Loss of Polycystin-1 Inhibits Bicc1 Expression during Mouse Development

Bicc1 is a mouse homologue of Drosophila Bicaudal-C (dBic-C), which encodes an RNA-binding protein. Orthologs of dBic-C have been identified in many species, from C. elegans to humans. Bicc1-mutant mice exhibit a cystic phenotype in the kidney that is very similar to human polycystic kidney disease. Even though many studies have explored the gene characteristics and its functions in multiple species, the developmental profile of the Bicc1 gene product (Bicc1) in mammal has not yet been completely characterized. To this end, we generated a polyclonal antibody against Bicc1 and examined its spatial and temporal expression patterns during mouse embryogenesis and organogenesis. Our results demonstrated that Bicc1 starts to be expressed in the neural tube as early as embryonic day (E) 8.5 and is widely expressed in epithelial derivatives including the gut and hepatic cells at E10.5, and the pulmonary bronchi at E11.5. In mouse kidney development, Bicc1 appears in the early ureteric bud and mesonephric tubules at E11.5 and is also expressed in the metanephros at the same stage. During postnatal kidney development, Bicc1 expression gradually expands from the cortical to the medullary and papillary regions, and it is highly expressed in the proximal tubules. In addition, we discovered that loss of the Pkd1 gene product, polycystin-1 (PC1), whose mutation causes human autosomal dominant polycystic kidney disease (ADPKD), downregulates Bicc1 expression in vitro and in vivo. Our findings demonstrate that Bicc1 is developmentally regulated and reveal a new molecular link between Bicc1 and Pkd1.


Introduction
Studies of animal mutant models for polycystic kidney disease (PKD) and of human PKD patients have identified more than 20 genes whose mutations can induce PKD phenotypes [1,2,3]. The continued study of these cyst-associated genes and their products will help elucidate the disease mechanism of human inherited polycystic kidney diseases, such as autosomal dominant and recessive PKD (ADPKD and ARPKD).
Bicc1 is a mouse homologue of Drosophila Bicaudal-C (dBic-C), the orthologs of which are much conserved in many species, from C. elegans to humans [4,5,6,7,8]. Loss of dBic-C in Drosophila disrupts the direction of anterior follicle cell migration and affects anteriorposterior patterning, so that the resulting embryos lack heads and exhibit duplicated posterior segments instead [9,10]. The Xenopus homologue of Bicaudal-C (xBic-C) is one of the few molecules that, when microinjected ectopically, causes endoderm formation in the absence of mesoderm induction [11]. Knockdown of the zebrafish homologue of Bicaudal-C (zBic-C) induces cystic kidneys in vivo [12].
The gene product of Bicc1 (Bicc1) contains several conserved Nterminal KH domains and a conserved C-terminal SAM domain [13]. The KH domains bind target mRNAs [4]. Recent studies indicated that the KH domains enable Bicc1 to recruit specific miRNA precursors and associate with Dicer, to guide these nascent miRNAs to anchor their target mRNAs. The SAM domain is nonessential for mRNA binding, but it is required for the transfer of Bicc1-targeted mRNAs to P-body-associated AGO proteins for silencing [14]. Therefore, the gene product Bicc1 is thought to be an RNA-binding molecule that functions to regulate diverse proteins at the post-transcriptional level.
To study the functional role of Bicc1, some chemically-induced or natively occurring Bicc1-mutant mouse models (jcpk and bpk) have been identified and several gene-targeting mouse models have been established [5,6,15,16,17]. Although Bicc1 mutations in these models result from different mutant Bicc1 proteins, all the models exhibit cystic phenotypes in the kidney that are very similar to human polycystic kidney disease. These mouse models can provide insight into the functional roles of Bicc1 during mouse development. Even though gene expression of Bicc1 during mouse development has been previously reported [5,6,9,14], the developmental profiles of Bicc1 protein during mammalian development remain uncharacterized. Here we generated a polyclonal antibody against Bicc1 and used it to examine the distribution patterns of Bicc1 during mouse embryogenesis and organogenesis. In addition, we investigated the molecular relationship of Bicc1 to other human cystoproteins and discovered that loss of polycystin-1 (PC1), the gene product of Pkd1, whose mutation causes human ADPKD, reduced Bicc1 expression in vitro and in vivo. Our findings demonstrate that Bicc1 is developmentally regulated and indicate that its normal function may require normal PC1 expression.

Antibodies and reagents
A pGEX3 GST expression vector (Amersham) was used as the backbone for producing Bicc1 GST-fusion proteins. The cDNA of mouse Bicc1 (Q711-D858) was inserted in-frame into the vector. The mBicc1-fused plasmid was used to transform Rosetta host cells (Novagen) to produce the GST-fusion antigen. The antigen was subcutaneously injected into New Zealand white rabbits, at 0.5 mg per injection. The anti-mBicc1 serum (mBicc1p) was generated as reported in our previous studies [18]. All the antisera product and the preimmune serum (Pre-IM) were affinity-purified before performing experiments.

Mouse strains
Our Pkd2, Pkd1, and Pkhd1 mutant mice were described in detail previously [19,20,21]. All the mouse models used in this study were backcrossed (over 10 times) to the C57BL/6 inbred background. The animal protocol was approved by Vanderbilt University Institutional Animal Care and Use Committee (Permit Number: M/12/143).

Western blotting and quantitative PCR
Western blot analyses were performed using protocols similar to those described previously [18,22]. Briefly, proteins from cultured cells or tissues were extracted in lysis buffer (0.5% NP-40, 5% Sodium deoxycholate, 50 mM NaCl, 10 mM Tris-HCl (pH 7.5), 1% BSA), homogenized and centrifuged. Protein samples were solubilized in protein loading buffer and denatured by boiling. The samples were electrophoresed in 10% SDS-PAGE gels. The membranes were incubated with 5% milk at room temperature for one hour and blotted with mBicc1p antibody at room temperature for 4 hours and then were incubated with peroxidase-conjugate secondary antibodies (Sigma) and detected with enhanced chemiluminescence (ECL) (Amersham).
Quantitative PCR was performed using the iCycler iQ Real

Histology, immunofluorescence (IF), immunohistochemistry (IHC), and confocal microscopy
The detailed procedures used for histology, IF, and IHC staining were published previously [18,23]. For the microscopic analysis, a Zeiss Axioplan 2IE research microscope and a Zeiss Axiovert 200 inverted microscope system with 106, 206, and 406 objectives were used.

Renal epithelial cell lines and their cultures
Mouse inner medullary collecting duct (IMCD) cells were cultured using the conditions described in the American Type Culture Collection manual. IMCD cells with Bicc1-silencing (IMCD shC4C ) were reported in our previous study [24].
To generate Bicc1-overexpressed stable cell line (IMCD Bicc1 ), we initially used Bicc1-pCMV-Tag4 clone (Fig. 1B) as backbone to insert mouse Bicc1 ORF cDNA into LZRS-GFP vector (Addgene). Resulting LZRS-GFP-Bicc1-flag vector and pBABE-puro vector (Addgene) were co-transfected into HEK293 cells. At time points of cultured 48 and 72 hours, the viral-transfected supernatant was separately harvested and filtered with a 45 uM syringe filter. The 48 hour filtered supernatants were added on subconfluent cultured IMCD cells. After 24 hours, the infection was repeated with the 72 hour viral supernatants. One day later, puromycin was added for cell selection. Through a week Puromycin-selected culture, the remaining cells were resuspended and seeded on 100 mm culture plates (Costar) at a cell density of 1000 per plate. Once Puromycinselected colonies formed, single colonies were picked and expend the colony into 24-well plates (Costar). A cloned cell line with Bicc1-overexpression, IMCD Bicc1 , was identified by RT-PCR and confirmed by western blot analysis.
The cells from null-Pkd1 and its littermate-derived wildtype mice were generated by an approach similar to that reported in our previous studies [23,25]. Briefly, the kidneys from E16.5 Pkd1 2/2 and wildtype littermate embryos with the C57BL/6 congenic background were finely minced with a scalpel, and the minced tissue was incubated with 0.5% collagenase type IV at 37uC for 45 minutes and pipetted vigorously. The undigested tissue was removed by filtration through a 40 mm mesh filter. The remaining single cells and small organoids were washed three times with PBS containing 5 mM glucose. The cells were then incubated with 10 mg/ml biotinylated Dolichos biflorus agglutinin (DBA) (Vector, B-1035) at 4uC for 60 minutes. The cells were then washed again with PBS followed by incubation with 50 ml CELLectin Biotin binder Dynabeads (Dynal Biotech) at 4uC for 30 minutes. Since Dynabeads are superparamagnetic polystyrene beads, the incubated mixtures were then washed twice with PBS containing 5 mM glucose, using a magnetic rack. The cells were eluted with release buffer (Dynal Biotech) and were plated on 24well dishes with LHC-9 Medium (Gibco) under 5% CO 2 at 37uC overnight. The cells were then transferred to culture medium containing 5 units/ml murine IFN-c (Peprotech Inc.), which was changed every other day, and placed in a 33uC incubator for at least 10 cell passages. The culture medium was then switched to 10% FCS DMEM/F12 (1:1) (Gibco), and the cells were cultured for at least another 10 cell passages under the same culture conditions. Since the isolated collecting duct cells were not further cloned, the pool of cells (pool cells) was used for the current study. (D) Western analysis of duplicated protein lysates from wildtype, Bicc1-silenced (IMCD shC4C ), and Bicc1-overexpressed IMCD cells (IMCD Bicc1 ) were blotted with the mBicc1p antibody. Compared to the wildtype control, the immunoreactivity was significantly increased in Bicc1-overexpressed IMCD cells and was almost not detected in the Bicc1-silenced cells. (E) Normalized quantitative analysis of the densitometry values of the western analyses. Compared to wildtype IMCD and Bicc1-silenced IMCD (IMCD shC4C ) cells, Bicc1-overexpressed IMCD (IMCD Bicc1 ) cells showed significantly increased Bicc1 expression (*P,0.05). (F) Immunohistochemistry (IHC) staining of Bicc1 protein in the kidneys of E18.5 Bicc1 2/2 and its wildtype littermates. Positive staining (arrows) were showed in the wildtype kidney (a), while no obvious positive staining was seen in the E18.5 Pkd1 2/2 pool cells with PC1-CT expression were produced by pSico Lentiviral vector system (Addgene) in which the entire human PKD1 COOH-terminus was (V4102-R4278) constructed. Recombinant viruses were packaged and amplified on a large scale in HEK293 cells. Lentiviral particles were purified and infected Pkd1 2/2 pool cells following the introduction of the manufacturer.
Before being used in any cell-based assays, all the established cell lines were cultured under non-permissive conditions (37uC without IFN-c) for at least three days to turn off the SV40 transgene activity.

Statistics
All biochemical assays were repeated at least three times. Statistical analysis was performed where appropriate using the Student's t-test or one-way analysis of variance (ANOVA) followed by Tukey's Multiple Comparison Test. Differences with P-values ,0.05 were considered statistically significant.

Generation and characterization of anti-Bicc1 antiserum
A purified mouse Bicc1-GST fusion protein (mBicc1), encoding residues Q711 to D858 of Bicc1, was used to produce a rabbit polyclonal antiserum (mBicc1p) against the Bicc1 gene product, Bicc1. The antiserum was generated and affinity-purified as described in our previous studies [18].
To test the specificity of mBicc1p, we performed western blot analyses of the antigen using the Pre-IM, mBicc1p, or an anti-GST antibody. No immunoreactive band was seen with Pre-IM, although strong positive bands of the expected size (,43 kD) were detected with mBicc1p and the anti-GST antibody (Fig. 1A). The mBicc1p-positive band could be competed away with the mBicc1-GST antigen (data not shown), suggesting that the mBicc1p antiserum was specific for Bicc1.
To further confirm the specificity, we constructed a Bicc1expression vector Bicc1-pCMV-Tag4, in which a Flag-tag was fused in-frame with the full-length ORF Bicc1 cDNA (Fig. 1B). Western analyses showed that the transient transfection of HEK293 cells with Bicc1-pCMV-Tag4 yielded bands of the expected size (,110 kD) that were immunoreactive with the mBicc1p and anti-Flag antibodies, but not with Pre-IM (Fig. 1C). We also used the mBicc1p antiserum to detect Bicc1 in lysates of IMCD cells, the stably Bicc1-overexpressed cell line IMCD Bicc1 and the Bicc1-silencing stable cell line IMCD shC4C [24]. The ,110-kD Bicc1 band was readily detected in wildtype IMCD cells and highly strong band was showed in the IMCD Bicc1 cells, whereas almost no band was seen in the IMCD shC4C cells (Fig. 1D-E).
In addition, renal tissues from Bicc1 knockout mice were also used for testing the specificity of mBicc1p antiserum. Strong immunohistochemistry (IHC) staining can be observed in E18.5 wildtype kidney, while there is no positive staining in its Bicc1 2/2 littermate (Fig. 1F). These results strongly indicated that the mBicc1p antiserum was specific for Bicc1.

Bicc1 expression during mouse development
To elucidate the patterns of Bicc1 protein expression during embryogenesis and primary organogenesis in mammals, mBicc1p was used for IHC analyses in developing mouse tissues. We found clear positive staining in the epithelial cells of the neural tube on embryonic day 8.5 (E8.5) (Fig. 2A), and then in the myocardial wall of the heart at E9.5 (Fig. 2B). By E10.5, the epithelial cells in the primordial gut and immature hepatocytes in the liver exhibited positive staining in their cytosol (Fig. 2C-D). By E11.5, Bicc1 expression was seen in the epithelia of the main bronchi, aorta, and early ureteric bud and mesonephric tubules, as the bud penetrated into the metanephrogenic mesenchyme (Fig. 2E-G). Clear positive staining continued to be observed in the renal comma-shaped body and the S-shaped body by E12.5 (Fig. 2H). At this stage, strong positive immunoreactivity could also be seen in the pancreatic primordium and posterior root ganglions (Fig. 2I-J). At E16.5, strong Bicc1 expression was observed in fasciculated cells in the developing adrenal cortex and in olfactory cells (Fig. 2K-L). By E16.5, the Bicc1 expression appeared to decrease in the cardiomyocytes and primordial gut (data not shown). The finding of Bicc1 protein profiling highly coincides with the studies by Bicc1 mRNA in situ hybridization [5,6,9].

Expression of Bicc1 in the postnatal mouse kidneys
Using the mBicc1p antibody, we next performed IHC analyses of the kidneys in newborn mice. Bicc1-positive staining was predominantly observed in the developing tubules of the kidney (Fig. 2M). By the age of 1 month, the Bicc1 expression was concentrated at the juxta-medullary region, with relatively weak expression in the cortical and medullary regions (Fig. 2N). At 3 months, although staining pattern was similar to that seen at 1 month, increased Bicc1 expression was seen over the cortical region of the kidney (Fig. 2O). At 6 months, the Bicc1 expression also extended into the medullary region (Fig. 2P). We also examined the kidneys of 9-and 12-month-old mice with the same antibody, but found no significant Bicc1 expression changes. The staining was mainly seen in the cytoplasm of the positive cells, suggesting that Bicc1 has a cytosolic distribution in vivo ( Fig. 2M-P). Collectively, these results showed that Bicc1 expression is regulated during renal development.

Bicc1 is expressed in multiple nephron segments of the adult mouse kidney
The nephron segments with positive mBicc1p staining were identified by costaining with Lotus tetragonolobus lectin (LTL) for the proximal tubules (PT), Tamm-Horsfall glycoprotein (THG) for the medullary thick ascending limb of Henle (LH) and distal tubules (DT) [20,26], and anti-Aquaporin-2 (AQP2) antibodies for the collecting ducts (CD). We examined the immunocytochemical distribution of Bicc1 in paraffin-embedded sections of wildtype adult kidneys. The Pre-IM from the same rabbit used to generate mBicc1p was used as the negative staining control (Fig. 3A). Under these conditions, Bicc1 showed diffuse cytoplasmic staining and was strongly co-expressed with the marker for epithelial cells of the proximal convoluted tubules (Fig. 3B), and distal tubules (Fig. 3C). Relatively weak-positive labeling was also seen in the collecting ducts of the adult kidneys (Fig. 3D). These results suggested that, although it is detected in all the nephronic segments, Bicc1 is highly expressed in the epithelial cells of the proximal convoluted tubules, which are derived from the metanephros, coincides with the previous report that this protein is co-expressed with proximal convoluted tubules of the kidney [14].
To further validate this finding, we have restored the entire human PC1 COOH-terminus (PC1-CT) into Pkd1 2/2 cells by lentivirus system to determine if re-expression of PC1 can rescue the downregulation of Bicc1 in the null-Pkd1 cells. The western blots of cell lysates from the littermate-derived cell lines: wildtype, and Pkd1 2/2 , as well as Pkd1 2/2 cells transfected with the PC1-pSico vector (Pkd1 2/2 +PC1-CT) showed that downregulation of Bicc1 of Pkd1 2/2 cells can be significantly rescued (Fig. 4F-G). This result strongly demonstrates that downregulation of Bicc1 expression in the Pkd1 deficient cells is due to loss of PC1.
Finally, we have also performed western analyses from tissue lysates of E14.5 null-Pkd2, null-Pkd1 and null-Pkhd1 embryos, and their wildtype-littermates as well. Compared to wildtype, Bicc1 is also significantly downregulated in the null-Pkd1 tissues. However, there are no any Bicc1 expression changes between the tissue lysates with and without Pkd2 and Pkhd1 (Fig. 5B-C). These in vivo results provide further evidence that the loss of Pkd1 downregulates Bicc1 expression.

Discussion
Studies using chemically-induced or natively occurring Bicc1mutant mouse models (jcpk and bpk) and other Bicc1-gene-targeted mouse models have recently demonstrated that the disruption of Bicc1 can induce polycystic kidney disease with phenotypes very similar to human ADPKD [5,6,17]. Although gene expression of Bicc1 during mouse development has been previously reported [5,6,9,14], the characteristics of Bicc1 gene product, Bicc1, during embryogenesis and organogenesis in mammals has not been completely explored. In this study, we generated a new polyclonal antibody, mBicc1p, against the Bicc1 gene product, Bicc1, and used it to study the expression of Bicc1 during mouse development. By using this antibody, we also discovered that the loss of Pkd1 downregulates Bicc1 expression in vitro and in vivo, revealing a molecular relationship between Bicc1 and Pkd1, a known causal gene for human ADPKD. Our findings demonstrate that Bicc1 expression is developmentally regulated and that its normal function may require PC1 expression.
To investigate the functional role of Bicc1 during mammalian development, several studies have examined mouse Bicc1 gene expression patterns using in situ mRNA hybridization [6,9]. By whole-mount in situ hybridization, Bicc1 mRNA expression is first detected at the rostral tip of the primitive streak around E7.5 and in neural tissues at E8.5. At E13.5, strong Bicc1 mRNA labeling is detected in the bone, heart, and lung tissues. In mouse kidney development, Bicc1 mRNA is detected at the mesonephros and the first branch of the ureteric bud tree at E11.5, and then at the metanephros and the comma-and S-shaped bodies at E13.5. At birth, Bicc1 mRNA is mainly seen in the proximal tubules of the Polycystin-1 Regulates Bicc1 Expression PLOS ONE | www.plosone.org mouse kidney. These Bicc1 mRNA patterns and tissue distributions highly correspond to our findings with anti-Bicc1 antibody staining, further demonstrating the specificity of the mBicc1p antibody for Bicc1, and supporting protein expression profile described in this study.
Primary cilia are found on diverse cell types ranging from fibroblasts to epithelia [27]. These cilia are generally thought to function as extracellular sensors, for regulating cell behavior [28]. Many human and rodent cystoproteins, the mutants of which cause PKD phenotypes, have been reported to co-localize with the primary cilium/basal bodies of renal epithelia [1,29]. However, previous reports indicate that Bicc1 was distributed to the P-bodies of renal epithelial cells [5,6,14]. This novel subcellular localization of Bicc1 indicated that the primary cilium/basal bodies in renal epithelial cells may not be the only targeted organelle for cystogenesis of the kidney.
We recently established stable IMCD (mouse inner medullary collecting duct) cell lines, in which Bicc1 is silenced by short hairpin Polycystin-1 Regulates Bicc1 Expression PLOS ONE | www.plosone.org RNA (shRNA) expression. Our previous results showed that the inhibition of Bicc1 disrupts normal tubulomorphogenesis and induces the cystogenesis of IMCD cells grown in three-dimensional culture [24]. These cells also have a significant defect in Ecadherin-based cell-cell adhesion, along with abnormalities in actin cytoskeletal organization, cell-extracellular matrix interactions, ciliogenesis, and cell proliferation/apoptosis. Interestingly, these defects are also seen in epithelial cells lacking Pkd1 and Pkd2, mutations that can cause human ADPKD [23,30,31]. We therefore assume that the cystogenesis resulting from the downregulation of Bicc1 may be associated with a disruption in the normal Pkd1 or Pkd2 expression.
As predicted, a recent study demonstrated that the absence of Bicc1 in cells promotes miR-17's binding to the 39UTR of Pkd2 and disrupts the stability of the Pkd2 mRNA [6]. This finding indicates that Bicc1 acts as a posttranscriptional factor upstream of Pkd2 and reveals the molecular relationship between Bicc1 and Pkd2, a causal gene of human ADPKD. In the present study, we showed that another human ADPKD causal gene, Pkd1, is involved in the regulation of Bicc1 expression in vitro and in vivo. This finding indicates that Bicc1 is not only associated with Pkd2, but also with Pkd1 expression. The association between Bicc1 and polycystins implies that a disruption of Bicc1 induces cystic phenotypes through the polycystin pathway.
In summary, we have newly generated a polyclonal antibody, mBicc1p, that is specific for the Bicc1 gene product Bicc1. Using this antibody, we demonstrated the developmental profile of Bicc1 protein during embryogenesis and organogenesis in mammals. We found that the Bicc1 protein is expressed as early as E8.5 in the mouse neural tube and appears at the ureteric bud and nephronic tubules by E11.5. After birth, Bicc1 expression extends into the proximal tubules of the kidney, and thereby is predominately expressed in this nephron segment. Moreover, we discovered that normal Bicc1 expression requires PC1 expression. These findings together indicate that Bicc1 is a key protein for embryogenesis and Figure 5. Loss of PC1 downregulates Bicc1 expression in vivo. (A) Positive mBicc1p IHC staining (arrows) in the mouse embryonic kidneys. The E14.5 wildtype (a) and its littermate Pkd1 2/2 (b) kidneys were stained by mBicc1p antibody (n$3). In comparisons of kidney tissues with and without Pkd1, the null-Pkd1 kidney showed significantly reduced Bicc1 expression. However, there are no Bicc1 expressional difference between the agematched wildtype (c and e) and their littermate Pkhd1 2/2 (d) or littermate Pkd2 2/2 (f) kidneys, respectively. ''cy'' = renal cysts; ''G'' = glomerulus. (B) Compared to its wildtype littermate, western blot of duplicated tissue lysates from E14.5 embryos showed that loss of PC1 markedly downregulates Bicc1 protein level (left panel). A similar western blot showed equal immunoreactivities between the tissue lysates from E14.5 Pkd2 2/2 and its wildtype littermate (middle panel) and between the E14.5 Pkhd1 2/2 and its littermate wildtype embryos (n$3). b-actin for protein loading control. (C) Normalized quantitative analysis of densitometry values of the western analyses. Compared tissues with and without Pkd1, Pkd2 or Pkhd1, only null-Pkd1 tissue shows significantly reduced Bicc1 expression (*P,0.05). Bars: 50 mm in A. doi:10.1371/journal.pone.0088816.g005 Polycystin-1 Regulates Bicc1 Expression PLOS ONE | www.plosone.org organogenesis in mammals and uncover a new molecular link between Bicc1 and Pkd1, whose mutation causes human ADPKD.