mass spectrometry

mass spectrometry

Journal of Chromatography B, 1003 (2015) 35–40 Contents lists available at ScienceDirect Journal of Chromatography B journal homepage: www.elsevier...

741KB Sizes 0 Downloads 93 Views

Journal of Chromatography B, 1003 (2015) 35–40

Contents lists available at ScienceDirect

Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb

Profiling of volatile compounds in APCMin/+ mice blood by dynamic headspace extraction and gas chromatography/mass spectrometry Shoji Kakuta a , Shin Nishiumi b , Masaru Yoshida b,c , Eiichiro Fukusaki a , Takeshi Bamba a,d,∗ a

Department of Biotechnology, Department of Engineering, Osaka University, 2-1 Yamadaoka, Suita, Osaka 565-0871, Japan Division of Gastroenterology, Department of Internal Medicine, Kobe University Graduate School of Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan c Division of Metabolomics Research, Kobe University Graduate School of Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobe, 650-0017, Japan d Division of Metabolomics, Research Center for Transomics Medicine, Medical Institute of Bioregulation, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan b

a r t i c l e

i n f o

Article history: Received 21 March 2015 Received in revised form 30 August 2015 Accepted 2 September 2015 Available online 11 September 2015 Keywords: Metabolomics GC/MS Dynamic headspace extraction Volatile APCMin/+ mice

a b s t r a c t Various volatile compounds as well as hydrophilic compounds exist in the blood. For example, 2-alkenals, 4-hydroxy-2-alkenals, and ketoaldehydes have been reported as oxidized lipid-derived volatiles in blood. These specific volatiles have been associated with diseases; however, multi-volatile analyses have not been performed. In this study, volatile profiling of APCMin/+ mouse plasma by dynamic headspace extraction was performed for multi-volatile analysis. In total, 19 volatiles were detected in the plasma of mice, based on information regarding oxidized lipid-derived volatile compounds, and eight of these compounds differed significantly between normal and diseased mice. 2-Methyl-2-butanol and benzyl alcohol were previously unreported in blood samples. Furthermore, 3,5,5-trimethyl-2(5H)-furanone was only detected in normal mice. 5-Methyl-3-hexanone and benzaldehyde have been detected in subjects with gastrointestinal diseases and lung cancer, respectively. Therefore, volatile profiling can be used to detect differences between samples and to identify compounds associated with diseases. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Metabolomics methods have been widely used in recent studies. The target compounds in metabolomics studies are metabolites. Slight differences between samples can be detected based on exhaustive quantitative metabolite profiling, and complex biological variation can be examined using multivariate analysis [1]. Multi-marker profiling using metabolomics techniques has been applied to analyze samples that lack apparent phenotypic differences, such as food quality evaluations [2] or budding yeast lifespan predictions [3]. This technique has also been used for clinical diagnosis in the medical field. Researchers have attempted to distinguish cancer from control samples based on metabolic profiling of tumor cells

∗ Corresponding author at: Division of Metabolomics, Research Center for Transomics Medicine, Medical Institute of Bioregulation, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan. Fax: +81 92 642 6172. E-mail address: [email protected] (T. Bamba). http://dx.doi.org/10.1016/j.jchromb.2015.09.002 1570-0232/© 2015 Elsevier B.V. All rights reserved.

or urine samples from colorectal cancer patients; however, specific biomarkers have not been identified [4,5]. In a recent study, 2hydroxybutyratem aspartic acid, kynurenine, and cystamine were identified as biomarkers of colorectal cancer based on a multivariate analysis of hydrophilic metabolites in serum samples. A prediction model that included levels of these four metabolites enabled the detection of stage 0–2 colorectal cancer with 82.8% accuracy [6]. Volatile compounds also exist in the body, including hydrophilic metabolites, and some are known to be bioactive compounds. For example, 4-hydroxy-2-nonenal (4-HNE) enhances cyclooxygenase-2 gene expression and is associated with macrophage foam cells during the initial progression of arteriosclerosis [7]. Therefore, volatile compounds seem to act as secondary messengers to regulate biological activity as well as the formation of adducts of proteins or nucleic acids. Oxidized lipids produce many compounds, including aldehydes and carboxylic acids, as secondary products [8]. Oxidized lipid-derived volatiles are classified into three groups, 2-alkenals, 4-hydroxy-2-alkenals, and ketoaldehydes [9–14]. Various lipid compounds exist in

36

S. Kakuta et al. / J. Chromatogr. B 1003 (2015) 35–40

the blood [15], and volatile compounds are released from these oxidized compounds. Therefore, multivariate profiling analysis of volatile compounds has been a focus of recent studies [16]. APCMin/+ mice are characterized by familial adenomatous polyposis and eventually develop bowel cancer. Recently, volatile analysis of these mice was performed, and the metabolism of 4-HNE, a secondary oxidized lipid-derived product, was found to be higher in APCMin/+ cells than in normal cells [17]. The 4HNE metabolism in APCMin/+ cells is related to cell survival after exposure to toxic volatile compounds such as various aldehydes and promotes the development of bowel cancer. Moreover, DNA adducts with oxidized lipid compounds form in APCMin/+ mice [18]. However, a multi-volatile compound analysis has not been performed, even though variation in the levels of specific volatile compounds has been observed in biological samples. There are various extraction methods and liquid–liquid extraction method has been widely applied to prepare samples. It is able to used optimum solvent from various solvent and there are many applications in each solvent [19]. However, this extraction needs desolvation and concentration processes as well as extraction process and has been required complex time consuming process in whole extraction process [20]. These process leads to some problems such as low accuracy, contamination or loos of volatile compounds [19,21]. In addition, this method could extract hydrophilic compounds as well as volatile compounds simultaneously [22], and extraction efficiency would be changed due to interaction with various compounds. However, volatile compounds were existed at trace concentration in body and it needed concentration process before analysis. For volatile compound multi profiling analysis, it is important to extract and concentrate various volatiles from samples. Therefore, we have been focusing on a dynamic headspace extraction method for volatile compounds profiling. We reported a multi-volatile compound profiling system that uses dynamic headspace gas chromatography/mass spectrometry (DHS-GC/MS) [23]. This method involves dynamic extraction based on the purging of headspace by inert gas and the trapping of volatile compounds by using sorbents. This method effectively extracts volatile compounds by purging almost all the headspace for the sorbent trap [24,25]. Because various trace volatile compounds exist in the blood, this method is a suitable extraction technique for the analysis of multiple volatile compounds. Therefore, we attempted to apply this system to APCMin/+ mouse samples and analyzed the profile of multiple volatile compounds.

2. Experimental 2.1. Materials Methanol (HPLC grade) was purchased from Kishida Chemical Co., Ltd. (Osaka, Japan). Sodium chloride (pesticide residue polychlorinated biphenyls analysis grade), distilled water (HPLC grade), and 2,2 -azobis(2-amidinopropane) dihydrochloride (AAPH) were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Sodium phosphate monobasic, dihydrogen phosphate 12 hydrate, and citric acid monohydrate were purchased from Nacalai Tesque, Inc. (Kyoto, Japan). n-Heptadecane was purchased from Acros Organics (Geel, Belgium). 1,2-Diacyl-sn-glycero-3-phosphocholine (PC) standards, including 1-palmitoyl-2-stearonyl-sn-glycero-3-phosphocholine (16:0/18:0 PC, PSPC), 1-palmitoyl-2-oleoyl-snglycero-3-phosphocholine (16:0/18:1 PC, POPC), 1-palmitoyl-2-linoleoyl-sn-glycero-3-phosphocholine (16:0/18:2 PC, PLPC), 1-palmitoyl-2-arachidonyl-sn(16:0/20:4 PC, PAPC), and glycero-3-phosphocholine

1-palmitoyl-2-docosahexaenoyl-sn-glycero-3-phosphocholine (16:0/22:6 PC, PDPC) were purchased from Avanti Polar Lipids Inc. (Alabaster, AL, USA) as lipid standards. Mouse plasma was purchased from Kohjin Bio Co., Ltd. (Saitama, Japan). The plasma was pooled and originated from both male and female mice that were 8–12 weeks of age and sexually mature. For in vivo mouse plasma samples, APCMin/+ mouse plasma was used. This strain is a model of spontaneous familial adenomatous polyposis. The mouse plasma, n = 3, was sampled at 18 weeks of age. The control sample was prepared from C57BL/6J wild-type mouse at 18 weeks of age. 2.2. Oxidized lipid standard preparation The oxidized lipid standard compound was prepared by following a previously reported method [26]. Each lipid standard compound was dissolved in methanol at 20 mmol L−1 . Each lipid standard (100 ␮L) was placed into a 2 mL micro tube and dried under N2 gas stream. AAPH reagent was added to the tube and mixed. AAPH was dissolved in phosphate buffered saline at 0.5 mol L−1 . This solution was prepared just before use. The mixture was incubated at 1400 rpm for 4 h at 37 ◦ C under aerated conditions and used as the oxidized lipid standard sample. 2.3. Sample conditions Oxidized lipid standard (10 ␮L) and 20% sodium chloride solution (990 ␮L) were placed in an 18-mm screw neck vial (10 mL) and sealed by a screw cap with 1.3 mm silicone/PTFE septa (GERSTEL K.K., Tokyo, Japan). Sodium chloride was dissolved in distilled water, and the pH was adjusted to 3 using 0.05 mol L−1 citric acid monohydrate. For the mouse plasma analysis, the plasma sample (100 ␮L), 20% sodium chloride solution (900 ␮L), and 0.01 ng ␮L−1 internal standard (1 ␮L) were placed in a vial. The internal standard was n-heptadecane dissolved in methanol. 2.4. Extraction conditions Dynamic headspace extraction was performed with an MPS2-xt Multi-Purpose Sampler (GERSTEL K.K., Tokyo, Japan). The sample was incubated for 10 min at 80 ◦ C with agitation (500 rpm). The headspace was purged using inert helium gas at 10 mL min−1 for 6 min and extracted using a Tenax TA Sorbent Tube. The sorbent temperature was 25 ◦ C. After the extraction process, the sorbent was purged at 50 mL min−1 for 12 min to remove moisture. The sorbent temperature was 40 ◦ C. Transfer heater temperature was 150 ◦ C. The sorbent was moved to a Thermal Desorption Unit (TDU) (GERSTEL K.K., Tokyo, Japan) after the extraction process. This unit included a cooled injection system, CIS4 (GERSTEL K.K., Tokyo, Japan). The initial temperature of the TDU was 40 ◦ C for 0.2 min, followed by a 720 ◦ C min−1 ramp to 280 ◦ C, which was held for 10 min. Transfer temperature was 300 ◦ C and the desorption mode was splitless. The initial temperature of the CIS was 10 ◦ C for 0.5 min, followed by a 12 ◦ C s−1 ramp to 280 ◦ C, held for 25 min. 2.5. Analytical conditions GC/MS analysis was performed with an Agilent 7890A and 5975 Inert XL MSD (Agilent Technologies, Santa Clara, CA, USA). Volatiles were analyzed with a fused silica capillary with a 30 m × 0.25 mm inner diameter and a 0.25 ␮m film thickness CP-CIL 8CB Low Bleed Column (Varian Inc., Palo Alto, CA, USA). The flow rate of helium gas was 1 mL min−1 . The oven temperature was set 60 ◦ C for 5 min, followed by a 15 ◦ C min−1 ramp to 300 ◦ C, which was held for 5 min. The front inlet, transfer line, and ion source temperatures were

S. Kakuta et al. / J. Chromatogr. B 1003 (2015) 35–40

37

Table 1 Consideration of linear range, LOD and LOQ. Compound

m/z

RT (min)

slope

intercept

−1

(␮g L 1-Pentene-3-one 2-Hexenal 1-Octen-3-ol 2-Pentyl furan Octanal 2,4-Heptadienal 2-Ethyl-1-hexanol Benzyl alcohol 2-Octenal Nonanal 2,4-Decadienal a b

55 41 57 81 55 81 57 79 55 57 81

3.0 5.5 7.9 8.1 8.3 8.5 8.7 8.9 9.1 9.7 12.1

1.5 × 102 0.3 × 102 1.0 × 102 0.9 × 102 0.2 × 102 1.0 × 102 1.2 × 102 9.7 0.2 × 102 0.3 × 102 0.4 × 102

1.7 × 103 0.6 × 102 1.1 × 103 3.7 × 103 1.3 × 103 −0.9 × 103 1.6 × 104 0.6 × 103 0.9 × 103 0.3 × 102 0.4 × 102

LODa

Linear range )

1–1000 25–1000 1–1000 1–1000 1–1000 1–1000 1–1000 50–1000 1–1000 1–1000 10–1000

LOQb −1

R2

(␮g L

)

0.9970 0.9997 0.9998 0.9972 0.9910 0.9961 0.9919 0.9844 0.9994 0.9791 0.9890

2.0 1.2 8.4 0.2 × 102 0.3 × 102 7.3 0.6 × 102 2.4 × 102 0.3 × 102 0.9 × 102 0.1 × 102

(␮g L−1 ) 6.0 3.5 0.3 × 102 0.5 × 102 0.9 × 102 0.2 × 102 1.9 × 102 7.2 × 102 0.9 × 102 2.6 × 102 0.4 × 102

LOD = 3.3 × S.D./slope. LOQ = 10 × S.D./slope.

Table 2 Volatile analysis of normal mouse plasma samples. Compound

1-Butanol Pentanalb Pyridine 1-Pentanol 2-Hexenal 1-Hexanola,b 2-Heptanoneb Heptanal 2-Ethylhexanalb 2-Heptenal Benzaldehyde 1-Heptanolb 1-Octen-3-ol n-Caproic acid vinyl esterb 2-Pentylfuranb Octanal (E,E)-2,4-Heptadienalb 2-Ethyl-1-hexanolb (E)-Octenalb 2-Octen-1-ol 3,5-Octadien-2-onea,b Nonanal 2-Nonenal 4-Ethylphenola 1-Nonanol 2,4-Decadienalb a b

m/z

56 44 79 42 41 56 43 70 57 83 106 70 57 99 81 55 81 57 70 57 95 57 70 107 56 81

RT

Normal plasma (N = 3)

RSD

(min)

Average

SD

(%)

2.8 3.1 3.6 3.8 5.5 5.8 6.3 6.5 7.5 7.6 7.7 7.7 7.9 8.0 8.1 8.3 8.5 8.7 9.1 9.2 9.6 9.7 10.4 10.5 10.5 12.1

0.20 × 105 0.19 × 105 0.10 × 106 0.74 × 104 0.20 × 104 0.40 × 105 0.53 × 104 0.64 × 104 0.11 × 105 0.26 × 104 0.91 × 104 0.24 × 104 0.71 × 105 0.12 × 105 0.84 × 104 0.14 × 104 0.35 × 104 0.13 × 106 0.54 × 104 0.21 × 104 0.55 × 104 0.53 × 104 0.14 × 104 0.21 × 105 0.13 × 104 0.35 × 104

0.14 × 104 0.16 × 104 0.45 × 104 0.30 × 103 0.88 × 102 0.39 × 104 0.24 × 103 0.71 × 103 0.15 × 104 0.20 × 103 0.71 × 103 0.92 × 102 0.22 × 104 0.19 × 104 0.59 × 103 0.89 × 102 0.13 × 103 0.18 × 105 0.53 × 103 0.37 × 102 0.33 × 103 0.19 × 103 0.17 × 103 0.51 × 103 0.13 × 103 0.23 × 103

7.1 8.0 4.3 4.0 4.3 9.9 4.5 11.1 13.6 7.9 7.8 3.9 3.0 16.4 7.1 6.4 3.8 14.2 9.7 1.8 6.0 3.7 12.1 2.4 10.3 6.5

Annotated using the NIST library. Unreported compound.

250 ◦ C, 280 ◦ C and 230 ◦ C, respectively. The scan range was from m/z 40 to m/z 350. The septum purge flow was 3 mL min−1 and the split ratio was 1:3.

2.6. Identification of volatiles The data acquisition was performed with GC/MS ChemStation (Agilent Technologies), and the data were transformed to CDF format. The data were analyzed with GCMSsolution (Shimadzu). The detected peaks for oxidized lipids for the total ion current chromatograms of oxidized POPC, PLPC, PAPC, and PDPC were compared with that of oxidized PSPC, and volatiles derived from oxidized lipids were identified using the NIST library (NIST 11). For the mouse plasma analysis, the detected peaks were identified based on this volatile information. Unidentified specific peaks were defined as unknown compounds in this study.

2.7. Statistical analysis Each peak area was divided by the internal standard peak area. The relative peak area was presented as the mean ± standard deviation. Students and Welch’s t-tests were performed with Microsoft Excel 2010 and p-values of less than 0.05 were considered statistically significant.

3. Results and discussion 3.1. Linear range, LOD and LOQ Firstly, we prepared mixed volatile compounds standard and considered linearity, LOD and LOQ of DHS-GC/MS method. We selected 11 volatile compounds such as straight chain aldehydes, unsaturated chain aldehydes, alcohols, ketone, cyclic structured volatiles (Table 1.). Volatile standard compounds were dissolved in methanol and the concentration was set at 0, 0.1, 1, 10, 25, 50,

38

S. Kakuta et al. / J. Chromatogr. B 1003 (2015) 35–40

Fig. 1. Total ion current chromatograms of plasma from control mice and APCMin/+ mice (A: 0–5 min B: 5–10 min).

Table 3 Comparison of volatile profiles between control and APCMin/+ mouse plasma. Compound

m/z

b,*

2-Methyl-2-butanol Pyridine* Toluenea 1-Hexanol 5-Methyl-3-hexanonea,b,* 2-Heptanone Heptanal* 2-Butoxy ethylenea,* 3,5,5-Trimethyl-2(5H)-furanonea 4-Methyl-2-heptanonea 6-Methyl-3-heptenonea Benzaldehyde* 1-Octen-3-ola 2-Ethtl-1-hexanol Benzyl alcohola,b,* 2-Octen-1-ol (E,E)-2,4-Nonadien-1-ol Acetophenonea Formin acid, phenylmethyl ester*

59 79 91 56 57 43 70 57 43 43 57 106 57 57 79 57 79 105 91

RT

Average (N = 3)

SD (N = 3) Min/+

(min)

Control

APC

2.7 3.6 3.8 5.8 6.2 6.3 6.5 6.6 6.9 7.2 7.4 7.7 7.9 8.7 8.9 9.2 9.3 9.3 9.4

0.47 × 10 0.31 0.11 0.55 × 10−1 0.69 × 10−2 0.11 × 10−1 0.48 × 10−2 0.15 × 10−1 0.61 × 10−1 0.89 × 10−2 0.13 × 10−1 0.12 × 10 0.48 × 10−1 0.10 0.67 × 102 0.74 × 10−2 0.20 × 10−1 0.63 × 10−1 0.30 × 10−1 2

APCMin/+

Control

0.88 × 10 0.48 0.26 0.38 × 10−1 0.15 × 10−1 0.19 × 10−1 0.84 × 10−2 0.26 × 10−1 N.D. 0.13 × 10−1 0.80 × 10−2 0.29 × 10 0.10 0.14 0.21 × 103 0.17 × 10−2 0.18 × 10−1 0.44 × 10−1 0.85 × 10−1 2

0.18 × 10 0.62 × 10−1 0.30 × 10−1 0.42 × 10−1 0.35 × 10−2 0.47 × 10−2 0.17 × 10−2 0.33 × 10−2 0.31 × 10−1 0.47 × 10−2 0.85 × 10−2 0.42 0.24 × 10−1 0.23 × 10−1 0.28 × 10−2 0.40 × 10−2 0.14 × 10−1 0.33 × 10−1 0.14 × 10−1 2

0.16 × 102 0.86 × 10−1 0.16 0.92 × 10−2 0.32 × 10−2 0.44 × 10−2 0.19 × 10−2 0.50 × 10−2 N.D. 0.44 × 10−2 0.28 × 10−2 0.11 × 10−1 0.32 × 10−1 0.42 × 10−1 0.65 × 10−2 0.64 × 10−2 0.85 × 10−2 0.96 × 10−2 0.31 × 10−1

N.D.: Not detected. * p-values of less than 0.05 in Student’s or Welch’s t-test. a Annotated using the NIST library. b Unreported compound.

100, 500, 1000 ␮g L−1 . The volatile compounds mixture sample was analyzed 3 times. LOD and LOQ were calculated by a formula for the standard deviation and slope of the calibration curve [27]. The standard deviation was calculated by the area value at the lowest concentration in the linear range.

All volatile compounds were observed good linearity and value of LOD and LOQ were ␮g L−1 order. Based on this method, we attempted to do volatile profiling of normal mouse plasma sample and considered to acquire profile and repeatability of volatile compounds.

S. Kakuta et al. / J. Chromatogr. B 1003 (2015) 35–40

39

Fig. 2. Comparison of compound levels between control and APCMin/+ mice, showing significant differences. Closed bars: control, open bars: APCMin/+ .

3.2. Normal mouse plasma analysis Trace amounts of various volatile compounds are present in blood samples. Therefore, volatile profiling was used to analyze normal pooled mouse plasma samples. Volatile compounds in the blood were identified using a library of oxidized lipid-derived volatile compounds (S. Table 1). In total, 26 compounds were identified in the mouse plasma samples (Table 2). Various volatile compounds, including alcohols and ketones, were detected, as were aldehydes, in the mouse plasma samples. Some have been reported as biomarker candidates for various diseases. For example, 1-octen-3-ol is significantly elevated in liver cancer blood samples [28], and benzaldehyde is a biomarker candidate for lung cancer [29]. In addition, 12 volatile compounds that have not previously been reported in blood samples were identified [30]. The peak area for each volatile compound was calculated from the extracted ion chromatogram for the target m/z values (Table 2). The relative standard deviations for 20 volatile compounds were lower than 10%, while those of heptanol, 2-ethylhexanal, n-caproic acid vinyl ester, 2-ethyl-1-hexanol, 2-nonenal, and 1-nonanol were not. Based on the above results, the volatile profiling technique can be used to acquire information regarding unreported and known volatile compounds in biological samples, based on an oxidized lipid-derived volatile compound library. 3.3. Volatile profiling of plasma samples from APCMin/+ mice Next, volatile profiling of the plasma of disease model mice was performed. We used plasma from APCMin/+ mice, a model of familial adenomatous polyposis disease. In previous reports, the metabolism of 4-HNE, an oxidized lipid-derived product, was elevated in APCMin/+ mouse cells and other various volatile compounds were released in the body [17]. However, a volatile profiling analysis has not been reported, even though volatile-DNA adducts are potential biomarker compounds [31]. In this research, we compared the volatile profiles of control plasma samples with those of APCMin/+ samples. Each total ion current chromatogram is shown in Fig. 1, and the volatile compounds detected as well as the relative peak areas are shown in Table 3. In total, 19 volatile compounds were detected and eight differed significantly between the diseased and control mice (Fig. 2).

3,5,5-Trimethyl-2(5H)-furanone was only detected in the control sample (Fig 1B, RT = 6.9 min). Because this volatile was not released from the oxidized lipid standard, it likely had a different origin. This volatile compound is a biomarker candidate; however, little information is known about it. Hence, further studies should examine this volatile. The detection of 2-methyl-2-butanol and benzyl alcohol in blood samples has not been previously reported; however, we were able to detect these volatile compounds in our study, and their levels were found to differ significantly between diseased and control mice. 5-Methyl-3-hexanone has not been detected in blood samples, but was found in fecal samples and is a biomarker candidate for gastrointestinal diseases [32]. Benzaldehyde is a lung cancer disease marker candidate [29]. It is released from phenyl ethanolamine or reactions of 2,4-alkanedienals with phenylalanine [33,34]. Moreover, the relative peak area of benzyl alcohol was higher than that of benzaldehyde (Fig. 2). This volatile is metabolized to benzaldehyde via alcohol dehydrogenase [33], and the difference in volatile compound levels could be detected using the volatile profiling technique in this study. Therefore, differences in the volatile profiles between control and APCMin/+ mouse plasma samples were observed using DHSGC/MS. Specific volatile compounds existed in control samples and were not detected in APCMin/+ samples. The volatile profiles included information consistent with both unreported and previously identified volatile compounds. Volatile compound levels would be affected by disease states as well as metabolites, and significant differences in volatile compounds can be detected using volatile profiling techniques.

4. Conclusion In this study, volatile profiling enabled us to identify profile patterns that differ between the plasma of control mice and APCMin/+ mice. Because various volatile compounds are present at trace levels in biological samples, the DHS method was applied for effective extraction in the pretreatment process. First, volatile profiling of commercial pooled mouse plasma samples was considered. Based on comparisons with the oxidized lipid standard-derived volatile compound library, alcohols and ketones were identified in the samples, along with aldehydes, some of which have not been reported previously in plasma. Next, a

40

S. Kakuta et al. / J. Chromatogr. B 1003 (2015) 35–40

comparison of the volatile profile pattern between the plasma of control mice and APCMin/+ mice was considered. Specific volatile compounds, present only in control samples, were observed, and the samples showed significantly different patterns. 4-HNE or DNA adducts have been the focus of some previous studies [17,31], and various volatile compounds differ among samples. These findings may facilitate the identification of volatile biomarker candidates for distinguishing or predicting diseases. Acknowledgements This work was partially supported by a Grant-in-Aid for Young Scientists (A) (23686120). We greatly thank Mr. Hirooki Kanda and Ms. Teruyo Ieda (GERSTEL K. K.) and Mr. Toshiyuki Yamashita (Yamashita Science) for support in the volatile compounds analysis by DHS extraction with the MPS2-xt Multi-Purpose Sampler. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jchromb.2015. 09.002. References [1] E. Fukusaki, A. Kobayashi, Plant metabolomics: potential for practical operation, J. Biosci. Bioeng. 100 (2005) 347–354. [2] S. Yamamoto, K. Shiga, Y. Kodama, M. Imamura, R. Uchida, A. Obata, T. Bamba, E. Fukusaki, Analysis of the correlation between dipeptides and taste differences among soy sauces by using metabolomics-based component profiling, J. Biosci. Bioeng. 118 (2014) 56–63. [3] R. Yoshida, T. Tamura, C. Takaoka, K. Harada, A. Kobayashi, Y. Mukai, E. Fukusaki, Metabolomics-based systematic prediction of yeast lifespan and its application for semi-rational screening of ageing-related mutants, Aging Cell 9 (2010) 616–628. [4] E.C.Y. Chan, P.K. Koh, M. Mal, P.Y. Cheah, K.W. Eu, A. Backshall, R. Cavill, J.K. Nicholson, H.C. Keun, Metabolic profiling of human colorectal cancer using high-resolution magic angle spinning nuclear magnetic resonance (HR-MAS NMR) spectroscopy and gas chromatography mass spectrometry (GC/MS), J. Proteome Res. 8 (2009) 352–361. [5] Y. Qiu, G. Cai, M. Su, T. Chen, Y. Liu, Y. Xu, Y. Ni, A. Zhao, S. Cai, L.X. Xu, W. Jia, Urinary metabonomic study on colorectal cancer, J. Proteome Res. 9 (2010) 1627–1634. [6] S. Nishiumi, T. Kobayashi, A. Ikeda, T. Yoshie, M. Kibi, Y. Izumi, T. Okuno, N. Hayashi, S. Kawano, T. Takenawa, T. Azuma, M. Yoshida, A novel serum metabolomics-based diagnostic approach for colorectal cancer, PLoS One 7 (2012) e40459. [7] M. Kanayama, S. Yamaguchi, T. Shibata, N. Shibata, M. Kobayashi, R. Nagai, H. Arai, K. Takahashi, K. Uchida, Identification of a serum component that regulates cyclooxygenase-2 gene expression in cooperation with 4-hydroxy-2-nonenal, J. Biol. Chem. 282 (2007) 24166–24174. [8] M.R.M. Domingues, A. Reis, P. Domingues, Mass spectrometry analysis of oxidized phospholipids, Chem. Phys. Lipids 156 (2008) 1–12. [9] K. Uchida, Current status of acrolein as a lipid peroxidation product, Trends Cardiovasc. Med. 9 (1999) 109–113. [10] K. Warner, W.E. Neff, W.C. Byrdwell, H.W. Gardner, Effect of oleic and linoleic acids on the production of deep-fried odor in heated triolein and trilinolein, J. Agric. Food Chem. 49 (2001) 899–905. [11] T. Hayashi, K. Uchida, G. Takebe, K. Takahashi, Rapid formation of 4-hydroxy-2-nonenal, malondialdehyde, and phosphatidylcholine aldehyde from phospholipid hydroperoxide by hemoproteins, Free Radic. Biol. Med. 36 (2004) 1025–1033. [12] E.K. Long, M.J. Picklo Sr., Trans-4-hydroxy-2-hexenal, a product of n-3 fatty acid peroxidation: make some room HNE, Free Radic. Biol. Med. 49 (2010) 1–8. [13] L.J. Marnett, Lipid peroxidation–DNA damage by malondialdehyde, Mutat. Res. 424 (1999) 83–95.

[14] S.H. Lee, I.A. Blair, Characterization of 4-oxo-2-nonenal as a novel product of lipid peroxidation, Chem. Res. Toxicol. 13 (2000) 698–702. [15] T. Yamada, T. Uchikata, S. Sasamoto, Y. Yokoi, S. Nishiumi, M. Yoshida, E. Fukusaki, T. Bamba, Supercritical fluid chromatography/orbitrap mass spectrometry based lipidomics platform coupled with automated lipid identification software for accurate lipid profiling, J. Chromatogr. A 1301 (2013) 237–242. [16] A. Amann, B.L. de, W.M. Costello, J.S. iekisch, B.B. chubert, J.P. uszewski, N.R. leil, T.R. atcliffe, The human volatilome: volatile organic compounds (VOCs) in exhaled breath, skin, emanations, urine, feces and saliva, J. Breath Res. 8 (2014) 034001. [17] M. Baradat, I. Jouanin, S. Dalleau, S. Taché, M. Gieules, L. Debrauwer, C. Canlet, L. Huc, J. Dupuy, F.H.F. Pierre, F. Guéraud, 4-Hydroxy-2(E)-nonenal metabolism differs in APC+/+ cells and in APCMin/+ cells: it may explain colon cancer promotion by heme iron, Chem. Res. Toxicol. 24 (2011) 1984–1993. [18] M.V. Williams, W.H. Lee, M. Pollack, I.A. Blair, Endogenous lipid hydroperoxide-mediated DNA-adduct formation in min mice, J. Biol. Chem. 281 (2006) 10127–10133. [19] H. Tuulia, yotlainen, Critical evaluation of sample pretreatment techniques, Anal. Bioanal. Chem. 394 (2009) 743–758. [20] Z. Zhang, G. Li, A review of advances and new developments in the analysis of biological volatile organic compounds, Microchem. J. 95 (2010) 127–139. [21] G. Vas, K. Vékey, Solid-phase microextraction: a powerful sample preparation tool prior to mass spectrometric analysis, J. Mass Spectrom. 39 (2004) 233–254. [22] J.B. Quintana, I. Rodríguez, Strategies for the microextraction of polar organic contaminants in water samples, Anal. Bioanal. Chem. 384 (2006) 1447–1461. [23] S. Kakuta, T. Yamashita, S. Nishiumi, M. Yoshida, E. Fukusaki, T. Bamba, Multi-component profiling of trace volatiles in blood by gas chromatography/mass spectrometry with dynamic headspace extraction, Mass Spectrom. 4 (2015) A0034. [24] S. Mallia, E. Fernández-García, J.O. Bosset, Comparison of purge and trap and solid phase microextraction techniques for studying the volatile aroma compounds of three European PDO hard cheeses, Int. Dairy J. 15 (2005) 741–758. [25] J.S. Elmore, M.A. Erbahadir, D.S. Mottram, Comparison of dynamic headspace concentration on tenax with solid phase microextraction for the analysis of aroma volatiles, J. Agric. Food Chem. 45 (1997) 2638–2641. [26] J. Takebayashi, J. Chen, A. Tai, A method for evaluation of antioxidant activity based on inhibition of free radical-induced erythrocyte hemolysis, in: D. Armstrong (Ed.), Advanced Protocols in Oxidative Stress Ii, Methods in Molecular Biology, Humana Press, New York, 2010, pp. 287–296. [27] J. Mocak, A.M. Bond, S. Mitchell, G. Scollary, A statistical overview of standard (IUPAC and ACS) and new procedures for determining the limits of detection and quantification: application to voltammetric and stripping techniques, Pure Appl. Chem. 69 (1997) 297–328. [28] R. Xue, L. Dong, S. Zhang, C. Deng, T. Liu, J. Wang, X. Shen, Investigation of volatile biomarkers in liver cancer blood using solid-phase microextraction and gas chromatography/mass spectrometry, Rapid Commun. Mass Spectrom. 22 (2008) 1181–1186. [29] M. Ligor, T. Ligor, A. Bajtarevic, C. Ager, M. Pienz, M. Klieber, H. Denz, M. Fiegl, W. Hilbe, W. Weiss, P. Lukas, H. Jamnig, M. Hackl, B. Buszewski, W. Miekisch, J. Schubert, A. Amann, Determination of volatile organic compounds in exhaled breath of patients with lung cancer using solid phase microextraction and gas chromatography mass spectrometry, Clin. Chem. Lab. Med. 47 (2009) 550–560. [30] D.L. de, A.A. Costello, H.A. mann, C.F. l-Keteb, W.F. lynn, T.K. ilipiak, D.O. halid, N.M.R. sborne, A review of the volatiles from the healthy human body, J. Breath Res. 8 (2014) 014001. [31] H. Bartsch, J. Nair, Chronic inflammation and oxidative stress in the genesis and perpetuation of cancer: role of lipid peroxidation, DNA damage, and repair, Langenbecks Arch. Surg. 391 (2006) 499–510. [32] G.E. Garner, S. Smith, B.L. de, P.W. Costello, R.S. hite, C.S.J.P. pencer, N.M.R. robert, Volatile organic compounds from feces and their potential for diagnosis of gastrointestinal disease, Fed. Am. Soc. Exp. Biol. J. 12 (2007) 1675–1688. [33] M. Stafford, M.G. Horning, A. Zlatkis, Profiles of volatile metabolites in body fluids, J. Chromatogr. 126 (1976) 495–502. [34] R. Zamora, E. Gallardo, F.J. Hidalgo, Strecker degradation of phenylalanine initiated by 2,4-decadienal or methyl 13-oxooctadeca-9,11-dienoate in model systems, J. Agric. Food Chem. 55 (2007) 1308–1314.