Gold Nanoparticles and Imaging Mass Spectrometry: Double Imaging

Feb 3, 2010 - Latent fingerprint (LFP) detection is a top-priority task in forensic science. It is a simple and effective means for the identification...
0 downloads 16 Views 4MB Size
Anal. Chem. 2010, 82, 1589–1593

Letters to Analytical Chemistry Gold Nanoparticles and Imaging Mass Spectrometry: Double Imaging of Latent Fingerprints Ho-Wai Tang, Wei Lu, Chi-Ming Che,* and Kwan-Ming Ng* Department of Chemistry and Open Laboratory of Chemical Biology of the Institute of Molecular Technology for Drug Discovery and Synthesis, The University of Hong Kong, Pokfulam Road, Hong Kong SAR, China Latent fingerprint (LFP) detection is a top-priority task in forensic science. It is a simple and effective means for the identification of individuals. Development of nanomaterials which maximize the surface interaction with endogenous substances on the ridges to enhance the contrast of the fingerprints is an important application of nanotechnology in LFP detection. However, most developments in this area have mainly focused on the visualization of the physical pattern of the fingerprints and failed to explore the molecular information embedded in LFPs. Here, we have integrated certain distinctive properties of gold nanoparticles (AuNPs) with imaging mass spectrometry for both the visualization and molecular imaging of LFPs. Two contrasting colors (blue and pink), arising from different surface plasmon resonance (SPR) bands of the AuNPs, reveal the optical images of LFPs. The laser desorption/ionization property of the AuNPs allows the direct analysis of endogenous and exogenous compounds embedded in LFPs and imaging their distributions without disturbing the fingerprint patterns. The simultaneous visualization of LFP and the recording of its molecular images not only provide evidence on individual identity but also resolve overlapping fingerprints and detect hazardous substances. Latent fingerprint (LFP) detection is an important task in forensic science. Most detection techniques for the visualization of LFPs use staining agents interacting with endogenous compounds secreted on skin, including glycerides, fatty acids, and amino acids.1 However, direct analysis of chemicals embedded in LFPs without damaging the fingerprint pattern remains a challenging task. Derivatization with fluorescence tag(s) on metal nanoparticles is a commonly adopted procedure for probing chemical information in LFPs based on the specificity of * To whom correspondence should be addressed. K.-M. Ng: fax, (852) 2857 1586; e-mail, [email protected]. C.-M. Che: fax, (852) 2857 1586; e-mail, [email protected]. (1) Ramotowski, R. S. Composition Of Latent Print Residue. In Advances In Fingerprint Technology, 2nd ed.; Lee, H. C., Gaensslen, R. E., Eds.; CRC Press: Boca Raton, FL, 2001. 10.1021/ac9026077  2010 American Chemical Society Published on Web 02/03/2010

antigen-antibody interactions.2 However, the application of specific interactions on trace amounts of fingerprint samples usually requires prior knowledge of target chemicals/biochemicals. Tagfree techniques, such as desorption electrospray ionization mass spectrometry (DESI-MS)3 and matrix-assisted laser desorption/ ionization mass spectrometry (MALDI-MS),4 have been reported for the chemical analysis/imaging of LFPs. However, these methods did not have the capability of visualizing the optical image of LFPs prior to the molecular imaging and thus are suitable for the analysis of LFPs at predefined/known positions. Development of a double imaging technique for both the visualization of the physical pattern and the recording of the chemical distribution in LFPs is of great practical interest. Nanomaterials have been attracting considerable attention in diverse disciplines due to their large surface area to volume ratio, as well as various distinctive physical and/or chemical properties. A key characteristic is the superior performance of nanomaterials in enhancing the sensitivity and/or specificity in measurement sciences.5,6 When gold metal is nanosized, it exhibits distinctive optical properties due to the surface plasmon resonance (SPR).7,8 The color of gold nanoparticles (AuNPs) can be tuned over a wide spectral range from blue to red by changing the particle size, shape, and their surrounding molecules. This fascinating property has been used for colorimetric sensing, surface-enhanced Raman scattering, and bioimaging.9 Another unique property of AuNPs is the laser desorption/ionization property. Irradiation of AuNPs with a UV laser can induce rapid heating in a small volume of the nanoparticles. This property can facilitate the efficient desorption/ ionization of molecules adsorbed on AuNPs. In this work, we have first integrated the dual properties of gold nanoparticles (AuNPs), (2) Hazarika, P.; Jickells, S. M.; Wolff, K.; Russell, D. A. Angew. Chem., Int. Ed. 2008, 47, 10167–10170. (3) Ifa, D. R.; Manicke, N. E.; Dill, A. L.; Cooks, R. G. Science 2008, 321, 805. (4) Wolstenholme, R.; Bradshaw, R.; Clench, M. R.; Francese, S. Rapid Commun. Mass Spectrom. 2009, 23, 3031–3039. (5) Murray, R. W. Anal. Chem. 2009, 81, 1723. (6) Wilson, R. Chem. Soc. Rev. 2008, 37, 2028–2045. (7) Kelly, K. L.; Coronado, E.; Zhao, L. L.; Schatz, G. C. J. Phys. Chem. B 2003, 107, 668–677. (8) Ni, W.; Kou, X.; Yang, Z.; Wang, J. ACS Nano 2008, 2, 677–686. (9) Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Van Duyne, R. P. Nat. Mater. 2008, 7, 442–453.

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

1589

i.e., surface plasmon resonance and laser desorption/ionization properties, for visualization of LFPs from macroscopic to molecular scales (Supporting Information, Figure S1). EXPERIMENTAL SECTION Chemicals and Materials. Fatty acids (palmitic acid, oleic acid, and stearic acid) and verapamil hydrochloride were all purchased from Sigma-Aldrich (St. Louis, MO) with purity not less than 95%. HPLC grade methanol and dichloromethane were purchased from Lab-Scan (Thailand). HPLC grade water was purchased from Tedia (Fairfield, OH). Analytical reagent grade ethanol was from Sasma (The Netherlands). The substrates used for LFP deposition, including glass coverslips, transparent photocopy plastic film, and white paper (60 g/m2), had a thickness of less than 0.2 mm. The plastic film was washed with methanol, and glass slides were washed with dichloromethane followed by methanol before use. Preparation of Latent Fingerprints. All LFPs were voluntarily provided by a Chinese male. To prepare sebum-rich LFPs, the donor wiped his thumb on his forehead for about 10 s, and then pressed his thumb on the desired substrates gently for about 10 s. The LFP donor’s thumb was first cleaned with an ethanol-soaked tissue (Kimwipes from Kimberly-Clark, Irving, TX) several times prior to wiping on the forehead. In the preparation of the exogenous compound-containing LFP, the procedure was similar to that for preparing sebum-rich LFPs. The exogenous compound (5 mg of verapamil hydrochloride powder) was added and rubbed on the sebum-rich thumb for about 20 s prior to LFP deposition. In the preparation of the overlapped LFPs (i.e., LFP I at the bottom overlapped with LFP II on the top) containing different compounds, three fatty acid standards were employed. The LFP donor’s thumb was first cleaned with an ethanol-soaked tissue to avoid contamination. The thumb was then applied with a small amount of the mixture containing stearic acid and oleic acid (6 mg and 5 µL, respectively) for blotting LFP I on a glass substrate. Then, the thumb was cleaned thoroughly, and a small amount of another mixture containing palmitic acid and oleic acid (6 mg and 5 µL respectively) was applied for blotting LFP II on the top of LFP I. All LFPs were allowed to stand for 1 h under normal room conditions prior to the development by gold sputtering. Visualization of Latent Fingerprints by Gold Sputtering. To develop the LFPs, the LFP-containing substrates (glass, plastic, and paper) were put into the chamber of a sputter coater (SCD 005; Bal-Tec AG). The system uses high-purity argon gas (99.999%) as the sputtering gas and a high-purity gold target (99.99%; Ted Pella Inc., Redding, CA). The sputtering conditions were optimized to maximize the contrast of the optical images of the LFPs. The distance between the substrate and the target was 80 mm. The chamber was first evacuated to a pressure of 0.02-0.03 mbar and flushed with argon gas more than 3 times before gold sputtering. In the sputtering period, the sputtering electrical current was set at 12 mA, and the chamber pressure was maintained at approximately 0.04-0.05 mbar by adjusting the flow of the argon gas. The optimal sputtering time was 130 s. The optical images of the developed LFPs were recorded by using a digital camera equipped with a macro lens under an electronic flash (EOS 400D with EF 100 mm f/2.8 Macro USM and Speedlite 430 EX II, Canon, Japan). 1590

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

Figure 1. Visualization of LFPs blotted on nonporous and porous substrates: plastic (a), glass (b), and paper (c) by gold sputtering (at a sputtering time of 130 s), two contrasting colors (d), pink and blue, are observed in the ridges and grooves of the fingerprints, respectively. Transmission electron microscopic examination of the distribution, size, and morphology of the AuNPs deposited (at a sputtering time of 80 s). The AuNPs homogenously distributed in the ridges of the fingerprints are smaller (∼3 nm) and in a pseudospherical shape (e), while the gold nanoislands formed in the groove of the fingerprints are at least 3 times larger (∼10 nm) and are in an irregular shape (f). The two contrasting colors are attributed to the two SPR bands in the visible wavelength (λmax at 592 nm (g) and λmax at 660 nm (h) determined at a sputtering time of 120 s), with an increase of the sputtering time of the AuNPs increasing the absorbance of the two SPR bands and thus enhancing the contrast of the image.

To characterize the morphology and size of the AuNPs deposited in the ridges and grooves of LFP, a LFP was blotted onto a transmission electron microscopy grid (Formvar/Carbon 400 mesh Cu Grid; SPI Supplies, West Chester, PA) and developed by gold sputtering, followed by examination under a transmission electron microscope (Tecnai G2 20 S-TWIN; FEI, Hillsboro, OR). The gold sputtering time was adjusted to 80 s to reduce the amount of AuNPs deposited into the LFP and thus avoid the interference of overlapped AuNPs on the observation of their size and morphology.

Figure 2. Molecular imaging of LFPs by gold nanoparticle-assisted laser desorption/ionization mass spectrometry. Schematic diagram showing the development of an LFP using gold sputtering (a), direct detection of endogenous chemicals in the LFP upon N2 laser irradiation (b), and molecular imaging of the LFP based on spatial distributions of different endogenous compounds (c).

To record the visible absorption spectra of the AuNPs deposited in the ridges and grooves of the LFP, AuNPs were sputtered onto clean glass coverslips (to mimic the grooves without the fatty substances) and onto the coverslips smeared with a thin layer of sebum (mimicking the ridges with the fatty substances). The visible absorption spectra of the AuNPs deposited on the glass coverslips at different sputtering times were recorded using a UV-Vis-NIR spectrometer (Lambda 900; PerkinElmer, Waltham, MA) from 350 to 800 nm with a scanning stepsize of 0.5 nm. Molecular Imaging of Latent Fingerprints by Imaging Mass Spectrometry. All of the imaging mass spectrometry (IMS) experiments were performed by using a Voyager-DE STR matrixassisted laser desorption ionization time-of-flight mass spectrometer (MALDI-TOF MS, Applied Biosystems, Foster City, CA) equipped with a nitrogen laser (337 nm, maximum pulse energy: 300 µJ). The AuNP-deposited-LFP sample was mounted onto a standard stainless steel MALDI plate using electrical conductive tapes (9713 XYZ-Axis; 3M, St. Paul, MN). All the IMS experiments were performed in the reflectron mode, and the instrumental parameters were set as follows: accelerating voltage, 20 000 V; grid voltage, 85%; guide wire, 0.05; delay time, 500 ns; mirror voltage ratio, 1.12; laser pulsing intensity, 2 600 arbitrary unit (AU) for negative mode/3 000 AU for positive mode; laser pulsing frequency, 3 Hz. At the laser pulsing intensity of 2 600 AU, the effective dimensions of the laser spots in the elliptical shape were found to be ∼50 µm × 150 µm on glass, ∼60 µm × 170 µm on plastic, and ∼80 µm × 200 µm on paper. The lateral movement resolution (i.e., center-to-center distance of the laser spot movement) for the acquisition of molecular images of the LFPs on glass, plastic, and paper were set at 60, 70, and 90 µm, respectively, to minimize the extent of overlapping of laser spots. The acquisition time of a mass spectrum at each spatial position of the LFP was

1 s. The mass spectra were acquired in the m/z range of 180-1000. At a resolution of 60 µm, a LFP with an area of 1.3 × 1.8 cm2 would be imaged in about 18 h. The instrument was controlled via the Voyager Control Panel system software 5.10.2, and the IMS experiments were carried out via the MMSIT software (version 2.2.1, Norvatis, Basel, Switzerland). The IMS data and molecular images of the LFPs were processed and exported using BioMap software (version 3.7.5, Norvatis). To obtain good quality LFP molecular images on paper, additional gold sputtering (sputtering distance, 53 mm; sputtering current, 30 mA; sputtering time, 10 s) was required to enhance the electrical conductivity of the LFPs, while the LFPs on glass and plastic did not require additional gold sputtering. Accurate mass measurements of endogenous chemicals in the LFPs were performed by using a QSTAR-XL quadrupole time-offlight tandem mass spectrometer (Applied Biosystems/MDS Sciex, Ontario, Canada) equipped with a MALDI source and a nitrogen laser (337 nm). To desorb chemicals in the LFPs without saturating the MCP detector, the laser pulse energy was set at 40 µJ at a pulsing frequency of 40 Hz. Ions were then mass analyzed by the TOF mass analyzer and detected by a microchannel plate (MCP) detector set at 2.6 kV. The mass spectra were acquired in the m/z range of 50-1000. The mass resolution of the MS system was greater than 5 000 (full width at half-height), and the mass scale was calibrated to achieve the mass accuracy of less than 10 ppm. Gold cluster ion peaks were employed for the internal mass calibration. The vacuum pressure in the quadrupole mass analyzer and TOF mass analyzer were recorded as ∼3.7 × 10-5 and ∼3.3 × 10-7 Torr, respectively. Instrumental control and data acquisition were performed via the Analyst 1.4 software. Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

1591

SAFETY CONSIDERATIONS Verapamil is an L-type calcium channel blocker. Its powder should be handled in a chemical fume hood to avoid inhalation. Fingers should be washed thoroughly with water after contacting the verapamil powder. RESULTS AND DISCUSSION Colloidal AuNPs have been employed for the development of LFPs by immersing the samples in AuNPs solution.10 Gold deposition by using the vacuum metal deposition method has also been reported for the development of LFPs.11 Here, we found that AuNPs generated from argon ion sputtering aggregate on the ridges and grooves of LFPs in two different forms, exhibiting two contrasting colors (i.e., pink on ridges and blue on grooves, Figure 1), allowing the clear visualization of LFPs on different substrates, including plastic, glass, and paper. The AuNPs homogenously deposited in the ridges and grooves of LFPs exhibited different extents of aggregation. The interaction/entrapment of the AuNPs in the ridges by a mixture of endogenous compounds, such as fatty acid components, might prevent the AuNPs from aggregation and thus favor the formation of AuNPs of smaller size and pseudospherical shape (Figure 1e), whereas the gold nanoislands formed in the grooves were of irregular shape and larger size (Figure 1f). The pseudospherical gold nanoparticles with smaller aspect ratio of ∼1 in the ridges and the irregular gold nanoislands with aspect ratio of >2 in the grooves exhibited different absorption bands in the visible region (Figures 1g and 1h). The developed fingerprints are stable for more than 1 month. In contrast, an optical image of an LFP developed by using silver nanoparticles (another noble metal nanoparticles with significant SPR property) was unstable and became blurred immediately on coming into contact with moisture/air upon leaving the vacuum chamber of the sputtering system (data not shown). Deriving chemical information directly from an LFP is helpful in obtaining more information relevant to individual identity. For instance, composition of endogenous substances in LFPs, such as lipid composition, has been reported to be relevant to the age and sex of individuals.1 In addition, the detection of exogenous compounds within an LFP, such as hazardous or illicit substances, would be useful for forensic investigations. We found that AuNPs is an effective medium for sensitive detection of endogenous compounds embedded in LFPs by using laser desorption/ ionization mass spectrometry. Abundant ions at m/z 227, 241, 253, 255, 281, and 283 were clearly revealed in the mass spectrum upon N2 (337 nm) laser irradiation of a gold nanoparticle-coated LFP, as shown in Figure 2b. In terms of accurate mass measurement, the m/z values of the ions (at m/z 227.2025, 241.2188, 253.2188, 255.2343, 281.2508, and 283.2653) match the m/z values of deprotonated myristic acid ([myristic acid - H]- at m/z 227.2011), deprotonated pentadecanoic acid ([pentadecanoic acid - H]at m/z 241.2168), deprotonated palmitoleic acid ([palmitoleic acid - H]- at m/z 253.2168), deprotonated palmitic acid ([palmitic acid - H]- at m/z 255.2324), deprotonated oleic acid ([oleic acid - H]- at m/z 281.2481), and deprotonated stearic acid ([stearic acid - H]- at m/z 283.2637), which have been (10) Stauffer, E.; Becue, A.; Singh, K. V.; Thampi, R.; Champod, C.; Margot, P. Forensic Sci. Int. 2007, 168, e5-e9. (11) Jones, N.; Stoilovic, M.; Lennard, C.; Roux, C. Forensic Sci. Int. 2001, 115, 73–88.

1592

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

Figure 3. Derivation of various information from molecular images of LFPs recorded using gold nanoparticle-assisted laser desorption/ ionization mass spectrometry. Detection of an exogenous substance, verapamil, embedded in an LFP (a) at a laser pulsing intensity of 3 000 AU and in the positive ion mode. Resolving the physical domains and time domains of the two overlapped fingerprints by their respective molecular images (b) recorded at a laser pulsing intensity of 2 600 AU and in the negative ion mode: the optical images of overlapped LFPs were unresolved (i); a specific chemical image recorded at m/z 283 arising from the fingerprint I (ii) and another specific chemical image recorded at m/z 255 arising from the fingerprint II (iii) allowed the unambiguous separation of the overlapped fingerprints and the differentiation of their overlapping sequence.

commonly found in LFPs.1 The spatial distribution of endogenous fatty substances can be used to construct the molecular images of the LFP (Figure 2c) which are identical to the optical image. In addition, we have demonstrated that the presence of a trace amount of a hazardous substrate, such as verapamil (a calcium channel blocker for treatment of cardiovascular disease) in an LFP, can be directly detected upon N2 laser irradiation. Reconstruction of the image at m/z 453 ([verapamil - H]+) reveals its contact area in the fingerprint (Figure 3a). Another advantage of breaking an optical image into molecular images is the capability of separating two overlapped LFPs (physical domains) based on the presence of specific compounds in the respective molecular images of the overlapped LFPs (Figure 3b). Upon inspection of the two resolved LFPs, we found that the first fingerprint at the bottom of the overlap (fingerprint I shown in Figure 3b,ii) displays minor interrupting patterns in the overlapping region due to the masking of the second fingerprint on the

top (fingerprint II shown in Figure 3b,iii), which could be used to differentiate the overlapping sequence (time domains) of the overlapped LFPs. CONCLUSIONS The method described in this work allows the development of an optical image of LFP in the dry state and is able to preserve chemical information embedded in the fingerprint for sequential analysis by imaging mass spectrometry. The different forms of AuNPs exhibiting different SPR properties allow clear visualization of LFPs by inspection with naked eyes. The laser desorption/ ionization property of AuNPs makes it an effective medium for direct analysis of small molecules within LFPs and the generation of molecular images. The complementary nature of the double images of LFPs, i.e., optical and molecular images, provides physical pattern of fingerprints for individual identification and reveals chemical information within fingerprints for forensic investigations. When considering the electrical conductivity of AuNPs previously reported for scanning electron microscopic (12) Garner, G. E.; Fontan, C. R.; Hobson, D. W. J. Forensic Sci. Soc. 1975, 15, 281–288.

(SEM) analysis of fingerprints at microscopic scale,12 we conceive that AuNPs could act as an effective medium for bridging different imaging techniques from macroscopic (naked eye inspective) to microscopic (SEM examination) levels and even to molecular analysis (mass spectrometric imaging). ACKNOWLEDGMENT We thank Mr. Frankie Y. F. Chan of the Electron Microscope Unit of The University of Hong Kong for assisting in the TEM measurements. This project is supported by the Area of Excellence Scheme (Grant AoE/P-10/01) administered by the University Grant Council (Hong Kong SAR, China) and the Strategic Research Theme on Drug administered by the University of Hong Kong. SUPPORTING INFORMATION AVAILABLE Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org. Received for review November 13, 2009. Accepted January 14, 2010. AC9026077

Analytical Chemistry, Vol. 82, No. 5, March 1, 2010

1593