Design and site-directed compartmentalization of gold nanoclusters within the intrasubunit interfaces of ferritin nanocage
- 172 Downloads
Protein nanocages have emerged as popular nanocarriers for either drug delivery or biotemplates for the preparation of nanomaterials. However, only three interfaces, namely exterior surface, intersubunit and inner cavity, have been used as reaction sites for the above purposes with all known protein nanocages. On the other hand, how to control the site of Au NCs formed within a targeted protein template while maintaining the functionality of protein itself remains challenging.
In this work, inspired by compartmentalization in living systems, we firstly come up with the conception of “intrasubunit interfaces”, located within subunit of protein nanocage. We built a new, specific compartment for fabrication of gold nanoclusters by genetic modification of the inherent ferroxidase center located within four-α-helix bundle of each ferritin subunit. This newly built compartment not only realizes the site-directed synthesis of gold nanoclusters but also has no effect on the functionality of ferritin itself such as encapsulation by its inner cavity. These redesigned composites can be further applied as fluorescent imaging agent and carriers for preparation of hybrid nanomaterials.
The designing strategy of intrasubunit interfaces opens a new way for future applications of cage-like proteins.
KeywordsCompartmentalization Intrasubunit interface Interface redesign Gold nanoclusters Mitochondrial ferritin
Among these protein nanocages, the ferritin nanocage is unique in that it exhibits high selectivity for human cancer cells which overexpress two kinds of receptors for ferritin [9, 10]. Thus, it has emerged as a class of drug delivery vehicles and imaging agents. Ferritin ubiquitously occurs in animals, plants and bacteria, and plays a significant role in iron storage and detoxification . It is a nearly spherical 24-subunits-protein with exterior diameter about 12 nm and a hollow cavity of 8 nm . Owing to its cage-like morphology and highly symmetrical structure, ferritin displays outstanding thermal and chemical stability and it is remarkably amenable to pH induced dis- and re-assembly [13, 14]. Like other protein nanocages, ferritins also have four chemically distinct interfaces where can be manipulated in order to impart functionality by design (Fig. 1a). The intersubunit interfaces in ferritin consist of C2, C3, C4, and C3–C4 as shown in Additional file 1: Figure S1. Among these intersubunit interfaces, C3 and C4 are mainly responsible for diffusion of small molecules into ferritin cavity [15, 16], while C2 plays a key role in protein shell assembly . The C3–C4 is the most largest intersubunit interface in ferritin, which controls protein geometry based on our recent studies [18, 19]. To extend the utilization of ferritin nanocage, protein exterior surface has been frequently modified so that it can be endowed with new functionalities such as specific targeting ability or luminescence [20, 21, 22]. As popular as the exterior surface, the inner cavity of ferritin has been explored to encapsulate various cargoes such as simple metal ions and complicated drugs [23, 24, 25]. However, so far, there have been no reports focusing on the intrasubunit interface of protein nanocages including ferritin prior to this study.
On the other hand, gold nanoclusters (Au NCs) are being actively pursued as a novel kind of fluorescent materials due to their potentially low toxicity, well-defined compositions, bright luminescence and facile surface functionalization [26, 27, 28]. After bovine serum albumin (BSA) was first exploited to be embedded with Au NCs in 2009 , substantial proteins have been used as biotemplates for Au NCs synthesis [30, 31, 32, 33]. However, how to control the site and size of Au NCs formed within a targeted protein template while maintaining the functionality of protein itself remains challenging. Herein, we take the compartmentalization concept to re-design the intrasubunit interface of ferritin to create a new compartment for fabrication of fluorescent Au NCs, while keeping the inherent protein cavity functional. Consequently, we can incorporate binary nanomaterials into one ferritin nanocage, namely small Au NCs buried within the intrasubunit and large inorganic or organic nanoparticles (4–6 nm) encapsulated within the cavity, which provide a new way for construction of hybrid nanomaterails.
Materials and methods
cDNA encoding the amino acid sequence of mature HuFtMt was cloned into the pET-3a (Novagen) and verified by DNA sequencing. Mutagenesis of the HuFtMt cDNA was performed with the fast site-directed mutagenesis kit from Tiangen Biotech (Beijing, China). The mutated HuFtMt and wild-type HuFtMt were purified as follows. The E. coli strain BL21(DE3) which contained the expression plasmid was grown at 37 °C. Protein expression was likewise induced with 200 μM of isopropyl β-d-1-thiogalactopyranoside after the cell density reached an absorbance of 0.6 at 600 nm. Then the cells were harvested by centrifugation after 8 h of induction and resuspended in 25 mM Tris–HCl (pH 8.0) to a concentration of 10 g fresh weight bacteria per liter, followed by disruption by sonication. The supernatant of the resulting crude extract was collected by centrifugation and fractionated by 60% saturation of ammonium sulfate. The pellet was resuspended in 25 mM Tris–HCl (pH 8.0) and dialyzed against the same buffer. The protein solution was applied to an ion-exchange column, followed by gradient elution with 0–0.3 M NaCl. Finally, the protein solution was concentrated and purified on a gel filtration column, equilibrated with 25 mM Tris–HCl and 150 mM NaCl (pH 8.0). Protein concentrations were determined according to the Lowry method with bovine serum albumin as standard.
Synthesis of ferritin-stablized Au NCs
Ferritin stablized Au NCs were synthesized as follows. 200 μL of 10 mM aqueous HAuCl4 solution (pH value tuned to 7.0) was added to a 200 μL Tris–HCl solution containing protein (50 mg/mL) with vigorous stirring at 37 °C. 5 min later, pH value was adjusted to around 12.8 with further vigorous stirring for 20 min. After that, the mixture solution was incubated at 37 °C overnight. To increase the storage stability while terminating the reaction, 1.0 M acetic acid was added drop-wise until the pH value reached ~ 7.0. Resulting products were stored in dark at 4 °C for further use.
Characterization of ferritin mutants and ferritin-stabilized Au NCs
UV–vis absorption spectra was recorded by a UV Spectrophotometer (Virian, 50 Bio, USA), and fluorescence studies in a sealed cuvette were carried out with a Fluorescence Spectrometer instrument (Virian, Palo Alto, USA). Fluorescence images were obtained under UV-light. Transmission electron microscope (TEM) images were obtained at 80 kV using a Hitachi S-5500 microscope. High-resolution transmission electron microscope (HRTEM) images were made by a JEM 2100F microscope, and elementary analysis was carried out with Horiba INCA 450 energy dispersive X-ray analysis spectroscopy. The oxidation state of the Au clusters was determined by X-ray Photoelectron Spectroscopy (XPS). Narrow-scan XPS spectra of Au 4f were deconvoluted by the XPSPEAK software (Version 4.1) using adventitious carbon to calibrate the binding energy of C1s (284.5 eV).
Crystallization, data collection and structure determination
Purified proteins were concentrated to 10 mg/mL in a buffer consisting of 10 mM Tris–HCl at pH 7.6 and 1 mM HAuCl4. Crystals of wild-type HuFtMt and ΔC complexed with Au3+ were obtained using the hanging drop vapor diffusion method by mixing equal volumes of the complex sample and mother liquid, which was composed of 0.1 M Bicine-NaOH (pH 8.5) and 2 M MgCl2. Crystals of 2Cys-ΔC and 3Cys-ΔC complexed with Au3+ were obtained using the hanging drop vapor diffusion method by mixing equal volumes of the complex sample and mother liquid, which was composed of 0.1 M Tris–HCl (pH 8.5) and 3.4 M 1,6-hexalene glycol. Diffraction data of the crystal were collected at SSRF to resolutions of after flash cooling with 25% glycerol as a cryo-protectant. Data were processed, merged and scaled with the HKL-3000 (HKL Research). Data processing statistics are shown in Additional file 1: Table S1. The data was determined by molecular replacement using coordinates of wild-type Mfn (PDB code 1R03) as an initial model using the MOLREP program in the CCP4 program package. Structural refinement was conducted using the Refmac 5 program and PHENIX software. The structure was rebuilt using COOT, which made the model manually adjusted. Figures of protein structures were prepared using the PyMOL program.
In vitro and in vivo bioimaging studies
HepG2 cells were cultured in DMEM supplemented with 10% (vol/vol) FBS, 2 mM l-glutamine and 1% penicillin/streptomycin and incubated at 37 °C with 5% CO2. Briefly, cells were seeded at a density of 3 × 105 on the cell culture dish and pre-incubated for 24 h. Then the cells were treated with RGD decorated 3Cys-ΔC and 3Cys-ΔC-Au NCs for 3 h at 37 °C with 5% CO2 and then washed by PBS three times. After that, cells were fixed by 4% paraformaldehyde for 20 min at room temperature. Last, The confocal laser scanning microscopy was used to image the cells.
Wild-type C. elegans were cultured in the nematode growth medium (NGM) agar plates containing E. coli OP50 bacteria (OP50 medium) as food supplement at 20 °C. L4 stage worms were seeded in the NGM agar plates supplemented with 3Cys-ΔC as negative group and with 3Cys-ΔC-Au NCs as experimental groups. After cultivation for different times, worms were transferred onto microslide smeared with circled albolene and centered 20 mM NaN3 solution to fix the worms and then observed by laser-scanning confocal microscopy. The images of the 3Cys-ΔC-stabilized Au NCs treated sample and negative control samples were collected in the range of 575–620 nm with 488 nm as excitation wavelength.
Iron cores formed within 3Cys-ΔC-stabilized Au NCs
Iron cores within 3Cys-ΔC-stabilized Au NCs were prepared by aerobically adding 500 Fe(II) per protein shell to 3Cys-ΔC-Au NCs solutions in five successive increments of 100 Fe(II) at intervals of 30 min in 50 mM MOPS at pH 7.0, followed by standing overnight. After iron loading, the sample was incubated at 4 °C in dark overnight because the iron mineralization was slower with ferritin mutants where the structure of ferroxidase center had been greatly altered.
Preparation of curcumin-loaded 3Cys-ΔC-Au NCs
Curcumin-loaded 3Cys-ΔC-Au NCs were prepared as previously reported with curcumin encapsulation by human H-chain ferritin. Briefly, for a 1 mL system of alkaline as-prepared 3Cys-ΔC-Au NCs (1 mg/mL in 0.15 M NaCl), a 200 μL solution of curcumin, freshly solubilized in 0.1 M sodium hydroxide (2 mM), was added. Immediately, the pH value was lowered to 7.5 using 1 M HCl. Resulting solution was stirred at room temperature for 2 h to promote the assembly of the protein. Finally resultant products were run through a NAP-5 column to remove free curcumin.
Thermal stability and photostability measurements
Both of curcumin encapsulated 3Cys-ΔC-Au NCs and free curcumin were heated at 60 °C in a dark tube. The degradation kinetic of curcumin was carried out by scanning UV–vis spectra from 250 to 800 nm at interval times. In the photostability experiment, the sample for photo degradation studies was placed under an incandescent lamp of 18 W/YH, at a distance of 10 cm at 25 °C. the degradation kinetic of curcumin was carried out by also scanning UV–vis spectra from 250 to 800 nm at interval times. As curcumin is poorly soluble in water, a solution of curcumin dissolved in ethanol at 10 mM was used as stock solution and then diluted to same intensity at ~ 425 nm with curcumin encapsulated 3Cys-ΔC-Au NCs to act as a control sample.
Results and discussion
Design of ferritin-templated Au NCs
In this work, we selected the relatively rare reported human mitochondrial ferritin (HuFtMt) nanocage as the protein model based on the following considerations: (1) it is a naturally occurring homopolymer which can simplify the design challenge; (2) its sequence and crystal structural information are available to guide genetic manipulations; and (3) it could be a promising vehicle for tumor imaging and drug delivery due to its nonimmunogenicity. Of particular interest is the site-directed synthesis method of Au NCs (Fig. 1). This involves genetically modifying the ferroxidase center buried within the intrasubunit interface into a compartment which is suitable for growth of Au NCs. To our knowledge this method has never previously been used to ‘trap’ any NCs and could be implemented to fabricate NCs within the intrasubunit interface of multisubunit protein architectures. More importantly, we demonstrate that incorporation of Au NCs into the intrasubunit interface imparts highly luminescent property to ferritin while having no effect on both the formation of inorganic nanomaterials (iron cores) and encapsulation of bioactive molecules (curcumin) within the cavity (Fig. 1e).
Au NCs are typically less than 2 nm in size which approaches the Fermi wavelength of the conduction electrons. Although the cavity of ferritin serves as an ideal nanocontainer for synthesis of various nanoparticles [3, 4], it is not suitable for preparation of Au NCs because the size of the cavity is around 8 nm in diameter which is too large for such preparation. On the other hand, once the cavity is occupied by Au NCs, the ability of ferritin to encapsulate guest molecules will be greatly compromised. Similarly, it is also difficult to control the size of Au NCs on the exterior surface of ferritin. Although the pore size of various ferritin channels (3- or 4-fold channels) located at the intersubunit interfaces (Additional file 1: Figure S1) is similar to that of Au NCs, occupation of these channels by Au NCs will inhibit diffusion of metal ions into the cavity for mineralization. By investigating all interfaces of HuFtMt, we found that the ferroxidase center buried within the intrasubunit interface appears to meet all kinds of requirements for fabrication of Au NCs. Firstly, the ferroxidase center is located in the middle of four α-helix bundle of each subunit, which is some 7 Å from the inside surface of the protein shell and ~ 12 Å from the exterior surface in a region of considerable hydrophobic character, providing a size-constrained reaction platform for Au NCs synthesis. Secondly, when Au NCs were generated at the ferroxidase center, they would not affect the utilization of the ferritin cavity.
Genes encoding the designed proteins (ΔC, 2Cys-ΔC, and 3Cys-ΔC) were constructed and cloned into an expression vector, Escherichia coli (E. coli) BL21 cells, respectively. After E. coli cells expressing the proteins were lysed, the proteins were purified by a combination of gel and ion-exchange chromatography. As expected, SDS-PAGE analyses revealed that all these three mutants consist of one type of subunit (Additional file 1: Figure S3A). Native PAGE of these proteins exhibited a single band, indicating that they were purified to homogeneity (Additional file 1: Figure S3B). As designed, these mutants are indistinguishable from the wt protein in terms of assembly, structure, and morphology based on fluorescence and transmission electron microscopy (TEM) analyses (Additional file 1: Figure S3C and S3D).
It is believed that a small nanocluster in protein scaffolds are formed by binding of metal ions, followed by nucleation [37, 38]. Therefore, we firstly determined whether Au ions can specifically bind at the modified ferroxidase center. To this end, we obtained the crystals through co-crystallization of three mutants ΔC, 2Cys-ΔC and 3Cys-ΔC with Au3+, respectively. ΔC shared a similar method with wt HuFtMt. However, for 2Cys-ΔC and 3Cys-ΔC, initial co-crystallization trials gave only small crystals, but large single crystals were eventually obtained, which are suitable for X-ray diffraction studies. Subsequently, we solved the crystal structure of these three samples at a resolution of higher than 2.3 Å. As expected, no Au ions binding was observed with ΔC mutant (Fig. 2A2 and B2) because the inherent binding residues Cys102 and Cys130 in wt ferritin had been replaced by Ala. Differently, the crystal structure of 2Cys-ΔC shows that one gold atom is coordinated by Cys27 and Cys62 in a linear geometry, occupying the original A site of the ferroxidase site, and this Au ion was buried within four α-helix bundle of each subunit (Fig. 2B3). There are totally 24 Au ions occurring at each protein shell as shown in Fig. 2A3. In contrast, two gold atoms are bound at the modified ferroxidase center of each 3Cys-ΔC subunit. One atom, named Au1, was located at the site A of the ferroxidase center, which is coordinated by Cys27 and Cys62 also in a linear geometry, while another gold atom, Au2, interacted with Cys61, nearby site B (Fig. 2B4). Consequently, each 3Cys-ΔC molecule can accommodate up to 48 Au ions (Fig. 2A4). Thus, these crystal structures validated our re-design of the ferroxidase center in atomic detail, thereby supporting the utility of the intrasubunit interface strategy for creating the specific Au binding sites.
Subsequently, we determined the ability of these three mutant proteins and wt HuFtMt as biotemplates to synthesize Au NCs, respectively, following our reported method by mixing protein and HAuCl4 at basic conditions and a further incubation at 37 °C overnight . Consistent with the above crystal structure results, no fluorescence was observed with ΔC due to lack of Au ions binding in protein molecule (Fig. 2C2). In contrast, the 2Cys-ΔC-templated Au NCs emit a weak fluorescence under the same experimental condition (Fig. 2C3), which has an UV–vis absorbance at 390 nm, which is the characteristic surface plasmon resonance peak of small metal NCs (Fig. 2C3, inset). Among these three mutants, the as-prepared 3Cys-ΔC-stabilized Au NCs emit the strongest red fluorescence at 650 nm (Fig. 2C4). Thus the crystal structure ensure the Au NCs were synthesized within the subunit, namely intrasubunit.
Characterization of 3Cys-ΔC-stabilized Au NCs
Imaging ability of 3Cys-ΔC-stabilized Au NCs
Fluorescent imaging in vivo is of essential importance for practical applications, and thus we further investigated the bioimaging of 3Cys-ΔC-templated Au NCs by utilizing Caenorhabditis elegans as an animal model (C. elegans is a kind of well-studied nematode with well-defined anatomy) [42, 43]. A series of worms at different cultivating times were fixed on glass slides and observed by laser confocal mircoscopy. As shown in Fig. 5c, the worms treated with two negative control samples (only E. coli or the complex of E. coli and HuFtMt mutant) showed no luminescence. Oppositely, upon treatment of the worms with the ferritin-Au NCs added E. coli, there existed enriching fluorescence along with the worms’ morphology from 30 to 50 h. After 70 h, the Au NCs excreted gradually, with luminescence decreasing (Additional file 1: Figure S9). For C. elegans, with short life-cycles, the ferritin-Au NCs composite exhibits fluorescence with high stability and comparably longer storage time. To verify the synthesized dye distribution, we detected a worm sample at the fluorescent peak by confocal microscopy. According to Fig. 5c, the ferritin-Au NCs were clearly observed in the intestines, with luminescence gathered. Differently, worms fed with only E. coli, complex of 3Cys-ΔC and E. coli showed no luminescence at all. On the other hand, worms were free living upon treatment with 3Cys-ΔC-stabilized Au NCs (Fig. 5d), indicating that the as-prepared Au NCs are safe in nature. These results demonstrate that 3Cys-ΔC-stabilized Au NCs could be absorbed by cells and C. elegans, while maintaining their stable fluorescence in vitro and in vivo even against complex environment. Thus, 3Cys-ΔC-stabilized Au NCs could be developed into a new class of imaging agents.
Utilization for hybrid nanomaterials
In conclusion, we built a specific compartment for fabrication of Au NCs by re-designing the intrasubunit interfaces of ferritin nanocage, resulting in the site-directed synthesis of Au NCs with protein as a biotemplate. The as-prepared Au NCs in protein shell hardly interferes with the ability of ferritin to either grow iron cores or encapsulate bioactive molecules within the protein cavity. Thus, we can incorporate two different types of nanoparticles into a single protein nanocage. These findings open a new way to prepare ‘all-in-one’ nanoparticles or nanostructures. Since the intrasubunit interface occurs in nearly all of simple proteins or multisubunit protein architectures, our engineering approach should, in principle, be applicable to these proteins. This would lead to the generation of a variety of fluorescent hybrid nanomaterials.
GZ designed the study, supervised the project. JZ conceived and carried out experiments and analysed data. BZ, XZ and PA assisted in the characterization of ferritin and nanoclusters. All authors wrote the final manuscript. All authors read and approved the final manuscript.
The authors gratefully acknowledge the National Natural Science Foundation of China (No. 31730069) and the Initiative Postdocs Supporting Program of China (No. BX201700284).
The authors declare that they have no competing interests.
- 41.Maity B, Abe S, Ueno T. Observation of gold sub-nanocluster nucleation within a crystalline protein cage. Nat Commun. 2017;6:93.Google Scholar
- 46.Ma PA, Xiao HH, Li XX, Li CX, Dai YL, Cheng ZY, Jing XB, Lin J. Rational design of multifunctional upconversion nanocrystals/polymer nanocomposites for cisplatin (IV) delivery and biomedical imaging. Adv Mater. 2013;25:4849–905.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.