Gold Biomineralization on Bacterial Biofilms for Leaching of Au3+ Damages Eukaryotic Cells.

Heavy metals not only pollute the environment but also are health and environmental hazard. Bacteria constitute inexpensive and eco-friendly material to eliminate and recycle heavy metals via biomineralization and biosorption. However, the effect of metal biomineralization in bacterial biofilms on the ecological balance of bacteria and infectious diseases is unclear. This study aimed to explore the interaction between a eukaryotic cell line HEK293T and mineralized Escherichia coli, using a model of gold biomineralization on E. coli biofilms (E. coli-Au). In our present model, bacterial activity was not disrupted and bacterial adhesion and invasion were enhanced. E. coli-Au invaded the cytoplasm and nuclei of HEK293T cells and damaged them via intracellular growth and multiplication. The present findings indicate that metal biomineralization in bacterial biofilms for leaching of heavy metal ions is hazardous to eukaryotic cells and even human health.

Introduction

Heavy-metal pollution resulting from anthropogenic factors including technological advancements is one of the most prominent environmental concerns today,[1,2] threatening human health and the ecosystem via exposure to heavy metals including Pd, Pt, Ag, Au, and Ru.[3,4] Methods for metal ion leaching from aqueous solutions primarily includes physical, chemical, and biological technologies.[5−7] Various biological materials, especially bacteria, have received increasing attention for heavy metal leaching and recovery owing to their optimal performance, low cost, and high availability.[8−10] Relatively inexpensive and eco-friendly alternatives to conventional methods have allowed for enhanced biomineralization and biosorption technology.[3,11] Nonetheless, large-scale industrial wastewater treatment still seems a distant reality.[9] However, one major concern is that these new technologies to counter heavy-metal pollution may result in certain side effects including invasive bacteria and some new infectious diseases.

Extracellular polymeric substances (EPS) are the major components of biofilms, primarily comprising polysaccharides and proteins.[12,13] Furthermore, Gram-positive bacteria contain up to a 20–80 nm peptidoglycan layer.[14] EPS essentially protects bacteria from environmental stresses including antibiotic agents, desiccation, extreme temperature, predation, high salinity, and pH conditions.[15,16] Bacteria have various mechanisms to reduce and biomineralize metal ions using organic molecules and functional groups in EPS to decrease their toxicity.[17−22] The amino, thiol, carboxyl, and phenolic groups in bacterial biofilms play crucial roles in binding and effectively immobilizing heavy-metal ions and reducing them to elementary substances (zerovalent) or zerovalent nanoparticles.[23−25] Bacterial biomineralization can also enrich toxic metals under certain conditions, with potentially calamitous effects on human health.[26] Moreover, whether metal biomineralization in bacterial biofilms disrupts the ecological balance of bacteria and introduces new diseases is a major potential concern.

Bacteria are one of the major components of the biosphere and their number, greatly exceeding that of all plants and animals, is approximately 5 × 1030.[27] They are extremely adaptable to natural environments, with high survival rates in diverse habitats. They grow in the deepest oceans, the driest deserts, the coldest habitats, and the most saturated salt brines. Bacterial surfaces play a critical role in interactions between bacteria and their natural environment.[28]

However, no study has reported whether bacteria widely used for heavy metal leaching can alter its biological characteristics and threaten human health. Bacteria in the human body, including the gut microflora, and numerous skin-dwelling bacteria outnumber human cells by a factor of 10:1.[29] Furthermore, several species of bacteria are pathogenic and cause infectious diseases, including cholera, syphilis, anthrax, leprosy, bubonic plague, and so on. Therefore, to study and understand the relationship between bacterial biomineralization and eucaryotic cell and biology has become the problem that brook no delay.

Compared with other metals, gold is rare, inert, and nonessential and does not form free ions in aqueous solution under surface conditions. It has been widely used in many high-tech fields, such as electronic, aerospace, chemical, medical, and so on. Moreover, the role of microorganisms as main drivers of metal mobility and mineral formation under Earth surface conditions is now widely accepted and bacterial biofilms are deemed to be associated with secondary gold grains from two sites.[18,30] Recently, many papers have reported that the biological ways of synthesizing metal nanoparticles via using microorganism have become an effective alternative.[31,32] Gold nanoparticles have attracted the highest amount of attention owing to their dynamic properties at the nanoscale and have been well-used in biomedical, bioimaging, cosmetics, pharmaceuticals, drug delivery, genetic engineering, and cancer treatment.[33−35] Hence, gold biomineralization on bacterial biofilms was easy to establish and obtain.

In this study, we generated a model of gold biomineralization on Escherichia coli biofilms (E. coli-Au) to investigate changes in bacterial biological characteristics and its effect on eukaryotic cells.

Results and Discussion

Model of Gold Biomineralization on E. coli Biofilms

Compared to most other metals, gold (Au) is rare, inert, and nonessential to bacteria and does not form free ions in aqueous solutions at surface conditions, while zerovalent gold nanoparticles are promising candidates for biomedical and environmental applications because of their chemical inertness, pronounced biocompatibility, and characteristic photoelectric effect.[36,37] It is of paramount importance to develop clean, nontoxic, and environmentally benign synthetic technologies for gold nanoparticles.[38−40] Proteins biomineralize metal nanoparticles for various biomedical applications, such as biosensing, biolabeling, bioimaging, and mimicking enzyme function, such as bovine serum albumin, fibrinogen, α-lactalbumin, lysozyme, and cytochrome c.[41,42] There are numerous similarities in the natural properties of proteins and peptide. Some studies reported that specific amino acid sequences of designed peptides harvested to biomineralize metal nanoparticles, such as thiol-containing amino acids possessing both reduction and coordination ability with metal ions and tyrosine (Y) and tryptophan (W), are efficient reducing groups.[43] The peptide cysteine–cysteine–tyrosine was designed to directly reduce Au ions into zerovalent Au through the phenolic group of tyrosine under alkaline conditions without requiring additional reducing agents and capture the Au atom using the −SH group of cysteine.[44]

In this study, we also generated alkaline conditions with E. coli and Au3+ to reduce Au ions into zerovalent Au and captured Au nanoparticles (Au NPs) by bacterial biofilms, as shown in Scheme . First, Au ions (HAuCl4) were added dropwise into the E. coli suspension. After stirring well, the final pH of the solution was adjusted to ∼10 with NaOH and the phenolic group of tyrosine (Tyr) residues, which are present on the surface protein of E. coli, was converted into a negative phenolic ion that can reduce Au ions to Au atoms. Thereafter, E. coli was discarded via centrifugation and the resultant Au atoms in the supernatant solution aggregated to form Au NPs (Scheme A). With time, the solution gradually became black from yellow (Figures and S2). Au NPs were obtained via high-speed centrifugation, and Na+, OH–, and other ions were eliminated. Finally, Au NPs were mixed again in the E. coli suspension. At this point, Au NPs were captured by E. coli (Scheme B).

Scheme 1

(A) and (B) Schematic Representation of the Surface Modifications of E. coli with Au Nanoparticles (NPs; E. coli-Au)

(A) Introduction of NaOH led to the conversion of the phenolic group of tyrosine on the cell wall of E. coli to a phenoxide ion, which can oxidize Au ions. The resultant Au atoms aggregate into Au NPs. (B) Au NPs are coated by the SH groups of the proteins in the cell wall of E. coli to form E. coli-Au. (C) HEK293T cells were incubated with E. coli-Au for 3, 6, and 24 h. E. coli-Au can be adsorbed and can penetrate the host cell plasma membrane, eventually invading the cytoplasm and nucleus and retaining bacterial proliferation and damaging the nucleus.

Figure 1

(A) Photomicrographs of E. coli (right) and E. coli-Au (left) harvested via centrifugation. (B) Size distribution of E. coli (red curve) and E. coli-Au (black curve) and inset at the top-right corner shows its statistical data (n = 3). (C) ζ-Potential of negative charges on E. coli (red curve) and E. coli-Au (black curve), and the inset at the top-right corner shows its statistical data (n = 3). (D) Field-emission scanning electron microscopy images of synthesized E. coli-Au.

Characterization of Gold Biomineralization in E. coli Biofilms

The morphology of E. coli and E. coli-Au were examined via optical microscopy (Figure S2). The size of E. coli-Au decreased significantly, compared with the control group (E. coli). Dynamic light scattering (DLS) measurement was used to further examine the change in size. The results were consistent with the aforementioned results (Figure B). After attracting Au NPs, the size of E. coli reduced from 1328 ± 267.5 to 658.7 ± 100.4 (n = 3, top-right corner of Figure B) mostly because the osmotic pressure at the E. coli habitat increased. Furthermore, the surface charges of E. coli and E. coli-Au were also analyzed via DLS. We observed that both E. coli and E. coli-Au were negatively charged (see Figure C). In comparison, E. coli harbored a more negative charge (E. coli vs E. coli-Au: −47.4 ± 2.36 vs −33.9 ± 3.56; Figure C and top-right corner of Figure B). Since Au NPs coated the bacterial surface, E. coli-Au with more positive charge had a decreased negative charge and their size also decreased, which is beneficial for the adhesion between cytomembranes and E. coli.

A field-emission scanning electron microscope (FESEM) was utilized to explore the Au core size distribution on the surface of E. coli. The mean size of Au cluster was 10–20 nm (Figure D). Moreover, the size of E. coli was approximately 600–700 nm, consistent with the results of DLS. In Figure D, small Au NPs were evenly distributed on the extracellular surface of E. coli, e.g., at the capsule, pilus, cell wall, or cytomembrane. The cell wall is the outermost layer and primarily comprises phospholipids, proteins, and lipopolysaccharides.[21,30] These macromolecular assemblies can provide some activating groups to capture Au NPs. However, we did not obtain an associated database to support these results, and hence the search for E. coli-Au will continue.

Biological Property of Au NPs Modifying E. coli Biofilms

During the synthesis of E. coli-Au, it should first be considered whether chemical reactions disrupted the growth and genetic material of bacteria. The Ampr-enhanced green fluorescent protein (EGFP)-plasmid was transformed into E. coli, and agarose gel electrophoresis was performed to confirm the plasmid stability in E. coli-Au. As shown in Figure A, the plasmid in lane 3 was not degraded in the E. coli-Au group and remained in high concordance with the control group. On comparison with the other two groups, E. coli + Au (E. coli + HAuCl4) and E. coli-Au + OH– (E. coli + HACl4 + NaOH), the corresponding band of the plasmid was not observed on the agarose gel (lanes 4 and 5; Figure A). HAuCl4 could degrade the plasmid, along with the acidic environment. To assess the growth activity of E. coli, four groups were cultured overnight in the LB medium supplemented with ampicillin, with agitation at 37 °C (Figure B). E. coli adsorbing Au NPs yet underwent proliferation, which was similar to that of individual E. coli upon measuring the absorbance value at 600 nm (n = 3, see Figure C). Because of the degradation of the plasmid or genetic material of bacteria, the proliferation of E. coli in E. coli + HACl4 + NaOH and E. coli-Au + NaOH group did not occur (Figure B).

Figure 2

(A) Agarose gel electrophoresis to assess plasmid stability after three different chemical reactions. (B) Images of Luria-Bertani (LB) culture medium supplemented with ampicillin and shaken overnight after inoculation of E. coli underwent treatment with three different chemical reactions. (C) Statistical data for the measurement of the optical density (OD) at 600 nm of LB culture medium (n = 3).

Localization of E. coli-Au in HEK293T Cells

The changes in size and surface charges of E. coli-Au are supposed to be beneficial to bind and penetrate the cell membrane in theory. E. coli was labeled with Texas Red (Texas Red-E. coli; Figure 5S), which yielded a red fluorescence to track the localization of E. coli in HEK293T cells. A certain amount of E. coli-Au incubated with HEK293T cells does not debilitate the activity of cells, and the cells retain their normal morphology in 3 h. To verify the interaction between E. coli-Au and HEK293T cells, confocal laser-scanning microscopy (CLSM) was used to track Texas Red in live HEK293T cells. Before the CLSM analysis, the HEK293T cells were incubated with E. coli and E. coli-Au for 3 h, and the final density of E. coli in the cell culture medium was 5 × 107 cells/mL. As shown in row 1 of Figure , no Texas Red stained E. coli were observed within the HEK293T when cells were treated with unbound E. coli. However, red fluorescence was observed in cells upon incubation with E. coli-Au in row 2 of Figure . This result confirmed our hypothesis that E. coli-Au can bind and penetrate the cell membrane.

Figure 3

Confocal laser scanning microscopy image of HEK293T cells incubated with E. coli (row 1) and E. coli-Au (row 2) for 3 h. Cell nuclei are marked with DAPI (blue) and E. coli are marked with Texas Red (red).

After 6 h of co-incubation, we observed red fluorescence in the cytoplasm of HEK293T cells and observed that E. coli-Au invaded the HEK293T cells (row 1 of Figures A and 4). Unexpectedly, some E. coli-Au penetrated the karyotheca and translocated to the nucleus (three-dimensional image in row 1 of Figures A and 4). However, the EGFP gene was not expressed because green fluorescence was not observed in the cytoplasm (Figure ). To further verify whether EGFP plasmid was translocated into the cytoplasm by E. coli-Au, polymerase chain reaction (PCR) was performed to examine the presence of the EGFP gene in the HEK293T cells. E. coli, E. coli + Au, and E. coli-Au were incubated with HEK293T cells and the medium was supplemented with ampicillin to eliminate nonresistant bacteria. Only in the E. coli-Au group, the EGFP gene was specifically amplified (line 5 of Figure B).

Figure 5

(A) Confocal laser scanning microscopy (CLSM) three-dimensional image of HEK293T cells incubated with E. coli-Au (row 1) for 6 h. CLSM image of HEK293T cells incubated with E. coli-Au (row 2) for 24 h. Cell nuclei are marked with DAPI (blue) and E. coli are marked with Texas Red (red). (B) Polymerase chain reaction analysis of the EGFP plasmid in cells incubated with three different types of E. coli for 6 h. (C) Cell viability of HEK293T cells incubated with E. coli-Au and E. coli for different durations: 0, 3, 6, 12, and 24 h.

Figure 4

Confocal laser scanning microscopy image of HEK293T cells incubated with E. coli-Au for 12 h. Cell nuclei are marked with DAPI (blue), E. coli are marked with Texas Red (red), and EGFP yielded a green fluorescence.

Biological membranes are selectively permeable barriers in bacteria and cells and separates them into dependent units of life activities. Nonetheless, it is difficult to enable intracellular translocation of common bacteria by more than 500 nm. However, certain intranuclear bacteria that can invade the cellular control center of eukaryotes have been identified in protists, arthropod, marine invertebrate, and mammalian hosts.[45] Recently, nanotechnology has been used to deliver certain bacteria harboring DNA, drugs, and signaling molecules into the cytoplasm for application in medicine and bioengineering.[46−48]

Disruption of Eukaryotic Cells by E. coli-Au

When E. coli-Au was further cultured with HEK293T cells for 24 h, nuclear fragmentation occurred and E. coli proliferated via CLSM (row 2 of Figure A). After HEK293T cells were incubated with E. coli-Au and E. coli for different durations (3, 6, 12, and 24 h), a Cell Counting Kit-8 (CCK-8) assay was performed to assess the cancer cell viability (see Figure C). In culture media, the original densities of E. coli-Au and E. coli were set as 5 × 107 cells/mL and, after a 3 h incubation, old medium was replaced with new medium supplemented with ampicillin. The cell viability ratio decreased to 24.42% in 24 h, consistent with the results of CLSM.

The majority of bacteria in the body are harmless. However, some pathogenic obligate intracellular bacteria pose a substantial public health concern owing to infections and intracellular replication, including Coxiella spp., pathogenic Chlamydia spp., and arthropod-transmitted members of the Rickettsiales.[49] Hence, it is imaginable that normal extracellular microflora induced to invade eukaryotic host cells could result in new diseases.

Conclusions

In this study, we developed a model of gold biomineralization on E. coli biofilms (E. coli-Au) to investigate its effect on eukaryotic cells. The adhesive force between them increased markedly, E. coli-Au penetrated the plasma membrane of HEK293T cells and even invade its cytoplasm and nucleus. Ultimately, HEK293T cells were damaged via intracellular growth and multiplication of invasive E. coli. Briefly, bacteria can potentially eliminate and cycle heavy metal ions via biomineralization and biosorption; however, it is important to acknowledge their potential hazards.

Materials and Methods

Material

HAuCl4·3H2O, NaOH, ethanol, dimethyl sulfoxide (DMSO), HCl, and HNO3 were obtained from Beijing Chemical Reagent Co. (China); Texas Red was from Thermo Fisher Scientific; Triton X-100 was from Sigma-Aldrich; cell culture media was from Hyclone; and fetal bovine serum was from Invitrogen. All solutions were prepared using ultrapure water (18.25 MΩ·cm) from a Millipore system. All glassware was washed with aqua regia (HCl/HNO3, 3:1 [v/v]) and then rinsed with ultrapure water and ethanol.

Bacterial Culture and Staining

E. coli expressing EGFP and anti-ampicillin were cultured in Luria-Bertani medium at 37°C with agitation at 200 rpm. Bacterial density was monitored spectrophotometrically at 600 nm (OD600).

Two milligrams of Texas Red was dissolved in 1.5 mL of DMSO. E. coli were harvested via centrifugation at 10 000g and resuspended in 100 μL of normal saline to obtain 1 × 107E. coli cells; 10 μL of Texas Red solution was then added. After 10 min of co-incubation at room temperature, E. coli were harvested via subsequent centrifugation at 10 000g. Thereafter, Texas Red staining of E. coli was traced and located via red fluorescence excitation at 561 nm.

Model of Gold Biomineralization on E. Coli Biofilms

For leaching of Au3+, 5.0 × 107E. coli collected and washed with phosphate-buffered saline (PBS) for three times were added to 5 mL of normal saline and then 0.1 mL of HAuCl4 aqueous solution (10 mM) was gradually added to the E. coli solution (1 × 108 cells/mL) in a 5 mL vial under vigorous stirring in an ice-bath. After fully mixing, 400 μL of NaOH solution (0.5 M) was added within 20 s to yield a final pH of 10. This solution was continuously stirred for 5 min in an ice-bath, and E. coli were removed via 10 000g centrifugation. This supernatant liquor was collected and allowed to stand for 30 min and zero-valent Au was obtained via centrifugation at 10 000g for 1 min at room temperature. The pellet was cleaned thrice to eliminate free ions.

Zero-valent Au and a suspension of 1 × 108E. coli were mixed and allowed to stand at room temperature for 10 min to gold biomineralization on bacterial biofilms. Thereafter, this mixture was centrifuged at 10 000g for 1 min at room temperature to harvest E. coli surface-modified with Au nanoparticles (NPs; E. coli-Au).

Characterization of E. coli-Au

The size distribution and ζ-potential of E. coli and E. coli-Au were determined via dynamic light scattering (DLS; particle sizing systems) analysis using the NICOMP 380/ZLS, and the data were analyzed using ZPW388 software. Morphology of E. coli-Au was examined using a field-emission scanning electron microscope (FESEM) (S-4800, Hitachi, Japan, operated at 10 kV).

Agarose Gel Electrophoresis

Plasmids were extracted by small alkali lysis. After growth for the night, E. coli were collected via subsequent centrifugation at 10 000g and washed with PBS for three times. Then, 100 μL of 0.5 M NaOH was mixed with E. coli for 5 min to release EGFP plasmids from E. coli.

Plasmid stability was assessed via agarose gel electrophoresis. Hundred microliters of the samples (5 × 107E. coli, E. coli + HAuCl4 + NaOH, E. coli + HAuCl4, E. coli-Au) and 100 μL of 0.5 M NaOH were mixed for 5 min. Six microliters of mixed liquor was subjected to 1% agarose gel with 0.05 mg/mL ethidium bromide in TAE buffer (pH = 7.4). The plasmid was electrophoresed at 90 V for 30 min. Fluorescence signal intensities were measured and analyzed using C1000Manager Software (BioRad, Hercules, CA).

Cell Culture

Human embryonic kidney cell line, HEK293T, was cultured in Dulbecco’s modified Eagle’s medium high-glucose medium cell culture supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin solution in a humidified incubator at 37 °C and 5% CO2.

Confocal Microscopy

Cellular uptake and intracellular distribution of E. coli-Au was visualized using a confocal laser-scanning microscope. HEK293T cells were seeded on a 35 mm glass-based dish at a density of 5 × 105 cells per well for 24 h. Old medium was replenished with fresh medium comprising E. coli and E. coli-Au at a density of 5 × 107 cells/mL of E. coli for 3, 6, and 24 h. Thereafter, staining with DAPI was performed to assess the cellular uptake. DAPI, EGFP, and Texas Red were excited at 405, 488, and 561 nm, respectively.

Polymerase Chain Reaction (PCR)

To detect E. coli DNA in HEK293T cells incubated with E. coli, E. coli + Au, and E. coli-Au, DNA was extracted using 1% Triton X-100 and the EGFP gene was amplified via PCR. The primers for EGFP were with the forward primers: 5′-TGAACCGCTCGAGCTGAAGGG-3′ and the reverse: 5′-TCCAGCAGGACCATGTGATCGC-3′ (Sangon). PCR conditions were as follows: 94 °C, 5 min; (94 °C, 30 s; 57 °C, 30 s; 72 °C, 30 s) × 20 cycles; 72 °C, 8 min. PCR products were assessed via electrophoresis on a 2% agarose gel.

Cell Viability Assay

Cell viability was evaluated using a Cell Counting Kit-8 system (CCK-8) in accordance with the manufacturer’s instructions (Dojindo Laboratory). Cells were seeded in 96-well plates at a density of 1 × 104 cells per well. HEK293T cells were treated with E. coli-Au and E. coli at a density of 5 × 106 for 3 h; thereafter, old medium was replaced with fresh medium supplemented with ampicillin. After 0, 3, 6, 12, and 24 h, 10 μL of CCK-8 solution was added before incubation for 1 h at 37°C. The absorbance was measured using a microplate reader at 450 nm. The formula for cell viability assay was that cell viability (%) = [(As – Ab/Ac – Ab)] × 100%, where As is the experimental group, Ac is the control group, and Ab is the blank group. All data are presented as mean ± SEM percentages compared to the OD values of untreated cells.

Statistical Analysis

All data are presented as the mean standard error of the mean from at least three independent experiments. Statistical significance was discerned using Student’s t-test.