Autosomal dominant polycystic liver disease (PCLD) is caused by mutations of either PRKCSH or Sec63, two proteins associated with the endoplasmic reticulum (ER). Both proteins are involved in carbohydrate processing, folding and translocation of newly synthesized glycoproteins. It is postulated that defective quality control of proteins initiates endoplasmic reticulum-associated degradation (ERAD), which disrupts hepatic homeostasis in patients with PRKCSH or Sec63 mutations. However, the precise molecular mechanisms are not known. Here, we show that over-expression or depletion of PRKCSH in zebrafish embryos leads to pronephric cysts, abnormal body curvature and situs inversus. Identical phenotypic changes are induced by depletion or over-expression of TRPP2. Increased PRKCSH levels ameliorate developmental abnormalities caused by over-expressed TRPP2, whereas excess TRPP2 can compensate the loss PRKCSH, indicating that the proteins share a common signaling pathway. PRKCSH binds the C-terminal domain of TRPP2, and both proteins co-localize within the ER. Furthermore, PRKCSH interacts with Herp, and inhibits Herp-mediated ubiquitination of TRPP2. Our findings suggest that PRKCSH functions as a chaperone-like molecule, which prevents ERAD of TRPP2. Dysequilibrium between TRPP2 and PRKCSH may lead to cyst formation in PCLD patients with PRKCSH mutations, and thereby account for the overlapping manifestations observed in PCLD and autosomal dominant polycystic kidney disease.
Autosomal dominant polycystic liver disease (PCLD) is a rare hereditary disorder characterized by slowly progressive cyst formation confined to the liver. Epithelial cysts originate in the biliary duct probably at young age. However, the disease typically remains asymptomatic until the fifth to sixth decade of life, when the cysts disseminate throughout the liver, causing a mass effect that impinges on surrounding organs. PCLD is a heterogenetic disease; mutations in two different genes, encoding the human protein kinase C substrate 80K-H (PRKCSH) (1,2) or Sec63 (3), have been identified in PCLD patients, but at least one other locus appears to be involved in this disease (3). Both PRKCSH and Sec63 reside in the endoplasmic reticulum (ER). The ER localization of these two proteins is in striking contrast to the ciliary localization of most proteins mutated in polycystic kidney diseases. Hence, it has been postulated that mutations of either PRKCSH or Sec63 cause aberrant processing of proteins, and thus, PCLD has been categorized as a disorder of co-translational protein processing (4).
PRKCSH interacts with the catalytic α-subunit of glucosidase II (5), an enzyme that removes α1,3-linked glucose from newly synthesized glycoproteins to generate monoglucosylated core oligosaccharides that are recognized by chaperone proteins calnexin and calreticulin [reviewed in (6)]. Binding to these two ER chaperones is essential for protein maturation; aberrant proteins may undergo another round of glucosylation catalyzed by UDP-glucose:glycoprotein glucosyltransferase, or be selected for ER-associated degradation (ERAD). Glucosidase II later cleaves the remaining glucose residue to release the glycoprotein from its ER chaperones, facilitating the exit of proteins from the ER. PRKCSH is a multifunctional molecule. Originally identified as a weak substrate of protein kinase C, PRKCSH serves as a down-stream target of fibroblast growth factor receptor activation (7) and functions as a receptor for advanced glycosylation end-products (AGE) (8). Although a C-terminal HDEL sequence should retain PRKCSH in the ER, partial proteolysis into a truncated form enables the protein to leave the ER and traffic to the plasma membrane, intracellular vesicles and nucleus (9,10). The role of PRKCSH in PCLD is poorly understood. Epithelial cells lining the cysts in patients with PCLD express little if any PRKCSH (11), consistent with a two-hit process whereby an inherited germline mutation is followed by the somatic inactivation of the remaining functional PRKCSH gene. Lack of functional glucosidase II is postulated to impede the processing of newly synthesized glycoproteins, leading to retro-translocation and cytosolic ERAD of misfolded proteins. ER receptors initiate the unfolded protein response to restore ER homeostasis, but apoptotic cell death will prevail if the adaptive response fails [reviewed in (12)]. But this hypothesis does not address how the apoptotic death of very few PRKCSH-deficient liver cells leads to cystic disease, especially in an organ of such tremendous regenerative capacity. It is also unclear why patients with autosomal dominant polycystic kidney disease (ADPKD) often develop liver cysts, and PCLD patients share some of the extrarenal manifestations observed in ADPKD (13). In ADPKD, mutations of either PKD1 or PKD2, encoding for polycystin-1 and TRPP2, are responsible for progressive cyst formation in the kidney. TRPP2 contains a carboxy-terminal ER retention motif that interacts with PACS-2, a connector protein of the secretory pathway. PACS-2 couples TRPP2 to COPII and retrieves TRPP2 from the cis Golgi back to the ER (14). Interaction with the ubiquitin-like protein Herp has recently been reported to target TRPP2 for ERAD-dependent degradation (15).
We report now that the over-expression or knockdown of PRKCSH in zebrafish embryos causes developmental changes that are indistinguishable from those induced by the over-expression or depletion of zebrafish TRPP2. Concomitant expression of TRPP2 rescues developmental defects caused by a lack of PRKCSH, and conversely, co-expression of PRKCSH rescues the phenotypic changes caused by ectopic TRPP2. Interaction of PRKCSH with Herp antagonizes the Herp-mediated ubiquitination of TRPP2, suggesting that a precise balance between both proteins is required to maintain normal tissue homeostasis.
Depletion of PRKCSH in zebrafish embryos results in pronephric cysts
To investigate the role for PRKCSH in cystic disease, we targeted endogenous zebrafish PRKCSH using a morpholino antisense oligonucleotide (MO) directed against the translational start codon. The PRKCSH target sequence fused to GFP demonstrated that this MO efficiently interferes with PRKCSH mRNA translation (Supplementary Material, Fig. 1). At 55 hours post fertilization (hpf), PRKCSH-deficient embryos showed dysmorphic changes and dorsal curvature of the anterior–posterior body axis reminiscent of other gene mutations associated with pronephric cysts (16) (Fig. 1A). Histochemistry of the zebrafish pronephros with knockdown of PRKCSH confirmed the formation of pronephric cysts (Fig. 1B). Interestingly, over-expression of PRKCSH resulted in phenotypic changes nearly indistinguishable from the changes caused by the loss of PRKCSH (Fig. 1A and B). Abnormal PRKCSH levels disturbed the normal left-right (L-R) asymmetry of the body axis, causing situs inversus (Fig. 1C, Supplementary Material, Fig. 2). Furthermore, injection of either PRKCSH MO or mRNA resulted in the dose-dependent appearance of dysmorphic embryos with situs inversus and pronephric cysts (Fig. 2). These findings suggest that normal PRKCSH levels are crucial for normal zebrafish pronephros development.
Genetic interaction between PRKCSH and TRPP2
The phenotype following knockdown of PRKCSH was remarkably similar to that observed after depletion of TRPP2 (Supplementary Material, Fig. 3). TRPP2 is a member of the TRP family of calcium-permeable cation channels that is mutated in ∼15% of patients with ADPKD (17). We previously demonstrated that imbalance of TRPP2 levels, following MO-mediated depletion or over-expression, results in defective body axis, pronephric cyst formation and situs inversus (18). Accordingly, we investigated whether TRPP2 and PRKCSH genetically interact. Co-injection of PRKCSH mRNA (8 and 16 pg) partially reversed the abnormalities caused by over-expression of TRPP2, including dysmorphic changes, pronephric cysts and heart situs (Fig. 3A–C). Interestingly, abnormalities caused by knockdown of PRKCSH were also ameliorated by co-injection of TRPP2 mRNA (Fig. 3D–F). By partially rescuing the phenotype of over-expressed TRPP2, PRKCSH appears to function as a chaperone in a buffering capacity, curtailing the consequences of too much TRPP2. MO-mediated depletion of one protein augmented the dysmorphic changes and cyst formation caused by MO-mediated depletion of the other protein, further supporting a genetic interaction between TRPP2 and PRKCSH (Fig. 4). In addition, co-injection of PRKCSH mRNA rescued the defects caused by the MO-mediated depletion of TRPP2 (Fig. 5). These observations suggest that a balance between these proteins is crucial for normal pronephros development.
Physical interaction between PRKCSH and TRPP2
PRKCSH was recently reported to interact with the carboxy-terminal domain of TRPV5 (19); we therefore tested for interaction between PRKCSH and TRPP2. TRPP2 co-immunoprecipitated with PRKCSH in transiently transfected HEK 293T cells, but not with a control protein (F.GFP) (Fig. 6A). The interacting region in TRPP2 could be mapped to the carboxy-terminal domain: this domain fused to an artificial integral-membrane protein sIg7 consisting of the CD5-leader sequence, constant domains of human IgG and the transmembrane domain of CD7 (sIg7.TRPP2-C) (20) immunoprecipitated PRKCSH. In contrast, a similar fusion, containing the carboxy-terminal domain of Nephrin (sIg7.Nephrin-C) failed to immobilize PRKCSH (Fig. 6B). We observed selective interaction of PRKCSH with other TRP channels; PRKCSH associated with TRPV1, but only very weakly with TRPV4 and TRPC4, and not with TRPC6 (Fig. 6C and D). PRKCSH and TRPV5 have been reported to interact at the plasma membrane (19); however, co-staining with the ER-specific marker protein BAP31 revealed that both TRPP2 and PRKCSH are retained in the ER (Supplementary Material, Fig. 4A and B). In addition, TRPP2 co-localized with endogenous PRKCSH in HEK 293T cells (Supplementary Material, Fig. 4C).
PRKCSH protects TRPP2 against Herp-dependent ubiquitination and ERAD
To further elucidate the functional relationship between PRKCSH and TRPP2, we depleted endogenous PRKCSH by RNA interference in Madin–Darby canine kidney (MDCK) cells, a well-characterized tubular epithelial cell line. To avoid the potential pitfalls of clonal selection, we used lentiviral-based system to generate a polyclonal cell line with tetracycline-inducible knockdown of PRKCSH. Following 4–6 days of tetracycline treatment, levels of PRKCSH were undetectable (Supplementary Material, Fig. 5C). Noticeably, the decline of PRKCSH proteins levels resulted in a reproducible decrease in TRPP2 levels (Fig. 7A). Tetracycline treatment of control cells, expressing an shRNA directed against luciferase, did not affect TRPP2 levels (Fig. 7A). Herp has recently been shown to target TRPP2 for ERAD (15). Whereas Herp reduced the steady-state levels of TRPP2, co-expression of PRKCSH normalized TRPP2 levels (Fig. 7B). Co-immunoprecipitation revealed that Herp interacts with both TRPP2 and PRKCSH (Fig. 7C). However, in the presence of PRKCSH, Herp precipitated less TRPP2 (Fig. 8A). Similar results were obtained with an HA-tagged version of TRPP2 (data not shown), suggesting that PRKCSH restricts access of Herp to TRPP2, and thereby reduces Herp's ability to target TRPP2 for ubiquitin-dependent degradation. The proteasome inhibitor MG132 increased TRPP2 steady-state levels in MDCK cells after tetracycline-dependent knockdown of PRKCSH, suggesting that loss of PRKCSH increases the ubiquitin-dependent degradation of TRPP2 (Fig. 8B). As recently shown (15), co-expression of TRPP2 with Herp strongly increased the levels of ubiquitinated TRPP2 (Fig. 8C). The presence of PRKCSH nearly abrogated the Herp-mediated increase in TRPP2 ubiquitination, whereas PRKCSH.1338, a truncation derived from a patient with PCLD, had no effect on the Herp-mediated degradation. These findings support the hypothesis that intact PRKCSH protects TRPP2 against Herp-mediated degradation.
Autosomal dominant PCLD is a rare genetic disorder that arises from mutations in at least two gene products, PRKCSH and Sec63. The identification of these two proteins as the underlying cause for PCLD was unexpected, since most proteins involved in cyst formation localize to the primary non-motile cilium, a microtubular organelle attached to most body cells (21,22). Notably, PCLD and autosomal polycystic kidney disease (ADPKD) share common extrarenal manifestations, such as cerebral aneurysms and cardiac valve abnormalities. In addition, liver cysts often appear in patients with ADPKD. The many similarities between the two disorders are highly suggestive of a coinciding signaling cascade, although a shared pathway has not been identified.
Resident in the ER, PRKCSH encodes the non-catalytic β-subunit of glucosidase II. As a mannose 6-phosphate receptor-like subunit, PRKCSH facilitates the positioning the (1,3)α glucose for enzymatic cleavage. Removal of a glucose residue from the core N-glycan Glucosidase II creates a monoglucosylated binding site for the ER chaperone calnexin, an important step for subsequent protein folding. Incorrectly folded proteins are destined for another round of glucosylation or are targeted for ERAD. TRPP2 is an ERAD substrate; the ubiquitin domain-containing molecule Herp recognizes TRPP2, and facilitates the retro-translocation and ubiquitination of TRPP2 (15). Our findings demonstrate that PRKCSH is able to interfere with Herp-mediated TRPP2 degradation. The domain organization of PRKCSH shares the PRKCSH domain and several predicted coiled-coils with OS-9 (Supplementary Material, Fig. 6), another ER-associated protein, and both PRKCSH and OS-9 bind select members of the TRP family of ion channels (19,23). OS-9 preferentially binds to TRPV4 monomers, and acts as a chaperone-like molecule that appears to prevent ubiquitin-mediated degradation of newly synthesized TRPV4 during its assembly into tetrameric ion channel complexes. PRKCSH may act in an analogous manner, protecting TRPP2 assembly against premature Herp-mediated ERAD. In MDCK cells, expressing moderate levels of endogenous TRPP2, an interaction with either Herp or PRKCSH was not detectable, suggesting that only small amounts of misfolded TRPP2 associate with Herp and/or PRKCSH.
Current theory proposes that mutations of PRKCSH interfere with glucosidase II activity, preventing correct folding of essential proteins required for liver and bile duct homeostasis. This concept implies that ERAD-associated apoptosis causes cyst formation; however, it does not explain how the apoptotic loss of single PRKCSH-deficient cells leads to clonal proliferation and cyst formation (24). Our data are consistent with an alternative mechanism, whereby PRKCSH acts as a chaperone-like molecule, blocking the Herp/ERAD-mediated degradation of TRPP2. Mutations and loss of PRKCSH may compromise the ability of liver cells to assemble and maintain functional TRPP2 complexes. Since Herp/PRKCSH regulate TRPP2 at the ER level, distal TRPP2 localizations such as the plasma membrane or the cilium are probably affected by a dysfunctional PRKCSH. Active TRPP2 ion channels may not only be required during hepatic development, but also play a role during adaptive and regenerative responses. Thus, slowly progressive cystic liver disease, characteristic of PCLD, may result from inadequate TRPP2 levels, due to the failure of PRKCSH to protect TRPP2 against premature degradation in the ER, and explain the partially overlapping clinical manifestations between patients with PCLD and ADPKD.
MATERIALS AND METHODS
Reagents and plasmids
The following antibodies were used: anti-PRKCSH monoclonal antibody (BD Bioscience), TRPP2 antibody (nanoTools, Germany), anti-FLAG M2 and anti-actin (Sigma-Aldrich), anti-V5 (Serotec), anti-GFP (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and anti-HA 12CA5 (Roche Applied Science). Secondary horseradish peroxidase (HRP)-coupled antibodies against rabbit and mouse IgG were from Dako and GE Healthcare. Cy3-and Cy5-conjugated antibodies were from Jackson ImmunoResearch. The MF20 antibody (Developmental Studies Hybridoma Bank, Iowa) was used to stain cardiac sarcomeric myosin in zebrafish. MG132 (Calbiochem) was used at 5 µg/ml. The coding sequence of human PRKCSH protein was amplified from a human liver cDNA library. PRKCSH with N-terminal V5-tag or FLAG-tag was constructed in the pcDNA6 vector. The plasmids TRPV4-V5/His, TRPC4-V5/His, TRPC6-V5/His, TRPV1-V5/His, TRPP2-V5/His, slg7.TRPP2-C were previously described (20,23,25). A plasmid containing the full-length human Herp cDNA with an N-terminal c-Myc tag and a C-terminal FLAG tag (m.Herp.F) was kindly provided by X.Z. Chen. The HA-ubiquitin plasmid was kindly provided from D. Bohmann. The F9.PRKCSH.1338 truncation was generated according to the 1338–2A→G mutation of one of the PCLD families, using the following primer set (forward, reverse): 5′- CGCACGCGTATGCTGTTGCCGCTGCT GCTGCTG-3′; 5′- ATTGCGGCCGCTAGCCAAGGCTGGTGGGAGAGC-3′ (the reverse primer contains a stop codon at 1338 bp, followed by a NotI site).
Cell cultures and viral gene transfer
Human embryonic kidney (HEK 293T), HeLa cells and MDCK cells were grown in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum. Transient transfections were carried out using the calcium phosphate method or with FuGENE 6 reagent (Roche Applied Science). A lentiviral system was used to generate a tetracycline-inducible knockdown of PRKCSH in MDCK cells as recently described (26). Briefly, MDCK cells were first transduced with lentivirus encoding the tetracycline sensitive tTR-KRAB repressor and a dsRed reporter. A second transduction step followed with lentivirus encoding an shRNA against PRKCSH (sequence: 5′-AAGTTCAGCGCCATGAAGTAC-3′) or luciferase (sequence: 5′-CGTACGCGGAATACTTCGA-3′), both under the control of tTR-KRAB.
Co-immunoprecipitation and ubiquitination assays
Thirty-six hours after transfection, HEK 293T cells were lysed in immunoprecipitation buffer (1% Triton X-100, 1% sodium deoxycholate, 150 mm NaCl, 50 mm Tris, pH 8.0) supplemented with protease inhibitor mixture (Roche Applied Science). Lysates were cleared by centrifugation at 15 000g for 15 min at 4°C and the supernatants were incubated with 30 µl of anti-FLAG-M2 beads for 3 h. For co-immunoprecipitation of the slg7-TRPP2 C-terminus with PRKCSH, lysates were incubated with 30 µl of Protein G-Sepharose beads for 3 h. The beads were washed extensively with buffer, and the retained proteins were analyzed by western blot analysis. To detect ubiquinated proteins, transfected HEK 293T were lysed in buffer A (8 m urea, 100 mm NaH2PO4, 1% Triton X-100, 10 mm Tris, pH 8.0) at room temperature. The supernatant, obtained after centrifugation of the lysates at 75 000g, was incubated with Ni2+-nitrilotriacetic acid-agarose (Qiagen) for 1 h. The beads were washed twice with buffer A and twice with buffer B (same as A, except 0.5% Triton X-100, pH 6.3). Bound proteins were eluted with buffer C (same as A, except 0.1% Triton X-100 and pH 4.5).
Cells were fixed in 3.7% paraformaldehyde in PBS for 10 min, permeabilized with 0.05% Triton X-100 and blocked in PBS containing 1% horse serum. Immunostainings were performed sequentially with appropriate primary antibodies and fluorescently labeled secondary antibodies. All images were obtained with an LSM 510 confocal microscope (Zeiss, Germany).
Zebrafish lines and manipulations
Wild-type AB, ABTL or TLEK zebrafish were maintained and raised as described (18). Dechorionated embryos were kept at 28.5°C in Danieau's solution with or without 0.003% PTU (1-Phenyl-2-thiourea, Sigma) to suppress pigmentation and staged according to somite number (som), or hpf. Morpholino antisense oligonucleotide (MO) and mRNA injections were performed as previously described (18). Briefly, wild-type embryos at the 1- to 2-cell stage were microinjected with MO (Gene Tools LLC), or in vitro transcribed capped RNA (mMessage Machine kit, Ambion), diluted in 200 mm KCL, 0.1% Phenol Red and 10 mm HEPES, pH 7.5. The MO for PRKCSH (PRKCSH MO) was designed to target the translational start site (5′-TCAACAATAGATGCACGCAGGTCAT-3′). To synthesize capped RNA, human wild-type PRKCSH was subcloned into pCS2+ MXN, linearized with Nsi1, and transcribed with SP6 polymerase. The vector pSGEM with human TRPP2 was linearized with Pac1; T7 polymerase was used for synthesis of capped RNA. Rescue experiments were performed by co-injecting 368 pg of human TRPP2 mRNA with the human PRKCSH mRNA, or by co-injecting 5 pmol PRKCSH MO with TRPP2 mRNA. The 5′-UTR of zPRKCSH, which included the zPRKCSH MO target sequence, was amplified from wild-type zebrafish embryos, using the following primer set: forward primer: 5′-CGGCTCGAGGTTCCTGCATTTGGGCATCAT-3′; reverse primer: 5′-CGGTCTAGAC TTTTCTCCTCTTCCTTTCTTGG-3′. The 5′-UTR-zPRKCSH was subcloned into pCS2+, yielding zPRKCSH-eGFP. After linearization with Nsi1, SP6 polymerase was used to synthesize capped RNA. Wild-type zebrafish embryos at the 1- to 2-cell stage were microinjected with 5′-UTR-zPRKCSH-eGFP mRNA alone or in combination with the zPRKCSH MO. For whole-mount immunofluorescence, zebrafish embryos were fixed in Dent's solution (80% methanol, 20% DMSO) at 4°C for 2 days. After blocking with 10% normal goat serum (NGS) (Sigma) at room temperature for 2 h, the primary antibody was incubated in 10% NGS containing monoclonal antibody MF20 (1:100) (Development Studies Hybridoma Bank) at 4°C overnight. After washing and blocking, the larvae were incubated with secondary antibody 1:250 donkey-anti-mouse Cy3 (Jackson ImmunoResearch and Molecular Probes) at 4°C overnight. After additional washing the embryos were examined using a Leica MZ16 stereomicroscope.
In situ hybridization and whole-mount histology
For in situ hybridization, the proteinase K treatments were 3 min for embryos at 24 hpf, 9 min for 48 hpf, and 14 min for 72 hpf-stage embryos. The fkd2 (forkhead-2) (kindly proved by W. Driever) antisense RNA was generated using an Xho1-linearized template and T3 RNA polymerase. The myl7 (myosin, light polypeptide 7) (kindly provided by W. Driever) antisense RNA was generated using a Not1-linearized template and T7 RNA polymerase. The Ins (preproinsulin) (kindly provided by W. Driever) antisense RNA was generated using an Xba1 linearized template and T7 RNA polymerase. Embryos were hybridized with digoxigenin-labeled riboprobes. Anti-DIG AP (1:2000) and the NBT/BCIP substrate (Roche Diagnostics GmbH, Mannheim) were used to detect the probe. After the color reaction was stopped, embryos were cleared in 100% glycerol and photographed using a stereomicroscope (Leica MZ16). For histology analysis, embryos at 55 hpf were fixed in 4% PFA in PBS, embedded with Technovit 7100 resin (Kulzer, Germany), and sectioned at 5 µm sections. Sections were stained in methylene blue/azure II.
This work was supported by grants of the Deutsche Forschungsgemeinschaft (SFB 746, G.W.).
We thank the members of the Renal Division for helpful discussions. Our special thanks go to Prof. Dietrich Goetze and the Athenaeum Foundation for Culture and Science for awarding a fellowship to H.G.
Conflict of Interest statement. None declared.
- situs inversus
- congenital abnormality
- endoplasmic reticulum
- molecular chaperones
- polycystic kidney, autosomal dominant
- quality control
- translocation (genetics)
- signal pathway
- balance impairment
- signal transduction pathways
- polycystic liver disease
- endoplasmic reticulum degradation pathway
- binding (molecular function)