable of binding one or more ligands in the extracellular compartment. Supportive evidence of this notion includes the finding that polycystin-1—containing epithelia specifically bind type I collagen (21).
The most distal specifiable domain in the extracellular portion of the polycystin-1 protein is a 1000 amino acid sequence that shows high homology to the sea urchin receptor for egg jelly (REJ) (11). Recently, additional family members of REJ have been identified in the human testis (9). Although the function of this domain in the polycystin-1 protein remains obscure, it should be noted that in sea urchin it is involved in the acrosome reaction during which, after the attachment of the sperm, a calcium flux is generated and leads to remodeling of the sperm membrane to facilitate binding. It is possible, therefore, to speculate that there is a similar calcium influx role for this portion of the membrane protein in polycystin-1.
Transmembrane Domain
Hydropathy plot and computer-assisted analyses identified 9 to 11
transmembrane regions from amino acid 3075 to 4104 (or 4014) with intervening
intracellular and extracellular loops
(8,14;
C. R. Burrow, personal communication). The uncertainty in the literature,
including suggestions that there are as few as seven transmembrane domains, is
a reflection of the limitation of computer algorithms and emphasizes the need
for direct biochemical protein purification and crystallization studies to
elucidate definitively the structure. Nevertheless, computer-assisted analysis
carried out since the elucidation of the cDNA and predicted amino acid
sequence has provided strong clues and allowed hypotheses relating to the
structure and function of the protein to be constructed and tested. One such
strong hypothesis was that polycystin-1 would prove to be a membrane protein;
this has been borne out by several immunolocalization studies
(15,16,17,21,22,23,24,25,26).
As yet unknown and untested are the significance of motifs present in the intervening loops between transmembrane domains 1 and 11. Of potential interest are a tyrosine at amino acid position 3614 (14) in an intracellular loop between membrane spans 5 and 6 and a proline-rich sequence at 3524 in an intracellular loop between membrane spans 3 and 4, consistent with a consensus sequence for a src homology 3 (SH3) binding site (27). Both of these suggest possibilities for signal transduction function of the protein.
Intracellular Domain
The C-terminal 200 (or 226) amino acids of polycystin-1 form the most
distal, intracellular portion of the molecule and contain numerous potential
phosphorylation sites, some of them in well-characterized consensus sequences
of known signaling molecules. A putative role for polycystin-1 in
intracellular signal transduction is highly predicted for this portion of the
molecule because of the presence in an intracellular C-terminal portion of the
protein of four tyrosines, one in a consensus sequence for an SH2
domain-containing protein binding site
(28). In addition, among the
several serines and threonines contained within this 200 amino acid portion,
there are two consensus RSSR sequences in the human protein at 4162 to 4167
and 4250 to 4253, which are consistent with target sites for phosphorylation
by protein kinase A (PKA) or C (PKC)
(28). Other putative domains
known to be involved in signaling functions include a proline-rich sequence
consistent with a putative WW site
(29) and a heterotrimeric G
protein activation sequence
(30). Also of interest, the
C-terminal portion of polycystin-1 contains a putative coiled-coil region
(human amino acids 4193 to 4248), which is a motif of bundles of
-helices wound into a superhelix with knobs into holes packing
geometry, first identified in 1953 by Pauling and Cory in fibrous proteins and
later also in the dimerization elements of the leucine zipper class of
transcription factors (see reference
31 for review). Coiled coils
are also found in the vesicle fusion molecules SNARES, synaptobrevin, and
SNAP, which play an important role in membrane protein trafficking. This
allows the speculation that mutations that truncate this portion of
polycystin-1 protein might be responsible for mispolarization of membrane
proteins demonstrated in ADPKD epithelia
(16,32,33,34,35).
The presence of this type of coiled-coil motif in the C-terminal portion of
polycystin-1 (and polycystin-2) strongly predicted protein-protein
interactions, and experimental analysis using yeast 2-hybrid and
overexpression techniques have been used to confirm this possibility. For
instance, the coiled-coil region of polycystin-1 has been shown to bind
polycystin-2 in vitro at least in cell-free systems or after
transfection and expression at an unphysiologically high level
(36,37).
Therefore, these studies show a capability for these two polycystins to bind,
but proof will await comprehensive immunolocalization studies in
vivo, which have begun and suggest that although polycystin-1 and
polycystin-2 occasionally co-localize, they often do not
(25,26).
|
Function of Polycystin-1 |
|---|
|
TopAbstractIntroductionStructure of Polycystin-1Function of Polycystin-1Regulation of Polycystin-1Structure of Polycystin-2Function of Polycystin-2ConclusionsReferences |
|---|
Expression
The tissue and cellular and subcellular distribution of proteins provides a
first line of interpretation of putative protein function. Several studies
have used anti-polycystin-1 antibodies
(15,16,17,21,22,23,24,38,39)
and cDNA as probes
(7,16).
Despite some discrepancies probably relating to individual monoclonal and
polyclonal antibodies raised against various epitopes and peptide sequences in
the large protein, the following general conclusions can be drawn:
polycystin-1 has a fairly wide tissue distribution, being highly expressed in
kidney, brain, liver, pancreas, heart, intestine, and several cell lines. Of
particular interest, there is strong evidence of developmental regulation of
polycystin-1 in the kidney
(23,24,38,40).
The functional significance of developmental regulation is borne out by the
perinatal lethality of targeted mouse PKD1 mutants with an 836
residue truncation due to enormously cystic kidneys
(41). Even more compelling is
the report that heterozygotes for this targeted disruption of polycystin-1
developed cystic kidneys in mid-life, which models the typical pattern of
disease seen in human ADPKD
(42). Other mutations in the
PKD1 gene that lead to a larger truncation lead to embryonic
lethality, as do those in PKD2
(43,44).
Taken together, these studies prove that polycystin-1 plays an essential role
in the development and differentiation of normal kidney architecture and
function.
The cellular and subcellular distribution patterns of polycystin-1 in the human and mouse kidney help to provide clues to determine its function. Polycystin-1 is highly concentrated in the epithelia of the ureteric bud structures in early metanephric kidneys, where it is seen on cell membranes associated with focal densities at the basal membrane in contact with the extracellular matrix of the blastema (16,21) (Figure 2A). Of great interest and significance is that in the adult kidney, not only is there downregulation of polycystin-1 content on a per-gram protein basis (21), but also the localization is predominantly at the cell-cell junctions localized to the opical/lateral aspect of the cells of the medullary collecting duct epithelia (Figure 2B).
|
Of importance for functional studies is that we have derived cell strains and lines from normal human fetal renal collecting tubule (HFCT) epithelia that endogenously express high levels of normal polycystin-1, as well as adult normal human kidney collecting tubule cell strains that express low levels (45,46). Using these cell cultures derived from normal epithelia, it is possible to mirror the two types of expression pattern seen in human kidneys in vivo. In subconfluent cultures, where the predominant interactions are between cells and their matrix, there is a focal adhesion pattern of distribution of polycystin-1 at the basal membrane of cells at the interface between the cell contact and the extracellular matrix (Figure 2C). By contrast, in confluent cultures where cell-cell interactions predominate, the distribution pattern of polycystin-1 is predominantly lateral/apical and shows the typical "chicken-wire" appearance (Figure 2D). These studies suggest that during development and maturation, the distribution of polycystin-1 shifts from the initial localization in cell-matrix focal adhesions to the later cell-cell association junctions at the apical aspect of the lateral membranes (47,48). Paradigms exist for proteins that are important and highly expressed in renal development being retained after development to control differentiation (e.g., the Wilms tumor protein). This also seems to be the case for polycystin-1 in normal kidneys, although it has been shown that expression levels and patterns in adult kidneys can also be increased and altered by injury, cytokines, and disease, including acute tubular necrosis and renal tumorigenesis.
Cell Biology of ADPKD
Much has been learned about the cell biologic abnormalities associated with
ADPKD in the kidney (see references
16,47,49,50,51
for reviews). Major defects of increased proliferation and apoptosis, ion and
fluid secretion, membrane protein mispolarization, cell-matrix and adhesion
abnormalities, and underlying persistent fetal gene expression have been
demonstrated. Although the individual abnormalities seem diverse, they all are
consistent with the role of polycystin-1 as critical in the pleiotropic
mechanisms of renal organogenesis, which involve an exquisite coordination of
proliferation and apoptosis of renal progenitor cells and tubule epithelia,
migration through attachment and disassociation of the ureteric bud epithelium
through the metanephric mesenchyme and matrix, differentiation and
polarization of membrane proteins in the maturing renal tubule epithelia, and
acquisition of ion and fluid-absorptive properties of the maturing
nephron.
To coordinate these events and produce a normal kidney, it is clear that numerous genes must be regulated strictly. Recent results have shown that the underlying mechanisms of mispolarized epithelial growth factor receptor (34) and NaK-ATPase membrane proteins (32) in ADPKD kidneys are due to the failure to switch off fetal gene transcription of ErbB2 and the ß-2 subunit of NaK-ATPase, respectively (52,53,54). This has led to the working hypothesis that ADPKD results from the failure of polycystin-1 to regulate gene transcription appropriately as a result of an interference with the activation of normal signal transduction cascades in response to ligand interactions with polycystin-1 at the cell surface (Figure 3). According to this model, in normal conditions during development, polycystin-1 receives information from the extracellular compartment and via phosphorylation events in its C-terminal signals intracellularly to the nucleus regulating gene transcription appropriately. In ADPKD, the lack of appropriate C-terminal—mediated signal transduction cascades would lead to inappropriate regulation of gene transcription and abnormal kidney development. A precedent for this type of matrix-induced signal transduction exists, of course, with regard to the matrix receptors of the integrin family. After interaction with extracellular ligands, integrins cluster and nucleate a group of proteins of the focal adhesion complex, which then activate by phosphorylation via protein tyrosine kinases, such as focal adhesion kinase (FAK) and src, downstream signaling cascades that impact on active gene regulatory protein (AP-1)-dependent gene transcription (55,56,57,58,59)
|
Binding Proteins
Several lines of evidence suggest that polycystin-1 can and does bind to
other proteins to form a multiprotein complex at the membranes of cells.
Initial studies using in vitro cell-free systems and fusion proteins
(yeast 2-hybrid) approaches as well as transfection and forced overexpression
in HEK293 cells suggested that polycystin-1 might bind to polycystin-2 and
possibly other cytoskeletal proteins in cells
(36,37).
Although these types of study are highly unphysiologic and prone to false
positives, they provide a rapid means of generating hypotheses.
Immunohistochemical analysis of polycystin-1 and polycystin-2 in a variety of
tissues has confirmed the co-expression of these proteins
(25,26).
An in-depth physiologic analysis has been carried out using cell lines that
endogenously express normal levels of natural polycystin-1 protein. Double
immunofluorescence labeling studies showed that immediately after plating HFCT
at low density on type I collagen, polycystin-1 co-localized in a completely
overlapping manner with the collagen receptor 2ß1-integrin and the
focal adhesion proteins vinculin and paxillin
(21). The sites of
co-distribution were at focal adhesion-like accumulations in the focal plane
of the basal cell membrane where it contacted the matrix. In a large series of
studies in which cell extracts were sedimented over sucrose density gradients,
13 fractions that were collected and analyzed by Western immunoblot and
immunoprecipitations established that polycystin-1 not only co-distributes but
also co-associates with several "classical" proteins of focal
adhesion complexes that form in cells after attachment to matrix via integrins
(21,48,60).
Taken together, the results of several studies show that the protein present
in polycystin-1 complexes include the structural and actin-binding proteins
vinculin, talin, tensin, and -actinin; the adaptor proteins paxillin
and p130cas; and the signaling kinases c-src and FAK
(Figure 4A). Polycystin-1
association with this group of proteins is prevalent shortly after cells
attach to type I collagen matrix in a subconfluent density, which models the
state of ureteric bud cells during development.
|
Focal adhesions are sites of nucleation of intracellular proteins on the cytoplasmic face of the cell membrane and provide a link between the extracellular matrix and the actin cytoskeleton. These have been shown recently to be dynamic structures of varying sizes and regulate cell-matrix interactions, motility, and signal transduction via a complex set of phosphorylation and dephosphorylations of component proteins (61,62,63,64). A critical component of focal adhesions is the nonreceptor tyrosine kinase FAK, which is phosphorylated in response to matrix adhesion via integrin and then regulates phosphorylation of its downstream substrate targets such as paxillin. FAK function has been shown to be determined by its targeting and recruitment to the focal adhesion complex via binding of its C-terminal domain focal adhesion-targeting sequence (65). In light of these findings, it was of great interest to demonstrate that although polycystin-1—containing focal-adhesion multiprotein complexes assembled at the membranes of ADPKD epithelial cells containing a 1227 amino acid truncation, one protein that is always absent from the ADPKD complexes was FAK (21,48). Because, as in all cells, it has been shown that there is free FAK in the cytoplasm of ADPKD epithelial cells, this suggests that the mutant polycystin-1 is specifically unable to recruit FAK to its multiprotein complexes, which is expected to have drastic consequences on downstream signaling.
Confluent normal renal epithelial cells are a model for the state found in
normal adult renal epithelia, because polycystin-1 is localized at the
apical/lateral cell-cell borders as in normal collecting tubule epithelia
in vivo. When co-localization, sedimentation, and precipitation
analyses were carried out on normal confluent cells, it was interesting to
note that in addition to some focal adhesion protein interactions,
polycystin-1 showed major interactions with proteins characteristic of the
cell-cell adherens-type junctions, notably, E-cadherin and ß-, -,
and -catenin
(48,66)
(Figure 4B). The cell-cell
adherens-type junctions in renal epithelia are formed at the apical-lateral
aspect of cells, just below the tight junctions. E-cadherin molecules from
adjacent cells form zipper-like attachments via homophilic calcium-dependent
interactions. Each cadherin molecule has a single membrane span, and the
highly conserved intercellular domain interacts with ß-catenin, which
binds to -catenin and perhaps plakoglobin (-catenin) to form a
link to the actin cytoskeleton via interactions with the actin-binding
proteins -actinin or vinculin. In addition to its role in adhesion and
assembly of adherens junctions, ß-catenin has been shown to play and
important role in signaling in the wnt pathway and subsequent
regulation of gene transcription. Cellular levels of ß-catenin are
downregulated constitutively by a component of the wnt pathway
glycogen synthase kinase (GSK)-3ß via phosphorylation found in complexes
with adenomatous polyposis coli and axin. When the wnt pathway is
activated, GSK-3ß is inactivated, allowing stabilization of
ß-catenin, which then translocates into the nucleus and forms a complex
with T-cell-specific transcription factor/lymphoid enhancer binding factor
(TCF/LEF) transcription factors and activates dependent gene transcription
(56,67,68).
Thus, ß-catenin has independent roles in adhesion and signal
transduction/gene transcription, and, not surprising, plays an important role
in many developmental processes
(69). This allows the
speculation that the same might be true in renal development via its
association with polycystin-1. Further studies suggest that polycystin-1 can
bind and activate heterotrimeric G proteins at least in vitro
(30) and that polycystin-1 can
inhibit the degradation of the regulator of G protein signaling protein, RGS7
(70), suggesting a putative
role for polycystin-1 in signal transduction.
Signal Transduction
Because the C-terminal portion of the polycystin-1 protein not only
interacts with many other proteins but also contains many putative
phosphorylation and proline-rich sites indicative of signaling function,
several studies have been carried out to identify signal transduction cascades
that are activated by polycystin-1. For these types of studies, it is
necessary to transfect into cells the C-terminal portion of polycystin-1 and
for it to be expressed at the cell membrane, its normal site of action. Walz
and colleagues were the first to succeed in this type of study using
heterologous constructs in which the C-terminal amino acid sequence of human
polycystin-1 was attached to CD16 and CD7 or CD5 sequences to secure insertion
into the plasma membrane (71),
which, of course, are not present in the normal polycystin-1 protein.
In an attempt to produce a more homologous system, we used constructs with
the C-terminal amino acid sequence of human polycystin-1 inserted into the
membrane via either a myristoylation signal (found in c-src) or an additional
sequence of the last five natural transmembrane domains of polycystin-1 and a
preprotrypsin leader sequence for membrane insertion. Both of these constructs
were shown to insert into the plasma membrane of transfected cells as detected
by FLAG or green fluorescence protein reporter tags. Using this approach we
have been able to confirm the findings that polycystin-1 can stimulate AP-1
dependent gene transcription via the c-Jun terminal kinase pathway. We also
found that this effect was stimulated by cyclic adenosine monophosphate. Most
interesting is that it has also been demonstrated that these effects can be
abrogated by dominant negative constructs of the cytoskeletal effector small
guanosine triphosphatases (GTPases) rac-1 and cdc-42 as well as by the
calcium-dependent PKC
(71).
In addition to the AP-1-dependent class of genes that is often related to proliferation control, Walz's group showed that TCF-dependent gene transcription can be regulated by polycystin-1 C-terminal via the wnt pathway (72). Polycystin-1 inhibits the phosphorylation of intermediate GSK-ß, which in turn leads to the accumulation and stabilization of ß-catenin, which complexes with the transcription factor TCF and leads to transcriptional activation of TCF-responsive genes. Although these studies should be viewed with some caution because of the heterologous nature of the constructs used and the forced overexpression of the proteins, the correlation with independent studies showing co-sedimentation and co-immunoprecipitation of polycystin-1 with ß-catenin in endogenously expressing cells (48,73) lends credence to the theory that modulation of the Wnt-pathway maybe an important route through which polycystin-1 regulates differentiation-related genes (Figure 5).
|
|
Regulation of Polycystin-1 |
|---|
|
TopAbstractIntroductionStructure of Polycystin-1Function of Polycystin-1Regulation of Polycystin-1Structure of Polycystin-2Function of Polycystin-2ConclusionsReferences |
|---|
-, and -catenin complexes. A
regulatory role for the actin cytoskeleton in formation of both types of
multiprotein complex was therefore hypothesized and has been demonstrated
using co-localization and co-sedimentation analyses
(48,74).
Cytochalasin D, which depolymerizes actin microfilament, altered cellular
architecture and completely disrupted the polycystin-1-containing multiprotein
complexes in subconfluent (focal adhesion-complexed) and confluent (cell-cell
adherens-complexed) cells. Because colchicine had no effects under the
identical conditions, this strongly implicates a role for the actin
cytoskeleton in the control of polycystin-1-containing multiprotein complex
formation.
Phosphorylation
There are several cytoplasmic phosphorylation sites on the polycystin-1
protein (see Figure 1).
Phosphorylation of the endogenous protein in normal HFCT cells was
demonstrated directly by immunoprecipitation with polycystin-1 antibody
followed by Western immunoblot analysis with antiphosphotyrosine and vice
versa (21). In these
studies, there was also a suggestion that ADPKD epithelia with truncation
mutations in polycystin-1 showed reduced levels of polycystin-1
phosphorylation on tyrosine. A different approach has also implicated a
regulatory role for tyrosine phosphorylation in polycystin-1-focal adhesion
multiprotein complex formation because the specific inhibitor tyrphostin has
been shown to abolish polycystin-1-FAK association. This was demonstrated by
sedimentation of cellular proteins over sucrose density gradients followed by
co-immunoprecipitation analyses of the separated fractions. After tyrphostin
treatment of cells, polycytsin-1 no longer co-immunoprecipitated with FAK
(48). By contrast, in
tyrphostin-treated HFCT cells, polycystin-1 associated with ß-catenin,
suggesting a potential for differential regulation of polycystin-1 complex
partners by tyrosine phosphorylation.
To analyze the specific sites of phosphorylation in the C-terminal cytoplasmic portion of polycystin-1, investigators have analyzed C-terminal polycystin-1 fusion proteins in vitro in the human (28) or the mouse (75). Using site-directed mutagenesis, Li et al. (28) analyzed all four tyrosines (Y4110, Y4118, Y4127, Y4237) and the serines in the two PKA/PKC RSSR consensus sequences (S4165/6; S4252/3) by in vitro kinase assays. With the use of either purified or recombinant enzymes as well as endogenous cell lysates, it has been established that c-src, FAK, PKA, and a novel kidney cyclic adenosine monophosphate-dependent kinase protein kinase X can phosphorylate the polycystin-1 C-terminal (28,76). This suggests that both tyrosine and serine/threonine phosphorylation may be important mediators of polycystin-1 function. Most important, the site-directed mutagenesis of single residues of the human protein has identified Y4237 as the specific target site for s-src and S4252 as the specific target site for PKA.
Calcium
Independent lines of evidence have suggested that calcium might play a role
in the regulation of polycystin function. The extracellular domain of
polycystin-1 contains C-lectin and REJ homology domains, and the intracellular
C-terminal of polycystin-2 contains an EF hand, all related to
calcium-dependent binding or ion influx. In addition, polycystin-2 can bind a
transient receptor channel protein-1 (TRPC-1), which might allow calcium ion
influx and intracellular signaling, and AP-1-dependent gene transcription can
be modulated by the calcium-dependent kinase PKC. Most compelling,
perhaps, is the ability of polycystin-1 to form multiprotein complexes with
the calcium-dependent E-cadherin; our recent functional studies implicated a
direct role for calcium in the regulation of the type of
polycystin-1-multiprotein complex associations. Increased calcium
concentrations in the media of confluent normal HFCT cells induced increased
polycystin-1-E-cadherin-ß-catenin association and decreased
polycystin-1-FAK associations
(48). This suggests that local
calcium concentrations at the cell membrane may regulate differential
distribution of polycystin between cell-matrix (integrin-FAK) and cell-cell
(E-cadherin-ß-catenin) contacts; the former contacts are predominant
during development and migration of the ureteric bud through the metanephric
blastema, and the latter are predominant in adult, fully differentiated
collecting ducts.
|
Structure of Polycystin-2 |
|---|
|
TopAbstractIntroductionStructure of Polycystin-1Function of Polycystin-1Regulation of Polycystin-1Structure of Polycystin-2Function of Polycystin-2ConclusionsReferences |
|---|
|
Function of Polycystin-2 |
|---|
|
TopAbstractIntroductionStructure of Polycystin-1Function of Polycystin-1Regulation of Polycystin-1Structure of Polycystin-2Function of Polycystin-2ConclusionsReferences |
|---|
Expression analyses show that polycystin-2 expression is widespread in many tissues and that it is developmentally regulated in the kidney with highest levels in distal tubules (25,26). It is interesting that the renal cysts derived in PKD2 knockout mice also seem to derive from predominantly distal segments of the nephron (44). Most recent results suggest a complex relationship between polycystin-2 and polycystin-1 because although there are some areas of overlap of expression, there are many areas in which polycystin-2 is expressed but polycystin-1 is not, suggesting independent functions. Also, at the subcellular level, although some similar tubular cell types can express both proteins, the basal localization of polycystin-2 in collecting ducts does not overlap with the apical/lateral localization of polycystin-1 (26).
It may be hypothesized that polycystin-2 could have more than one function, dependent on its association with polycystin-1 protein or not. Elucidation of its true function will await further studies that use transfected or endogenously expressing cell lines. A single such study of signal transduction potential showed that polycystin-2, like polycystin-1, can activate AP-1-dependent gene transcription via c-Jun terminal kinase but also via mitogen-activated protein kinase and that this can be modulated by the calcium-independent PKC epsilon (71), again suggesting that polycystin-2 might have an impact on similar downstream targets as polycystin-1 but via different intermediate pathways. Furthermore, PKD2 has been shown to interact with Hax-2, a protein that forms links with the actin cytoskeleton (79), showing further overlap with polycystin-1 targets.
|
Conclusions |
|---|
|
TopAbstractIntroductionStructure of Polycystin-1Function of Polycystin-1Regulation of Polycystin-1Structure of Polycystin-2Function of Polycystin-2ConclusionsReferences |
|---|
200
amino acid intracellular C-terminal domain with several phosphorylation sites
in known signaling consensus. Cellular and subcellular distribution,
developmental regulation, and genetic studies in mice with targeted mutations
show that polycystin-1 is essential for normal renal development and that its
primary site of action is at the basal membrane focal adhesions of the
ureteric bud epithelium. Normal renal development of the metanephric kidney is
the result of reciprocal interactions between the migrating and branching
ureteric bud and the metanephric blastema composed of mesenchyme and matrix.
Several components of the focal adhesion protein complex have been implicated
and are necessary for normal migration of cells, including FAK, p130Cas,
paxillin, and c-src
(64,80,81,82).
It now seems that polycystin-1 is part of this complex also and that it, too,
is critical for normal migration of the ureteric epithelium through the matrix
of the metanephric blastema. In addition, activation of polcystin-1 seems to regulate at least two sets of genes: AP-1-dependent, which are often involved in regulation of proliferation, and wnt-responsive TCF/LEF genes, also involved in differentiation. These processes, of course, are critical during development, and upstream mutations in components such as polycystin-1 is expected to disrupt gene transcription related to nephrogenesis and differentiation. This is consistent with the findings that ADPKD cellular defects are a consequence of inappropriate persistent fetal gene expression of critical differentiated membrane proteins (49,52,53,54,83).
Because mutations in polycystin-2 cause a disease in humans with an
identical phenotype, although maybe a slower progression, it is suggested that
these genes work in the same pathway. These conclusions have been verified by
the similar renal cystic phenotypes seen in the knockout mice for
PKD1 and PKD2. A new paradigm for the understanding of
genetic diseases with similar phenotypic outcomes has been established in
studies of the genetic basis of muscular dystrophy, in which it has been shown
that mutations in dystrophin, sarcoglycans, and laminin-2 all give rise to a
type of muscular dystrophy
(84,85).
Of particular interest, these proteins form a complex that links the
extracellular matrix to the actin cytoskeleton. Clear parallels therefore
exist with polycystin-1, which forms a complex with focal adhesion proteins to
link the extracellular matrix through the cell membrane to the actin
cytoskeleton. If the muscular dystrophy paradigm is correct, then mutations in
other proteins of the polycystin-1 complex also should lead to cystic kidney
disease. This has been established for four components: polycystin-1,
polycystin-2, nephrocystin, and tensin
(Figure 4A, denoted by
*)
(4,7,41,44,86,87).
Furthermore, an additional protein of the focal adhesion complex
-actinin-4 has been implicated as a genetic cause of familial focal
segmental glomerulosclerosis
(88).
If polycystin-1-containing complex function is critical to providing a link between the extracellular matrix and the actin cytoskeleton to lead to development of a normal kidney, it is predicted that the organization and integrity of this cytoskeleton is required. Synthesis and organization of the actin cytoskeleton are under the control of several proteins of the Ras superfamily, including the small GTPases rac, rho, and cdc-42. Of interest, dominant negative rac-1 and cdc-42 abrogated polycystin-1-dependent AP-1 activation (71). This family of small GTPases also plays critical roles in cell polarity and adhesion during cell migration (89). Further evidence for the involvement of the actin cytoskeleton in normal polycystin-1 function is the complete disruption of polycystin-1-containing complexes by cytochalasin D, showing the necessity of an intact actin cytoskeleton for function in a multiprotein complex (48).
The current body of evidence suggests that polycystin-1 functions normally at the cell membrane by forming multiprotein complexes and receives input from ligand interactions in the extracellular matrix (or on other cells), becomes phosphorylated, and regulates a phosphorylation cascade of activation of intracellular signaling through its C-terminal intracellular portion. Truncation mutations that fail to signal appropriately seem to fail to recruit the normal complement of proteins, lacking FAK, and subsequently fail to downregulate fetal gene transcription and thereby allow the postnatal expression of fetal proteins associated with the cell biology and pathophysiology of ADPKD, notably altered matrix adhesion, ion and fluid secretion, and increased epithelial cell proliferation. Our view is that polycystin-1 acts as a type of master regulator of cell function through its ability to assemble large signaling complexes and activate several signaling pathways.
|
References |
|---|
|
TopAbstractIntroductionStructure of Polycystin-1Function of Polycystin-1Regulation of Polycystin-1Structure of Polycystin-2Function of Polycystin-2ConclusionsReferences |
|---|
2-ß1 integrin in focal clusters in adherent
renal epithelia. Lab Invest 79:1311
-1323, 1999[Medline]
This article has been cited by other articles:
 |
|
 |
|
  D. Olteanu, M. B. Hovater, and E. M. Schwiebert Intraluminal autocrine purinergic signaling within cysts: implications for the progression of diseases that involve encapsulated cyst formation Am J Physiol Renal Physiol, January 1, 2007; 292(1): F11 - F14. [Full Text] [PDF]
|
|
|
|
 |
|
 |
, phosphorylation.