ABSTRACT
Polychlorinated aromatic compounds represent a large group of industrial and byproduct compounds which are resistant to chemical and biological degradation and highly toxic. QSAR analysis was performed on 74 molecules of three classes of polychlorinated aromatic compounds (polychlorinated dibenzo-p-dioxin (PCDDs), polychlorinated dibenzofuran (PCDF) and polychlorinated biphenyl (PCB)). QSAR models were also developed using each of the three series. Genetic function algorithm (GFA) approach was used to generate 5 models, for the three series all together and for each of the three series. The models with the highest statistical significance were selected as the best. For the three series all together: (Model-1: R2 = 0.9673, R2adjusted = 0.9592, R2cv = 0.9402, R2pred. = 0.7209, F-test = 118.48, LOF = 0.4377). For the series PCDDs: (Semi empirical- LOF = 0.5881, R2= 0.9516, R2adj.= 0.9389, R2cv= 0.9090, F-value = 74; DFT- LOF = 0.3877, R2 = 0.9680, R2adj= 0.9596, R2cv = 0.9518, F-value = 115.07). For the series PCDFs: (LOF = 0.2955, R2 = 0.9229, R2adj. = 0.9091, R2cv = 0.8885, F-value = 67.03).For the series PCBs: (LOF = 0.4582, R2 = 0.9186, R2adj. = 0.8734, R2cv = 0.8086, F-value = 20.31). From the models generated using the three series all together, it seems that polarizability, SP-7, ETA_Epsilon_5, GRAVH_3, and MOMI-R contributed positively to the toxicity of these compounds while MaxHBint5, ETA_dApha_B, ETA_Epsinlon-2, n5Ring and GRAV_2 contribute negatively. From the comparison of the models generated using DFT and semi-empirical and based on their statistical parameters, semi-empirical (AM1) has slightly better predictive power than DFT (BLYP/6-31G*). The robustness and applicability of the models were established by internal and external validation techniques. These validated modelsbring important insight to aid the prediction and identification of other toxic polychlorinated aromatic compounds.
viii
TABLE OF CONTENTS
Cover page ……………………………………………………………………………………………………………………… i
Title page ………………………………………………………………………………………………………………………. ii
Declaration ……………………………………………………………………………………………………………………. iii
Certification ………………………………………………………………………………………………………………….. iv
Dedication ……………………………………………………………………………………………………………………… v
Acknowledgement …………………………………………………………………………………………………………. vi
Abstract ……………………………………………………………………………………………………………………….. vii
List of Tables ………………………………………………………………………………………………………………. xiv
List of Figures ……………………………………………………………………………………………………………… xvi
List of Abbreviations ……………………………………………………………………………………………………. xix
CHAPTER ONE …………………………………………………………………………………………………………… 1
1.0INTRODUCTION…………………………………………………………………………………………………….. 1
1.1.0 Sources of Polychlorinated Aromatic Compounds ………………………………………………………. 2
1.1.1 Incineration sources ………………………………………………………………………………………………… 3
1.1.2 Combustion sources ………………………………………………………………………………………………… 6
1.1.3 Industrial sources ……………………………………………………………………………………………………. 7
1.1.4. Reservoir sources …………………………………………………………………………………………………… 8
1.2.0 Source for Human Exposure …………………………………………………………………………………… 11
1.3.0 Toxicity of PolychlorinatedAromatic Compounds …………………………………………………….. 11
1.4.0 Economic Impact of Polychlorinated Aromatic Compounds ………………………………………. 13
1.5 Statement of the Problem ………………………………………………………………………………………….. 14
1.6 Justification to the Research ……………………………………………………………………………………… 15
1.7 Research Question: …………………………………………………………………………………………………. 17
1.8 Research Hypothesis ………………………………………………………………………………………………… 17
ix
1.9 Theoretical Frame Work …………………………………………………………………………………………… 17
1.10. Aim and Objectives……………………………………………………………………………………………….. 19
1.7 Scope and Limitations of the Study ………………………………………………………………………….. 20
CHAPTERTWO …………………………………………………………………………………………………………. 22
2.0LITERERATUREREVIEW ……………………………………………………………………………………. 22
2.1 Number of Possible Isomers ……………………………………………………………………………………… 24
2.2 Toxicity of Polychlorinated Aromatic Compounds ………………………………………………………. 29
2.2.1 The concept of toxic equivalency factors …………………………………………………………………. 30
2.2.2 Evolution of the toxicity equivalence methodology …………………………………………………… 31
2.2.3 Toxicological Effects on experimental animals and in vitro test systems ……………………… 39
2.2.4 Carcinogenic effects on experimental animal ……………………………………………………………. 40
2.2.5 Toxicological effects in humans ……………………………………………………………………………… 41
2.2.6 Carcinogenic effects in Human ……………………………………………………………………………….. 41
2.2.7 Population groups at higher exposures …………………………………………………………………….. 42
2.2.8 Critical organs, tissues and effects …………………………………………………………………………… 43
2.3.0 QSAR ………………………………………………………………………………………………………………….. 44
2.3.1 Short history of QSAR and molecular descriptors …………………………………………………….. 45
2.3.2 Chemometrics and QSAR modeling ………………………………………………………………………… 48
2.3.3 Specific QSAR Approaches ……………………………………………………………………………………. 53
2.3.3 Molecular descriptors…………………………………………………………………………………………….. 57
2.4 Previous QSAR work on polychlorinated aromatic compounds …………………………………….. 70
2.4.1 Docking and 3D-QSAR studies on the Ah receptor binding affinities of polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs)………………………… 71
2.4.2 Binding affinities (AhR) of polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) study combining DFT and QSAR Results ……………………………….. 72
x
2.4.3 QSAR for toxicities of polychlorodibenzofurans, polychlorodibenzo-1, 4-dioxins, and polychlorobiphenyls ………………………………………………………………………………………………………. 73
2.4.4 QSAR Studies on polychlorinated aromatic compounds using topological descriptors ….. 75
CHAPTERTHREE ……………………………………………………………………………………………………… 77
3.0 MATERIALSANDMETHODS ………………………………………………………………………………. 77
3.1 Materials ………………………………………………………………………………………………………………… 77
3.1 Methods………………………………………………………………………………………………………………….. 77
3.2.0 Data Collection …………………………………………………………………………………………………….. 78
3.3 Molecular Optimization ……………………………………………………………………………………………. 79
3.3.1 Optimization with spartan 14. V.1.1.0 ……………………………………………………………………… 97
3.4 Calculation of molecular descriptors ………………………………………………………………………….. 97
3.4.1 Computing molecular descriptors using Spartan 14v.1.1.0 software ………………………….. 100
3.4.2 Computing molecular descriptors using Padel descriptor tool kit. ……………………………… 110
3.5 Calculation of Molecular Descriptors using Mathematical Formulae ……………………………. 111
3.6 Selection and Elimination of Molecular Descriptors …………………………………………………… 121
3.7 Data Set Division …………………………………………………………………………………………………… 122
3.8 Correlation Analysis using Material Studio 7.0 ………………………………………………………….. 126
3.9 Model Building using Genetic Functional Approximation (GFA) ………………………………… 126
3.9.1 Procedure for Building GFA Model ………………………………………………………………………. 132
3.10 Model Validation …………………………………………………………………………………………………. 130
3.10.1 Internal validation parameters……………………………………………………………………………… 136
3.10.2 External validation parameters ……………………………………………………………………………. 142
3.11 Multicollinearity among Descriptors: Variance Inflation Factor (VIF) ……………………….. 144
3.12 Euclidean Based Applicability Domain (AD) ………………………………………………………….. 144
3.13 Procedure for Computing Euclidean Distance using DTC-Euclidean Program …………….. 145
xi
CHAPTER FOUR ……………………………………………………………………………………………………… 150
4.0 RESULTS ……………………………………………………………………………………………………………. 150
4.1 Results of QSAR Modeling of pEC50 for (PCDDs, PCDFs and PCBs). ……………………….. 150
4.1.1 GFA Generated models of pEC50 for PCDDs, PCDDFs and PCBs all together ………….. 150
4.1.2 Statistical/Validation parameters for the generated models ……………………………………….. 150
4.1.3 Comparison of the observed and predicted pEC50 …………………………………………………… 151
4.1.4 Univariate analysis of the toxicity (pEC50) data ……………………………………………………… 154
4.1.5 External validation parameters ……………………………………………………………………………… 160
4.1.5.2 Euclidean base applicability domain for the selected model (model-1) ……………………. 160
4.1.5.3 Golbraikh and Tropsha‘s parameters for the selected model (model-1) ……………………. 160
4.1.6 Linear relationship of observed and the predicted pEC50 values for all the series ………… 160
4.1.7 Plot of residual versus actual toxicity (pEC50) values ………………………………………………. 160
4.2.0 Results of QSTR modeling for pEC50 for all the series together (PCDDs, PCDFs, and PCBs). ……………………………………………………………………………………………………………………….. 161
4.2.1 Five (5) generated models by GFA ………………………………………………………………………… 152
4.2 Results of QSAR modeling of pEC50 for (PCDDs) ……………………………………………………. 168
4.2.1 GFA Generated models of pEC50 for PCDDs Using DFT and SE approaches ……………. 168
4.2.2 Correlation matrix for DFT and SE approaches ………………………………………………………. 168
4.2.3 Statistical/Validation parameters for the five models generated for DFT and SE approaches………………………………………………………………………………………………………………………………….. 168
4.2.4 Comparison of the observed and predicted pEC50 …………………………………………………… 172
4.2.5 Univariate analysis of the of the toxicities (pEC50) data For DFT and SE approaches …. 175
4.2.4 Linear relationship of observed pEC50 values with the predicted values …………………….. 175
4.2.6 Plot of residual versus actual toxicity values for DFT and SE calculated descriptors …… 175
4.3 Results of QSAR modeling of pEC50 for (PCDDs) ……………………………………………………. 175
4.4.0 Results of QSAR modeling of pEC50 for PCDFs …………………………………………………….. 183
xii
4.4.1 GFA Generated models of pEC50 for PCDFs ………………………………………………………….. 183
4.4.2 Statistical/Validation parameters for the generated models ……………………………………….. 183
4.4.3 Comparison of the observed and predicted pEC50 …………………………………………………… 183
4.4.4 Univariate analysis of the toxicity (pEC50) data ………………………………………………………. 183
4.4.6 Linear relationship of observed and the predicted pEC50 values for PCDFs ………………… 183
4.4.7 Plot of residual versus actual toxicity (pEC50) values ………………………………………………. 184
4.5.0 Results of QSAR modeling of pEC50 for PCBs ……………………………………………………….. 185
4.5.1 GFA Generated models of pEC50 for PCBs …………………………………………………………….. 185
4.5.2 Statistical/Validation parameters for the generated models ……………………………………….. 185
4.5.3 Comparison of the observed and predicted pEC50 …………………………………………………… 185
4.5.4 Univariate analysis of the toxicity (pEC50) data ………………………………………………………. 192
4.5.6 Linear relationship of observed and the predicted pEC50 values for PCBs ………………….. 192
4.5.7 Plot of residual versus actual toxicity (pEC50) values ………………………………………………. 192
CHAPTERFIVE ……………………………………………………………………………………………………….. 199
5.0DISCUSSIONOFRESULTS ………………………………………………………………………………….. 199
5.1 Geometry Optimization and Calculation of Molecular Descriptors ………………………………. 199
5.2 GFA Generated Model for PCDDs, PCDDFs and PCBs all together ……………………………. 199
5.2.1 Significance of the descriptors in model 1 ………………………………………………………………. 200
5.3 GFA Generated Models for the pEC50 of PCDDs ………………………………………………………. 201
5.3.1 Significance of the descriptors in the models for PCDDs …………………………………………. 202
5.4 GFA Generated Models for the pEC50 of PCDFs ……………………………………………………….. 203
5.4.1 Significance of the descriptors in the models for PCDFs ………………………………………….. 204
5.5 GFA Generated Models for the pEC50 of PCBs …………………………………………………………. 204
5.5.1 Significance of the descriptors in the models for PCBs ……………………………………………. 206
xiii
CHAPTERSIX ………………………………………………………………………………………………………….. 207
6.0 SUMMARY, CONCLUSIONANDRECOMMENDATIONS………………………………….. 207
6.1 Summary of the Findings ………………………………………………………………………………………… 207
6.2 Conclusion ……………………………………………………………………………………………………………. 208
6.3 Recommendation …………………………………………………………………………………………………… 208
REFERENCES …………………………………………………………………………………………………………… 210
xiv
CHAPTER ONE
1.0 INTRODUCTION
1.1 Background to the study Since the middle of the 20th century there has been an increasing concern about the exposure ofhumans and wildlife to certain xenobiotic that were released into the environment due to diverse anthropogenic activities. One group of environmental toxicants that is of particular interest with regard to potential environmental health effects are polychlorinated aromatic compounds (PACs). These ubiquitous compounds are hydrophobic, lipophilic and resistant to biological and chemical degradation, properties that impart persistency and a propensity to bio-accumulate and biomagnify to concentrations that can cause harmful effects (Eichbaum et al., 2014; Giesy et al., 1994; Larsson et al., 2013; Poland et al., 1982; Song et al., 2006; Vries et al., 2006).
Polychlorinated aromatic compounds are a group of compounds comprising polycyclic aromatic hydrocarbons with two or more aromatic rings and one or more chlorine atoms attached to the ring system. Polychlorinated aromatic compounds can be divided into two groups: chloro-substituted Polycyclic Aromatic Hydrocarbons(PAHs), which have one or more hydrogen atoms substituted by a chlorine atom, and chloro-addedChlorinated Polycyclic Aromatic Hydrocarbons(ClPAHs), which have two or more chlorine atoms added to the molecule(Nilssonetal., 1993).They are products of incomplete combustion of organic materials. They have many congeners, and the occurrences and toxicities of the congeners differ (Kitazawaet al., 2016). ClPAHs are hydrophobic compounds and their persistence within ecosystems is due to their low water solubility (Cernigliaetal., 1992). They are structurally
2
similar to other halogenated hydrocarbons such as polychlorinated dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs), and polychlorinated biphenyls (PCBs). ClPAHs in the environment are strongly susceptible to the effects of gas/particle partitioning, seasonal sources, and climatic conditions (Ohuraet al., 2008). Polychlorinated aromatic compounds such as polychlorinated dibenzo–p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), polychlorinated biphenyls (PCBs), as well as other related halogenated aromatic compounds are trace amounts of undesired impurities (by products) in the manufacture of other chemicals such as chlorinated phenols and their derivatives, chlorinated biphenyl ethers, and polychlorinated biphenyls (PCBs), and combustion of chlorine containing materials under some conditions. These compounds are also referred to as unintentionally produced Persistent Organic Pollutants (UPOP) (WHO, 1988).
1.2 Sources of PolychlorinatedAromatic Compounds
Polychlorinated aromatic compounds are generated by combustion of organic compounds. They enter the environment from a multiplicity of sources and tend to persist in soil and in particulate matter in air. Environmental data and emission sources analysis for polychlorinated aromatic compounds reveal that the dominant process of generation is by reaction of polycyclic aromatic hydrocarbon (PAHs) with chlorine in pyrosynthesis (Ohura, 2007).
Polychlorinated aromatic compounds (PCBs and Dioxins) have commonly been detected in tap water, fly ash from an incineration plant for radioactive waste, emissions from coal combustion and municipal waste incineration, automobile exhaust, snow, and urban air (Nilsson
3
et al., 1993). PCBs and Dioxins have also been detected in electronic wastes, workshop-floor dust, vegetation, and surface soil collected from the vicinity of an electronic waste (e-waste) recycling facility and in surface soil from a chemical industrial complex (comprising a coke-oven plant, a coal-fired power plant, and a chloro-alkali plant), and agricultural areas in central and eastern China (Maet al., 2009). In addition, the combustion of polyvinylchloride and plastic wrap made from polyvinylidene chloride result in the production of polychlorinated aromatic compounds, suggesting that combustion of organic materials including chlorine is a possible source of environmental pollution (Wanget al., 2003).
1.2.1 Incineration sources
Incinerationis the largest source of dioxinrelease into the environment. According to National Renewable Energy Laboratory (1999), the majority of dioxins are predominantly from municipal solid waste incineration.
From literature, PVC produces HCl upon combustion that is almost quantitatively related to its chlorine content (Steiglitz et al., 1988). However, extensive studies in Europe indicate that the chlorine found in emitted dioxins is not derived from HCl in the flue gases, instead, most dioxins arise in the condensed solid phase by the reaction of inorganic chlorides with graphitic structures in char-containing ash particles and the reaction was catalyzed by copper (Steiglitz et al., 1988). Studies of household waste burning indicate consistent increases in dioxin generation with increasing PVC concentrations (Costneret al., 2005). According to the EPA dioxin inventory, landfill fires are likely to represent an even larger source of dioxin to the environment. A survey of international studies consistently identifies high dioxin concentrations in areas
4
affected by open waste burning and a study that looked at the homologue pattern found the sample with the highest dioxin concentration was typical for the pyrolysis of PVC (Costneret al., 2005). Other EU studies indicate that PVC likely accounts for the overwhelming majority of chlorine that is available for dioxin formation during landfill fires(Costneret al., 2005). The next largest sources of dioxin in the EPA inventory are medical and municipal waste incineraton (Beychok , 1987). For instance a study of commercial-scale incinerators showed no relationship between the PVC content of the waste and dioxin emissions (Rigoet al., 1995). Other studies have shown a clear correlation between dioxin formation and chloride content and indicate that PVC is a significant contributor to the formation of both dioxin and PCB in incinerators (Katamiet al., 2002; Wagneret al., 1993; Thornton, 2002). Several researchers have described their mechanism of formation dioxins and dioxin-like compounds. Overall, it is observed that the emission of dioxins and furans into the environment can be explained mainly by two principal surface catalytic processes: i) Formation from precursors and ii) formation by de novo synthesis (Altwickeret al., 1996). An informative review of the formation and mechanism of dioxins from municipal solid waste incineration was reported (Tuppurainenet al., 1998). It was observed that several past studies demonstrated the presence of significant quantities of dioxins and dioxin precursors in municipal solid waste around 50 ng( Toxic equivalent) I-TEQ/kg (Abadet al., 2002).
Gullett et al. (1990a; 1990b; 1991a; 1991b; 1992) studied the formation mechanisms through extensive combustion research at EPA, and verified the observations of Vogg et al. (1987). It was proven that CDDs and CDFs could be ultimately produced from low temperature
5
reactions (i.e., 350°C) between Cl2 and a phenolic precursor, combining to form a chlorinated
precursor, followed by oxidation of the chlorinated precursors (catalyzed by a copper catalyst
such as copper chloride) as in examples (1) and (2), below.
(1) The initial step in forming dioxin is the formation of chlorine from HCl in the
presence of oxygen (the Deacon process), as follows (Vogg et al., 1987; Bruce et al., 1991):
2 2 2 2
2HCl 1 O H O Cl
(2) Phenolic compounds adsorbed on the fly ash surface are chlorinated to form the
dioxin precursor, and the dioxin is formed as a product from the breakdown and molecular
rearrangement of the precursor. The reaction is promoted by copper chloride acting as a catalyst
(Vogg et al., 1987; Dickson and Karasek, 1987; Gullett et al., 1992):
(a) ( ) 2 Phenol Cl chlorophenol dioxin precursor
(b) 2-chlorophenol + ½ O2 ————> dioxin + Cl2
On the other hand, Eklund et al. (1986) observed the high temperature formation of a
large variety of chlorinated toxic compounds, including CDDs and CDFs, from precursors during
a simple experiment in which phenol was oxidized with HCl at 550°C. One milligram of phenol
was placed in a quartz tube reactor with an aqueous solution (10μL) of HCl and heated at 550°C
for 5 minutes. Trichlorobenzene, dichlorophenol, dichlorobenzofuran, tetrachlorobenzene,
trichlorophenol, and tetrachlorophenol were identified as major products formed.
Monochlorobenzene, chlorophenol, dichlorobenzene, tetrachloropropene, pentachloropropene,
trichlorobenzofuran, tetrachlorodibenzofuran, trichlorodibenzodioxin, tetrachlorodibenzodioxin,
6
hexachlorodibenzodioxin, hexachlorodibenzofuran, pentachlorobenzene, pentachlorobiphenyl, and pentachlorodihydroxycylohexane were seen as minor products. Trace species formed included: monochlorodibenzofuran, pentachlorodibenzofuran, pentachlorodibenzodioxin, octachlorodibenzofuran, and octachlorodibenzodioxin. (Eklund et al. 1986) hypothesized that chlorinated organic compounds can be produced from phenols, acids, and any chlorine source in the hot post-combustion region (e.g., just exit to the furnace). The reaction was seen as very sensitive to HCl concentration 1.2.2 Combustion sources Cement kilns The switch to burning hazardous waste as fuels for cement kilns has created problems for individuals and organizations. About 16% of the facilities burn hazardous wastes as auxiliary fuel. Some limited data suggest that PCDD/PCDF levels in clinker dust and stack emissions of these kilns may be significantly higher than the kilns which do not burn hazardous waste (Abad et al., 2004; Eduljee et al., 1999). Wood burning A number of studies have found dioxins in the emissions and ash/soot from wood fires in non-industrial situations (Stanmore et al., 2004). According to the European Emission Inventory, wood combustion is at present one of the most important air emission sources of dioxins (Quaßet al., 2000). In an appealing review paper, it is reported that the dioxin emission from wood burning is about 945 g (Toxic equivalent) I-TEQ/year (Lavric et al.,2004). Diesel vehicles
7
A very scanty literature available on this source showed that dioxins could be emitted from diesel vehicles. In fact, researchers from Sweden and Norway have shown that dioxins are emitted from diesel vehicles (Marklundet al., 1990; Oehme et al., 1991). As these studies depend on the fuel used in a particular country more studies are required in order to reach a conclusive estimate. Crematoria Crematoria procedures can be a ready source of organic material and chlorine, and hence are possible source of dioxin emission (Alcocket al., 1999). Inventory estimates rate this source as 0.3% of European output and 0.24% of US output (USEPA, 1998). Coal-fired utilities Although emission of dioxins compared to the wood burning are less, they are numerous, large in size and their high stacks indicate that they could impact verylarge areas (Chenet al., 2004; Harradet al., 1991). Considering the large scale usage the importance of these facilities is very much unknown.
1.2.3 Industrial sources
Pulp and paper mills
The manufacture of bleached pulp and paper has in the past resulted in dioxin releases to water, land and paper products. These compounds can be formedthrough the chlorination of naturally occurring phenolic compounds such as those present in wood pulp (Rappe et al., 1987). It is
8
reported that the waste generated from a pulp mill of China produces dioxin concentration of 300 pg/l I-TEQ (Zheng et al., 2001). Metals industry The metallurgical processes such as high temperature steel production, smelting operations, and scrap metal recovery furnaces are found to be typical sources of dioxins (Anderson et al., 2002). Processes in the primary metals industry, such as sintering of iron ore, have also been identified as potential sources (Ciepliket al., 2003; Wanget al., 2003). In several countries the annual release of dioxins is estimated to be 500–4000 g I-TEQ (Anderson et al., 2002). Chemical manufacturing PCDDs and PCDFs can be formed as by-products from the manufacture of chlorinated compounds such as chlorinated phenols, PCBs, phenoxy herbicides, chlorinated benzenes, chlorinated aliphatic compounds, chlorinated catalysts and halogenated diphenyl ethers (Öberget al., 1992; Oberg et al., 1993; Sidhu et al., 2002). 1.2.4 Reservoir sources
The persistent and hydrophobic nature of these compounds causes them to accumulate in soils, sediments, landfill sites, vegetation and organic matter. They have potential for redistribution and circulation of dioxins in the environment. The dioxin compounds in the reservoirs‖ can be redistributed and circulated in the environment by dust or sediment suspension and transport (Kjelleet al., 1995; Rotard et al., 1994). The major reservoir sources include:
9
a. Biological processes: The action of microorganisms on chlorinated phenolic compounds results in the formation of dioxins under certain environmental conditions (Siewerset al., 1994).
b. Photochemical processes: Dioxins like OCDD (1,2,3,4,5,6,7,8,9-octachloro-dibenzodioxin), HPCD (1,2,3,4,5,6,7,8-heptachloro-dibenzodioxin) formation occurs by photolytic radical reactions of pentachlorophenol (Baker et al., 2000; Tysklindet al., 1993).
c. Accidental sources: The incidents of dioxin release at Seveso, Italy and Yusho Japan could be considered as an accidental release of dioxins into the atmosphere. Further, forest fires and volcanoes also come under this category (Clementet al., 1991; Ruokojärvi et al., 2000).
d. Miscellaneous sources: Miscellaneous sources include formation of dioxins in FBC (Fluidized Bed Combustion) boilers, thermal oxygen cutting of scrap metal at demolition sites, power generation, PVC in house fires, Kraft liquor boilers, laboratory waste, drum and barrel reclaimers, tire combustors, carbon reactivation furnaces and scrap electric wire recovery facilities, etc. (Anthonyet al., 2001; Carrollet al., 1996; Menzelet al., 1998). The summary of the various sources of polychlorinated aromatic compounds is presented in Figure 1.1 according to Weber et al. (2008).
The main new (de novo) sources of PCDD/Fs are combustion processes, such as burning of waste and metal smelting,and refining. In Europe, the Baltic Sea is an important sink ofPCBs and dioxins. However, recent studies have revealed a major problem at localized spots,
10
Figure 1.1 Emission sources of polychlorinated aromatic compounds \
11
due to theproduction and use of chlorophenolsfor impregnation of timber. In the most contaminated regions the concentration of PCDDs and PCDFs in soil and sediments appears to be incredibly high. An unpredictable source is old transformers and capacitors, each of which may contain several kilograms of PCBs and hundreds of milligrams of PCDD/Fs.
1.2.5 Source for Human Exposure
Food is the major source for human exposure to PCBs and dioxins, especially fatty foods: dairy products (butter, cheese, fatty milk), meat, egg, and fish. Food of animal origin accounts for 95 % of total exposure. The current average body burden of dioxins is about 5–50 ng/kg (as WHOTEq in fat; pg/g = ng/kg) or 100–1000 ng (WHO-TEq) per person which is close to the lowest concentrations possibly causing health effects. Some subgroupswithin the society (e.g., nursing babies and people consuming plenty of fish) may be exposed to higher than average amounts of these compounds and are thus at greater risk. Dioxin concentrations have been screened in five WHO international studies, and in Central Europe the concentrations have decreased in breast milk from about 40 ng/kg (as TEq in milk fat) in 1987 to below 10 ng/kg in 2006. PCBs have decreased at about the same rate. The decrease in environmental concentrations is due to cessation of PCB use and improved incineration technology.
1.2.6 Toxicity of PolychlorinatedAromatic Compounds
When several Polychlorinated Aromatic compounds appear in mixture (Fiedler et al., 1990) (as they occur in the environment), and when as usual they have the same mechanism of action, their cumulated toxic effect could be found by a relationship involving partial toxicities, reminiscent of Dalton‘s Law of partial pressures: total equivalent toxicity is the sum of the
12
product of their concentration with their Toxic Equivalent Factor (TEF) value (Van der Berg et al., 1994). Polychlorinated aromatic compounds (Dioxin-like compounds) is one of the most toxic chemicals known. A report released for public comment on September 1994 by the US Environmental Protection Agency (USEPA, 1994) clearly describes dioxin and dioxin-like compounds as a serious public health threat in the 1960s. According to the EPA report, not only does there appear to be not safelevel of exposure to dioxin, but levels of dioxin and dioxin -like chemicals have been found in the general US population that is at or near levels associated with adverse health effects. The EPA report confirmed that dioxin is a cancer hazard to people, that exposure to dioxin could also cause severe reproductive and developmental problems (at levels 100 times lower than those associated with its cancer causing effects); and that dioxin could cause immune system damage and interfere with regulatory hormones. These organic pollutants constitute a group of lipophilic chemically, stable environmental contaminants with low volatility which have been identified in fatty tissues of animals and humans. Several of these polychlorinated aromatic compounds have been shown to exert a number of common toxic responses similar to those observed for 2, 3, 7, 8 tetrachlorodibenzo- p- dioxin (TCDD). These include thermal toxicity, immunotoxicity, endocrine toxicity and carcinogenicity/tumour promotion.
Some of the effects of polychlorinated aromatic compounds biological effects including hepatoxicity, endocrine effects, immunotoxicity, body weight loss, teratogenecity,carcinogenicity and the induction of enzymes such as ary hydrocarbon
13
hydroxylate [AHH] and 7-ethoxyresorufin o-deethylase[EROD] in various organisms (chovoncovaet al., 2005). There is strong evidence suggesting a common mechanism of action of 2, 3, 7, 8 – TCDD and related compounds, based on a binding of these compounds to a specific receptor (the aryl hydrocarbon – receptor) (Walker et al 1991). Some of polychlorinated biphenyl (PCB) also have dioxin–like properties and are included as part of dioxin and dioxin–like compounds with 1 to 10 chlorine atoms attached to biphenyl molecule. There are about 209 possible PCB congeners although only 130 were found in commercial PCB mixture (Van den Berg et al, 2006).
1.2.7 Economic Impact of Polychlorinated Aromatic Compounds
Information on effects in humans predominantly comes from high exposure in industrial settings or accidents and intoxication incidents, e.g. the contamination of rice oil with PCBs that contained comparatively high concentrations of PCDFs in Japan and Taiwan in 1968 and 1979, respectively. The effect observed are generally in agreement with what could be expected from animal experimentation, although the most typical human effect of high exposure to PCDDsand PCDFs seems to be the dermatological response chlorace Subtle endocrinological effect e.g. modulation of thyroid hormone and testosterone levels in plasma, and decreased glucose tolerance and subtle neurological effect have been observed in humans, but their clinical significance is currently unknown (Huisman et al., 1995).
The production and use of PCBs have been discontinued in most countries, but large amounts remain in electrical equipment, plastic products, buildings (e.g. plastic carpeting, sealing materials), and in the environment. Because PCBs are considered problem waste, their disposal is expensive, and may sometimes lead to attempts to dispose of them by mixing them to
14
other waste products. Two well-established environmental accidents leading to human suffering have occurred, called Yusho (Japan) and YuCheng (Taiwan) poisonings. In both cases rice oil was contaminated andcaused a number of health effects. Later contamination incidents have not lead to clear health consequences. 1.3 Statement of the Problem Polychlorinated aromatic compounds are persistent and widespread environmental contaminants which can cause a great diversity of biological effects including hepatoxicity, endocrine effects, immunotoxicity, body weight loss, teratogenecity,carcinogenicity and the induction of enzymes such as ary hydrocarbon hydroxylate [AHH] and 7-ethoxyresorufin o-deethylase[EROD] in various organisms. (chovoncovaet al., 2005). These environmental pollutants, which persist worldwide and are implicated in a variety of adverse health effects, are metabolized by cytochrome P450s to their hydroxylated forms. The formation of the hydroxylated form of these environmental pollutants may lead to their detoxication, however, these metabolites have been shown to be retained in blood, liver and adipose tissues (safe et al, 1985).
Polychlorinated aromatic compounds are ubiquitously environmental contaminants .Some of them, notably TCDD (2,3,7,8-tetrachlorodibenzo-p-dioxin) belong to the most toxic synthetic compounds known. They are very stable against chemical and microbiological degradation and therefore persistent in the environment. They are fat-soluble and thus tend to bioaccumulate in tissues, lipid and in the food chain. These factors increase their potential harzards to humans and animals. Polycholorinated aromatic compounds (PCBs) which are
15
considered problem waste are present in large amount in electrical equipments, plastic products, buildings (plastic carpenting, sealing materials) and in the environment. The Quantitative Structure-Toxicity Relationship (QSTR) models for predicting the toxicity of polychlorinated compounds can be used to screen out the most toxic compounds of polychlorinated compounds and other related compounds. 1.4 Justification to the Research Because of the persistent and toxic effect of polychlorinated aromatic compounds which are released into the enviroment by chemical manufacture, industrial and agricultural uses, and disposal following domestic application, information regarding their toxicity and fate is required. However, assessing the fate and toxicity of large number of these chemicals using conventional methods using classical in vivo animal testing becomes expensive and time consuming, hence, there is need for an alternative method. Consequently, in the area of computer – aided toxicity prediction, quantitative structure activity relationships (QSARs) have been seen asan attractive method for toxicity and fate assessment. (Knight et al., 2002).Therefore, this research is initiated to help:
a. reduce the requirement for lengthy and expensive animal tests,
b. reduces animal use,
c. reduces /eliminates pains and discomfort to animals,
d. promotes savings in the cost of product development and
e. ultimately promotes greener chemistry to increase efficiency and eliminate waste.
The summary of the need for this research to be carried out is presented in Figure 1.2.
16
Figure1.2: The Significance of QSAR (Gajewicz et al, 2012).
17
1.5 Research Question
Structural features of molecules (molecular descriptors) are function of the activity/toxicity of the molecules. This research intends to address the questions: (i) What are the structural features (molecular descriptors) responsible for the toxicity of polychlorinated aromatic compounds. (ii) How could tocixity of polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), polychlorinated biphenyls (PCBs) and other related compounds be predicted using QSAR approach?
1.6 Research Hypothesis
.This research is guided by the following null and alternative hypotheses. Null hypothesis (H0): The biological activities of new untested and even non-synthesized chemicals (polychlorinated aromatic compounds) are dependent of their structural properties (descriptors). Alternative Hypothesis (H1):The biological activities of new untested and even non-synthesized chemicals (polychlorinated aromatic compounds) are independent of their structural properties (descriptors).
1.7 Theoretical Frame Work of the study
Quantitative Structural-Activity Relationship (QSAR) is a mathematical model used to relate known activity of a congeneric series of compounds to their structure or properties to
18
predict other compounds‘ unknown activity. QSAR is based on the assumption that the biological activity of a new or untested chemical can be inferred from the molecular structure, or properties of similar compounds whose activities have already been assessed. When the relationship is developed with;non continuous or categoramrical data, it is called Structural Activity relationship (SAR) and continuous or quantitative data it is called a (QSAR) (Schultz et al., 2003). The study of the quantitative relationship between toxicity and molecular structure (QSTR/QSAR) is an important research area in computational chemistry and has been widely used in the prediction of toxicity and other biological activities of organic compounds (Katritzky et.al, 2000 and Katritzky et.al, 2001). This kind of study develops a method for the prediction of the activity under investigation of new compounds that have not been synthesized. It can also identify and describe important structure features of moleculesthat are relevant to variations in molecular properties, thus gain some insight into structural factors affecting molecular properties. To develop a QSAR model, the following steps are usually involved; data collection, molecular geometry optimization, molecular descriptors generation, descriptors selection, model development and finally model performance evaluation.
The reliability of the developed QSAR model can be tasted by comparing the end point values of the model to predict experimentally determined end point values of similar chemicals (training set chemicals). If the predictions are poor, one can restart the model development by using different descriptors, or noting training set compounds, whose predicted values deviate greatly from the experimental values. This can help to identify compounds that act by a different mechanism of action. If the predictions are good, one can define the selection criteria and limits
19
of the model and then make toxicity predictions to other chemicals that meet the selection criteria, and were not used during model development (Walker et al., 2003).
1.8 Aim and Objectives
The aim of this research is to develop a Quantitative Structure-Toxicity relation(QSTR) model using Genetic function approximation (GFA) for the prediction of the toxicity of polychlorinated aromatic compounds (PCDDs, PCDF and, PCBs) solely on basis of chemical structures.The aim of this research was achieved through the following objectives:
(i) Selecting the data set of the molecules (polychlorinated aromatic compounds)
(ii) Drawing the molecular structure of the compounds and determining the energy of minimization and optimization for the molecules using Density Functional Theory and Semi empirical method;
(iii)Computation of molecular descriptors (numerical values that characterize properties of the molecules) using the softwares: Spartan ―14‖, PaDel descriptor and theoretical calculations.
(iv) Conducting elimination and selection techniques of variables (descriptors).
(v) Dividing the data set into training set and test set.
(vi) Modeling the relationship between molecular descriptors and the activities of the training set using Genetic function approximation (GFA);
(vii) Validating the models using internal and external validation techniques (vii)Selecting the best model that can give correct prediction.
20
1.9 Scope and Limitations of the Study
In this study, polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans(PCDFs), and polychlorinated biphenyls (PCBs) are considered to represent a large group of industrial and byproduct compounds which are resistance to chemical and biological degradation and highly toxic. This research involves subjecting series of PCDDs, PCDFs, and PCBs compounds to Quantitative Structural-TOXICITY Relationship (QSTR) studies computationally using different molecular descriptors. The dependent variable in this study represents log of (EC50) values (EC50 – median effective concentration during bioassay). Multivariate regression analysis will show how much well the toxicity can be modeled. This work will therefore contribute in the identification of those compounds which are very toxic and polluting our environment. The limitations of this research are as follows: If there is a measurement error in the experimental data, it is very likely that false correlations may arise If the training dataset is not large enough, the data collected may not reflect the complete property space. Consequently the QSAR results cannot be used to confidently predict the most likely compounds of the best toxicity. There are many successful applications of this method but one cannot expect the QSAR approach to work well all the time. QSAR models are easy to build also very easy to get it wrong.
Multiple Linear Regression analysis (MLR) has the following limitations:
21
MLR required normally distributed, independent and 100% relevance descriptors. This means that each descriptor is assumed the 100% relevance for explanation of the ‗cause‘ of the measured endpoint. When the number of variables is greater than the number of observations, the MLR will not yield a unique solution (Korhonen, 2007).
22
IF YOU CAN'T FIND YOUR TOPIC, CLICK HERE TO HIRE A WRITER»
