Free
Biochemistry and Molecular Biology  |   June 2014
Identification of the Cellular Mechanisms That Modulate Trafficking of Frizzled Family Receptor 4 (FZD4) Missense Mutants Associated With Familial Exudative Vitreoretinopathy
Author Affiliations & Notes
  • Reham M. Milhem
    Department of Pathology, College of Medicine and Health Sciences, United Arab Emirates University, Al-Ain, United Arab Emirates
  • Salma Ben-Salem
    Department of Pathology, College of Medicine and Health Sciences, United Arab Emirates University, Al-Ain, United Arab Emirates
  • Lihadh Al-Gazali
    Department of Pediatrics, College of Medicine and Health Sciences, United Arab Emirates University, Al-Ain, United Arab Emirates
  • Bassam R. Ali
    Department of Pathology, College of Medicine and Health Sciences, United Arab Emirates University, Al-Ain, United Arab Emirates
  • Correspondence: Bassam R. Ali, Department of Pathology, College of Medicine and Health Sciences, United Arab Emirates University, PO Box 17666, Al-Ain, United Arab Emirates; bassam.ali@uaeu.ac.ae
Investigative Ophthalmology & Visual Science June 2014, Vol.55, 3423-3431. doi:10.1167/iovs.14-13885
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Reham M. Milhem, Salma Ben-Salem, Lihadh Al-Gazali, Bassam R. Ali; Identification of the Cellular Mechanisms That Modulate Trafficking of Frizzled Family Receptor 4 (FZD4) Missense Mutants Associated With Familial Exudative Vitreoretinopathy. Invest. Ophthalmol. Vis. Sci. 2014;55(6):3423-3431. doi: 10.1167/iovs.14-13885.

      Download citation file:


      © 2016 Association for Research in Vision and Ophthalmology.

      ×
  • Supplements
Abstract

Purpose.: Fifteen missense mutations in the frizzled family receptor 4 (FZD4) reported to cause familial exudative vitreoretinopathy (FEVR) were evaluated to establish the pathological cellular mechanism of disease and to explore novel therapeutic strategies.

Methods.: The mutations were generated by site-directed mutagenesis and expressed in HeLa and COS-7 cell lines. Confocal fluorescence microscopy and N-glycosylation profiling were used to observe the subcellular localization of the mutant proteins relative to wild-type (WT). Polyubiquitination studies were used to establish the involvement of the proteasome. Culturing at reduced temperatures and incubation in the presence of chemical compounds were used to enhance mutant protein processing and exit out of the endoplasmic reticulum (ER).

Results.: Confocal fluorescence microscopy of the mutants showed three distinct subcellular localizations, namely, a plasma membrane pattern, an ER pattern, and a mixed pattern to both compartments. Confocal fluorescence microscopy and N-glycosylation profiling established the predominant ER localization of P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D mutants. Coexpression of these mutants with WT FZD4 showed the inability of the mutants to trap WT FZD4. Culturing the expressing cells at reduced temperatures or in the presence of chemical agents directed at ameliorating protein misfolding resulted in partial rescue of trafficking defects observed for M105T and C204Y mutants.

Conclusions.: Defective trafficking resulting in haploinsufficiency is a major cellular mechanism for several missense FEVR-causing FZD4 mutants. Our findings indicate that this trafficking defect might be correctable for some mutants, which may offer opportunities for the development of novel therapeutics approaches for this condition.

Introduction
Familial exudative vitreoretinopathy (FEVR) (Online Mendelian Inheritance in Man 133780) is a developmental anomaly characterized by incomplete or no vascularization of the peripheral retina. 1 Patients with FEVR exhibit highly variable expression of the disease, ranging from asymptomatic to complete blindness. Familial exudative vitreoretinopathy is genetically heterogeneous and can be inherited in a dominant manner via frizzled family receptor 4 (FZD4), low-density lipoprotein receptor–related protein 5 (LRP5), or tetraspanin 12 (TSPAN12) genes. This condition can also be inherited in a recessive manner via mutations in the LRP5 gene or in an X-linked fashion through mutations in the Norrin (NDP) gene. 2 However, the autosomal dominant inheritance is by far the most common cause of FEVR. Mutations in FZD4, a wingless-type mouse mammary tumor virus integration site family (Wnt) receptor, have been shown to cause premature arrest of retinal angiogenesis in patients with FEVR. 3 In addition, mutations in this protein have been linked to retinopathy of prematurity, a condition with phenotypic overlap with FEVR. 4  
The seven-pass transmembrane FZD4 protein has a highly conserved extracellular cysteine–rich domain frizzled (CRD-FZ) (Fig. 1). This N-terminal extracellular CRD is conserved among frizzled family members and determines binding specificity for Wnt ligands. The intracellular domains contain threonine-X-valine PDZ-binding and lysine-threonine-X-X-X-tryptophan disheveled association sites. 5 The Norrin/FZD4 proteins are involved in the wingless (Wnt) signaling pathway controlling a transcriptional program that regulates endothelial growth and maturation throughout the course of retinal vascular development. 68  
Figure 1
 
Illustration of FZD4 protein domain structure. Frizzled family receptor 4 is a seven-pass transmembrane protein with a signal peptide sequence found between amino acid positions 1 and 36/37, a CRD-FZ domain highlighted in green found at amino acid positions 42 through 167, a frizzled region spanning most of the protein within amino acid positions 210 through 514, a KTXXXW motif found at amino acid positions 499 through 504, and a PDZ motif located close to the C-terminal at amino acid positions 535 through 537. The two potential N-glycosylation sites are indicated by arrows at amino acid positions 59 and 144. Missense FEVR-causing mutations studied in this article are labeled across the protein. Missense mutations that traffic abnormally are shown with red stars (P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D).
Figure 1
 
Illustration of FZD4 protein domain structure. Frizzled family receptor 4 is a seven-pass transmembrane protein with a signal peptide sequence found between amino acid positions 1 and 36/37, a CRD-FZ domain highlighted in green found at amino acid positions 42 through 167, a frizzled region spanning most of the protein within amino acid positions 210 through 514, a KTXXXW motif found at amino acid positions 499 through 504, and a PDZ motif located close to the C-terminal at amino acid positions 535 through 537. The two potential N-glycosylation sites are indicated by arrows at amino acid positions 59 and 144. Missense FEVR-causing mutations studied in this article are labeled across the protein. Missense mutations that traffic abnormally are shown with red stars (P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D).
To date, 58 different FEVR pathogenic mutations have been reported for FZD4, of which 31 are missense changes scattered throughout the protein structure. 9 All the mutations except R417Q have been shown to cause the disease in the heterozygous state. 10 However, patients have been diagnosed as having more severe forms of FEVR when both alleles of FZD4 are mutated. 4  
Studies on the cellular localization of FZD4 pathogenic mutations and their functional implications are scarce. A deletion of two nucleotides that led to a frameshift and created a stop codon at residue 533 (L501fsX533) resulted in a defective PDZ-binding motif and hence loss of FZD4 function in the calcium/calmodulin-dependent protein kinase II and protein kinase C signaling pathway. 11 An FZD4 haploinsufficiency effect was postulated to be the causative defect underlying FEVR in a patient hemizygous for FZD4. 12 Because of the nature of the missense mutations and the importance of the CRD domains for proper protein folding, 13 we hypothesized that some of the missense mutations in FZD4 might lead to protein misfolding and failure to pass the stringent endoplasmic reticulum (ER) quality control machinery. Proteins that fail this machinery are targeted for degradation by ER-associated protein degradation (ERAD). Protein misfolding is the underlying cause of numerous highly debilitating disorders, ranging from Alzheimer disease to cystic fibrosis. 1316  
In this article, we examine trafficking of 15 FZD4 missense mutations causing FEVR, shown in Figure 1. We characterized the mutations for their cellular processing, subcellular localization, and polyubiquitination profiles and compared them with wild-type (WT). We coexpressed the mutant constructs with WT construct to determine if a dominant-negative effect contributes to the disease pathology. We also evaluated culturing the expressing cells at reduced temperatures or in the presence of chemical compounds in an attempt to rescue trafficking defects of some mutants. Our data suggest trafficking defects as a major disease mechanism for nine of the studied mutations: P33S, 17 G36N, 18 H69Y, 19 M105V, 4 M105T, 18 C181R, 19 C204Y, 20 C204R, 21 and G488D. 4  
Methods
Chemicals and Reagents
Human embryonic kidney (HEK 293T), HeLa, and COS-7 cells were obtained from ATCC (Manassas, VA, USA). Bovine serum albumin was obtained from Fisher Scientific (Loughborough, UK). Dulbecco's modified Eagle's medium (DMEM) and PBS were obtained from Invitrogen (Grand Island, NY, USA). Fetal bovine serum (FBS), l-glutamine, and penicillin-streptomycin were obtained from GIBCO (Grand Island, NY, USA). The liposomal transfection reagent FuGENE HD was purchased from Promega (Madison, WI, USA). The following were purchased from Sigma-Aldrich Corp. (St. Louis, MO, USA): endoglycosidase H (Endo H) (A0810), protease inhibitors (S8830, SIGMAFAST protease inhibitor cocktail tablets), phosphatase inhibitor cocktail (P0044), phosphatase inhibitor cocktail 2 (P5726), β-mercaptoethanol (BME) (M7154), glycerol (G5516), thapsigargin (T9033), dimethyl sulfoxide (DMSO) (5879), curcumin (C7727), and dithiothreitol (D9779). Turbo Pfu DNA polymerase was obtained from Stratagene (La Jolla, CA, USA). A bicinchoninic acid protein (BCA) assay kit (No. 23227) and sepharose A/G beads (No. 20421) were obtained from Pierce (Rockford, IL, USA). Immunofluor mounting media were purchased from Dako (Carpinteria, CA, USA). 
Antibodies, their dilutions, and sources included the following antibodies for immunofluorescence (IF) and Western blot (WB): mouse FLAG epitope tag (DYKDDDDK) (anti-FLAG) monoclonal antibody (1:1000, IF and WB; Sigma-Aldrich Corp.), rabbit anti-FLAG antibody (1–2 μg/mL immunoprecipitation [IP], F7425; Sigma-Aldrich Corp.), rabbit anti-calnexin polyclonal antibodies (1:50, IF; Santa Cruz, Dallas, TX, USA), mouse monoclonal anti–ubiquitin (Ub) antibody (1:1000, WB, U0508; Sigma-Aldrich Corp.), AlexaFluor 568 goat anti-mouse IgG (1:200; Molecular Probes, Grand Island, NY, USA), and AlexaFluor 488 goat anti-rabbit IgG (1:2000; Molecular Probes). Secondary antibodies for WB were anti-rabbit IgG (whole molecule)-peroxidase antibody produced in goat (A0545, 1:30,000; Sigma-Aldrich Corp.), anti-mouse IgG (whole molecule)-peroxidase antibody produced in rabbit (A9044, 1:50,000; Sigma-Aldrich Corp.), and Pierce ECL plus WBsubstrate (32132; Thermo Scientific, Rockford, IL, USA). 
Construction of the FLAG-Tagged FZD4 Construct
The FLAG-tagged version of WT FZD4 was generated by sequential site-directed mutagenesis to introduce the eight amino acids of the FLAG tag (DYKDDDDK) in two cycles using PCMV6-sport-FZD4 plasmid (a kind gift from Jeremy Nathans, The Johns Hopkins University, Baltimore, MD, USA) as a template. In the first cycle, four amino acids (DYKD) of the FLAG tag were introduced before the stop codon of the cDNA. In the second cycle, the resulting construct was used as a template, and the remaining four amino acids (DDDK) were introduced. Similarly, the human influenza hemagglutinin (HA) epitope tag (YPYDVPDYA) version of WT FZD4 was generated by site-directed mutagenesis to introduce the nine amino acids of the HA tag (YPYDVPDYA) in two cycles. The sequences of the primers used are listed in Supplementary Table S1
Generation of the Missense FZD4 Mutants
The 15 missense mutants were introduced by site-directed mutagenesis with Turbo Pfu DNA polymerase (Stratagene) and the C-terminally FLAG-tagged WT FZD4 construct as a template. Primers were generated using the software PrimerX (http://www.bioinformatics.org/primerx/ [in the public domain]) and are listed in Supplementary Table S1
DNA Sequencing
Deoxyribonucleic acid sequencing was performed using the dideoxy Sanger method by fluorescent automated sequencing on the ABI 3130xl genetic analyzer (Applied Biosystems, Foster City, CA, USA). Sanger sequencing is based on the selective incorporation of chain-terminating dideoxynucleotides by DNA polymerase during in vitro DNA replication. The sequence data were analyzed using sequencing analysis ABI software version 5.3 (Applied Biosystems). ClustalW2 (http://www.ebi.ac.uk/Tools/msa/clustalw2/ [in the public domain]) was used for sequence alignments. 
Cell Culture and Transfection
HeLa, COS-7, and HEK 293T cells were cultured in DMEM supplemented with 10% FBS, 2 mM l-glutamine, and 100 U/mL penicillin-streptomycin at 37°C with 5% carbon dioxide. For confocal fluorescence microscopy (CFM), cells were grown on sterile coverslips in 24-well tissue culture plates. Transfection was performed after 24 hours of seeding using the liposomal transfection reagent FuGENE HD reagent (Promega) according to the manufacturer's instructions. For each well, 1 μg WT FZD4 or the mutant constructs were used. Forty-eight hours after transfection, the cells were fixed and processed for CFM as described in the next section. For WB, 1 day before transfection cells were seeded on six multiwells (35 mm in diameter) and grown to 70% to 80% confluency. Transfection was performed using FuGENE HD transfection reagent (Promega) according to the manufacturer's instructions. In this case, each well was transfected with 2 μg WT FZD4 or mutant constructs' DNA. 
Immunofluorescence and CFM
For IF, coverslip-grown HeLa or COS-7 cells were washed with PBS and fixed by cold methanol at −20°C for 4 minutes. Fixed cells were then washed in PBS three times and incubated in blocking solution (10% BSA in PBS) for 1 hour at room temperature. The fixed cells were then incubated for 1 hour at room temperature with either the primary monoclonal antibodies (anti-FLAG, anti-HA, etc.) alone or coincubated with anti-calnexin polyclonal antibodies. After washing with PBS, the cells were incubated with the appropriate fluorescently labeled secondary antibodies for 1 hour at room temperature and then washed several times with PBS and mounted in Immunofluor medium (Dako). We used enhanced green fluorescent protein-Harvey rat sarcoma viral oncogene homolog (EGFP-HRAS) as a marker for the plasma membrane (PM). In this case, the EGFP-HRAS construct was cotransfected with FZD4 constructs. For the dominant-negative experiments, WT FZD4 HA-tagged construct was cotransfected at a 1:3 ratio (2 μg total) with the individual FLAG-tagged mutants. Data were acquired using CFM (Nikon, Tokyo, Japan). For presentation purposes, images were pseudocolored as either red (FZD4) or green (calnexin and EGFP-HRAS), contrast enhanced, and merged using ImageJ version 1.47 (http://rsbweb.nih.gov/ij [in the public domain]). All images presented are single sections in the z-plane. Colocalizations were quantitated when required using ImageJ version 1.47. 
Protein Extraction and IP
Immunoprecipitation was used to isolate the expressed FZD4 proteins from whole-cell lysates of HEK 293T cells. Forty-eight hours after transfection, culture media were removed, and cells were gently washed twice with ice-cold PBS. After washing, 150 μL ice-cold radioimmunoprecipitation assay buffer (RIPA) buffer (R0278; Sigma-Aldrich Corp.) containing protease inhibitors was added to the 35-mm culture dishes and incubated on ice for 5 minutes. Cells were then scraped off the plate and transferred to microcentrifuge tubes and kept on ice for 15 minutes. The cell lysates were then centrifuged at 1000g for 15 minutes at 4°C. The supernatants were then transferred to a new tube without disturbing the pellet. Proteins were quantified with the BCA assay kit (Pierce) according to the manufacturer's instructions. One microliter anti-FLAG antibodies was added to the supernatants and incubated at 4°C on a rotator for 1 to 3 hours, and then 10 μL protein sepharose A/G (preequilibrated in the corresponding IP lysis buffer) was added and incubated at 4°C for 2 hours to capture the immune complexes. The tubes were then centrifuged at 2500g for 30 seconds at 4°C, and the supernatants were carefully removed, washed once with RIPA buffer, and recentrifuged, and the supernatants were removed. 
Endo H Digestion
Following IP, FZD4 was eluted from the sepharose A/G beads using 0.2 M glycine buffer at pH 2.6. Fifty microliters glycine buffer was added to the beads and incubated for 10 minutes with frequent agitation before centrifugation. The eluates were pooled and neutralized by adding equal volumes of Tris at pH 8.0. The beads were then neutralized by washing two times with 150 μL lysis buffer (without detergent) and pooled with eluates. Five microliters sodium phosphate buffer (250 mM at pH 5.5) was added to 15 μL of each eluate with 1.25 μL denaturation solution (containing 2% SDS and 1 M BME). Two microliters Endo H enzyme was added to each sample and incubated at 37°C for 4 hours. The treated samples were then subjected to SDS-PAGE and WB using anti-FLAG antibodies as described below. 
WB Analysis
Frizzled family receptor 4 IPs were treated with 2× Laemmli buffer containing 8% BME. Thirty micrograms protein from control (untransfected cells), WT FZD4, and mutant preparations was subjected to 8% SDS-PAGE and transferred to nitrocellulose membranes. After 30 minutes of blocking in 3% nonfat dry milk (NFM) in Tris-buffered saline (TBS) (at pH 8) 0.05% (w/v) Tween 20 (T), nitrocellulose blots were incubated for 1 hour with anti-FLAG antibodies in 3% NFM TBS-T. Following washing with TBS-T, membranes were then incubated with rabbit anti-mouse IgG-peroxidase antibody in TBS-T containing 3% NFM for 1 hour and developed by Pierce ECL plus WB reagents (Thermo Scientific). All blots were viewed with Typhoon FLA 9500 ImageQuant TL software (GE Healthcare Sciences, Uppsala, Sweden). 
Anti-Ubiquitin WB
Following IP and SDS fractionation on an 8% gel, the immunoblots were stripped with stripping buffer, washed three times with TBS-T, and then autoclaved as previously described. 22 The blots were then blocked with 1% NFM in TBS-T (0.05% w/v Tween 20). The TBS contained (per liter) 8 g sodium chloride, 0.2 g potassium chloride, and 3 g Tris base and was adjusted to pH 7.4. Anti-Ub antibody was diluted 1:1000 in 1% NFM in TBS-T and incubated overnight at 4°C. The blots were washed three times with TBS-T and incubated with the secondary antibodies for 1 hour in 1% NFM TBS-T and processed as described above. 
Results
Seven of the 15 FZD4 Mutants Are Predominantly Localized to the ER
Fifteen FEVR-causing missense mutations shown in Figure 1 were generated, and their localization was assessed in both HeLa and COS-7 cell lines using IF CFM following transient expression. The data for the COS-7 cell line are not shown but were always consistent with the HeLa cell line results. In both cell lines, the ER lectin chaperone calnexin was stained to establish the colocalization with the ER network. In contrast to WT FZD4 construct, seven (P33S, G36N, H69Y, M105T, C204R, C204Y, and G488D) of the 15 tested mutants exhibited high degrees of colocalization with calnexin, suggesting mislocalization away from the plasma membrane in a perinuclear and reticular pattern, suggestive of the ER (Fig. 2A [panels B–J are compared with panel A of WT FZD4]). Figure 2B shows quantification of the degree of colocalization of mutants with calnexin compared with WT. Two mutants (M105V and C181R) showed partial ER localization with some residual protein localizing to the PM (Fig. 2A [compare with panels F, G]). The localization of the other six FEVR-causing mutants (W335C, R417P, R417Q, T445P, G492R, and S497P) resembled WT FZD4 on the plasma membrane (Supplementary Fig. S1). 
Figure 2
 
Colocalization of several FZD4 missense mutants with the ER marker in HeLa cells. Confocal IF images of WT FZD4 and FZD4 mutants (P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D) show the colocalization of these mutants with calnexin, an ER marker. A FLAG epitope expressed WT FZD4 protein, and mutants are seen in red on the left of each panel, and calnexin is shown in green in the middle image of each panel. The image on the right is the merging of the two signals. (A) Shown is WT FZD4 localization to the plasma membrane, which is distinct from that of calnexin. In contrast, as shown in (BJ), the above missense mutants are largely localized to an intracellular compartment that is perinuclear and reticular in nature and colocalizing with calnexin, suggestive of an ER localization. However, C181R and M105V seem to be partially localized to the ER, with some localization to the PM. Cells were transfected for 48 to 72 hours and then processed for IF as described in the Methods section. (B) Colocalization with calnexin was measured as a function and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments (mean [SE] of each experiment). Data (arbitrary units) are expressed as the mean (SE). ***P < 0.001 compared with WT control. Scale bar: 25 μm. Magnification: ×100.
Figure 2
 
Colocalization of several FZD4 missense mutants with the ER marker in HeLa cells. Confocal IF images of WT FZD4 and FZD4 mutants (P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D) show the colocalization of these mutants with calnexin, an ER marker. A FLAG epitope expressed WT FZD4 protein, and mutants are seen in red on the left of each panel, and calnexin is shown in green in the middle image of each panel. The image on the right is the merging of the two signals. (A) Shown is WT FZD4 localization to the plasma membrane, which is distinct from that of calnexin. In contrast, as shown in (BJ), the above missense mutants are largely localized to an intracellular compartment that is perinuclear and reticular in nature and colocalizing with calnexin, suggestive of an ER localization. However, C181R and M105V seem to be partially localized to the ER, with some localization to the PM. Cells were transfected for 48 to 72 hours and then processed for IF as described in the Methods section. (B) Colocalization with calnexin was measured as a function and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments (mean [SE] of each experiment). Data (arbitrary units) are expressed as the mean (SE). ***P < 0.001 compared with WT control. Scale bar: 25 μm. Magnification: ×100.
To further assess the subcellular localization, WT FZD4 and the 15 FZD4 FEVR-causing mutant constructs were cotransfected in HeLa cells with EGFP-HRAS plasmid, the latter acting as a plasma membrane marker. The same seven mutants (P33S, G36N, H69Y, M105T, C204R, C204Y, and G488D) exhibited a distinct perinuclear reticular pattern that is clearly different from the EGFP-HRAS, establishing mislocalization away from the plasma membrane (Figs. 3B–E, 3H–J). A dual pattern of ER and PM localization for M105V and C181R was once again observed (Figs. 3F, 3G). The data in HeLa and COS-7 cell lines suggest that this mislocalization of the FEVR-causing FZD4 mutants is not cell-type specific. 
Figure 3
 
Exclusion of the seven missense FZD4 mutants from the plasma membrane in HeLa cells. The CFM images of WT FZD4 and FZD4 mutants P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D (red in the images on the left side of each panel) show their localization in HeLa cells relative to HRAS, a plasma membrane marker (shown in the green image in the middle of each panel). The FLAG epitope–expressed WT FZD4 protein in (A) indicates colocalization, with HRAS indicating plasma membrane localization. However, (BJ) show that the mutants are largely excluded from the PM and localized to the perinuclear region, implying ER localization. C181R and M105V seem to be partially retained in the ER, with some PM localization, validating the data shown in Figure 2. Cells were transfected for 48 to 72 hours and then processed for IF. Scale bar: 25 μm. Magnification: ×100.
Figure 3
 
Exclusion of the seven missense FZD4 mutants from the plasma membrane in HeLa cells. The CFM images of WT FZD4 and FZD4 mutants P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D (red in the images on the left side of each panel) show their localization in HeLa cells relative to HRAS, a plasma membrane marker (shown in the green image in the middle of each panel). The FLAG epitope–expressed WT FZD4 protein in (A) indicates colocalization, with HRAS indicating plasma membrane localization. However, (BJ) show that the mutants are largely excluded from the PM and localized to the perinuclear region, implying ER localization. C181R and M105V seem to be partially retained in the ER, with some PM localization, validating the data shown in Figure 2. Cells were transfected for 48 to 72 hours and then processed for IF. Scale bar: 25 μm. Magnification: ×100.
N-Glycosylation Profiling Was Used to Verify the ER Mislocalization of the Mutants Observed by CFM
To further assess the subcellular localization of the ER-retained mutants, Endo H sensitivity and resistance in vitro assays on the expressed proteins were carried out. Frizzled family receptor 4 protein has two potential N-glycosylation sites in its extracellular domain (Fig. 1, arrows). The principle of this assay is that the carbohydrate moieties of the ER-localized glycoproteins are cleavable by Endo H, whereas the post-ER species are resistant due because of further remodeling of their N-glycans in the Golgi complex. Endo H cleaves after the first asparagine-linked N-acetylglucosamine (N-glycan) of high-mannose and hybrid polysaccharides. Therefore, one can expect that the mutants retained in the ER will have their N-glycans cleaved when treated with Endo H. Proteins that have reached the Golgi complex would be resistant to Endo H treatment. Upon Endo H treatment, P33S, G36N, H69Y, M105V, M105T, C204Y, and C204R mutants showed complete conversion of the high-molecular-weight band into a lower-molecular-weight band, suggesting that these mutants are still in their immature state (Fig. 4A). On the other hand, WT FZD4 showed resistance to this treatment (Fig. 4A). Interestingly, the C181R mutant gave two bands, one higher-molecular-weight band similar to the mature-form WT FZD4 and a smaller-molecular-weight band that suggested incomplete ER retention of this protein (Fig. 4A). Quantification of these data is shown in Figure 4B. This observation may indicate that the change in amino acid may cause slow trafficking of this mutant to the PM. Interestingly, the M105V mutant, which previously appeared by IF to show mixed localization between the ER and PM, resulted in one immature band. This may suggest that expression of the M105V mutant protein on the PM is somewhat minimal. It was notable that Endo H–treated and Endo H–untreated P33S and G488D samples showed strong dimerization (Fig. 4A). This behavior of FZD4 has been observed previously. 2325  
Figure 4
 
N-glycosylation profiles for the ER-retained mutants and WT FZD4. In whole-cell lysates (WCL), WT FZD4–expressed protein has been seen as a monomer, dimer, and multimers at the top of an SDS-PAGE gel (Supplementary Fig. S4). Wild-type FZD4 and mutants have been pulled down with mouse anti-FLAG antibodies, followed by sepharose A/G bead sedimentation as described in the Methods section. Upon elution, 20 μg protein was treated and then digested with Endo H for 5 hours with 60 mM dithiothreitol and 8 M urea (v/v) as described in the Methods section. P33S, M105T, and G488D mutants show an immature FZD4 protein, with lower-molecular-weight bands compared with WT FZD4 and its treated counterpart, suggestive of possible folding defects. The monomer's molecular weight of FZD4 is approximately 60 kDa. Mutants P33S, G36N, H69Y, M105T, C204R, C204Y, and G488D are quantitatively cleavable with Endo H, establishing their predominant localization in the ER. M105V and C181R have been shown to have a dual pattern by CFM. However, when the M105V mutant was subjected to Endo H treatment, a single lower-molecular-weight band was observed. On the other hand, two bands, one mature and one immature, were observed for C181R. Furthermore, P33S and G448D formed dimers and probably multimers on Endo H treatment at a temperature of 37°C. Molecular mass standards are indicated in kDa. A bar chart in (B) shows the percentages of Endo H–resistant and Endo H–sensitive species for each sample in (A). Wild-type is almost exclusively resistant to Endo H, whereas all the mutants (with the exception of C181R) are exclusively Endo H sensitive. Quantification was carried out by Typhoon FLA 9500 ImageQuant TL software (GE Healthcare Life Sciences).
Figure 4
 
N-glycosylation profiles for the ER-retained mutants and WT FZD4. In whole-cell lysates (WCL), WT FZD4–expressed protein has been seen as a monomer, dimer, and multimers at the top of an SDS-PAGE gel (Supplementary Fig. S4). Wild-type FZD4 and mutants have been pulled down with mouse anti-FLAG antibodies, followed by sepharose A/G bead sedimentation as described in the Methods section. Upon elution, 20 μg protein was treated and then digested with Endo H for 5 hours with 60 mM dithiothreitol and 8 M urea (v/v) as described in the Methods section. P33S, M105T, and G488D mutants show an immature FZD4 protein, with lower-molecular-weight bands compared with WT FZD4 and its treated counterpart, suggestive of possible folding defects. The monomer's molecular weight of FZD4 is approximately 60 kDa. Mutants P33S, G36N, H69Y, M105T, C204R, C204Y, and G488D are quantitatively cleavable with Endo H, establishing their predominant localization in the ER. M105V and C181R have been shown to have a dual pattern by CFM. However, when the M105V mutant was subjected to Endo H treatment, a single lower-molecular-weight band was observed. On the other hand, two bands, one mature and one immature, were observed for C181R. Furthermore, P33S and G448D formed dimers and probably multimers on Endo H treatment at a temperature of 37°C. Molecular mass standards are indicated in kDa. A bar chart in (B) shows the percentages of Endo H–resistant and Endo H–sensitive species for each sample in (A). Wild-type is almost exclusively resistant to Endo H, whereas all the mutants (with the exception of C181R) are exclusively Endo H sensitive. Quantification was carried out by Typhoon FLA 9500 ImageQuant TL software (GE Healthcare Life Sciences).
The Seven ER-Retained Mutants Are Polyubiquitinated
To determine if the ER-retained mutants are targeted for ERAD, HEK 293T cells were transiently transfected with WT or mutant FZD4 plasmids. The expressed proteins underwent IP and were subjected to SDS-PAGE and probed with anti-FLAG antibodies (Fig. 5B), stripped, and reprobed with anti-Ub antibodies. Immunostaining by anti-Ub antibodies showed a striking smear of multiple Ub-mutant FZD4 conjugates at the top of the resolving gel (Fig. 5A). Conjugates with multiple Ub moieties in the form of branched chains appear to be associated with the seven FZD4 mutants to a much greater extent compared with WT (Fig. 5C). These data suggest that the seven ER-retained mutants are probably targeted for degradation by the ERAD machinery. 
Figure 5
 
Polyubiquitination pattern of WT FZD4 and ER-retained FZD4 mutants. (A) Immunoprecipitations of WT FZD4, P33S, G36N, H69Y, M105T, C204Y, C204R, and G488D mutants were run on SDS-PAGE, blotted on nitrocellulose, and probed with anti-FLAG antibodies. The blots were then washed overnight with TBS-T, stripped, autoclaved, and then reprobed with anti-Ub monoclonal antibodies as described in the Methods section. The samples of the seven ER-retained mutants show very-high-molecular-weight smears, suggesting polyubiquitination. Wild-type FZD4 also seems to be ubiquitinated, albeit to a much lesser extent. (B) Shown is an immunoblot of IP WT FZD4, P33S, G36N, H69Y, M105T, C204Y, C204R, and G488D with anti-FLAG antibody before stripping for polyubiquitination studies. (C) Quantification of the polyubiquitination blot presented above is shown. Considerably higher levels of ubiquitination are observed for the mutants compared with WT. A negative IP is presented for comparison. Quantification was carried out by Typhoon FLA 9500 ImageQuant TL software (GE Healthcare Life Sciences).
Figure 5
 
Polyubiquitination pattern of WT FZD4 and ER-retained FZD4 mutants. (A) Immunoprecipitations of WT FZD4, P33S, G36N, H69Y, M105T, C204Y, C204R, and G488D mutants were run on SDS-PAGE, blotted on nitrocellulose, and probed with anti-FLAG antibodies. The blots were then washed overnight with TBS-T, stripped, autoclaved, and then reprobed with anti-Ub monoclonal antibodies as described in the Methods section. The samples of the seven ER-retained mutants show very-high-molecular-weight smears, suggesting polyubiquitination. Wild-type FZD4 also seems to be ubiquitinated, albeit to a much lesser extent. (B) Shown is an immunoblot of IP WT FZD4, P33S, G36N, H69Y, M105T, C204Y, C204R, and G488D with anti-FLAG antibody before stripping for polyubiquitination studies. (C) Quantification of the polyubiquitination blot presented above is shown. Considerably higher levels of ubiquitination are observed for the mutants compared with WT. A negative IP is presented for comparison. Quantification was carried out by Typhoon FLA 9500 ImageQuant TL software (GE Healthcare Life Sciences).
The Effect of Reducing Temperature on Trafficking of the Seven ER-Retained Mutants Is Shown
Interestingly, deletion of phenylalanine at position 508 (ΔF508) in the cystic fibrosis transmembrane conductance regulator (CFTR) is mainly retained in the ER as a result of misfolding; however, when a small amount of the ΔF508 CFTR protein is expressed on the cell surface, it retains significant chloride channel conductance. 26,27 Such results provided insight and proof of principle that the molecular machinery directing folding and export could have an important role in both understanding the molecular basis for disease and targeting therapeutic intervention. 26,2830  
In addition, it has been shown that cell membrane expression of functional ΔF508 CFTR can be achieved by culturing the expressing cells at lower temperatures. 31,32 Low-temperature incubation of affected tissues in CFM is not a viable therapeutic option. However, incubation at lower temperatures may provide proof of principle that the kinetic and/or thermodynamic aspects of the secretory protein folding can be manipulated and therefore prevent targeting of the mutant proteins to the ERAD machinery. Of the seven ER-retained mutants, M105T and C204Y showed partial rescue of trafficking to the PM by 57% and 46%, respectively, upon culturing the expressing cells at 27°C (Figs. 6B–E). 
Figure 6
 
Effect of low-temperature incubation and chemical compounds on M105T and C204Y localization. (A) Shown are CFM images of reduced temperature and chemical compound incubation of WT FZD4 acting as a control. (B, C) Shown are CFM images of M105T and C204Y mutants across the three treatments (culturing at 27°C, 7.5% glycerol, and 0.1% DMSO, respectively). (D) Using ImageJ version 1.47, quantification of colocalization with the ER marker (expressed as percentage) of the IF pattern of the expressed FZD4 proteins is shown. At 27°C, WT colocalized with the ER marker by approximately 3% only, whereas M105T and C204Y colocalized by 43% and C204Y by 54%. With 7.5% glycerol, M105T colocalized by 50% and C204Y by 85%. With 0.1% DMSO, M105T colocalized with calnexin by 32% and C204Y by 95%. Colocalization with calnexin was measured as a function of and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments (mean [SE] of each experiment). (E) Shown is quantification of the amount of FZD4 proteins colocalizing with HRAS, a PM marker. At 27°C, M105T and C204Y colocalized by 57% and 46%, respectively, whereas in the presence of 7.5% glycerol, M105T colocalized by 50% and C204Y by 32%. The M105T mutant showed partial plasma membrane distribution when treated with 0.1% DMSO by approximately 68% and C204Y by approximately 5%. One microgram WT FZD4 or the mutant constructs was used for transfection experiments. For coexpression studies, 1 μg for each construct was cotransfected together. Colocalization with HRAS was measured as a function of and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments. Scale bar: 25 μm. Data (arbitrary units) are expressed as the mean (SE). ***P < 0.001.
Figure 6
 
Effect of low-temperature incubation and chemical compounds on M105T and C204Y localization. (A) Shown are CFM images of reduced temperature and chemical compound incubation of WT FZD4 acting as a control. (B, C) Shown are CFM images of M105T and C204Y mutants across the three treatments (culturing at 27°C, 7.5% glycerol, and 0.1% DMSO, respectively). (D) Using ImageJ version 1.47, quantification of colocalization with the ER marker (expressed as percentage) of the IF pattern of the expressed FZD4 proteins is shown. At 27°C, WT colocalized with the ER marker by approximately 3% only, whereas M105T and C204Y colocalized by 43% and C204Y by 54%. With 7.5% glycerol, M105T colocalized by 50% and C204Y by 85%. With 0.1% DMSO, M105T colocalized with calnexin by 32% and C204Y by 95%. Colocalization with calnexin was measured as a function of and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments (mean [SE] of each experiment). (E) Shown is quantification of the amount of FZD4 proteins colocalizing with HRAS, a PM marker. At 27°C, M105T and C204Y colocalized by 57% and 46%, respectively, whereas in the presence of 7.5% glycerol, M105T colocalized by 50% and C204Y by 32%. The M105T mutant showed partial plasma membrane distribution when treated with 0.1% DMSO by approximately 68% and C204Y by approximately 5%. One microgram WT FZD4 or the mutant constructs was used for transfection experiments. For coexpression studies, 1 μg for each construct was cotransfected together. Colocalization with HRAS was measured as a function of and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments. Scale bar: 25 μm. Data (arbitrary units) are expressed as the mean (SE). ***P < 0.001.
The Effect of Incubation of M105T and C204Y With Different Compounds That Are Directed at Promoting Folding and Expression Is Shown
Different compounds have been shown to rescue folding of mutant proteins. 3335 The IF pattern of M105T and C204R mutants when cultured in the presence of 7.5% glycerol showed escape from ER retention by approximately 50% and 32%, respectively (Figs. 6B–E). This observation indicates that glycerol enhances protein processing and exits from the ER to the cell surface, although its trafficking is expected to be slower than WT. In addition, M105T and C204Y mutants were treated with 0.1% DMSO (Fig. 6C), 10 μM thapsigargin, or 1 μM curcumin (data not shown). The M105T mutant showed partial PM distribution when treated with 0.1% DMSO by approximately 32% (Figs. 6D, 6E). C204Y showed a lower pattern of rescue to the PM compared with M105T (Figs. 6D, 6E). Both M105T and C204Y showed no rescue with differing concentrations with either thapsigargin or curcumin. The other ER-retained mutants showed no rescue to the PM with any of the examined treatments. 
Discussion
Vascular diseases of the retina are a major cause of impaired vision and blindness. It has become evident that a large number of diseases with very different pathologies share a common framework of protein misfolding accompanied by degradation and/or aggregation of these misfolded proteins. 14,36 Inherited diseases that harbor missense mutations may affect the folding or stability of proteins. For some of these types of mutant proteins, it might be possible to improve folding, stabilize the native structure, or suppress aggregation using chemical compounds that mimic chaperone effects. 
In this article, we have shown how several FEVR-causing mutations lead to the retention of the mutant proteins by the stringent ER quality control machinery. The ER chaperones have been shown to detect misfolded proteins, unmodified or orphaned subunits of protein complexes and target them for degradation by ERAD. 3741 Our study shows that P33S, G36N, H69Y, M105T, C204R, C204Y, and G488D FZD4 mutants are exclusively retained in the ER. Interestingly, all these mutations are found in the extracellular portion of the protein except for the G488D mutation, which is found on the seventh transmembrane domain (Fig. 1). 
A closer look at each mutation's physicochemical properties and PolyPhen-2 (http://genetics.bwh.harvard.edu/pph2/ [in the public domain]) and SIFT (http://sift.jcvi.org/ [in the public domain]) server predictions sheds light on the predicted effects of some of the mutations on FZD4 protein (Table). For example, the P33S mutation is expected to cause inefficient or no cleavage of the signal peptide; therefore, the P33S mutant is likely to be stuck on the ER membrane. On the other hand, the C204R and C204Y mutations result in the disruption of a vital cysteine disulfide bond. Smallwood et al. 8 have shown that the binding of Norrin to the CRD domain of FZD4 extends beyond the CRD's 114 amino acids to include residue C204. Any effect beyond the CRD on Norrin binding is likely to represent just an effect on FZD4 folding and not direct contact with Norrin. Zhang et al. 5 demonstrated how the C204R mutation fails to bind Norrin. The retention of both of these proteins in the ER may explain the inability of the protein to bind to its ligand. The glycine at position 488 is also highly conserved in orthologs and paralogs, 4 and the change to an aspartic acid is expected to be deleterious. 
Table
 
Summary of the Physicochemical Properties of the FEVR-Causing FZD4 Mutations Shown in This Study to Be Mislocalized to the ER
Table
 
Summary of the Physicochemical Properties of the FEVR-Causing FZD4 Mutations Shown in This Study to Be Mislocalized to the ER
Mutation PolyPhen-2 SIFT Charge Polarity Hydrophobicity Size Score
P33S None 0.47, T Equal Increase Increase Decrease 23
G36N 0.004, N 0.62, T Equal Increase Decrease Increase 32
H69Y 0.029, N 0.08, T Decrease Equal Increase Equal 49
M105T 0.997, D 0.47, T Equal Increase Decrease Decrease 32
M105V  0.66, T 0.93, D Equal Equal Equal Decrease 16
C181R 0.033, N 0.52, T Increase Increase Decrease Increase 77
C204R 1, D 0.02, D Increase Increase Decrease Increase 77
C204Y 1, D 0.04, D Equal Increase Equal Increase 85
G488D 0.06, T 1, D Increase Increase Decrease Increase 28
Interestingly, the M105V and C181R mutations showed a dual pattern (Figs. 2F, 2G, 3F, 3G). This partial ER retention behavior has been reported for other cysteine missense mutations. 42 The location of M105V and C181R is significant because they are within the CRD of the FZD4 extracellular domain and are highly conserved in other species. P33S and G488D show mainly dimers, multimers, and oligomers on an SDS-PAGE gel, which has been previously observed with FZD4 protein (Fig. 4; Supplementary Fig. S4), 2325 In addition, heating of FZD4 protein has been reported to cause the formation of aggregates, 24 which was seen in our experiments. 
Interestingly, the mutations found to traffic abnormally are mainly found in the extracellular domain of FZD4, where the CRD region is located (Fig. 1). The CRDs of frizzled proteins have been shown to form homo-oligomers and/or hetero-oligomers within the ER. 23,43 Endoplasmic reticulum–associated protein degradation results in loss of function of the mutant protein as a result of being degraded before reaching its final destination. Frizzled family receptor 4 proteins dimerize in the ER, and Norrin dimers have been shown to interact with FZD4 dimers. 44  
Therefore, we wanted to examine the possible dominant-negative effects of the mutant proteins on WT FZD4. However, coexpression of WT FZD4 with individual ER-retained mutants failed to show the retention of WT protein in the ER or its dimerization with any of the mutants (Supplementary Figs. S2, S3). Wild-type protein was expressed as usual on the plasma membrane when coexpressed with the mutants (Supplementary Fig. S2). Therefore, we can conclude that the studied ER-retained mutants do not trap WT within the ER and hence have no dominant-negative effects. In addition, IP of the mutant proteins failed to pull down WT FZD4 (Supplementary Fig. S3). This implies that the mutant protein may adopt conformations that do not allow dimerization with WT. These data suggest that WT function might not be affected by the coexpression of a mutant allele. A reported hemizygous mutation has shown haploinsufficiency and therefore loss of function of FZD4 as the underlying disease mechanism.12 Our data support the haploinsufficiency theory for this disease.12  
All but one mutation are found in the heterozygous state. It is likely that the phenotypic variability observed in heterozygous individuals within the same family may result from the effects of modifier genes alongside an alteration in the gene dosage, resulting in allelic insufficiency. The ER retention for reported compound heterozygote patients carrying two different mutations (H69Y and G488D4) can explain why a more severe phenotype is observed compared with single heterozygous mutations. Our data further illustrate the complexity of FEVR and provide a better understanding of the genotype–phenotype correlations for some mutants. However, the exact effects of the studied mutants on retinal vascularization and angiogenesis remain to be fully established. 
In addition, the non–ER-retained mutants could cause FEVR by forcing FZD4 to take up conformations that do not permit binding to its ligands or downstream targets. For example, Norrin, a cysteine knot protein and a ligand of FZD4, shows wide variation in the arrangement of monomers within the dimer and the surfaces that interact with receptors or other binding proteins. 8,44 This may also affect ligand specificity. Interestingly, dimerization of frizzled is emerging as a possible mechanism for transduction specificity. 45  
An essential outcome of ubiquitination is targeting to the 26S proteasome, a specific signal that is responsible for the degradation of many cellular proteins, including numerous ER-retained proteins upon export from the ER to cytosol. 39,40 Most proteasome substrates are tagged not by a single Ub (monoUb) but by a polyubiquitin (polyUb) chain. It is generally thought that polyUb chains longer than four Ub molecules are the preferred signal for efficient recognition and degradation by 26S proteasomes. 46 The FZD4 polyubiquitination pattern is that of a polyUb chain increasing the molecular weight of FZD4 to approximately 200 kDa (Fig. 5). Once bound by proteasomes, substrate conjugates are deubiquitinated, unfolded, and subsequently degraded. 36,47 Loss of function of FZD4 mutants caused by proteasomal degradation of the misfolded protein further reveals the mechanism of certain FZD4 mutations that cause retinal dysfunction. 
Interestingly, incubation at lower temperatures of 27 to 28°C 31 rescued the M105T mutant and partially rescued the C204Y mutant, as is observed with previous studies on CFTR. 31,32 Glycerol, a small synthetic chemical that acts as an osmolyte, increases the hydration layer and the strength of the intramolecular hydrophobic bonding of the protein in the ER, which consequently allows the free movement of proteins, so as to prevent aggregation. 48 Dimethyl sulfoxide acts by probably increasing protein synthesis of the mutant or by overwhelming the quality control system. Thapsigargin, an inhibitor of the ER sarco(endo)plasmic reticulum calcium adenosine triphosphatase calcium pump, has been shown to increase cytosolic calcium, resulting in enhanced rescue of mutant proteins. 49 Curcumin, a nontoxic natural constituent of turmeric spice affects the calcium adenosine triphosphaatases 50 found on the ER plasma membrane by inhibiting their ability to maintain a high ER calcium level. This, in turn, affects the ability of ER molecular chaperones to target the misfolded protein for ERAD, hence allowing the mutant protein to exit the ER. It would be interesting to look into more compounds that will inevitably provide the groundwork for designing a pharmacological chaperone to further salvage these mutations from their trafficking defect. 
The present work represents an initial step in delineating the cellular mechanisms underlying some FEVR-causing missense mutations. Our findings also support efforts for rescuing trafficking of other missense mutations that cause human diseases. It will be of great interest to determine whether modulating this pathway could represent a therapeutic approach to human retinal vascular disease. 
Supplementary Materials
Acknowledgments
The authors thank Jeremy Nathans (The Johns Hopkins University, Baltimore, Maryland, United States) for providing FZD4 cDNA, members of the United Arab Emirates University Biochemistry Department for their helpful discussions, and Saeed Tariq for CFM training. RMM thanks the United Arab Emirates University for providing a fellowship grant. 
Supported by the United Arab Emirates University Grants 31M010, 31M092, and NP/13/22. 
Disclosure: R.M. Milhem, None; S. Ben-Salem, None; L. Al-Gazali, None; B.R. Ali, None 
References
Pendergast SD Trese MT Liu X Shastry BS. Study of the Norrie disease gene in 2 patients with bilateral persistent hyperplastic primary vitreous. Arch Ophthalmol . 1998; 116: 381–382. [PubMed]
Norrie G. Causes of blindness in children. Acta Ophthalmology . 1927; 5: 357–386. [CrossRef]
van Nouhuys CE. Signs, complications, and platelet aggregation in familial exudative vitreoretinopathy. Am J Ophthalmol . 1991; 111: 34–41. [CrossRef] [PubMed]
Kondo H Hayashi H Oshima K Tahira T Hayashi K. Frizzled 4 gene (FZD4) mutations in patients with familial exudative vitreoretinopathy with variable expressivity. Br J Ophthalmol . 2003; 87: 1291–1295. [CrossRef] [PubMed]
Zhang K Harada Y Wei X An essential role of the cysteine-rich domain of FZD4 in Norrin/Wnt signaling and familial exudative vitreoretinopathy. J Biol Chem . 2011; 286: 10210–10215. [CrossRef] [PubMed]
Ye X Wang Y Nathans J. The Norrin/Frizzled4 signaling pathway in retinal vascular development and disease. Trends Mol Med . 2010; 16: 417–425. [CrossRef] [PubMed]
Ye X Wang Y Cahill H Norrin, frizzled-4, and Lrp5 signaling in endothelial cells controls a genetic program for retinal vascularization. Cell . 2009; 139: 285–298. [CrossRef] [PubMed]
Smallwood PM Williams J Xu Q Leahy DJ Nathans J. Mutational analysis of Norrin-Frizzled4 recognition. J Biol Chem . 2007; 282: 4057–4068. [CrossRef] [PubMed]
Stenson PD Ball EV Mort M Human Gene Mutation Database (HGMD): 2003 update. Hum Mutat . 2003; 21: 577–581. [CrossRef] [PubMed]
Kondo H Qin M Tahira T Uchio E Hayashi K. Severe form of familial exudative vitreoretinopathy caused by homozygous R417Q mutation in frizzled-4 gene. Ophthalmic Genet . 2007; 28: 220–223. [CrossRef] [PubMed]
Lemma V D'Agostino M Caporaso MG A disorder-to-order structural transition in the COOH-tail of Fz4 determines misfolding of the L501fsX533-Fz4 mutant [serial online]. Sci Rep . 2013; 3: 2659. Available at: http://www.nature.com/srep/2013/130916/srep02659/full/srep02659.html. Accessed April 15, 2014.
Li P Zhang HZ Huff S Karyotype-phenotype insights from 11q14.1-q23.2 interstitial deletions: FZD4 haploinsufficiency and exudative vitreoretinopathy in a patient with a complex chromosome rearrangement. Am J Med Genet A . 2006; 140: 2721–2729. [CrossRef] [PubMed]
Ali BR Jeffery S Patel N Novel Robinow syndrome causing mutations in the proximal region of the frizzled-like domain of ROR2 are retained in the endoplasmic reticulum. Hum Genet . 2007; 122: 389–395. [CrossRef] [PubMed]
Aridor M. Visiting the ER: the endoplasmic reticulum as a target for therapeutics in traffic related diseases. Adv Drug Deliv Rev . 2007; 59: 759–781. [CrossRef] [PubMed]
Chen Y Bellamy WP Seabra MC Field MC Ali BR. ER-associated protein degradation is a common mechanism underpinning numerous monogenic diseases including Robinow syndrome. Hum Mol Genet . 2005; 14: 2559–2569. [CrossRef] [PubMed]
Ali BR Xu H Akawi NA Trafficking defects and loss of ligand binding are the underlying causes of all reported DDR2 missense mutations found in SMED-SL patients. Hum Mol Genet . 2010; 19: 2239–2250. [CrossRef] [PubMed]
MacDonald ML Goldberg YP Macfarlane J Samuels ME Trese MT Shastry BS. Genetic variants of frizzled-4 gene in familial exudative vitreoretinopathy and advanced retinopathy of prematurity. Clin Genet . 2005; 67: 363–366. [CrossRef] [PubMed]
Toomes C Bottomley HM Scott S Spectrum and frequency of FZD4 mutations in familial exudative vitreoretinopathy. Invest Ophthalmol Vis Sci . 2004; 45: 2083–2090. [CrossRef] [PubMed]
Omoto S Hayashi T Kitahara K Takeuchi T Ueoka Y. Autosomal dominant familial exudative vitreoretinopathy in two Japanese families with FZD4 mutations (H69Y and C181R). Ophthalmic Genet . 2004; 25: 81–90. [CrossRef] [PubMed]
Nikopoulos K Venselaar H Collin RW Overview of the mutation spectrum in familial exudative vitreoretinopathy and Norrie disease with identification of 21 novel variants in FZD4, LRP5, and NDP. Hum Mutat . 2010; 31: 656–666. [CrossRef] [PubMed]
Nallathambi J Shukla D Rajendran A Namperumalsamy P Muthulakshmi R Sundaresan P. Identification of novel FZD4 mutations in Indian patients with familial exudative vitreoretinopathy. Mol Vis . 2006; 12: 1086–1092. [PubMed]
Swerdlow PS Finley D Varshavsky A. Enhancement of immunoblot sensitivity by heating of hydrated filters. Anal Biochem . 1986; 156: 147–153. [CrossRef] [PubMed]
Kaykas A Yang-Snyder J Heroux M Shah KV Bouvier M Moon RT. Mutant Frizzled 4 associated with vitreoretinopathy traps wild-type Frizzled in the endoplasmic reticulum by oligomerization. Nat Cell Biol . 2004; 6: 52–58. [CrossRef] [PubMed]
Xu Q Wang Y Dabdoub A Vascular development in the retina and inner ear: control by Norrin and Frizzled-4, a high-affinity ligand-receptor pair. Cell . 2004; 116: 883–895. [CrossRef] [PubMed]
Robitaille J MacDonald ML Kaykas A Mutant frizzled-4 disrupts retinal angiogenesis in familial exudative vitreoretinopathy. Nat Genet . 2002; 32: 326–330. [CrossRef] [PubMed]
Amaral MD. Therapy through chaperones: sense or antisense? Cystic fibrosis as a model disease. J Inherit Metab Dis . 2006; 29: 477–487. [CrossRef] [PubMed]
Mendes F Wakefield J Bachhuber T Establishment and characterization of a novel polarized MDCK epithelial cellular model for CFTR studies. Cell Physiol Biochem . 2005; 16: 281–290. [CrossRef] [PubMed]
Devesa I Fernandez-Ballester G Ferrer-Montiel A. Targeting protein-protein interactions to rescue Deltaf508-cftr: a novel corrector approach to treat cystic fibrosis. EMBO Mol Med . 2013; 5: 1462–1464. [CrossRef] [PubMed]
Hanrahan JW Sampson HM Thomas DY. Novel pharmacological strategies to treat cystic fibrosis. Trends Pharmacol Sci . 2013; 34: 119–125. [CrossRef] [PubMed]
Rowe SM Verkman AS. Cystic fibrosis transmembrane regulator correctors and potentiators [ serial online]. Cold Spring Harb Perspect Med . 2013; 3. Available at: http://perspectivesinmedicine.cshlp.org/content/3/7/a009761.long. Accessed April 15, 2014.
Denning GM Anderson MP Amara JF Marshall J Smith AE Welsh MJ. Processing of mutant cystic fibrosis transmembrane conductance regulator is temperature-sensitive. Nature . 1992; 358: 761–764. [CrossRef] [PubMed]
Li C Ramjeesingh M Reyes E The cystic fibrosis mutation (delta F508) does not influence the chloride channel. Nat Genet . 1993; 3: 311–316. [CrossRef] [PubMed]
Brown CR Hong-Brown LQ Biwersi J Verkman AS Welch WJ. Chemical chaperones correct the mutant phenotype of the delta F508 cystic. Cell Stress Chaperones . 1996; 1: 117–125. [CrossRef] [PubMed]
Tachikawa M Kanagawa M Yu CC Kobayashi K Toda T. Mislocalization of fukutin protein by disease-causing missense mutations can be rescued with treatments directed at folding amelioration. J Biol Chem . 2012; 287: 8398–8406. [CrossRef] [PubMed]
Gautherot J Durand-Schneider AM Delautier D Effects of cellular, chemical, and pharmacological chaperones on the rescue of a trafficking-defective mutant of the ATP-binding cassette transporter proteins ABCB1/ABCB4. J Biol Chem . 2012; 287: 5070–5078. [CrossRef] [PubMed]
Guerriero CJ Brodsky JL. The delicate balance between secreted protein folding and endoplasmic reticulum–associated degradation in human physiology. Physiol Rev . 2012; 92: 537–576. [CrossRef] [PubMed]
Ellgaard L Helenius A. Quality control in the endoplasmic reticulum. Nat Rev Mol Cell Biol . 2003; 4: 181–191. [CrossRef] [PubMed]
Brodsky JL McCracken AA. ER protein quality control and proteasome-mediated protein degradation. Semin Cell Dev Biol . 1999; 10: 507–513. [CrossRef] [PubMed]
Kostova Z Wolf DH. For whom the bell tolls: protein quality control of the endoplasmic reticulum and the ubiquitin-proteasome connection. EMBO J . 2003; 22: 2309–2317. [CrossRef] [PubMed]
Plemper RK Wolf DH. Retrograde protein translocation: ERADication of secretory proteins in health and disease. Trends Biochem Sci . 1999; 24: 266–270. [CrossRef] [PubMed]
Sommer T Wolf DH. Endoplasmic reticulum degradation: reverse protein flow of no return. FASEB J . 1997; 11: 1227–1233. [PubMed]
Rajan S Eames SC Park SY In vitro processing and secretion of mutant insulin proteins that cause permanent [serial online]. Am J Physiol Endocrinol Metab . 2010; 298: E403–E410. Available at: http://ajpendo.physiology.org/content/298/3/E403.long. Accessed April 15, 2014. [CrossRef] [PubMed]
Dann CE Hsieh JC Rattner A Sharma D Nathans J Leahy DJ. Insights into Wnt binding and signalling from the structures of two Frizzled cysteine-rich domains. Nature . 2001; 412: 86–90. [CrossRef] [PubMed]
Ke J Harikumar KG Erice C Structure and function of Norrin in assembly and activation of a Frizzled 4-Lrp5/6 complex. Genes Dev . 2013; 27: 2305–2319. [CrossRef] [PubMed]
Carron C Pascal A Djiane A Boucaut JC Shi DL Umbhauer M. Frizzled receptor dimerization is sufficient to activate the Wnt/beta-catenin pathway. J Cell Sci . 2003; 116: 2541–2550. [CrossRef] [PubMed]
Thrower JS Hoffman L Rechsteiner M Pickart CM. Recognition of the polyubiquitin proteolytic signal. EMBO J . 2000; 19: 94–102. [CrossRef] [PubMed]
Nakatsukasa K Brodsky JL. The recognition and retrotranslocation of misfolded proteins from the endoplasmic reticulum. Traffic . 2008; 9: 861–870. [CrossRef] [PubMed]
Sato S Ward CL Krouse ME Wine JJ Kopito RR. Glycerol reverses the misfolding phenotype of the most common cystic fibrosis mutation. J Biol Chem . 1996; 271: 635–638. [CrossRef] [PubMed]
Robben JH Sze M Knoers NV Deen PM. Rescue of vasopressin V2 receptor mutants by chemical chaperones: specificity and mechanism. Mol Biol Cell . 2006; 17: 379–386. [CrossRef] [PubMed]
Egan ME Glockner-Pagel J Ambrose C Calcium-pump inhibitors induce functional surface expression of Delta F508-CFTR. Nat Med . 2002; 8: 485–492. [CrossRef] [PubMed]
Petersen TN Brunak S von Heijne G Nielsen H. SignalP 4.0: discriminating signal peptides from transmembrane regions. Nat Methods . 2011; 8: 785–786. [CrossRef] [PubMed]
Figure 1
 
Illustration of FZD4 protein domain structure. Frizzled family receptor 4 is a seven-pass transmembrane protein with a signal peptide sequence found between amino acid positions 1 and 36/37, a CRD-FZ domain highlighted in green found at amino acid positions 42 through 167, a frizzled region spanning most of the protein within amino acid positions 210 through 514, a KTXXXW motif found at amino acid positions 499 through 504, and a PDZ motif located close to the C-terminal at amino acid positions 535 through 537. The two potential N-glycosylation sites are indicated by arrows at amino acid positions 59 and 144. Missense FEVR-causing mutations studied in this article are labeled across the protein. Missense mutations that traffic abnormally are shown with red stars (P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D).
Figure 1
 
Illustration of FZD4 protein domain structure. Frizzled family receptor 4 is a seven-pass transmembrane protein with a signal peptide sequence found between amino acid positions 1 and 36/37, a CRD-FZ domain highlighted in green found at amino acid positions 42 through 167, a frizzled region spanning most of the protein within amino acid positions 210 through 514, a KTXXXW motif found at amino acid positions 499 through 504, and a PDZ motif located close to the C-terminal at amino acid positions 535 through 537. The two potential N-glycosylation sites are indicated by arrows at amino acid positions 59 and 144. Missense FEVR-causing mutations studied in this article are labeled across the protein. Missense mutations that traffic abnormally are shown with red stars (P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D).
Figure 2
 
Colocalization of several FZD4 missense mutants with the ER marker in HeLa cells. Confocal IF images of WT FZD4 and FZD4 mutants (P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D) show the colocalization of these mutants with calnexin, an ER marker. A FLAG epitope expressed WT FZD4 protein, and mutants are seen in red on the left of each panel, and calnexin is shown in green in the middle image of each panel. The image on the right is the merging of the two signals. (A) Shown is WT FZD4 localization to the plasma membrane, which is distinct from that of calnexin. In contrast, as shown in (BJ), the above missense mutants are largely localized to an intracellular compartment that is perinuclear and reticular in nature and colocalizing with calnexin, suggestive of an ER localization. However, C181R and M105V seem to be partially localized to the ER, with some localization to the PM. Cells were transfected for 48 to 72 hours and then processed for IF as described in the Methods section. (B) Colocalization with calnexin was measured as a function and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments (mean [SE] of each experiment). Data (arbitrary units) are expressed as the mean (SE). ***P < 0.001 compared with WT control. Scale bar: 25 μm. Magnification: ×100.
Figure 2
 
Colocalization of several FZD4 missense mutants with the ER marker in HeLa cells. Confocal IF images of WT FZD4 and FZD4 mutants (P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D) show the colocalization of these mutants with calnexin, an ER marker. A FLAG epitope expressed WT FZD4 protein, and mutants are seen in red on the left of each panel, and calnexin is shown in green in the middle image of each panel. The image on the right is the merging of the two signals. (A) Shown is WT FZD4 localization to the plasma membrane, which is distinct from that of calnexin. In contrast, as shown in (BJ), the above missense mutants are largely localized to an intracellular compartment that is perinuclear and reticular in nature and colocalizing with calnexin, suggestive of an ER localization. However, C181R and M105V seem to be partially localized to the ER, with some localization to the PM. Cells were transfected for 48 to 72 hours and then processed for IF as described in the Methods section. (B) Colocalization with calnexin was measured as a function and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments (mean [SE] of each experiment). Data (arbitrary units) are expressed as the mean (SE). ***P < 0.001 compared with WT control. Scale bar: 25 μm. Magnification: ×100.
Figure 3
 
Exclusion of the seven missense FZD4 mutants from the plasma membrane in HeLa cells. The CFM images of WT FZD4 and FZD4 mutants P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D (red in the images on the left side of each panel) show their localization in HeLa cells relative to HRAS, a plasma membrane marker (shown in the green image in the middle of each panel). The FLAG epitope–expressed WT FZD4 protein in (A) indicates colocalization, with HRAS indicating plasma membrane localization. However, (BJ) show that the mutants are largely excluded from the PM and localized to the perinuclear region, implying ER localization. C181R and M105V seem to be partially retained in the ER, with some PM localization, validating the data shown in Figure 2. Cells were transfected for 48 to 72 hours and then processed for IF. Scale bar: 25 μm. Magnification: ×100.
Figure 3
 
Exclusion of the seven missense FZD4 mutants from the plasma membrane in HeLa cells. The CFM images of WT FZD4 and FZD4 mutants P33S, G36N, H69Y, M105T, M105V, C181R, C204R, C204Y, and G488D (red in the images on the left side of each panel) show their localization in HeLa cells relative to HRAS, a plasma membrane marker (shown in the green image in the middle of each panel). The FLAG epitope–expressed WT FZD4 protein in (A) indicates colocalization, with HRAS indicating plasma membrane localization. However, (BJ) show that the mutants are largely excluded from the PM and localized to the perinuclear region, implying ER localization. C181R and M105V seem to be partially retained in the ER, with some PM localization, validating the data shown in Figure 2. Cells were transfected for 48 to 72 hours and then processed for IF. Scale bar: 25 μm. Magnification: ×100.
Figure 4
 
N-glycosylation profiles for the ER-retained mutants and WT FZD4. In whole-cell lysates (WCL), WT FZD4–expressed protein has been seen as a monomer, dimer, and multimers at the top of an SDS-PAGE gel (Supplementary Fig. S4). Wild-type FZD4 and mutants have been pulled down with mouse anti-FLAG antibodies, followed by sepharose A/G bead sedimentation as described in the Methods section. Upon elution, 20 μg protein was treated and then digested with Endo H for 5 hours with 60 mM dithiothreitol and 8 M urea (v/v) as described in the Methods section. P33S, M105T, and G488D mutants show an immature FZD4 protein, with lower-molecular-weight bands compared with WT FZD4 and its treated counterpart, suggestive of possible folding defects. The monomer's molecular weight of FZD4 is approximately 60 kDa. Mutants P33S, G36N, H69Y, M105T, C204R, C204Y, and G488D are quantitatively cleavable with Endo H, establishing their predominant localization in the ER. M105V and C181R have been shown to have a dual pattern by CFM. However, when the M105V mutant was subjected to Endo H treatment, a single lower-molecular-weight band was observed. On the other hand, two bands, one mature and one immature, were observed for C181R. Furthermore, P33S and G448D formed dimers and probably multimers on Endo H treatment at a temperature of 37°C. Molecular mass standards are indicated in kDa. A bar chart in (B) shows the percentages of Endo H–resistant and Endo H–sensitive species for each sample in (A). Wild-type is almost exclusively resistant to Endo H, whereas all the mutants (with the exception of C181R) are exclusively Endo H sensitive. Quantification was carried out by Typhoon FLA 9500 ImageQuant TL software (GE Healthcare Life Sciences).
Figure 4
 
N-glycosylation profiles for the ER-retained mutants and WT FZD4. In whole-cell lysates (WCL), WT FZD4–expressed protein has been seen as a monomer, dimer, and multimers at the top of an SDS-PAGE gel (Supplementary Fig. S4). Wild-type FZD4 and mutants have been pulled down with mouse anti-FLAG antibodies, followed by sepharose A/G bead sedimentation as described in the Methods section. Upon elution, 20 μg protein was treated and then digested with Endo H for 5 hours with 60 mM dithiothreitol and 8 M urea (v/v) as described in the Methods section. P33S, M105T, and G488D mutants show an immature FZD4 protein, with lower-molecular-weight bands compared with WT FZD4 and its treated counterpart, suggestive of possible folding defects. The monomer's molecular weight of FZD4 is approximately 60 kDa. Mutants P33S, G36N, H69Y, M105T, C204R, C204Y, and G488D are quantitatively cleavable with Endo H, establishing their predominant localization in the ER. M105V and C181R have been shown to have a dual pattern by CFM. However, when the M105V mutant was subjected to Endo H treatment, a single lower-molecular-weight band was observed. On the other hand, two bands, one mature and one immature, were observed for C181R. Furthermore, P33S and G448D formed dimers and probably multimers on Endo H treatment at a temperature of 37°C. Molecular mass standards are indicated in kDa. A bar chart in (B) shows the percentages of Endo H–resistant and Endo H–sensitive species for each sample in (A). Wild-type is almost exclusively resistant to Endo H, whereas all the mutants (with the exception of C181R) are exclusively Endo H sensitive. Quantification was carried out by Typhoon FLA 9500 ImageQuant TL software (GE Healthcare Life Sciences).
Figure 5
 
Polyubiquitination pattern of WT FZD4 and ER-retained FZD4 mutants. (A) Immunoprecipitations of WT FZD4, P33S, G36N, H69Y, M105T, C204Y, C204R, and G488D mutants were run on SDS-PAGE, blotted on nitrocellulose, and probed with anti-FLAG antibodies. The blots were then washed overnight with TBS-T, stripped, autoclaved, and then reprobed with anti-Ub monoclonal antibodies as described in the Methods section. The samples of the seven ER-retained mutants show very-high-molecular-weight smears, suggesting polyubiquitination. Wild-type FZD4 also seems to be ubiquitinated, albeit to a much lesser extent. (B) Shown is an immunoblot of IP WT FZD4, P33S, G36N, H69Y, M105T, C204Y, C204R, and G488D with anti-FLAG antibody before stripping for polyubiquitination studies. (C) Quantification of the polyubiquitination blot presented above is shown. Considerably higher levels of ubiquitination are observed for the mutants compared with WT. A negative IP is presented for comparison. Quantification was carried out by Typhoon FLA 9500 ImageQuant TL software (GE Healthcare Life Sciences).
Figure 5
 
Polyubiquitination pattern of WT FZD4 and ER-retained FZD4 mutants. (A) Immunoprecipitations of WT FZD4, P33S, G36N, H69Y, M105T, C204Y, C204R, and G488D mutants were run on SDS-PAGE, blotted on nitrocellulose, and probed with anti-FLAG antibodies. The blots were then washed overnight with TBS-T, stripped, autoclaved, and then reprobed with anti-Ub monoclonal antibodies as described in the Methods section. The samples of the seven ER-retained mutants show very-high-molecular-weight smears, suggesting polyubiquitination. Wild-type FZD4 also seems to be ubiquitinated, albeit to a much lesser extent. (B) Shown is an immunoblot of IP WT FZD4, P33S, G36N, H69Y, M105T, C204Y, C204R, and G488D with anti-FLAG antibody before stripping for polyubiquitination studies. (C) Quantification of the polyubiquitination blot presented above is shown. Considerably higher levels of ubiquitination are observed for the mutants compared with WT. A negative IP is presented for comparison. Quantification was carried out by Typhoon FLA 9500 ImageQuant TL software (GE Healthcare Life Sciences).
Figure 6
 
Effect of low-temperature incubation and chemical compounds on M105T and C204Y localization. (A) Shown are CFM images of reduced temperature and chemical compound incubation of WT FZD4 acting as a control. (B, C) Shown are CFM images of M105T and C204Y mutants across the three treatments (culturing at 27°C, 7.5% glycerol, and 0.1% DMSO, respectively). (D) Using ImageJ version 1.47, quantification of colocalization with the ER marker (expressed as percentage) of the IF pattern of the expressed FZD4 proteins is shown. At 27°C, WT colocalized with the ER marker by approximately 3% only, whereas M105T and C204Y colocalized by 43% and C204Y by 54%. With 7.5% glycerol, M105T colocalized by 50% and C204Y by 85%. With 0.1% DMSO, M105T colocalized with calnexin by 32% and C204Y by 95%. Colocalization with calnexin was measured as a function of and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments (mean [SE] of each experiment). (E) Shown is quantification of the amount of FZD4 proteins colocalizing with HRAS, a PM marker. At 27°C, M105T and C204Y colocalized by 57% and 46%, respectively, whereas in the presence of 7.5% glycerol, M105T colocalized by 50% and C204Y by 32%. The M105T mutant showed partial plasma membrane distribution when treated with 0.1% DMSO by approximately 68% and C204Y by approximately 5%. One microgram WT FZD4 or the mutant constructs was used for transfection experiments. For coexpression studies, 1 μg for each construct was cotransfected together. Colocalization with HRAS was measured as a function of and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments. Scale bar: 25 μm. Data (arbitrary units) are expressed as the mean (SE). ***P < 0.001.
Figure 6
 
Effect of low-temperature incubation and chemical compounds on M105T and C204Y localization. (A) Shown are CFM images of reduced temperature and chemical compound incubation of WT FZD4 acting as a control. (B, C) Shown are CFM images of M105T and C204Y mutants across the three treatments (culturing at 27°C, 7.5% glycerol, and 0.1% DMSO, respectively). (D) Using ImageJ version 1.47, quantification of colocalization with the ER marker (expressed as percentage) of the IF pattern of the expressed FZD4 proteins is shown. At 27°C, WT colocalized with the ER marker by approximately 3% only, whereas M105T and C204Y colocalized by 43% and C204Y by 54%. With 7.5% glycerol, M105T colocalized by 50% and C204Y by 85%. With 0.1% DMSO, M105T colocalized with calnexin by 32% and C204Y by 95%. Colocalization with calnexin was measured as a function of and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments (mean [SE] of each experiment). (E) Shown is quantification of the amount of FZD4 proteins colocalizing with HRAS, a PM marker. At 27°C, M105T and C204Y colocalized by 57% and 46%, respectively, whereas in the presence of 7.5% glycerol, M105T colocalized by 50% and C204Y by 32%. The M105T mutant showed partial plasma membrane distribution when treated with 0.1% DMSO by approximately 68% and C204Y by approximately 5%. One microgram WT FZD4 or the mutant constructs was used for transfection experiments. For coexpression studies, 1 μg for each construct was cotransfected together. Colocalization with HRAS was measured as a function of and then compared with WT on multiple CFM image sections of at least 50 cells in three independent experiments. Scale bar: 25 μm. Data (arbitrary units) are expressed as the mean (SE). ***P < 0.001.
Table
 
Summary of the Physicochemical Properties of the FEVR-Causing FZD4 Mutations Shown in This Study to Be Mislocalized to the ER
Table
 
Summary of the Physicochemical Properties of the FEVR-Causing FZD4 Mutations Shown in This Study to Be Mislocalized to the ER
Mutation PolyPhen-2 SIFT Charge Polarity Hydrophobicity Size Score
P33S None 0.47, T Equal Increase Increase Decrease 23
G36N 0.004, N 0.62, T Equal Increase Decrease Increase 32
H69Y 0.029, N 0.08, T Decrease Equal Increase Equal 49
M105T 0.997, D 0.47, T Equal Increase Decrease Decrease 32
M105V  0.66, T 0.93, D Equal Equal Equal Decrease 16
C181R 0.033, N 0.52, T Increase Increase Decrease Increase 77
C204R 1, D 0.02, D Increase Increase Decrease Increase 77
C204Y 1, D 0.04, D Equal Increase Equal Increase 85
G488D 0.06, T 1, D Increase Increase Decrease Increase 28
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×