Chromium(VI) pollution has attracted increasing attention worldwide because of its potential carcinogenic effect on humans.1 This toxic chemical is usually produced by industrial activities. For example, chromate has been widely used in the chrome plating of metals, the manufacture of dyes/pigments, and in leather tanning. The disposal of industrial wastes that contain CrVI into the natural environment can lead to serious pollution problems to natural water.3 Irrigation with CrVI-polluted water can also cause extensive soil pollution as well as plant/crop contamination.4 More worryingly, CrVI can contaminate drinking water owing to water pollution and pose a serious health risk to the general population through water consumption. The World Health Organization (WHO) first proposed a guideline value of 50 ppb for CrVI in 1958, but later changed to the guideline for total chromium (including CrIII and CrVI) due to the difficulty of analyzing CrVI only.5 For the European Union (EU), the maximum chromium limit for drinking water is 50 ppb.6 Recently, the US Environmental Protection Agency (EPA) has released a draft of Toxicological Review of Hexavalent Chromium, which highlights the carcinogenic effect of CrVI.7 Currently, the US EPA is reviewing the human health assessment for CrVI to decide if a new drinking water regulation for CrVI or a revision to the current total chromium level (including CrIII and CrVI; maximum contaminant level=100 ppb) is warranted. Nevertheless, CrVI has become an important target in the monitoring of drinking water.
Extensive frontline CrVI monitoring is a good strategy for the control of CrVI pollution, but it is a challenging task for the following reasons: 1) effective CrVI monitoring requires the differentiation of CrVI from other common metals (including CrIIICrVI speciation; CrIII is much less toxic than CrVI), and 2) there is an urgent need for simple, sensitive, cost-effective, and specific sensing tools capable of performing on-site measurements of CrVI. Such tools can allow the establishment of an extensive monitoring network for CrVI over various natural water sources and industrial discharge sites. This extensive monitoring network can comprehensively monitor the CrVI levels at various regions and quickly provide an alert to any cases of CrVI pollution, so that immediate control and treatment can be conducted. Different analytical methods for CrVI have been developed in recent years. Ion chromatography coupling to post-column chemical modification has been the conventional method for CrVI detection.8 Inductively coupled plasma optical emission spectrometry (ICP-OES)9 and inductively coupled plasma mass spectrometry (ICP-MS),10 as well as atomic absorption spectrometry (AAS),11 can also detect CrVI if they are used in combination with chromium-chelating agents in sample preparations. HPLC coupling to cloud-point extraction12 and ionic-liquid extraction13 techniques have been reported for CrVI analysis. Electrochemical sensors,14 enzyme-based biosensors,15 ionophore-based optical sensors,16 and luminescent chemosensors17 have also been developed for CrVI sensing. Nanoparticle-based CrVI sensors have also been developed in recent years.18 Although advanced instruments can provide sensitive CrVI detection, such approaches often require well-trained operators, sophisticated and expensive instruments, as well as tedious sample treatment. These difficulties largely limit their extensive applications at the frontline of CrVI monitoring. To address the increasing need for CrVI monitoring in natural/drinking water, it is highly desirable to develop a rapid and sensitive sensor capable of performing on-site CrVI detection.
Herein, we describe the development of a new colorimetric nanosensor for CrVI from amyloid fibrils. Amyloid fibrils are the hallmarks of amyloid diseases, which involve the deposition of proteins or peptide fragments into water-insoluble aggregates under physiologically relevant conditions.19 These highly ordered protein aggregates are characterized by a cross-β sheet structure, which is formed through the formation of intermolecular hydrogen bonds between the peptide groups of individual polypeptide chains.19 These nanosized biomaterials are promising tools for metal sensing because of their advantageous properties: 1) amyloid fibrils are cost-effective and robust biomaterials, which can be easily prepared under green and mild aqueous conditions; 2) amyloid fibrils can carry charges along their structures owing to the presence of charged amino acids (e.g., Lys, Arg, Glu, and Asp) and can therefore act as effective binders to oppositely charged analytes (e.g., metal ions) for sensitive sensing purposes; and 3) amyloid fibrils can act as a convenient visual-based tool to probe the metal ions adsorbed onto their structures. The working principles of the amyloid-fibril-based CrVI nanosensor is outlined in Scheme 1.
In aqueous solution, CrVI exists mainly as HCrO4− and CrO42− under acidic and neutral/alkaline conditions, respectively.2 Amyloid fibrils of hen lysozyme, which are positively charged, are used as chromate binders. These positively charged fibrils are very useful for binding and concentrating CrVI that exists at trace levels for sensitive sensing purposes. If samples contain chromate, this negatively charged species would be adsorbed onto the fibrils through electrostatic attraction. Upon addition of acidified DPC, CrVI in the adsorbed chromate will be reduced to CrIII to form a pink complex with oxidized DPC,20 which can be easily visualized from the fibrils. This new colorimetric nanosensor is not only amenable to spectrophotometric CrVI analysis, but is also able to probe CrVI rapidly and conveniently in an instrument-free format for on-site CrVI monitoring purposes. Moreover, the nanosensor represents a good model for the development of a new class of rapid, cost-effective, and sensitive optical sensors from amyloid fibrils for other significant metals for environmental monitoring purposes.
Hen lysozyme has more positively charged amino acids (6 Lys and 11 Arg) than negatively charged amino acids (7 Asp and 2 Glu), and therefore, this protein usually carries a net positive charge under acidic and neutral conditions (pI=9.3). Thus, upon aggregation under denaturing conditions, hen lysozyme can form amyloid fibrils that carry positive charges on their structures. Amyloid fibrils of hen lysozyme were prepared under denaturing conditions (3 M guanidine hydrochloride in 20 mM potassium phosphate buffer (pH 6.3) at 50 °C).21 TEM measurements showed that hen lysozyme formed fibrillar structures under the denaturing conditions (Figure 1). The charge state of the hen lysozyme fibrils was examined by zeta-potential measurements. In general, particles with a net positive charge on their surfaces show a positive zeta-potential value, whereas those with a net negative charge exhibit a negative zeta-potential value. Figure 2 shows the zeta-potential values of hen lysozyme fibrils at different pH values. Hen lysozyme fibrils have positive zeta-potential values (+3.8–47.2 mV) over the pH range of 2.0–10.5; this indicates that they are positively charged within this pH range. At pH 11.0, the zeta-potential value of hen lysozyme fibrils becomes negative (−15.0 mV); this implies that hen lysozyme fibrils become negatively charged at this pH.
The ability of the amyloid-fibril-based nanosensor to detect CrVI in chromate under neutral conditions was investigated. At pH 7.0, CrVI exists mainly as CrO42−.2a Briefly, a series of CrVI solutions (0, 30, 50, and 80 ppb; pH 7.0) were incubated with hen lysozyme fibrils. The fibrils were collected by centrifugation and the solution fractions were discarded. DPC/sulfuric acid was then added to the fibrils. Without CrVI, the hen lysozyme fibrils appeared as white solids (Figure 3 a). The hen lysozyme fibrils, however, turned pink for the samples containing 30, 50, and 80 ppb CrVI; the pink color became more intense at higher CrVI concentrations (Figure 3 b–d). The whole sensing task, which only requires simple mixing of fibrils/sample/reagents, can be completed in 10 min. The pink fibrils were then resuspended in DPC/sulfuric acid and analyzed by UV-visible spectrophotometry. An absorption peak at λ=543 nm appeared in the UV-visible spectrum of the pink fibrils (Figure 4 a). For comparison, a similar UV-visible absorption spectrum was recorded from the reaction of free CrVI with DPC/sulfuric acid (Figure S1 in the Supporting Information), which indicated a characteristic absorption peak at about λ=540 nm; an observation attributed to the fact that CrVI could be reduced to CrIII by DPC to form a pink complex with oxidized DPC.20 These observations indicated that the hen lysozyme fibrils of the nanosensor could bind CrVI and colorimetrically probe the adsorbed CrVI with the aid of acidified DPC. We then determined the amount of CrVI adsorbed onto the fibrils by using ICP-MS; about 33, 63, and 102 ng of CrVI were adsorbed onto the fibrils after incubation with 30, 50, and 80 ppb CrVI solution (pH 7.0), respectively. These results are consistent with the observations that the nanosensor shows a stronger colorimetric response with a higher concentration of CrVI (Figure 3). The nanosensor can sensitively detect CrVI down to the 30 ppb level without the use of any instruments (Figure 3 b). This observation highlights the ability of the amyloid-fibril-based nanosensor to provide a rapid and convenient visual-based sensing platform for on-site CrVI monitoring.
The fibrils of the nanosensor are very useful for binding CrVI at the trace level and concentrating it in a smaller solution volume (through fibril centrifugation and resuspension) to improve the sensitivity of spectrophotometric CrVI detection. Figure 4 b shows the results of absorbance measurements (λ=543 nm) on the nanosensor against a series of CrVI solutions (0–0.75 ppm, pH 7.0). Upon the addition of DPC/sulfuric acid, the CrVI-adsorbed fibrils show a linear absorbance response curve in the concentration region of 0–0.75 ppm CrVI (Figure 4 b). Without the nanosensor, the absorbance response curve has a much lower slope than that of the nanosensor (Figure 4 b). These observations highlight the ability of the nanosensor to increase the sensitivity of spectrophotometric CrVI detection.
Similar colorimetric sensing studies were also performed on a series of CrVI samples (0, 30, 50, and 80 ppb) at pH 4.0 by using the nanosensor. At pH 4.0, CrVI exists mainly as HCrO4−.2a Figure S2 in the Supporting Information shows the hen lysozyme fibrils of the nanosensor collected after adsorption of CrVI at pH 4.0 and the addition of DPC/sulfuric acid. Without CrVI, the fibrils appeared as white solids (Figure S2 a in the Supporting Information). For the samples containing 30, 50, and 80 ppb of CrVI, the fibrils showed a pale-pink color (Figure S2 b–d in the Supporting Information). Absorbance measurements (λ=543 nm) were also performed on the nanosensor against a series of CrVI samples (0–0.75 ppm, pH 4.0). As shown in Figure S3 in the Supporting Information, the nanosensor gives a linear absorbance response curve in the concentration region of 0–0.75 ppm CrVI. Without the nanosensor, the absorbance response curve has a lower slope than that of the nanosensor (Figure S3 in the Supporting Information). In general, the nanosensor gives a much stronger colorimetric response towards CrO42− (pH 7.0) than that of HCrO4− (pH 4.0; Figures 3 and 3, 4 and Figures S2 and S3 in the Supporting Information).
We then studied the effects of the charge states of chromate and hen lysozyme fibrils on CrVI absorption. To this end, we investigated the adsorption capacity of hen lysozyme fibrils for CrVI at different pH values, because the charge states of chromate and hen lysozyme fibrils are pH dependent. Briefly, a series of CrVI samples (2.0 ppm) at pH 4.0, 7.0, and 11.0 were prepared and incubated with hen lysozyme fibrils (1.0 mg mL−1). After adsorption, the concentration of CrVI remaining in each solution was determined by absorbance measurements at λ=543 nm after the addition of DPC/sulfuric acid. The decrease in CrVI concentration was then used to determine the amount of CrVI adsorbed by one gram of hen lysozyme fibrils (adsorption capacity (qe); see the Experimental Section). Figure 5 shows the adsorption capacities of hen lysozyme fibrils at pH 4.0, 7.0, and 11.0. At pH 7.0, CrVI exists mainly as CrO42−, and hen lysozyme fibrils are positively charged (zeta potential=+37.6 mV; Figure 2). Under these neutral conditions, hen lysozyme fibrils have a relatively high adsorption capacity for CrVI (qe=1.37 mg g−1; Figure 5). At pH 4.0, hen lysozyme fibrils remain positively charged (zeta potential=+45.5 mV; Figure 2), but CrVI exists mainly as HCrO4−.2a The reduction in the negative charge of chromate leads to weaker CrVI adsorption by the hen lysozyme fibrils (qe=0.98 mg g−1, Figure 5), compared with the case at pH 7.0 (qe=1.37 mg g−1, Figure 5). This observation is consistent with the findings that the nanosensor gives a much stronger colorimetric response to CrO42− (pH 7.0) than that of HCrO4− (pH 4.0; Figures 3 and 4 and Figures S2 and S3 in the Supporting Information). If the pH increases from 7.0 to 11.0, CrVI remains mainly as CrO42−, but the hen lysozyme fibrils become negatively charged (zeta potential=−15.0 mV, Figure 2). This positive-to-negative change in charge results in a significant decrease in the adsorption capacity of the hen lysozyme fibrils for CrVI (qe=1.37 and 0.18 mg g−1 at pH 7.0 and 11.0, respectively, Figure 5). This result is consistent with the observations that the fibrils of the nanosensor do not show an observable pink color after the adsorption of CrVI at pH 11.0 and reaction with DPC/sulfuric acid (Figure S4 in the Supporting Information) and the absorbance response curve of the nanosensor (pH 11.0) is much weaker than that for pH 7.0 (Figure S5 in the Supporting Information). To further prove that the lysozyme fibrils of the nanosensor make use of their positive charges to bind to chromate, we conducted similar colorimetric experiments on aminopropyl-coated silica gels (particle size: 40–63 μm), which were used as CrVI adsorbents. Under neutral aqueous conditions, the amino groups on the silica gels become protonated. The silica gels exhibit a pink color after the adsorption of 100 ppb CrVI (pH 7.0) and reaction with DPC/sulfuric acid (Figure S6 in the Supporting Information); an observation that is similar to the colorimetric response of the nanosensor to CrVI (Figure 3). This observation supports the fact that the lysozyme fibrils of the nanosensor make use of their positive charges to bind to chromate through electrostatic attraction. Notably the nanosized lysozyme fibrils can sensitively detect CrVI down to the 30 ppb level, whereas microsized aminopropyl-coated silica gels can only respond to CrVI at the 100 ppb level under similar experimental conditions (both adsorbents: 1.5 mg; Figure 3 and Figure S6 in the Supporting Information). These observations highlight the significant value of hen lysozyme in the construction of positively charged protein nanofibers for effective CrVI adsorption and sensitive CrVI sensing purposes. Taking all of these observations together, the hen lysozyme fibrils of the nanosensor can adsorb CrVI by electrostatic attraction. This charge attraction enables the nanosensor to detect CrVI at trace levels through the binding and concentrating function of hen lysozyme fibrils towards chromate.
In the presence of NaCl and NaNO3, the adsorption capacity of hen lysozyme fibrils for CrVI at pH 7.0 decreases if the salt concentration increases from 0 to 500 mM; this is presumably due to the charge neutralization of Cl− and NO3− on hen lysozyme fibrils (Figure S7 in the Supporting Information). In general, the colorimetric response of the nanosensor to CrVI becomes weaker if the concentration of Cl− and NO3− increases (Figures S8 and S9 in the Supporting Information). Despite this, the nanosensor can still detect CrVI down to the 30 ppb level in the presence of 20 mM Cl− and NO3−; a concentration level higher than is environmentally relevant [Cl−] and [NO3−] in natural stream/groundwater22 (Figures S8 and S9 in the Supporting Information). Moreover, anion interference from Cl− and NO3− can be relieved if using more lysozyme fibrils as adsorbents to adsorb CrVI, as revealed by the stronger colorimetric response from the larger amount of lysozyme fibrils upon the addition of DPC/sulfuric acid (Figures S10 and S11 in the Supporting Information).
We then investigated the specificity of the amyloid-fibril-based nanosensor for CrVI. A series of samples with different metal ions (0.1 ppm K+, Ca2+, Cu2+, Mg2+, Cr3+, Na+, Pb2+, Ni2+, and Cd2+) were prepared, and hen lysozyme fibrils were added to each sample. The fibrils were then collected by centrifugation and DPC/sulfuric acid was added after removing the solution of metal. For comparison, the same colorimetric experiment was also performed on 0.1 ppm CrVI. As shown in Figure 6, the fibrils of the nanosensor only show an intense pink color with CrVI. For other metals, including CrIII, the fibrils do not show any observable color changes (Figure 6). These observations indicate that the nanosensor can detect CrVI with high specificity and perform CrIIICrVI speciation. The high specificity of the nanosensor for CrVI is contributed to by the combined use of positively charged hen lysozyme fibrils to adsorb chromate (negative species) and the CrVI-specific reagent, DPC, to probe the adsorbed CrVI.20
The CrVI level in drinking water has been a major health concern because drinking water is a major medium of CrVI exposure for the general population. Effective CrVI monitoring in natural/drinking is therefore of particular importance and this task can be achieved if a sensitive and convenient sensor is available for on-site CrVI testing. We investigated the applicability of the nanosensor to detect CrVI in tap water. Briefly, a series of tap-water samples were spiked with different concentrations of CrVI (30, 50, and 80 ppb). Hen lysozyme fibrils were then added, collected by centrifugation, and then DPC/sulfuric acid was added. For comparison, a tap-water sample without CrVI spiking was also analyzed with the nanosensor. As shown in Figure 7 a, the fibrils of the nanosensor do not show any color change with the CrVI-free tap-water sample. For tap-water samples with CrVI, the fibrils of the nanosensor turn pink, with the color becoming more intense with higher CrVI concentrations (30–80 ppb; Figure 7 b–d). Similar colorimetric CrVI testing was performed on river water with the nanosensor. Similarly, the fibrils of the nanosensor turned pink with river-water samples containing 30–80 ppb of CrVI (Figure S12 in the Supporting Information). These results indicate that the nanosensor can detect CrVI in real water samples down to the 30 ppb level in a simple visual-based format and highlight its potential use in the on-site monitoring of CrVI in drinking water, the maximum contaminant level of which for total chromium (including CrIII and CrVI) is 100 ppb or below (e.g., 100 ppb for the US EPA and 50 ppb for the EU); limits set on the assumption that the total chromium arises from 100 % CrVI, which is the more toxic form. In the case of wastewater, the nanosensor can also colorimetrically respond to 30–80 ppb of CrVI (Figure S13 in the Supporting Information). Unlike other advanced instrumental methods (e.g., ion chromatography,8 ICP-OES9 and ICP-MS10), the nanosensor does not require sophisticated and expensive instruments, tedious sample treatment, and skilled operators for CrVI testing. With these advantages, the nanosensor can play a critical role in the frontline monitoring of CrVI prior to the use of advanced instrumental methods for confirmatory CrVI testing in laboratories.
In summary, we have successfully developed a new colorimetric nanosensor for CrVI from the amyloid fibrils of hen lysozyme. This nanosensor, which makes use of hen lysozyme fibrils as CrVI binders and acidified DPC as the colorimetric probe, can specifically detect CrVI at the ppb level without sample pretreatment and the use of advanced instruments. These encouraging results highlight the potential use of the nanosensor in on-site CrVI monitoring. Our nanosensor also represents a good model to demonstrate that rapid, convenient, and sensitive metal sensors can be constructed from amyloid fibrils, which are promising bio-nanomaterials with the advantages of high robustness as well as easy preparation under green and mild aqueous conditions. In addition, amyloid fibrils can be fabricated into positively/negatively charged nanotubes; thus allowing them to act as strong binders to metal analytes through electrostatic attraction. This binding function is particularly useful for concentrating metal analytes at trace levels for sensitive metal-sensing purposes. Recent studies have also reported the development of CrVI sensors from gold- and silver-based nanoparticles and nanorods.18 Although such sensors can detect CrVI at the sub-micromolar level, they have some technical difficulties, such as the need for acquisition over a narrow acidic pH range (pH 1.0–2.0) for effective CrVI detection,18b,c susceptibility for pH-dependent agglomeration, interference from metal ions,18b and the need for sample pretreatment.18b With the advantages of robust structure, an easy and green method of preparation, and room for the fabrication of tailor-made nanotubes through protein engineering/chemical modifications, amyloid fibrils can be a good choice of biomaterial for nanosensor development and will play a promising role in the development of a new class of rapid, sensitive, and cost-effective nanosensors for other significant metals for environmental monitoring purposes.