Bivalent Copper Ions Promote Fibrillar Aggregation of KCTD1 and Induce Cytotoxicity

Potassium channel tetramerization domain containing 1 (KCTD1) family members have a BTB/POZ domain, which can facilitate protein-protein interactions involved in the regulation of different signaling pathways. KCTD proteins have potential Zn2+/Cu2+ binding sites with currently unknown structural and functional roles. We investigated potential Cu2+-specific effects on KCTD1 using circular dichroism, turbidity measurement, fluorescent dye binding, proteinase K (PK) digestion, cell proliferation and apoptosis assays. These experiments indicate that the KCTD1 secondary structure assumes greater β-sheet content and the proteins aggregate into a PK-resistant form under 20 μM Cu2+, and this β-sheet-rich aggregation with Cu2+ promotes fibril formation, which results in increased cell toxicity by apoptosis. Our results reveal a novel role for Cu2+ in determining the structure and function of KCTD1.


Results
Cu 2+ alters the structure of KCTD1. Approximately 24% of all proteins require metal ions for normal function 21 . KCTD1 contains a highly conserved BTB/POZ domain that binds metal ions. We investigated whether metal ions affect KCTD1 structure by monitoring the far-UV CD spectra of purified KCTD1 in the presence or absence of Cu 2+ /Zn 2+ . The secondary structure of purified KCTD1 in the absence of metal ions had spectral features typical of a protein containing significant alpha-helical structure, with well-defined minima at 208 and 222 nm. After 15 min incubation with varying Cu 2+ concentrations, the troughs around 208 and 222 nm were reduced, which is indicative of an increase in β -sheet structure (Fig. 1a). However, the KCTD1 secondary structure did not change in the presence of equivalent Zn 2+ concentrations (Fig. 1b). Therefore, we focused on determining how Cu 2+ affects KCTD1 structure and function.
To further characterize Cu 2+ -mediated changes in KCTD1 structure, we used 8-anilino-1-naphthalenesulfonic acid, a naphthalene derivative that undergoes an increase in fluorescence emission on interaction with hydrophobic sites. The bis-ANS fluorescence emission was significantly higher in the sample containing Cu 2+ than in control samples without Cu 2+ , and the increase was proportional to the Cu 2+ concentration (Fig. 2a,b). These results were consistent with those of CD analysis and indicated that Cu 2+ increased KCTD1 hydrophobicity. Cu 2+ triggers KCTD1 aggregation. The fluorescence emission analysis showed that the interaction of Cu 2+ with KCTD1 exposed a hydrophobic surface on the protein, which could promote protein aggregation. To test this, we measured the turbidity of KCTD1 after incubating at 37 °C for up to 80 min with and without Cu 2+ by a all-in-one Microplate Reader at OD420. The results indicated that few KCTD1 oligomers were present at the beginning of the incubation in PBS containing 0 or 20 μ M CuSO 4 . The turbidity increased over time during the 80 min incubation, and the amount of aggregates increased with increasing KCTD1 concentrations (Fig. 3a). Importantly, Cu 2+ enhanced protein aggregation and increased turbidity (Fig. 3a,b). Cu 2+ causes the formation of fibrillar KCTD1 aggregates. To investigate the nature of Cu 2+ -mediated KCTD1 aggregation, we evaluated binding of the common dye ThT to KCTD1. The intensity of ThT fluorescence emission (around 485 nm) drastically increases on binding to fibrils containing β -sheet structures, such as amyloid fibrils 22 . Therefore, we utilized ThT to detect the formation of KCTD1 fibrils in the presence or absence of 20 μ M CuSO 4 . There was a remarkable difference in ThT fluorescence intensity in samples with and without Cu 2+ (Fig. 4a,b). ThT fluorescence intensity at peak point was much higher for KCTD1 in the presence of Cu 2+ than that without Cu 2+ (Fig. 4b), which was similar to that observed for amyloids (Fig. 4a). This result indicated that Cu 2+ caused an increase in β -sheet formation, which caused KCTD1 aggregation and fibrillar formation.
Next, we analyzed KCTD1 morphology in the presence or absence of Cu 2+ by electron microscopy. In the absence of Cu 2+ , protein aggregates that formed after 30 min incubation were amorphous with irregular shapes (Fig. 5b,c), whereas no aggregates were detected before the incubation (Fig. 5a). By contrast, KCTD1 in the presence of Cu 2+ formed numerous fibrils of variable lengths (Fig. 5d,e). These results indicated that Cu 2+ induced the formation of KCTD1 fibrils. Cu 2+ -induced KCTD1 aggregates are more resistant to PK digestion. We determined whether KCTD1 aggregates could be cleaved by proteinase K. KCTD1 or BSA control were incubated with or without 20 μM CuSO 4 for 2 hours at 37 °C. The protein samples were then digested with protease K for 35 min and quenched by 2 mM PMSF. And then the treated samples were analyzed by SDS-PAGE gel. It is distinctly different in Fig. 6a,b that the KCTD1 with Cu 2+ had obvious the monomer and polymer; On the contrary, the polymer was indiscernible in the gel of the KCTD1 free Cu 2+ . KCTD1 fibrillar aggregates formed in the presence of Cu 2+ were resistant to SDS-dissolution and PK digestion under a range of PK concentrations (2.5, 5, 10, and 20 μ g/ml) (Fig. 6b), whereas KCTD1 aggregates formed in the absence of Cu 2+ were no SDS-resistant and easily digested by PK in a concentration-dependent manner (Fig. 6a). Although, BSA control samples were treated in the same condition as KCTD1, however, the BSA control was digested by PK irrespective of the presence or absence of Cu 2+ , after digestion, the same bands were observed in the SDS-PAGE gel (Fig. 6c,d). Although PK molecular weight was about 29 KD, which makes it diffcult to distinguish with KCTD1 (molecular weight about 29.4 KD) in the SDS-PAGE gel, Fig. 6e showed the different concentrations PK (2.5, 5, 10, and 20 μ g/ml) had no band from 25 to 35 KD in SDS-PAGE gel. And it also indicated the PK at experimental concentrations didn't affect the bands of KCTD1 digestion.
KCTD1 aggregation in the presence of Cu 2+ is more cytotoxic to cell. One mechanism for pathogenesis resulting from protein misfolding is due to cytotoxicity. To determine the effect of Cu 2+ -induced KCTD1  misfolding on cell viability, we performed the CCK-8 assay using the human HuH7 cell line. The addition of KCTD1 fibrillar aggregates formed in the presence of 20 μ M CuSO 4 to HuH7 cells caused much more severe cytotoxicity than that caused by KCTD1 aggregates formed without Cu 2+ (Fig. 7b). To exclude possible Cu 2+ alone effect on cell, the cytotoxicity of buffer with or without 20 μ M CuSO 4 on HuH7 cells was monitored with the same method, and the result indicated that the cytotoxicity had no relationship with Cu 2+ (Fig. 7a). The CCK-8 result indicated KCTD1 fibrillar aggregates could decrease cell proliferation strongly, then we considered whether the cell cytotoxic  was affected by the cell apoptosis. We performed the Hoechst33258 staining on HuH7 cells and HCT116 which were treated by KCTD1 protein with or without 20 μ M CuSO 4 . The result showed that the KCTD1 fibrillar aggregates with 20 μ M CuSO 4 induced more severe apoptosis than that induced by KCTD1 aggregates formed free CuSO 4 , and apoptosis increased in a concentration-dependent manner ( Fig. 8a-b,f-g). What's more, to further obseve the apoptotic cells induced by different treated KCTD1 protein, Annexin-V and PI dual stained HuH7 and HCT116 cells treated with or without 20 μ M Cu 2+ were analyzed by Flow Cytometry, the same results as Hochest33258 have validated that 20 μ M Cu 2+ induced fibrillar aggregates of KCTD1 surely contributed to no matter cell early apoptosis or cell total apoptosis (early apoptosis and late apoptosis) on HuH7 cell lines (Fig. 8c-e)and HCT116 cells (Fig. 8h-j). The cell cytotoxic results suggested that Cu 2+ affected not only KCTD1 structure, but also its function.

Discussion
Protein misfolding and aggregation is associated with a wide range of diseases. Several degenerative diseases are associated with prion-like proteins with similar properties as mammalian prion proteins. Although prion diseases are rare, the incidence of prion-like pathologies is high. Metal ions significantly enhance protein misfolding and aggregation, and are directly linked with protein misfolding diseases. Copper ions strongly promote amyloid formation in a variety of proteins. For example, Cu 2+ binding causes structural changes in prion proteins and has been implicated in prion misfolding and fibril formation 5,23 . Proteins self-assemble under increasing Cu 2+ ion concentrations, resulting in the formation of amyloid fibrils that are cytotoxic 12 .
This study showed that increasing Cu 2+ concentrations enhanced KCTD1 aggregation. KCTD1 aggregation was first determined by performing turbidity assays. Fluorescence emission assays using ANS and far-UV CD studies detected an increase in β -sheet structure that was dependent on KCTD1 and Cu 2+ concentration. The KCTD1 structural change indicated that Cu 2+ enhanced exposure of hydrophobic surface. The ThT spectra suggested that increasing Cu 2+ concentrations induced formation of fibrillar KCTD1 structures, which was confirmed by transmission electron micrographs.
KCTD1 contains highly conserved BTB domains that mediate protein-protein interactions. KCTD may act as an adaptor for the interactions between the Cul3 ubiquitin ligase, which is involved in protein ubiquitination/ degradation, and its substrates. Zhang et al. confirmed KCTD1 interacted with beta-catenin and enhanced the ubiquitination/degradation of beta-catenin in a concentration-dependent manner in HeLa cells. And our previous study verified that KCTD1 interacted with PrPC and suggested that KCTD1 could mediate PrP ubiquitination and degradation. Copper accumulates in amyloid plaques in neurological disorders 24 , and KCTD1 is strongly expressed in brain cells 20 . In vitro studies, excessive Cu 2+ would affect KCTD1 structure and lead to KCTD1 aggregation, which might be cause disease. Our study identified a potential mechanism for Cu 2+ -mediated effects on human physiology and pathology through the formation of KCTD1 fibrils.

Materials and Methods
Cloning. Construction of the expression vector pET30a-KCTD1 and protein purification were performed as described previously 25   (d,i) The early apoptosis ratio and (e,j) total apotosis ratio (early apoptosis and late apoptosis) was quantified by Flow J software. The data shown are the mean ± s.d. of 3 independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001; t-test.
performed with a 1-mm path-length cuvette at 1-nm intervals. Multiple scans were averaged and the background spectra (buffer alone) were subtracted to produce the collated spectrum. The ellipticity of the far-UV CD spectrum was analyzed using DICROPROT software (ICBP, CNRS, Lyon University, Lyon, France).

Fluorescent dye binding and emission intensity assays. The 5 mM thioflavin T (ThT; Sigma-Aldrich,
St. Louis, MO, USA) and 8-anilino-1-naphthalenesulfonic acid (ANS; Sigma-Aldrich) stock solutions were prepared in PBS and filtered with 0.22 μ m filters (Merck Millipore, Germany). KCTD1 sample was incubated for 2 h at 37 °C with or without CuSO 4 in 20 μ M PBS. Then, the samples were incubated with either 10 μ M ThT or 5 μ M ANS in the dark for 10 min at room temperature before measurements respectively. The emission spectra of ThT and ANS were recorded after excitation at 443 nm (excitation and emission slit widths, 5 nm) and 380 nm (excitation and emission slit widths, 10 nm) by Fluorescence spectrophotometer F-4500 (HITACHI, Japan), respectively. Each emission spectrum was the average of three scans. Background fluorescence was subtracted from the results. Turbidity measurement. Turbidity was measured to assess protein aggregation. Turbidity measurements were performed at 37 °C in flat-bottomed 96-well plates. The KCTD1 sample in PBS (pH 7.0) was first centrifuged at 12,000 rpm for 5 min at 4 °C, and the supernatant was transferred into a new tube. Then, 200 μ l PBS (pH 7.0) containing different KCTD1 concentrations (0.156-5.0 μ g/μ l) was added to each well in a 96-well plate, and 2 mM CuSO 4 was added to the sample so the final CuSO 4 concentration was 20 μ M. The samples were mixed immediately in the wells. The sample turbidities were monitored within 5 min by reading the absorbance at 420 nm in a Synergy H1 microplate reader (BioTek, USA) using a kinetic photometric assay protocol (interval time 5 min, 16 cycles with 1 s shaking before every cycle).

Transmission Electron Microscopy.
Electron micrographs were acquired with H-8100 transmission electron microscope (Hitachi, Tokyo, Japan) operated at an acceleration voltage of 100 kV. The aggregation of KCTD1 with or without 20 μ M CuSO 4 was prepared. The protein samples were placed on a carbon-coated copper grid and negatively stained with 2% aqueous phosphotungstic acid for 5 min.
Proteinase K digestion. Samples were prepared for PK digestion as follows: 0.8 μ g/μ l KCTD1 with or without 20 μ M CuSO 4 was heated to 37 °C for 2 h in 20 μ M PBS (pH 7.0), then the solution was equilibrated at room temperature (25 °C). 0.8 μ g/μ l BSA samples (as controls) with or without 20 μ M CuSO 4 were dealed with in the same way. The samples were digested with various concentrations of PK (0-20 μ g/ml) for 35 min at 37 °C. Digestion was quenched with 2 mM phenylmethanesulfonyl fluoride (PMSF). Protein digestion was evaluated by SDS-PAGE with a 15% polyacrylamide gel in Tris-glycine buffer (pH 8.3) containing 0.1% (w/v) SDS, followed by staining with Coomassie Brilliant Blue.
Cell cultures. Human Huh7 cells obtained from China Center for Type Culture Collection (CCTCC) were cultured in minimum essential medium (Hyclone, USA) supplemented with 10% fetal bovine serum (Gibco, USA). Cultured cells were grown at 37 °C in a humidified atmosphere containing 5% CO 2 .
Cell proliferation assay. Cells were plated onto a 96-well plate at a density of 4,000 cells per well in 200 μ l of fresh minimum essential medium containing 10% fetal bovine serum and incubated for 24 h at 37 °C in a humidified atmosphere containing 5% CO 2 . 0.8 μ g/μ l KCTD1 with or without 20 μ M CuSO 4 was heated to 37 °C for 2 h in 20 μ M PBS (pH 7.0), Then, 50 μ l aliquots of the KCTD1 samples were transferred to wells containing 150 μ l fresh Dulbecco's modified Eagle medium (DMEM) with fetal bovine serum. the control sample were 50 μ l PBS (20 μ M, pH 7.0). Cell viability of these cultures and controls were evaluated 24 or 48 h later using the cell counting kit-8 (CCK-8) reduction assay. Briefly, 10 μ l of CCK-8 (Beyotime Institute of Biotechnology, Haimen, Jiangsu, China) was added to each well and incubated for 1 h at 37 °C. The production of blue formazan crystals was determined by spectrophotometric measurement of absorbance values at 450 nm.
Cell apoptosis assay. Place a sterile coverslip on the 6 well plate, and seed Huh7 cells or HCT116 on the plate to 80% confluency. Different concentrational KCTD1 (0, 0.8, 1.6 μ g/ μ l) with or without 20 μ M CuSO 4 was heated to 37 °C for 2 h in 20 μ M PBS (pH 7.0), Then, 500 μ l aliquots of the KCTD1 samples were transferred to wells containing 1500 μ l fresh Dulbecco's modified Eagle medium (DMEM) with fetal bovine serum. After culturing for 36 hours, the cells were subsequently fixed, washed three times with phosphate buffered saline (PBS), and stained with Hoechst 33258 (Sigma-Aldrich) according to the manufacturer's protocol. Apoptotic features were assessed by analyzing chromatin condensation and by staining the fragments under an inverted fluorescent microscope (Olympus, Tokyo, Japan).
Flow Cytometry Assay. Flow cytometry assay was also used to quantify the survival, necrosis and apoptosis of HuH7 and HCT116 cell line. And the cells were treated as previously described as Hochest33258, After culturing for 36 hours, the cells were collected, washed with PBS for three times. And the cells were then filtered with FACS used filter method, then Annexin V-FITC and PI (BD Pharmingin, USA) were used to stain the cells according to manufacture instruction. The cells were then analyzed by FAC-Scan flow cytometry (BD FACSAriaIII, USA), the apoptosis ratio was quantified by Flow J software. The above experiment was repeated three times.