Homozygous inv mice lack a functional inversin protein and exhibit situs inversus plus severe cystic changes in the kidney and pancreas. Although the inversin sequence has provided few clues to its function, we and others have previously identified calmodulin as a binding partner. We now provide evidence that inversin interacts with the anaphase promoting complex protein Apc2. As expected of an Apc2 target, inversin possesses D-boxes and site-directed mutagenesis of the well-conserved D-box residues abrogates inversin–Apc2 interaction. An inversin-specific antibody reveals a dynamic expression pattern throughout the cell cycle and strong expression in the primary cilia of renal epithelium. Our data support a role for inversin in primary cilia and involvement in the cell cycle. Mutations of the proteins polaris, cystin and polycystin-2 which are expressed in renal epithelium primary cilia, lead to renal cystic changes. Aberrant cell proliferation is also involved in cyst development. The data reported here suggest that inversin may provide a link between these two mechanisms.
The inv (inversion of embryo turning) mouse is an insertional mutant with a complex phenotype: homozygotes exhibit consistent situs abnormalities (unlike other mouse models of laterality defects where situs is randomized) and also severe cystic changes of the kidney and pancreas (1). Of the various mouse mutants displaying laterality disturbance or polycystic kidneys, only two other mutants, orpk (2) and pkd2 (3) have been reported with the same type of complex phenotype as seen in the inv mouse.
Primary cilia provide a common link between these two phenotypic aspects. Various mouse laterality mutants have been shown to display nodal cilia with aberrant morphology or motility (2,4–7) and the primary cilia of the embryonic node have been envisaged to play a major role in left–right axis determination (8,9). Primary cilia of renal tubule principal cells, which project into the lumen, have also been implicated in renal cyst formation: in each of the orpk (10), pkd2 (11) and cpk (12) mutants the normal products of the mutated loci (polaris, polycystin-2 and cystin, respectively) localize to renal primary cilia. Polaris has been shown to be a component of the intraflagellar transport system that is essential for ciliogenesis and as a result, polaris-deficient mutants lack cilia, both in the kidney and at the embryonic node (2,13).
Of the mechanisms implicated in cyst development, abnormal proliferation of renal tubular cells is a consistent characteristic finding in both humans and animal models with polycystic kidneys (14–16). The link between cilia and aberrant cell proliferation has, however, not been clear.
Invs, the gene mutated in the inv mouse, specifies a 1062 amino acid protein, inversin, whose precise function remains unclear (17,18). At the N-terminal end are 16 ankyrin repeats spanning amino acids 13–557 and putative nuclear-localization sequences have been reported at positions 589–604 and 782–798 (18). Evolutionary analyses have shown strong conservation of the ankyrin repeat region and also of a lysine-rich central domain spanning amino acids 558–604 (19). A highly conserved IQ calmodulin binding domain was identified in the lysine-rich central domain and another such domain in the otherwise poorly conserved C-terminal region and yeast two-hybrid screens, plus confirmatory co-immunoprecipitation and mutagenesis studies have identified calmodulin as an inversin binding partner (19,20).
In the present study we report evidence for another binding partner for inversin: Apc2, a subunit of the anaphase promoting complex. We also report that in dividing MDCK-II renal cells inversin has a dynamic pattern of expression through the cell cycle and although not essential for ciliogenesis, it is strongly expressed in the primary cilia of renal inner medullary collecting duct cells. Our data lead us to speculate on how an involvement in the cell cycle could explain the link between inversin and cystic kidneys.
RESULTS AND DISCUSSION
Inversin binds to the Apc2 subunit of the anaphase promoting complex and inversin–Apc2 interaction is dependent on a conserved D-box motif
We previously reported that calmodulin was a binding partner for inversin in a yeast two-hybrid screen (19). This screen also provided evidence for interaction between inversin and the Apc2 subunit of the anaphase promoting complex (APC; also called cyclosome). The APC, which consists of at least 10 subunits in vertebrates, is responsible for the ubiquitination of cell regulators at the metaphase–anaphase and mitosis–G1 transitions (21). A conserved destruction box (D-box) is important for the ubiquitination-mediated destruction of most APC targets and has a consensus sequence of RxxLxxxxN/D/E (21,22).
Inspection of the available inversin sequences revealed the presence of two conserved D-boxes: an N-terminal D-box at amino acids 490–498, within the ankyrin repeat domain (D-box #1); and a C-terminal D-box spanning amino acids 907–915 (D-box #2; see Fig. 1). We tested the importance of the D-boxes for inversin–Apc2 interaction by using a directed yeast two-hybrid assay with an activation domain-linked Apc2 prey and baits of normal or D-box-modified inversin linked to a Gal4 binding domain. Modification of individual D-boxes involved sequential site-directed mutagenesis to convert both end residues of the D-box to alanine, a strategy known to abrogate the functions of D-boxes in other proteins (23–25). The net effect of modifying D-box #1 was to disrupt inversin–Apc2 interaction but the equivalent modification in D-box #2 did not abolish the interaction (Table 1). The differential D-box effects may relate to their respective amino acid environments: the sequence encompassing D-box #1 has been extremely highly conserved during evolution, whereas that of D-box #2 is much less conserved (Fig. 1).
Additional verification of the inversin–Apc2 interaction was obtained by co-immunoprecipitation studies. Specific polyclonal antiserum to inversin (19) and human APC2 were used to immunoprecipitate protein complexes from 14.5 day post coitum wild-type mouse embryo extracts. Western blotting of the immunoprecipitates with the reciprocal antibody resulted in bands of the expected molecular size (Fig. 2).
Inversin displays a dynamic expression pattern through the cell cycle and is strongly expressed in primary cilia but is not essential for ciliogenesis
The interaction with Apc2 suggested that inversin may have some involvement in the cell cycle. To investigate this further we sought to determine the intracellular localization of inversin at different stages in the cell cycle. Using a rabbit polyclonal anti-inversin antibody (19), we examined inversin expression by immunofluorescence staining of cultured MDCK-II cells, a canine kidney cell line. A generalized nuclear staining was evident prior to nuclear envelope breakdown, whereupon a dynamic pattern of expression was clear at different cell cycle stages (Fig. 3). Cells in early prophase showed strong expression at the centrosomes (Fig. 3A) but in metaphase and anaphase inversin localized to the spindle poles (Fig. 3B and C). In cells at late telophase undergoing cytokinesis (Fig. 3D) expression was observed at the midbody, a region of microtubule overlap. The observed dynamic expression of inversin during the cell cycle lends further credence to the interaction between inversin and Apc2 which has also been reported to be specifically expressed in the centrosome (26).
Given that primary cilia provide a link between left–right axis specification and cystic kidneys (see Introduction) we also sought evidence for expression of inversin in the primary cilia of renal cells. When cultured at high cell densities, the mouse inner medullary collecting duct cell line (mIMCD-3) produces primary cilia. Using an inversin-specific antibody we were able to show strong inversin expression in these cilia (Fig. 4A and D). Double staining experiments using an acetylated tubulin antibody as a reference marker for primary cilia (Fig. 4A–H) showed that all cilia expressed the inversin protein. Despite extensive co-localization of inversin and acetylated tubulin (yellow signal in Fig. 4C and F–H), inversin staining appeared to be more pronounced in the varicosities, swellings observed in the cilia. Videomicroscopic studies have shown that these ciliary swellings move along the cilium shaft (27).
Although inversin is strongly expressed in renal primary cilia varicosities, our scanning electron microscopy studies have shown that the length and gross morphology of renal cilia from homozygous inv mice appear normal (Fig. 4I and J). Inversin does not therefore appear to be required for structural assembly of the cilia. Instead, it may play a role in the recently proposed function of primary cilia as sensory organelles (28).
The link between inversin, the cell cycle and cyst formation
There are a number of factors in cyst formation including increased proliferation and hypersecretion. Increased proliferation is an early event in the process as the proliferative index is increased in non-cystic portions of renal tubules taken from patients with both autosomal dominant and autosomal recessive polycystic kidney disease (29). Although 95% of patients with ADPKD have a dominant germline mutation in either PKD1 or PKD2, cyst formation appears to be recessive at the cellular level (15,16). The observation that loss of heterozygosity for PKD1 or PKD2 is evident in 30–40% of cysts, has raised the possibility of mutations in a number of other genes encoding proteins which interact with polycystin 1 and 2 (16,30,31).
The possibility that inversin acts in the polycystin pathway has been strengthened by the recent finding that a Pkd2 knockout mouse shows randomized laterality as well as polycystic kidneys (3) and that polycystin 2, a calcium permeable ion channel, localizes to the primary cilia of MDCK cells (11). Our finding that inversin, like the proteins of other polycystic mouse models, also localizes to renal primary cilia raises the possibility that these cilia serve to regulate renal tubular cell proliferation.
Having long been dismissed as rudimentary structures, primary cilia are now the focus of much interest. These rod-like structures arise from centrosomes and are resorbed during mitosis, to be reassembled through interphase. Given the pivotal role of centrosomes in cell spindle formation, the cilia–centrosome link has been postulated to offer an ideal system in which cell division occurs in response to the external environment (32,33). Praetorius and Spring (28) have found that bending of primary cilia (by increased flow rate or suction) results in an influx of extracellular calcium followed by calcium release from inositol-1,4,5-triphosphate stores and subsequent spreading of the calcium signal to neighbouring cells. This culminates in profound hyperpolarization of the cell, and on the basis of these observations primary cilia have been proposed to act as sensors (28).
An increase in the rate of cell proliferation is a key event in the cystic process. Inversin interacts with Apc2, a component of the anaphase promoting complex and shows a dynamic expression pattern in mitotic cells. Inversin also interacts with calmodulin in a calcium-regulated manner and both inversin and polycystin 2, a calcium channel protein, localize to the primary cilium of renal tubular cells. We speculate that the link between inversin, polycystins 1 and 2, the cell cycle and calcium is that primary cilia serve as an environmental sensor for the centrosome, thereby regulating the cell cycle.
Yeast two-hybrid assays and site-directed mutagenesis
As detailed in (19), a yeast two-hybrid screen was conducted using an inversin cDNA-containing bait plasmid to identify interacting proteins from a 7 day post coitum (dpc) mouse cDNA library. After identifying positive interaction with Apc2, follow-up yeast two-hybrid assays involved directed screening for interaction between an activation domain-linked Apc2 prey and various Gal4 binding domain-linked inversin baits, some of which had been subject to site-directed mutagenesis of one or both D-boxes.
The same inversin bait plasmid as in the initial yeast two-hybrid screen was subjected to sequential rounds of mutagenesis using the Quick-Change mutagenesis kit (Stratagene) to create double mutations in the sequences specifying individual D-boxes. The mutagenesis design was to change the N-terminal conserved arginine and C-terminal conserved asparagine of each D-box to alanine in both cases, as has been described previously for other D-boxes (23–25). The oligonucleotides used to create the mutants are listed with the mutated bases shown in italics and by underlining:
D-box #1 (R→A): GACAAAGAGGGAGCCACAGCTTTGCAC
D-box #1 (N→A): GCACTGGTCCTGCGCCAATGGCTACCT
D-box #2 (R→A): AGGGAACGAAGCGCGAAAGAGCTGTTT
D-box #2 (N→A): TTTCGACGGAAAGCCAAGGCAGCGGCG
To generate a plasmid with both D-boxes mutated a clone containing mutated D-box #2 was digested with NheI and SalI and the 750 bp fragment released was ligated into a plasmid containing mutated D-box #1.
Soluble protein extracts were prepared by homogenizing a single 14.5 dpc mouse embryo in 1 ml IP buffer (50 mm Tris, pH 7.4, 150 mm NaCl, 1 mm EGTA, 1% NP-40 and 1× protease inhibitor cocktail). The soluble cell extracts were obtained by centrifugation at 15 000 g for 10 min at 4°C and the protein concentration was determined by the Bio-Rad DC protein colorimetric assay using bovine serum albumin as standard. Co-immunoprecipitation was performed as described previously (19). Two co-immunoprecipitation reactions were performed separately using an inversin antibody (19) and a commercially available rabbit polyclonal anti-Apc2 antibody (34,35). Western blotting was then performed with the corresponding antibody.
mIMCD-3 cells were grown to confluence on Lab-Tek® chamber slidesTM (Nalge Nunc International). Cells were fixed in 3% PFA for 10 minutes, permeabilized with 0.5% Triton X-100 for 10 minutes and then blocked with 5% BSA in PBS for 5 minutes. Blocked cells were incubated with rabbit polyclonal anti-inversin (1 : 100) and mouse monoclonal anti-acetylated tubulin (1 : 1000) (Sigma) diluted in blocking buffer overnight at room temperature. Cells were washed 3 times with PBS and then incubated with fluorescein conjugated goat anti-rabbit IgG (Vector Laboratories) and TRITC conjugated rabbit anti-mouse (Dako) diluted 1 : 100 in blocking buffer overnight at room temperature. Cells were washed 3 times in PBS and mounted using Vectashield® (Vector laboratories). The images were captured with a Zeiss LSM 510 confocal microscope using FITC and TRITC filter sets with sequential scanning. MDCK-II cells were grown on flat bottomed microplates for two weeks then fixed with 3% PFA for 10 minutes. Cells were permeabilized with 0.1% Triton X-100 for 15 minutes and then blocked with 3% horse serum in PBS for 15 minutes. Blocked cells were incubated with rabbit polyclonal anti-inversin (1 : 100) for 2 hours at room temperature then overnight at 4°C. Cells were washed with PBS, blocked for 15 minutes with 3% goat serum and then incubated with goat anti-rabbit IgG FITC (Sigma) (1 : 50) for 1 hour at room temperature. Cells were washed with PBS then incubated for 2 minutes only with TRITC conjugated wheat germ agglutinin (Vector Laboratories) (1 : 10) for use as a plasma membrane marker. Images were captured with a Leica TCS-NT confocal microscope using FITC and TRITC filter sets with sequential scanning.
Scanning electron microscopy
Kidneys were removed from 5 day old inv homozygote mice and wild-type littermates and immediately fixed in 2% glutaraldehyde in Sorenson's buffer pH7 overnight at 4°C. The kidneys were ethanol-dehydrated, placed in liquid nitrogen and then fractured on pre-chilled foil using forceps to expose the cilia. The fractured kidneys were washed sequentially in 90%, 95% and 100% ethanol for 1 hour each. The tissue was air dried using liquid CO2 and covered in gold. Tissue morphology was captured using a Cambridge S240 scanning electron microscope.
We would like to thank Trevor Booth (University of Newcastle-upon-Tyne) for his help with the scanning EM images. This work was supported by the Biotechnology and Biological Sciences Research Council, the British Heart Foundation, the Lily Ross Fund, the Medical Research Council and the National Kidney Research Fund.
To whom correspondence should be addressed at: Institute of Human Genetics, University of Newcastle upon Tyne, International Centre for Life, Central Parkway, Newcastle upon Tyne. NE1 3BZ, UK. Tel.: +44 1912418616; Fax: +44 1912418699; Email: firstname.lastname@example.org
The authors wish it to be known that, in their opinion, these two authors should be considered as joint First Authors.
|Inversin construct||Interaction with Apc2|
|Wild type inversin||+|
|Mutated D-box #1||−|
|Mutated D-box #2||+|
|Mutated D-box #1 and #2||−|
|Inversin construct||Interaction with Apc2|
|Wild type inversin||+|
|Mutated D-box #1||−|
|Mutated D-box #2||+|
|Mutated D-box #1 and #2||−|