Sensitive Detection of Aromatic Hydrophobic Compounds in Water and Perfluorooctane Sulfonate in Human Serum by Surface-Assisted Laser Desorption/Ionization Mass Spectrometry (SALDI-MS) with Amine Functionalized Graphene-Coated Cobalt Nanoparticles

Abstract

In this article, we describe the application of surface-assisted laser desorption/ionization mass spectrometry (SALDI-MS) with the use of amine functionalized graphene-coated cobalt nanoparticles (CoC–NH₂ nanoparticles) to analyse aromatic hydrophobic compounds that are known environmental contaminants, including polycyclic aromatic hydrocarbons (PAHs) and pentachlorophenol (PCP). Our results demonstrated that SALDI-MS can detect PCP, anthracene, and pyrene in water. In particular, the CoC–NH₂ nanoparticles proved to be an efficient means of capturing PCP in water because of the high adsorption capacity of the nanoparticles for PCP, which resulted in a detectability of 100 ppt. Furthermore, the CoC–NH₂ nanoparticles also functioned as an adsorbent for solid-phase extraction of perfluorooctane sulfonate (PFOS) from human serum, displaying good performance with a detectability of 10 ppb by SALDI-MS.

INTRODUCTION

Various carbon-based nanomaterials have been developed for sample extraction in biological and environmental analysis.^1–6) Recently, graphene-nanocomposite magnetic nanoparticles have been developed as an alternative to the application of carbon-based nanomaterials in the environment.^7–12) Typically, magnetic nanoparticles are dispersed into a sample solution to adsorb target compounds through the interaction of graphene with analytes. The nanoparticles are then collected rapidly and easily with an external magnetic field, and the adsorbed analytes are subsequently eluted with solvents for the identification of analytes of interest. These graphene-nanocomposite magnetic nanoparticles possess several advantages over traditional carbon-based nanomaterials due to several factors: larger surface areas, shorter diffusion routes of analytes, π-electron-rich structures, increased chemical stability and selectivity, and decreased extraction time, and sample throughput. However, regardless of the type of nanomaterial used, an elution step is still required to identify the adsorbed analytes on the nanoparticles. This process is often limited by analyte loss during elution from the analyte-binding solid and may require the significant consumption of organic solvents for elution. Therefore, the combination of magnetic nanoparticles with surface-sensitive mass spectrometric methods is useful for identifying the adsorbed analytes on nanoparticles without the need for elution.^13–18) Recently, we have demonstrated that amine functionalized graphene-coated cobalt nanoparticles (CoC–NH₂ nanoparticles) with an extremly thin (∼2 nm) layer of graphitic carbon (this points to the thin layer of graphitized carbon being interpreted as graphene), can be characterized by surface-assisted desorption/ionization mass spectrometry (SALDI-MS), and can be used as effective extraction sorbents of polyfluorinated compounds (PFCs) in water due to their high surface area and high-strength saturation magnetization.¹⁹⁾

Chlorophenol and polycyclic aromatic hydrocarbons (PAHs) are widespread environmental pollutants, and analytical methods are needed for their detection.^20,21) However, there are known problems with analysing these compounds by electrospray ionization mass spectrometry (ESI-MS) or by conventional matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) analysis, due to difficulties associated with protonation/deprotonation of the compounds. Recently, SALDI-MS analysis of PAHs with the use of graphene is reported.²²⁾ SALDI-MS allows the identification of PAHs based on their molecular masses and allows different species to be distinguished from one another. In the case of CoC–NH₂ nanoparticles, the interaction between the benzylamino groups or the graphite surface of Co nanoparticles with the aromatic compounds is possible.^22–24) Therefore, it was expected that CoC–NH₂ nanoparticles are effective for the sample extraction and SALDI-MS analysis. In this study, we have investigated the application of SALDI-MS using CoC–NH₂ nanoparticles for the detection of aromatic hydrophobic compounds of pentachlorophenol, anthracene and pyrene as model analytes of PAHs. In addition, we have also demonstrated that the SALDI-MS can detect perfluorooctane sulfonate in human serum, and recent reports suggests that PFOS might bioaccumulate in the human body.²⁵⁾

EXPERIMENTAL

Materials

Methanol, acetone, anthracene, pyrene, pentachlorophenol (PCP), and perfluorooctane sulfonate (PFOS) were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). All chemicals were used as received. Milli-Q purified water was used for experiments.

Fluorescence spectroscopy

The fluorescence (FL) spectra were recorded by a JASCO FP-6300 fluorometer (JASCO, Tokyo, Japan). The adsorption amount of pyrene by CoC–NH₂ nanoparticles that is denoted by Eq. (1) was estimated from the calibration curve based on the emission peak of pyrene in water at 390 nm in the fluorescence spectra.

SALDI-MS

Spectra were acquired in linear mode on an AXIMA-CFR time-of-flight mass spectrometer (Shimadzu/Kratos, Manchester, UK), with a pulsed nitrogen laser (λ=337 nm). The mass spectra of 100 different profiles of a single sample were averaged. The two-layer sample preparation method was employed for SALDI-MS with CoC–NH₂ nanoparticles in the case of no extraction treatment. First, a nanoparticle solution was prepared by dispersing nanoparticles in water by vortexing and then spotting 1 μL of solution onto a stainless steel plate, followed by drying. Then, a 1 μL portion of solution containing the analyte was spotted onto the plate.

Liquid chromatography-atmospheric pressure chemical ionization-mass spectrometry (LC-APCI-MS)

Mass spectra were obtained in an APCI negative-ion mode using an orbitrap mass spectrometer (Exactive, Thermo Fisher Scientific, CA). The electrospray voltage was 4.3 kV, and the flows of the nebulizer gas, auxiliary gas, and sweep gas were 12, 4, and 1 L min⁻¹, respectively. The capillary temperature was 250°C. An UltiMate3000 was used for HPLC analysis (Thermo Fisher Scientific, CA). The injection volume of the sample solution was 5 μL and the mobile phases, A and B, were H₂O and MeOH. The flow rate of mobile phases consisting of A : H₂O and B : MeOH was 0.2 mL min⁻¹. An Inertsil ODS-P column (2.1×150 mm, 5 μm, GL Sciences Inc., Tokyo, Japan) was used for reverse-phase chromatography with a gradient program of 50% B: 0–3 min, 50–95% B: 3–4 min, 95% B: 4–10 min, 95–50% B: 10–10.1 min, and 50% B: 10.1–15 min.

Preparation of CoC–NH₂ nanoparticles

CoC–NH₂ nanoparticles were prepared according to our previous papers.¹⁹⁾ Briefly, CoC nanoparticles were prepared via flame spray pyrolysis, which proceeds by the combustion of a suitable metal containing an organic precursor. The surface of the CoC nanoparticles was functionalized with benzylamine groups (CoC–NH₂) according to previously reported procedures.¹⁹⁾ The CoC–NH₂ nanoparticles can be also commercially purchased as “TurboBeads Amine” (TurboBeads Co.).

Adsorption amount

Initially, 50 μL of a 10 mg mL⁻¹ solution of CoC–NH₂ nanoparticles was dispersed into a 500 mL sample containing water and analytes. In the case of PCP and PAHs, the aqueous analyte solutions were prepared by mixing 1 mL acetone solution of analyte and 500 mL water. The mixture was stirred vigorously for 1 h to ensure a uniform distribution of sorbents in solution. Next, an Nd–Fe–B strong magnet was deposited at the bottom of the beaker to separate the CoC–NH₂ nanoparticles from the solution. The concentration of analyte in the supernatant was determined by spectrofluorophotometry for pyrene, and by LC-APCI-MS for PCP. The amount of analyte adsorbed per unit of nanoparticles (Q_e) was calculated as:   

(1)

where C_i and C_e are the initial and equilibrium concentrations (g L⁻¹), m is the weight of the adsorbent (g), and V is the volume of the solution (L).

Sample extraction of analytes in water with CoC–NH₂ nanoparticles followed by SALDI-MS detection

To separate the nanoparticles from a large amount of analyte solution, we constructed a separation method of CoC–NH₂ nanoparticles via a flow system (Fig. 1). In this method, 50 μL of a 10 mg mL⁻¹ solution of CoC–NH₂ nanoparticles were added to 500 mL aqueous solution including the analytes of interest. The mixture was stirred using a MMS-210 agitator (EYELA, Tokyo, Japan) for 1 h (Step 1). The resulting mixture was then passed into a separable flow optical cell (Type 49 Short part length Demountable, Starna, England) by a peristaltic pump. The analyte-bound nanoparticles were separated in the cell using an external magnetic field during the flow process, after which they were washed with water three times (Step 2). Finally, 1 μL of the aqueous solution including the analyte-bound nanoparticles (50 mg mL⁻¹) was deposited on a MALDI plate, followed by SALDI-MS analysis without the need for elution (Step 3). By using this system, we could easily separate CoC–NH₂ nanoparticles from litre volumes of solution. The CoC–NH₂ nanoparticles were used to enrich analytes from aqueous solution, followed by detection using SALDI-MS.

Fig. 1. Schematic picture of the flow system.

Sample extraction of analytes in human serum with CoC–NH₂ nanoparticles followed by SALDI-MS detection

A 50 μL aliquot of a 10 mg mL⁻¹ solution of CoC–NH₂ nanoparticles was dispersed into 1 mL of serum spiked with PFOS. The pH of the serum solution was adjusted to ∼1–2 by adding 100 μL of 2 M HCl. The addition of the acid caused the denaturing of HSA in human serum, resulting in the effective capture of PFOS by CoC–NH₂ nanoparticles. The mixture was stirred vigorously for 1 h to ensure that the sorbents were dispersed uniformly in solution. An Nd–Fe–B strong magnet was then deposited at the bottom of the solution to separate the CoC–NH₂ nanoparticles, which were removed, washed with water three times, and analysed by SALDI-MS without the need for elution.

RESULTS AND DISCUSSION

After PCP was extracted from the aqueous solution using CoC–NH₂ nanoparticles, it was analysed by SALDI-MS in negative ion mode (Fig. 2). The deprotonated ion peak [M−H]⁻ was observed at m/z 265 in the mass spectrum, and the isotopic pattern was consistent with that of PCP (Supplementary data, Figure S1). The signal to noise values (S/N) of these peaks were 83 for 1 ppb of PCP, 30 for 100 ppt, and 3 for 10 ppt. The detectability (semi-quantitative limit of detection) of PCP was approximately 100 ppt, which was lower than the drinking water limit of 1 ppb (=1000 ppt) set by the United States Environmental Protection Agency (USEPA). We also examined the detectability of PAHs (anthracene and pyrene) by SALDI-MS in positive ion mode, after extraction from the aqueous solution using CoC–NH₂ nanoparticles (Fig. 3). The radical cation peak of pyrene, [M⁺]^·, was observed at m/z 202 in the mass spectrum with a detectability of 1 ppb (=1000 ppt). A detectability of 1 ppb was obtained for anthracene (Supplementary data, Figure S2). The results indicated that the analysis of PCP was more sensitive than anthracene and pyrene in SALDI-MS. We identified two possibilities for the superior detection sensitivity of PCP compared to the PAHs: higher extraction efficiency of PCP by CoC–NH₂ nanoparticles and higher SALDI efficiency (i.e., higher ionization/desorption efficiency) of PCP.

Fig. 2. SALDI mass spectra of 1 ppb, 100 ppt, and 10 ppt PCP after extraction from aqueous solution. These mass spectra were obtained in negative ion mode at a laser power of 65 (a.u.).

Fig. 3. SALDI mass spectra of pyrene after the extraction from the aqueous solution of 10 ppb, 1 ppb, and 100 ppt. These mass spectra were conducted in positive ion mode at the laser power of 65 (a.u.).

To further investigate these factors, we compared the SALDI efficiency of PCP and pyrene by examining the detectability without applying the extraction process (i.e., direct deposition of analyte onto the CoC–NH₂ nanoparticles on the MALDI plate). The detectability of pyrene was observed to be 100 ppb, while the detectability of PCP was 1 ppm (=1000 ppb) (Fig. 4). This is contrast to the case of SALDI-MS with applying the extraction process. These results indicated that high pre-accumulation and concentration efficiency of PCP by CoC–NH₂ nanoparticles was responsible for the superior detectability by SALDI-MS. In fact, the adsorption amount from 20 ppb PCP in water including a 1 mg L⁻¹ mixture of CoC–NH₂ nanoparticles was estimated to be 33 μmol g⁻¹, while the adsorption amount from 20 ppb pyrene in water was 12 μmol g⁻¹. The adsorption amount of pyrene was estimated from the calibration curve based on the fluorescence spectra, while that of PCP was obtained from the calibration curve based on the LC APCI-MS (Supplementary data, Figure S3).

Fig. 4. SALDI mass spectra of (a) pyrene and (b) PCP without extraction from the aqueous solution. These mass spectra were obtained in (a) positive ion mode and (b) negative ion mode at the laser power of 65 (a.u.).

The unique chemical structure of the CoC–NH₂ nanoparticles may have contributed to the higher extraction efficiency of PCP in several ways. The graphene sheets may have strongly interacted with the PCP molecule.^22–24) The water-wettability of benzylamine groups on the CoC–NH₂ nanoparticles may also have enhanced the retention of PCP. It should be noted that better extraction efficiency (i.e., higher recovery ratio of PCP) was achieved when the amount of CoC–NH₂ nanoparticles was increased from 1 mg L⁻¹ to more than 20 mg L⁻¹. However, the increase in the amount of nanoparticles resulted in the decrease of the SALDI-MS peak intensities. Therefore, we determined the optimal concentration of CoC–NH₂ nanoparticles to be 1 mg L⁻¹ in the extraction process.

In a previous study, we demonstrated that the CoC–NH₂ nanoparticles can effectively enrich PFOS in water, following identification with SALDI-MS at high sensitivity (∼0.1 ng L⁻¹).¹⁹⁾ More recently, it has been reported that humans and wildlife are globally exposed to polyfluorinated compounds including PFOS, which is known to bioaccumulate in the human body at several tens of ppb in human serum.²⁵⁾ Therefore, application of the SALDI-MS technique for the detection of PFOS in human serum is beneficial and thus was performed in this study (Fig. 5). The detectability of PFOS by the SALDI-MS technique was determined to be 10 ppb in human serum. The relatively high detectability of PFOS even in a complex sample such as human serum may be useful for high-throughput screening of PFOS in the human body. The quantification of PFOS in human serum is important for demonstrating the practical use, but it was difficult to construct the calibration curve on the SALDI-MS for the quantification because of large variations in the peak intensities. Thus, the quantification of PFOS in human serum using the SALDI-MS still remains an open research problem.

Fig. 5. SALDI mass spectra of PFOS after the extraction from human serum of 50 ppb, 10 ppb, and 1 ppb. These mass spectra were conducted in negative ion mode at the laser power of 65 (a.u.).

It should be noted that the addition of the acids into human serum in the extraction process of PFOS was necessary to detect PFOS by the SALDI-MS, since non-covalent interactions occur between PFOS and serum albumin under normal physiological conditions.^26,27) It is common in bioanalysis to add an acid (often phosphoric acid) to release compounds associated with proteins. The association is broken due to shielding of charge interactions. Thus, the serum albumin-to-PFOS binding can disturb the extraction of PFOS by CoC–NH₂ nanoparticles under the acid condition. Our results suggested that the denaturing of serum albumin occurred under the acidic conditions of pH ∼1–2, which allowed for the extraction of PFOS in human serum with the use of CoC–NH₂ nanoparticles and subsequent detection by SALDI-MS.

CONCLUSION

We have demonstrated the detection of aromatic hydrophobic compounds of anthracene, pyrene, and PCP by SALDI-MS with the use of CoC–NH₂ nanoparticles. The nanoparticles were effective at capturing PCP in water because of the high adsorption capacity, which resulted in higher detectability (∼100 ppt). Furthermore, CoC–NH₂ nanoparticles were applied to the extraction of PFOS from human serum, displaying good performance with a detectability of 10 ppb by the SALDI-MS. The applicability of CoC–NH₂ nanoparticles in conjunction with SALDI-MS is demonstrated for the enrichment and detection of polycyclic aromatic hydrocarbons (PAHs) and pentachlorophenol (PCP) as well as PFOS. The method presented in our work has the potential to be applied to other compounds of interest in biological and environmental sciences.

Acknowledgements

We thank Yumiko Ueno at Kansai University for the measurements of LC-ESI-MS. This study was supported in part by the “Strategic Project to Support the Formation of Research Bases at Private Universities” with a matching Fund Subsidy from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan. This study was also partially supported by Grants-in-Aid for Scientific Research (Nos. 23360361, 23655074, and 22350040) from the Japan Society for the Promotion of Science (JSPS), and Network Joint Research Center for Materials and Devices (2013A20).

REFERENCES

• 1) M. Safarikova, I. Safarik. Magnetic techniques for the isolation and purification of proteins and peptides. Biomagn. Res. Technol. 2: 7, 2004.
• 2) Q. Liu, J. Shi, G. Jiang, Q. Liu. Application of graphene in analytical sample preparation. Trends Analyt. Chem. 37: 1–11, 2012.
• 3) J. Chen, J. Zou, J. Zeng, X. Song, J. Ji, Y. Wang, J. Ha, X. Chen. Preparation and evaluation of graphene-coated solid-phase microextraction fiber. Anal. Chim. Acta 678: 44–49, 2010.
• 4) H. Zhang, H. K. Lee. Plunger-in-needle solid-phase microextraction with graphene-based sol–gel coating as sorbent for determination of polybrominated diphenyl ethers. J. Chromatogr. A 1218: 4509–4516, 2011.
• 5) Q. Chang, S. Song, Y. Wang, J. Li, J. Ma. Application of graphene as a sorbent for preconcentration and determination of trace amounts of chromium(III) in water samples by flame atomic absorption spectrometry. Anal. Methods 4: 1110–1116, 2012.
• 6) Y. Wang, S. Gao, X. Zang, J. Li, J. Ma. Graphene-based solid-phase extraction combined with flame atomic absorption spectrometry for a sensitive determination of trace amounts of lead in environmental water and vegetable samples. Anal. Chim. Acta 716: 112–118, 2012.
• 7) V. Chandra, J. Park, Y. Chun, J. W. Lee, I. C. Hwang, K. S. Kim. Water-dispersible magnetite-reduced graphene oxide composites for arsenic removal. ACS Nano 4: 3979–3986, 2010.
• 8) G. Zhao, S. Song, C. Wang, Q. Wu, Z. Wang. Determination of triazine herbicides in environmental water samples by high-performance liquid chromatography using graphene-coated magnetic nanoparticles as adsorbent. Anal. Chim. Acta 708: 155–159, 2011.
• 9) Q. H. Wu, G. Y. Zhao, C. Feng, C. Wang, Z. Wang. Preparation of a graphene-based magnetic nanocomposite for the extraction of carbamate pesticides from environmental water samples. J. Chromatogr. A 1218: 7936–7942, 2011.
• 10) W. Wang, X. Ma, Q. Wu, C. Wang, X. Zang, Z. Wang. The use of graphene-based magnetic nanoparticles as adsorbent for the extraction of triazole fungicides from environmental water. J. Sep. Sci. 35: 2266–2272, 2012.
• 11) W. N. Wang, Y. P. Li, Q. H. Wu, C. Wang, X. H. Zang, Z. Wang. Extraction of neonicotinoid insecticides from environmental water samples with magnetic graphene nanoparticles as adsorbent followed by determination with HPLC. Anal. Methods 4: 766–772, 2012.
• 12) M. Tian, W. Feng, J. Yea, Q. Jia. Preparation of Fe₃O₄·TiO₂/graphene oxide magnetic microspheres for microchip-based preconcentration of estrogens in milk and milk powder samples. Anal. Methods 5: 3984–3991, 2013.
• 13) C. T. Chen, Y. C. Chen. Fe₃O₄/TiO₂ core/shell nanoparticles as affinity probes for the analysis of phosphopeptides using TiO₂ surface-assisted laser desorption/ionization mass spectrometry. Anal. Chem. 77: 5912–5919, 2005.
• 14) W. Y. Chen, Y. C. Chen. Affinity-based mass spectrometry using magnetic iron oxide particles as the matrix and concentrating probes for SALDI MS analysis of peptides and proteins. Anal. Bioanal. Chem. 386: 699–704, 2006.
• 15) H. Kawasaki, A. Tarui, T. Watanabe, K. Nozaki, T. Yonezawa, R. Arakawa. Sulfonate group-modified FePtCu nanoparticles as a selective probe for LDI-MS analysis of oligopeptides from a peptide mixture and human serum proteins. Anal. Bioanal. Chem. 395: 1423–1431, 2009.
• 16) Y. Iwaki, H. Kawasaki, R. Arakawa. Human serum albumin-modified Fe₃O₄ magnetic nanoparticles for affinity-SALDI-MS of small-molecule drugs in biological liquid. Anal. Sci. 28: 893–900, 2012.
• 17) S. Taira, Y. Sahashi, S. Shimma, T. Hiroki, Y. Ichiyanagi. Nanotrap and mass analysis of aromatic molecules by phenyl group-modified nanoparticle. Anal. Chem. 83: 1370–1374, 2011.
• 18) H. N. Abdelhamida, H. F. Wu. Multifunctional graphene magnetic nanosheet decorated with chitosan for highly sensitive detection of pathogenic bacteria. J. Mater. Chem. B 1: 3950–3961, 2013.
• 19) H. Kawasaki, K. Nakai, R. Arakawa, E. K. Athanassiou, R. Grass, W. J. Stark. Functionalized graphene-coated cobalt nanoparticles for highly efficient surface-assisted laser desorption/ionization mass spectrometry analysis. Anal. Chem. 84: 9268–9275, 2012.
• 20) D. G. Crosby. Enviromental chemistry of pentachlorophenol. Pure Appl. Chem. 53: 1051–1080, 1981.
• 21) K. H. Kim, S. A. Jahan, E. Kabir, R. J. C. Brown. A review of airborne polycyclic aromatic hydrocarbons (PAHs) and their human health effects. Environ. Int. 60: 71–80, 2013.
• 22) J. Zhang, X. Dong, J. Cheng, J. Li, Y. Wang. Efficient analysis of non-polar environmental contaminants by MALDI-TOF MS with graphene as matrix. J. Am. Soc. Mass Spectrom. 22: 1294–1298, 2011.
• 23) Q. Liu, J. Shi, L. Zeng, T. Wang, Y. Cai, G. Jiang. Evaluation of graphene as an advantageous adsorbent for solid-phase extraction with chlorophenols as model analytes. J. Chromatogr. A 1218: 197–204, 2011.
• 24) S. Zhanga, H. Niua, Z. Hua, Y. Caia, Y. Shiam. Preparation of carbon coated Fe₃O₄ nanoparticles and their application for solid-phase extraction of polycyclic aromatic hydrocarbons from environmental water samples. J. Chromatogr. A 1217: 4757–4764, 2010.
• 25) A. E. Loccisano, J. L. Campbell Jr., M. E. Andersen, H. J. Clewell III. Evaluation and prediction of pharmacokinetics of PFOA and PFOS in the monkey and human using a PBPK model. Regul. Toxicol. Pharmacol. 59: 157–175, 2011.
• 26) X. Zhang, L. Chen, X. C. Fei, Y. S. Ma, H. W. Gao. Binding of PFOS to serum albumin and DNA: Insight into the molecular toxicity of perfluorochemicals. BMC Mol. Biol. 10: 16, 2009.
• 27) M. Salvalaglio, I. Muscionico, C. Cavallotti. Determination of energies and sites of binding of PFOA and PFOS to human serum albumin. J. Phys. Chem. B 114: 14860–14874, 2010.