Bioanalytical strategies for the quantification of xenobiotics in biological fluids and tissues 3

Chapter Chapter Biomonitoring of environmental organic pollutants in human ovarian tumor cyst fluids samples using µ-SPE-GC-MS and HPLC-florescence principal component analysis detection and Chapter 3.1 Preface to Chapter To assess a possible etiological role of environmental organic pollutants in ovarian cancer development on ovarian cancer patients, concentrations of different groups of organic pollutants were measured in 20 malignant and benign ovarian cyst fluid samples of women with ovarian cancer. A total of 60 chemicals of six groups including heterocyclic aromatic amines, Low molecular weight organic acids, aromatic amines, N-nitrosamines, Polybrominated diphenyl ethers and halogenated flame retardants and organochlorine pesticides were assayed via porous membrane protected micro-solid-phase extraction followed by GC-MS detection and HPLCfluorescence detection. High performance liquid chromatography coupled with fluorescence detection was used to quantify heterocyclic aromatic amines and aromatic amines. Gas chromatography-mass spectrometry was used to quantify PBDEs and halogenated flame retardants, and LMW organic acids. Trace amounts of most of the chemicals were found both in benign and malignant cyst fluid samples. The trend in their concentration in benign and malignant samples was projected by principal component analysis using R program. The results reveal that the possible correlations in the concentration of chemicals with the malignancy of the ovarian tumour. 44 Chapter 3.2 Introduction Ovarian cancer is the fifth most common cancer among women worldwide and is the fourth most common cancer in Singapore [1]. It causes more deaths than any other type of female reproductive cancer. The risk for developing ovarian cancer appears to be affected by several factors. Exposure to endocrine disruptive xenobiotics is recognized as an important environmental risk factor associated with development of cancer. Global epidemiologic studies have indentified environmental and occupational chemicals as potential carcinogens. Many studies provide the direct association between these chemicals, especially EDCs with the development of different types of gynaecological cancer [2]. Many environmental chemical contaminants, which may also be the metabolic intermediates, particularly those that are lipophilic and of relatively low molecular weight, can accumulate in tissue and body fluids. The potential health effects of these contaminants on human are of great concern, making it important to carefully monitor their levels and trends. Many methods have been developed for the exposure to carcinogens in human, through the detection of carcinogens or their metabolic derivatives in body fluids. Biomonitoring studies, designed to assess the health implication of environmental chemicals, including carcinogens, are seriously negotiated by the lack of quantitative exposure data for individuals in exposed populations. Monitoring data on levels of compounds in environmental media often represent the average population exposure is therefore the only quantitative factor that can be estimated. In the present study, we evaluated the association level of wide range of environmental 45 Chapter contaminants in human ovarian cyst fluids with early stage (benign) and late stage (malignant) ovarian cancer. The groups of EDCs, which are well known carcinogens, studied were heterocyclic aromatic amines (HAAs), PBDEs, OCPs and Nnitrosamines. In addition, metabolic intermediates such as aromatic amines and low molecular weight (LMW) organic acids, which are potential xenobiotics, were studied as well. OCPs are persistant in nature and non biodegradable. Since they are highly lipophilic, they can bioaccumulate in fatty tissues getting up to metabolism through diet, especially foods of animal origin [3]. Being a chlorinated compound, organochlorine pesticides strongly mimic estrogen in the body. Due to their estrogenic activity, most OCPs were classified as “possibly carcinogenic to humans” (2B group); consequently they increase special attention in public health and epidemiology [4-6]. N-nitrosamines are classified as class 2A genotoxic chemical carcinogens and animal testing indicated mutagenic, carcinogenic and tetragonal effects. They occur in the human diet and in environment, and can be formed endogenously in the human body [7]. More than 90% of nitrosamines had shown to cause cancer in animals. It had also been reported that with exposure to endogenously formed N-nitrosamines, there is a higher risk of tumor [8]. PBDEs are flame-retardant chemicals that are added to plastics and foam products to make them difficult to burn [9]. They are environmentally widespread and human exposure to those compounds is logical. Many studies have been reported that PBDEs have endocrine disrupting properties suggesting their potential role in hormonally related cancers such as ovarian cancer [10-13]. Based on a study on mice 46 Chapter and rats, the US EPA has classified some of the PBDEs are a possible human carcinogen [14]. The carcinogenicity of aromatic amines and heterocyclic aromatic amines are well documented [15-17]. Being an important class of industrial and environmental chemical, aromatic amines easily entered into biota and human metabolism. Aromatic amines are converted in the hosted organism to arylhydroxamic acid or arylhydroxylamines derivatives which are thought to be the critical carcinogenic forms of those amines. These derivatives stimulate tumors, usually in tissues distance from the site of administration [18]. Cooking of proteinrich foods mainly from animal origin may stimulate the formation of a series of heterocyclic aromatic amines [19]. They have also identified in cigarette smoke condensate and diesel exhaust [20, 21]. LMW organic acids can be found in the environment naturally such as in rainwater or soil. They are important intermediate breakdown products between large biomolecules and the ultimate demineralization products CH4 and CO2 [22]. Determination of organic acid concentrations is crucial in body fluids since abnormal levels of organic acids in the blood (organic acidemia), urine (organic aciduria), and tissues can be toxic and can cause adverse health effects [23]. Moreover, it is significance to estimate the organic acid level variation in benign and malignant tumor cyst fluids as they have genotoxicity. Hence, monitoring these chemicals in ovarian tumor cyst fluids will be useful to explore the carcinogenicity of such chemicals. The objective of this study is to determine profile and quantify various xenobiotics (total sixty individual analyte of 47 Chapter six different groups) which mimic estrogens and estrogen metabolites from malignant and benign ovarian cyst fluid samples. Further, from the results, their potential associated with malignancy associated with ovarian tumors was investigated. The samples were preconcentrated using the micro-solid phase extraction (µ-SPE) which had proved to be a suitable technique for cyst fluid samples [25]. Wide choice of sorbents makes this technique versatile for variety of group of analytes. The determination was done by using liquid and gas chromatographic techniques. The obtained data was processed by principal component analysis (PCA) to simplify the complex data system with focus on concentration patterns and correlations. Measurements are made on twenty individual samples, they provide an indication not only of exposure to a given substance, but also of the amount absorbed and metabolically transformed to activated derivatives. No previous study has directly investigated the presence of these toxic chemicals in the ovarian cyst fluids of human patients with ovarian cancer. 3.3 Materials 3.3.1 Sample collection This study used 20 human cyst fluid samples, 10 from benign and 10 from malignant ovarian tumor patients between 19 and 66 years of age who were diagnosed at National University Hospital (NUH), Singapore. Cyst fluid obtained from benign and malignant ovarian tumor samples were collected following approval from the Domain Specific Review Board, National Health Group, Singapore. Samples were collected from patients post-operatively after getting their consent to use the samples for research purpose and immediately stored in −80◦C deep freezer until analysis. Regular safety considerations were put in place during the handling of cyst fluids. All 48 Chapter body fluids and solvents used in this project were decontaminated according to standard biohazard disposal protocols. All patients’ personal information was concealed to protect their identities. The pathological information of the samples is listed in Table 3.1. Table 3.1 Age, Tumour marker CA-125* and Pathology of the samples. Sample code B1 B2 B3 B4 B5 Age 35 60 44 42 65 Tumour marker CA-125 59 25.5 6000 serous adenocarcinoma late stage M10 66 777.0 serous cystadenocarcinoma borderline *CA-125, cancer antigen-125, is a protein that is found at levels in most ovarian cancer cells that are elevated, compared to normal cells. CA-125 is produced on the surface of cells and is released in the blood stream. 49 Chapter 3.3.2 Chemicals BDE -47, -49, -99, -153 and -154 were bought from AccuStandard (New Haven, USA) (Figure 3.1). PEB was obtained from Sigma-Aldrich (Wisconsin, MO,USA). HAAs compounds studied were purchased from Eckert & Ziegler CNL Scientific Resources (Valencia, CA, USA) (Figure 3.2). Aromatic amine compounds studied were bought from Fluka (neu-Ulm, Germany) (Figure 3.3). Oxalic acid, fumaric acid and citric acid were purchased from Sigma Aldrich (Milwaukee, USA) whereas lactic acid came from Fluka (Buchs, Switzerland) (Figure 3.4). 3methylglutaric acid, adipic acid and sebacic acid were purchased from Merck. OCPs were were purchased from Polyscience (Niles, IL, USA). Q3/2 Accurel polypropylene hollow fiber membrane was purchased from Membrana GmbH (Wuppertal, Germany). The solvents used for HPLC detection (HPLC-grade methanol, acetone, triethylamine) were obtained from Tedia Company, Inc. (Farfield, OH, USA). From Fisher Scientific (Loughborough, UK), HPLC-grade toluene, hexane, isooctane and dichloromethane were obtained. HPLC-grade acetonitrile and ACS-grade sodium acetate, glacial acetic acid, bis (trimethylsilyl) – trifluoroacetamide (BESTFA) were bought From Merck. Analytical grade Pyridine was obtained from J.T Baker (Philipsburg, NJ). Sodium chloride, sodium sulphate anhydrous and sodium hydroxide come from Goodrich Chemical Enterprise (Singapore). The water used was purified using a Milli-Q (Millipore, Bedford, MA, USA) water purification system. 50 Chapter Figure 3.1 Chemical Structures and abbreviated names of PBDEs and PEB. 51 Chapter Figure 3.2 Structures, names and abbreviated names of the heterocyclic aromatic amines. Figure 3.3 Structures and names of aromatic amines. 52 Chapter concentrations below LOQ were treated as missing values. PCA was performed on mean-centred and auto-scaled data using ‘R’ program (Vienna, Austria). 3.6 Results and discussion In this study, ten benign (B1 to B10) and ten malignant (M1 to M10) cyst fluids were analyzed to examine the association between organic pollutants exposure and ovarian cancer risk. A total of sixty prominent carcinogens were profiled in cyst fluids. Cyst fluids of the ovary were collected from the subjects with benign (control) and malignant ovarian (study) lesions and were analyzed to determine the chemicals. Table 3.4 presents complete data analyzed for various xenobiotic chemicals (concentration in µg L-1). The table only includes the chemicals which are present at least any one of the samples. This analysis highlights trends of different xenobiotic and their impact on severity of cancer. The trends of the chemicals in the two groups of tumor cyst fluid samples may be account for the source of the analyte and etiology of the tumor. In general, from the Table 3.4 we can see that, some analytes were found only in either malignant or benign samples group, particularly HAAS, and some group of chemicals show unusual trend in a few samples, i.e. M1, M7, M9 and B4. Considering the vastness and complexity of the data, it is difficult to visually observe the exact data patterns. Thus, we applied PCA to interpret and to express the data in such a way to highlight their similarities and differences. First, to select principal components which explain maximum variance, scree test was carried out by R program. The scree plot is a two dimensional graph with factors on the x-axis and eigenvalues on the y-axis. The scree plot (Figure 3.5) showed that two first principal components explained 62.61% of total variability of 59 Chapter the data and the remaining factors all have small eigenvalues. Portion of each two components was approximately 47% and 16% respectively. The greatness of these numbers influence good separation of samples and shows high valuable relations. If there would be correlations or similarities among samples, these components can provide suitable grouping and separate chemicals in distinct sample groups. Figure 3.5 Scree plot: the first two components are significant. Proportion of variance explained by the first component is 47.25%. Figure 3.6 Cluster plot: two clusters in 2D space (1) Malignant, (2) Benign. 60 Chapter Table 3.4 Concentration of organic pollutants in ovarian tumor cyst fluids (1) Concentration in µg L-1 , compounds * - Not detected Benign Malignant M5 M6 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 M1 M2 M3 M4 M7 M8 M9 M10 Oxalic acid 0.005 0.005 0.04 * 0.01 0.14 * * * 0.047 0.011 0.073 0.01 * * 0.014 0.01 0.023 0.018 * Citric acid 0.027 0.02 0.02 0.062 0.017 * * * 0.215 0.136 * 0.09 0.12 0.34 0.103 0.023 0.037 * 0.056 0.029 Fumaric acid * * * * * 0.056 * * * * * 0.011 * * * * * * * * 3-Methylglutaric acid * * * * * 0.79 * * * 0.217 * 0.17 0.015 0.005 0.095 * * * * * Adipic acid 0.059 0.06 0.03 0.006 0.041 * 0.25 * 0.154 0.065 0.017 0.19 0.31 0.09 0.041 0.044 0.054 0.08 0.25 * Sebacic acid * * * * * * * * * 0.091 * * * 0.03 * * * * 0.005 * Harman * * * * 0.03 * * 0.09 * * 0.13 * * * * * 0.08 * 0.09 * Norharman * * * * * * * * * * 0.88 * * * * * 0.05 * 0.08 * Trp-P-2 * * 0.21 * * * 0.25 * * * 0.54 * * * * * 0.05 * 0.64 * PhIp * * * * * 0.7 * * * * 0.74 0.73 0.61 0.46 0.66 0.81 0.54 0.34 1.35 0.53 Trp-P-1 * * * * * 0.9 * * * * * * 1.1 0.7 * 0.13 0.13 0.53 0.95 0.59 0.43 0.22 * 0.08 0.02 0.13 0.61 0.22 1.06 0.55 * * * * * 0.5 * * * * 0.01 * * 0.009 * * * 0.009 0.012 0.004 0.005 * 0.31 0.01 0.019 0.741 0.107 0.44 0.512 0.08 BDE 49 * * * 0.744 * 0.164 * * 0.108 * * 0.069 * * 0.021 * * 0.166 * * BDE 47 * 0.157 * 0.294 0.216 0.502 * * * * * * * 0.25 0.15 * * * * * BDE 99 * * * 0.76 0.099 * * * * * * 0.048 * * 0.13 * * * * * BDE 154 * * * * * * * 0.131 * * * * 0.07 0.141 * 0.122 * * 0.086 * BDE 153 0.24 * * 0.156 * * * * * * * * * * 0.156 * * * * 0.195 Organic acids Heterocyclic amines AαC PBDEs PEB 61 Chapter Table 3.5 Concentration of organic pollutants in ovarian tumor cyst fluids (2). Concentration in µg L-1 , compounds * - Not detected Benign Malignant Aromatic amines 0.4 0.25 0.31 * * * 0.03 0.06 * * * 0.21 0.2 * * 0.01 * 0.03 0.1 4-Chloroaniline 0.042 0.06 0.02 0.08 * * * * * * 0.11 * * * * 0.093 * 0.04 0.03 4-Bromoaniline 0.013 0.023 0.21 0.093 0.06 0.01 0.04 0.04 0.058 0.072 0.011 * 0.023 0.04 0.01 * * * 0.03 α-BHC 0.02 0.01 0.11 * 0.01 0.02 0.2 * 0.3 0.7 0.054 * 0.2 0.09 0.08 * * 0.5 0.1 Heptachlor 0.82 * 0.43 0.22 0.01 0.08 0.1 0.93 1.1 0.7 0.2 * 0.3 0.7 0.03 * * 0.09 * Aldrin 0.09 0.16 * * 0.01 * 0.5 0.1 0.06 0.1 * * * 0.01 * 0.3 0.2 * 0.3 * * * 0.01 * 0.3 0.2 * 0.3 0.7 * * * * 0.01 * 0.5 0.1 0.05 0.09 * * 0.07 0.03 * * * 0.02 * 0.02 0.1 0.31 0.09 0.15 0.043 0.01 0.07 0.03 * * * 0.003 0.1 * 0.02 0.16 * 0.041 0.21 * * * * 0.16 0.12 * 0.1 Endrin 0.5 * * * * 0.43 * 0.21 0.05 0.02 * * 0.21 * * * 0.25 * * 4,4'-DDT 0.03 * 0.09 0.023 0.01 0.01 * 0.05 0.026 0.03 0.21 0.02 0.01 0.11 0.27 0.013 0.02 0.19 0.09 * 0.11 0.12 0.17 0.15 0.21 0.42 0.28 0.23 0.21 0.11 0.12 0.3 0.19 0.33 0.22 0.19 0.31 0.02 NDMA * * * * * * * * * * 0.008 0.01 0.004 * * * * * * NMEA * * * 0.039 * * * * 0.017 * * * * 0.044 0.012 * * * * NDBA * * * * * * * * * * 0.02 0.016 * 0.084 0.069 0.007 * * * 3-Nitroaniline OCPs Heptachlor epoxide 4,4'-DDE Dieldrin Methoxychlore Nitrosamines 62 Chapter To explore more about the relationship between the two variables, a PCA biplot was generated. In a biplot, lines are used to reflect the variables of the dataset, and dots are used to show the observations. In the above biplot, the observations are the samples and the variables are the chemicals present in those samples. In a biplot, the length of the lines approximates the variances of the variables. The longer the line, the higher is the variance. Inferring from Figure 3.7, 3-nitroaniline (NA) has by far the highest variance among the variables in the biplot, while 4-4’-DDT (DDT) has the lowest. The angle between the lines approximates the correlation between the variables they represent. The closer the angle is to 90, or 270 degrees, the smaller the correlation. An angle of or 180 degrees reflects a correlation of or -1, respectively. The biplot in Figure 3.7 shows a strong relationship between the PhlP, Trp-P-1 (TP1) and PEB and a weak relationship between the PEB and 4, 4’-DDE (DDE). The correlation between the 3-nitroaniline (NA) and each of the other variables is negative. Over all, from the biplot, we can infer that the malignant samples are highly correlated to each other. The chemicals PhlP, Trp-P-1 (TP1) and PEB which are prominently present in the malignant samples are highly positively correlated. This is important in the view that, the synergic effect of these chemicals could provide significant influence in the malignancy of the tumor. On the other hand, benign samples are greatly scattered in the biplot, showing less significance in the chemical’s trend. Chapter Figure 3.7 PCA biplot of the first two principal components. OA- Oxalic acid, AA- Adipic acid, CA- Citric acid, PhlP- 2-Amino-1-methyl-6phenylimidazo- [4,5-b] pyridine , TP1- N-((2-substituted phenyl)-4,5-diphenyl-1Himidazol-1yl) (phenyl) methyl) substituted amine (Trp-P-1), AC- 2-amino-9Hpyrido[2,3-b]indole (AαC), PEB- 1,2,3,4,5-pentabromo-6-ethylbenzene, NA- 3Nitroanline, CA.14-Chloroaniline, BA4-Bromoaniline, BHCαHexachlorocyclohexane, HC- Heptachlor, AL- Aldrin, HCE- Heptachlor epoxide, DDE- p-dichlorodibenzodichloroethene, DDT- Dichlorodiphenyltrichloroethane, MC- Methoxychlore. For better understanding, the above PCA data was further subjected to detailed PCA analysis using BioplotGUI in which samples are represented as points and chemicals are represented as calibrated axes. As should be the case for all biplots, a unit aspect ratio is used to ensure that distances within the biplot are properly represented. In Figure 3.8, the points representing most of the samples i.e. M3, M6, M7, M8 and M9 lie ordered along a virtually straight line. In fact, the imagined line corresponds very closely to the biplot axis for PEB, PhlP, and Trp-P-1. The reason for this becomes obvious, by looking at the column of these chemicals (highlighted in green colour) in the Table 3.4 data set. The concentration of those chemicals are 64 Chapter orders larger in malignant than in benign samples. Similarly, the benign samples B7, B8 and B10 are closely positioned near the axes of 4-bromoaniline (BA) and heptacholore (HC), implies that these chemicals are in higher order in benign compare to malignant group (highlighted in orange colour in Table 3.4). However, most of the benign samples represented in such a way that, the axes are far away from their position in the biplot. This implies that no specific chemical has significance in benign samples. 0.2 0.0 NA. 0.0 0.1 0.00 0.0 0.15 B1 0.2 0.1 B2 B3 DDT DDE 0.10 0.1 B5 0.02 M1 0.05 0.0 M2 0.6 TP1 M5 0.10.02 0.20 0.2 M6 M9 B4 0.2 0.4 0.4 0.4 0.0 M7 M8 0.00 M3 0.2 0.0 0.08 0.2 0.06 B6 0.05 0.04 0.2 M4 0.10 0.2 0.2 B9 B8 0.03 B7 A.C 0.4 0.5 -2 0.6 Phlp PEB 0.12 0.2 0.10 0.6 M10 0.3 B10 0.15 CA.1 OA 0.1 0.25 0.3 MC AL -2 HC 0.3 AA BHC HCE CA BA 0.04 0.8 0.20 Figure 3.8 A predictive PCA biplot of the chemical data. 65 Chapter Figure 3.9 (A) PCA point predictivities of the centered, scaled chemicals data. (B) PCA axis predictivities of the centered, scaled chemicals data. The goodness of the biplot approximation depends on the axes and the points. As for the points, the `quality' of the PCA approximation is found from the R program. In this case, the quality 0.627 implies that 62.7% of the variation in the samples is accounted for by the first two principal components. For the above biplot (Figure 3.8), the point and axis predictivities also be calculated using the program options (Figure 3.9 (A) and (B)) [26]. Predictivities indicate how well individual points or axes are represented in various dimensions of the biplot. Generally, for a good PCA approximation, the points and axes are always appears above the diagonal in the unit square. The further to the right a point or axis appears, the better represented it is in the first (or horizontal) biplot dimension (principal component 1(PC1)). The closer to the top of the diagram, the better the point or axis is represented overall in the biplot, taking into account the contribution of both the first and the second (vertical) biplot dimension (principal component and 2). The marginal contribution of the second biplot dimension is indicated by the vertical distance between the diagonal line and the point or axis. This interpretation suggests that M1, M2 and M9 are relatively well represented in the first biplot dimension. B1, 66 Chapter B2, B3, B7, B10 are represented reasonably in the first dimension, but poorly in the second. M6 is poorly represented overall and B6 is the best represented sample overall. On the whole, most of the samples are well represented in both PCs, whereas, the chemicals are reasonably represented in PC and are not well represented in PC 1. However, the reliability of the generated biplot can be further checked by using error factor. Another measure of the goodness of the approximation is its relative absolute error, which may be calculated for any sample on any variable. The relative absolute error is defined to be the absolute difference between the predicted and actual values, expressed as a percentage of the range (max-min) of the actual values of the particular variable. By taking means over the samples, mean relative absolute errors can be obtained for the different variables. Table 3.6 shows the relative mean absolute error values of the chemicals ranging from 4.026% to 21.77%. Table 3.6 Relative absolute error % OA AA CA PhlP TrP1 AαC PEB 9.07 7.9 9.8 4.03 12.8 6.54 7.9 NA A.1 BA BHC 10.2 6.3 6.35 13.2 HC AL HCE DDE DDT MC 13.1 9.6 18.9 13.9 21.2 13.6 We also observed differences in chemicals concentrations within their group, which revealed a significant trend between benign and malignant samples. For instance, the potential carcinogenic forms of the heterocyclic aromatic amines PhlP and Trp-P-1 are present in all of the malignant samples but are not at all present in the benign group. Similarly, AαC is present almost all benign samples (except one), but is not present in the malignant group (except the one at trivial quantity). This trend is clearly visible in Figure 3.5. Comparatively, most of the heterocyclic amines are present in malignant group than benign samples. Especially, in samples M1, M7, M9, 67 Chapter five out of six HAAs are present in a significant quantity (Figure 3.6). The flame retardant PEB is widely present in both samples, however, PBDEs are not present in all samples (only present [...]... generated In a biplot, lines are used to reflect the variables of the dataset, and dots are used to show the observations In the above biplot, the observations are the samples and the variables are the chemicals present in those samples In a biplot, the length of the lines approximates the variances of the variables The longer the line, the higher is the variance Inferring from Figure 3. 7, 3- nitroaniline... monitoring (SIM) was used for the spectrometer For nitrosamines, the helium carrier gas was maintained at a constant flow of 0.5mL min−1 Injection volumes of 3mL were used The injection port was held at 230 ˚C and used in the splitless mode, applying a pressure pulse of 40 psi The GC temperature was programmed as follows: start temperature of 70˚C (held 3min) and increase to 140˚C at 15˚C min−1, then... above the diagonal in the unit square The further to the right a point or axis appears, the better represented it is in the first (or horizontal) biplot dimension (principal component 1(PC1)) The closer to the top of the diagram, the better the point or axis is represented overall in the biplot, taking into account the contribution of both the first and the second (vertical) biplot dimension (principal... biplot of the chemical data 65 Chapter 3 Figure 3. 9 (A) PCA point predictivities of the centered, scaled chemicals data (B) PCA axis predictivities of the centered, scaled chemicals data The goodness of the biplot approximation depends on the axes and the points As for the points, the `quality' of the PCA approximation is found from the R program In this case, the quality 0.627 implies that 62.7% of the. .. malignant samples Similarly, the heterocyclic aromatic amine, AαC, the OCP, heptachlor and the aromatic amine, 3- bromoaniline were prominent in most of the benign samples The clinical importance of their elevated levels in malignant samples in terms of their roles in benign to malignant transformation can be further studied by more number and variety of samples Specially, though the individual chemicals were... the variation in the samples is accounted for by the first two principal components For the above biplot (Figure 3. 8), the point and axis predictivities also be calculated using the program options (Figure 3. 9 (A) and (B)) [26] Predictivities indicate how well individual points or axes are represented in various dimensions of the biplot Generally, for a good PCA approximation, the points and axes are... is important in the view that, the synergic effect of these chemicals could provide significant influence in the malignancy of the tumor On the other hand, benign samples are greatly scattered in the biplot, showing less significance in the chemical’s trend Chapter 3 Figure 3. 7 PCA biplot of the first two principal components OA- Oxalic acid, AA- Adipic acid, CA- Citric acid, PhlP- 2-Amino-1-methyl-6phenylimidazo-... M3, M6, M7, M8 and M9 lie ordered along a virtually straight line In fact, the imagined line corresponds very closely to the biplot axis for PEB, PhlP, and Trp-P-1 The reason for this becomes obvious, by looking at the column of these chemicals (highlighted in green colour) in the Table 3. 4 data set The concentration of those chemicals are 64 Chapter 3 orders larger in malignant than in benign samples... extraction, the device was taken out of the sample solution, dried thoroughly with lint free tissue and placed in a 500 µL autosampler vial for desorption The analytes were desorbed from the 54 Chapter 3 device to the solvent by 8 min ultrasonication Then µ-SPE device was removed from the desorption vial and the extract was kept in a water bath at 60˚C for 20 min Finally, 2 µL of extract was injected into the. .. differ in benign and malignant samples, the individual group of 71 Chapter 3 chemicals shows a trend in their levels in both samples The goodness of the PCA biplot and the absolute error were measured as well to make sure the reliability of the statistical results This pilot study with moderate number of samples reveals the correlation of various carcinogens with respect to their levels in benign and . spectrometer. For nitrosamines, the helium carrier gas was maintained at a constant flow of 0.5mL min −1 . Injection volumes of 3mL were used. The injection port was held at 230 ˚C and used in the splitless. usually in tissues distance from the site of administration [18]. Cooking of protein- rich foods mainly from animal origin may stimulate the formation of a series of heterocyclic aromatic amines. follows: initial temperature 50 ◦ C, held Chapter 3 58 for 2 min, then increased by 10 ◦ C min −1 to 30 0 ◦ C and held for 3 min. OCP standards and samples were analyzed in selective