Activity Prediction of Various Herbicides against Honey Bee, Avian, and Multiple Human Leukemia, CNS, Ovarian, Prostate Cancer Cell Lines

Herbicides classify as chemicals targeting specific biochemical pathways in plants and may influence human or animal health according to their chemistry, concentration, environment, biological target and others. With safety concern, International Agency for Research on Cancer (IARC) classified herbicides and their metabolites as fetal developments may be a consequence of enzymatic inhibition or other mechanisms. Thirty phytotoxins were subjected to online pkCSM website, as a Quantitative Structure Activity Relationship (QSAR) prediction activity against honey bee, avian, and multiple human Leukemia, CNS, Ovarian, Prostate Cancer cell lines. Prediction outcomes were varied and influenced by chemical structure of each tested herbicide. Sulfentrazone having evidence of human non-carcinogenic character (Group E) had hepatotoxicity prediction and cancer cell lines activity less than 5 of Leukemia, CNS, Ovarian, and Prostate. Also, it had CYP1A2 inhibition, negative response of p-glycoprotein, Ames, skin sensitization, renal OCT2, and hERG. All above characters beside low intestinal absorption and Blood-Brain Barrier (BBB) presented encouraging online funding as more structurally safe having active – multiple toxicological and cellular interactions. Simetryn and Simazine that have the same core structure except (-SCH3) group replaced with chloro group gave semi


Introduction
In general, herbicides are chemicals that target plant biochemical pathways by their toxicological mechanisms. These high toxicities on plants may be with lower or higher effects on animal or human. Chemical structure, concentration, environmental conditions (time, temperature), plant (or animal or human) classification and chemistry, and others are mainly impact factors on herbicidal destroying effects upon living creatures [1]. These phytotoxins are working with variable degrees in growth inhibition or destroying weed/crop to control plantation instead of hand weeding or using nonspecific chemicals [2].
Their mainly actions are focusing on enzymes such as carboxylase, oxidase, dioxgenase, synthase, and others together with photosynthesis processes inhibitions. It is known that agricultural activities need herbicides but these chemicals have serious impacts on environment (soil, plant, food, animal, atmosphere, seawater, freshwater, rain, other water sources), human resources and public human health. In human life section, these phytotoxins are actively and generally employed on cell cycle-growth with particular effect(s) on liver, intestine, skin, lung, and / or kidney encouraging cell-organ damage. Banning of these toxics in a great degree is due to metabolism and permanent damage or disorder effect that is more seriously with chronic exposure [3].
Many microorganisms in environment are capable to metabolize herbicide(s) then access, accumulate in human through oral, dermal, and / or pulmonary path then elimination out of body may be occurred with easily detection by advanced analytical instruments. With safety concern by international agencies like the International Agency for Research on cancer (IARC), herbicides and their metabolites are classified according to benignityinjury categories. For example glyphosate is probably carcinogenic to human [4]. Even with minimum residual possibilities of these agricultural-vegetation materials in foods and living organisms, fetal developments may be a consequence of enzymatic inhibition or minimizing of essential bio-synthetic materials such as amino acids [5].
To have a deep ranking about herbicide toxicity and carcinogenicity in both severity and risk issues, many international organizations like World Health Organization (WHO), The European Food Safety Authority (EFSA), US Environmental Protection Agency (EPA), and /or IARC classified, studied and reported these chemicals even at final level of non-corresponding statements. Finding a link between herbicide trace in food and cancer is critical questionnaires with scientific evidences. Presence of incomparable conclusions based upon collected outcomes (Tables), test regulation, and exposure scenarios motivated researchers and these authorized organizations to category chemicals (including herbicides and their metabolites) according to cancer classification, Globally Harmonized System (GHS), Occupational Safety and Health Administration (OSHA), and other safety and hazard tabulations [6][7][8][9][10][11].
According to our knowledge, mathematical prediction in herbicidal toxicity and safety subjects is not observed yet by researchers especially in Adsorption-Distribution, Metabolism, Excretion, Toxicity (ADMET) and various cancers. Here, a new Iraqi study was designed to investigate the influence of thirty herbicides having various chemical-physical-biological properties on Pharmacokinetics, Crop, and Leukemia -CNS -Ovarian -Prostate Cancers by employing online websites belong to pkCSM -http://biosig.unimelb.edu.au/biosig/ -University of Melbourne -Australia as a Quantitative Structure Activity Relationship (QSAR) prediction.

Experimental Procedure
Herbicides: thirty herbicidal materials were chosen for this study (Table 1   CropCSM prediction according to website [13].

Results and Discussion
Thirty herbicides were selected to test their QSAR through various predictors with the help of isomeric SMILES notations and [12] as free online website (Table 1.). The easy steps of all [12,13,14] predictions were very helpful in proceeding of this research article. More than thirty predicted characters (Tables (2-7)) are summarized (Tables 8, 9, & 10)    Caco-2 (cancer coli-2) is a human colon cancer cell line, intensively studied for the expression of enzymes and transportation proteins, bioactive passive diffusion via intestinal epithelial pathways, cytotherapy and food bioactive (drugs or formulation) effect on cancer due to the reproducibility and simplicity of this system [15,16].
In general, Caco-2 permeability may be influenced by paracellular permeabilityhydropholicity relationship [17]. So, herbicide with low hydrophilic character had high Caco-2 in vitro monolayer model (Tables 2 & 3)   Many in vitro and /or in vivo toxicological studies were following managed guidelines to evaluate chemical hazard to human and environment especially in oral absorption issues and related consequences [18][19][20][21]. Outputs of these studies permit conventional safety assessment with high transparency basing on science to protect human -animal health and leading the way to new methods companied by consuming less time-cost-resource findings [22].
Always, natural and synthesized chemicals need safety evaluation that approve using by human according to scientific approaches related to Animal/ Human/ Environment Reduction-Refinement and Replacement (AHE-3Rs) [23]. From these essential points, this in Silico study targeted many toxicological points related to herbicides effects.
One of these studied toxicological points was intestinal adsorption that had the same sequence of variation as in Caco-2 from (V) to (P11). In particular observations, replacement of chloro group in (S6) with (SCH3) group gave a little raise in this character but this replacement beside the addition of two methyl groups decreased this character as in (P7 and P6).
Oxygen presence (P7) gave more intestinal absorption compared to sulfur presence (P6), and absence of two methyl groups (S5) increased this important character compared to (P6) that may be related to polar surface area, hydrogen bonding, organic-aqueous phase solubility, atom negativity, and other related reasons (Figure 1., Tables 2-3).
Also, twenty seven of these tested herbicides exhibited high intestinal absorption (with more than 87%) presenting 90% of them that associated with their molecular characterizations.
The lowest intestinal absorption values were in S9 (51.056%), T5 (56.457%) and P2 (64.197%) that have the same core structure where replacement of (O-CH2-CH2-Cl) with (CH2-CH2-CF3) pending group gave a raise with more than 7% while structural changing between S9 and T5 gave a (5.411%) as difference between S9 and T5 ( Figures 5 & 6., Tables 2-3). These structural reasons can be summarized as permeabilityhydropholicity relationship that has been mentioned in Caco-2 explanation. Skin is a major human barrier consisting of multi-layers with various morphological functions. Its permeation is a rate-limited nature controlling chemical (or drug) administration, discovery, and development processes. So, skin as a primary protection shield of human, an controller of losing water, and a homeostatic barrier can be influenced by chemical(s) through penetration (in the body) or remaining out of the body (non-penetration). These dermal ways are examples of infection and medication that limit chemical diffusion in bloodstream or its stability in gastrointestinal tract. As it is known, in Silico modeling provides non-ethical economical methodology of natural, synthesized, and hypothetical compound(s) with limited time-consuming steps. Permeation of any chemical can be occurred by diffusion, trans-or inter-cellular route according to its molecular polarity character and other physicochemical properties [24,25].
Here, skin permeability was influenced by functional group presence or absence such as replacement of chlorine atom with the (SCH3, OCH3), hydrogen atom with Cl-CH=CH2 or CH3, or carbonyl (C=O) group with thiocarbonyl (C=S) group ( Figure 1) and (Tables 2 & 3).
In another comparison of P2, S9, and T5 that have the same core structure, skin sensation did not affected by replacement of chloro group in ( (Tables 2 & 3). In contract to the above note, presence of (Cl-CH=CH2), carbonyl or thiocarbonyl, methyl group in V and S2 had a remarkable note related to their high skin sensation values ( Figure 1) and (Tables 2 & 3). P-Glycoprotein (P-gp) [26,27] is a special carrier belongs to Adenosine TriPhosphate (ATP) binding cassette transporter having various interaction nature with chemical(s) classified as substrates / inducers / inhibitors. Some chemicals are bi-or tri-mixed type of classification beside one type of acting. For example, the famous antimalarial alkaloid (Quinidine) is a tri-mixed p-gp acting as substrate, inhibitor, and inducer. In another site, Artemisin is acting as non-transported inducer, non-competitive inhibitor, and non-substrate to this carrier.
There are many natural and pharmaceutics achieving P-gp inhibition like herbs (garlic, green tea, rosemary), fruits (grape, orange), natural constituents (glycoside, flavonoid, terpenoid), anticancer drug (doxorubicin), and antibiotic (erythromycin) [27,28]. To achieve inhibitory mechanism, there are various strategies including:  Inhibition of ATPase that affects protein pump sensitization.  Interacting with polar heads in the lipid bilayer and changing bonding forces.  Inhibition of transport function by reacting with cysteine located in P-gp transmembrane through thiol group.  Decreasing of P-gp phosphorylation leading to inhibition of protein kinase.
So in general outlook, P-glycoprotein acts as a multidrug resistance protein or cell protector against toxic material found in brain, intestine, kidney, pancreas, liver, and/or testis and a secretionelimination controller that may failure drug treatment. P-Glycoprotein may prohibit this protein system (inhibitor class) or merging drug to protein (Substrate class) [29].
p-gp substrate may have electron donor atom or group ((>C=O, -O-, -S-, R-F, -NR3, -N=) with spatial separation or three hydrogen bond acceptor with more spatial separation. In another site, hydroxyl, primary or secondary amine may decrease passive diffusion. Substrate that revealed non-effective transport (net efflux) has hydrogen bond acceptor with lower rate of transport (flipping) than active (flopping) [27,30].
Interfering with the transporter leads to inhibit P-gp giving Yes response for p-Glycoprotein inhibition type I which was presenting only with T1, T6, P1, P2, and P8 beside Yes response for p-Glycoprotein inhibition type II was only with P1.
Central Nerve System (CNS) is bio-controlling environment in mammals and well developed organisms that effectively works with biological barriers. One of the most important protectors is Blood-Brain Barrier (BBB) that penetrate spinal cord and brain from bome agent(s) and exogenous material(s) and circulating back to the blood (for potentially harmful compound(s)) as well as delivering useful material(s) to brain from the blood. These dynamic regulating functions are highly responding for CNS induction and maintenance at various pathological and physiological conditions [31].
pH, calcium, magnesium, potassium, and water soluble nutrients beside their metabolites are essentials to human where they regulated by BBB via low passive permeability while fat soluble substances are transported by specific selective systems such as proteins or polypeptides to achieve their activities. Blockage mechanism of endogenous and xenobiotics minimizes CNS exposure to these substances. Excogenous compounds like herbicides and their metabolites may access / blocked CNS through the primary dynamic gate BBB [27,31]. Changing in BBB values are mainly related to selective semipermeable BBB nature beside large surface area, negative surface charge, diffusion distance (or uniform thickness) from BBB to CNS through intracellular and intercellular paths, and structural -physicochemical properties of target molecule(s). Lipophilic, ionization, and polarity of atom(s) or group(s) regulate intra-passage of cell(s) while diffusion controls inter-movement. These regulators in both BBB and CNS membranes manage stabilityspecificity base of chemical(s) in human body toward drug designdevelopment research gates [32]. In various cases, changing in attached functional group(s) varied BBB values (Figures 1, 6  Cytochromes P450 (CYPs) are enzymes binding to cell membrane having heme structure. They had been found in human and other mammals with advanced identification of fifty-seven isoforms. In clinical studies, CYP3A4, CYP2D6, CYP2C19, and CYP2C9 perform drug or xenobiotic metabolism, drug -drug competing inhibition, and drug with high dose causing accumulated drug -plasma toxicity. So with these and other functions, CYPs are primer drug design and safety risk alerts. Relationship between food components, drug(s), and various CYP450 were studied and reviewed showing that some dietary components had possible effect on CYPs such as garlic, pepper, peppermint, grapefruit, orange, and others [33].
Major tested herbicides exhibited high intestinal absorption (with more than 87%) presenting 90% of them as mention above. By oral intake of plant parts and products having herbicidal molecules or their metabolites, these absorbed xenobiotic molecules may be transported to the liver leading to more number of enzymatically converted water-soluble compound(s) from fat-soluble compound(s) [31][32][33][34].
From Tables (1, 2 Figures (1 & 6), and above (Yes/ No response) observations, it can be noticed that presence of heteroatoms in chemical structures of the tested herbicides had a great influence upon CYP activity especially oxygen atom to nitrogen atom ratio. Increasing this ratio in P11, P10, P9, P8, and T1 gave repeated Yes inhibition predictions towards CYP1A2, CYP2C19, CYP2C9, and CYP3A4 enzymes.
Other tested factor in this in Silico study was total clearance which is referring to" the body capacity to eliminate the target by all needed mechanisms according to volume of blood per time (mL/minute)". So, total clearance of a molecule is pharmacokinetic key related in sum of liver or kidney elimination steps which valuable issue in drug discovery and development and xenobiotic (or its metabolic product(s)) filtration -secretion -reabsorption routes in mammals.
Physicochemical properties of target molecule and its interactions with tissue and/or other molecule(s) provide a preclinical comprehensive view of molecular deposition or biochemical pathways especially in renal section. In Silico, in vitro, and in vivo studies presents potential knowledge in understanding transportation process via both renal and non-renal systems related to molecular physicochemical properties (i.e., hydrophilic and hydrogen bonding).
This excretion character identified that these herbicides (or their metabolics) by entering mammal body including human may be removed through biliary, hepatic, and renal mechanism [35][36][37]. Most of the herbicides under online prediction process exhibited total clearance values below 0.6 ml/min/kg. P11 and T7 gave the highest values while P4 and T8 presented the lowest number (negative values) (Tables (2 and 3  OCT2 (Organic Cation Transporter 2, electrogenic transporter) is an active sodium independent transport belongs to one of three specified gene families and targets molecule in cationic form into the filtrate by selective overlapping in renal secretion process to the renal lumen driven by the negative charge of the internal site of the cell membrane of kidney cortex in human and rat [38,39,40,41]. Hydrophilic character of ionic compound(s) at physiological pH encourages net secretion by kidney after transportation process from blood and from the cell [42]. Its initial controlling steps of cationic molecule secretion is important in determining kinetics and dynamic of this molecule especially in drug-drug interaction, ligand complexity of toxin or other organic molecule through binding site and affinity character between transporter, substrate, and target molecule (endogenous or xenobiotic) [43,44]. In this study, only T6 gave Yes response to renal organic cation transporter 2 ( Figure (9  According to Chemical Abstract Service (CAS) [45] and American Chemical Society beside other global authorities, more than 200 million substances belonged to more than 100, 000 types were registered with a critical note that every two minutes, a new substance was introduced to the registration system. This note needs a huge work of toxicity -mutagenicity tests beside other characterization systems. Mutagenicity as an irreversible permanent oncogenic process may led to cancer generation even with lower exposure level (lower than threshold), permitted daily exposure (PDE), and acceptance daily intake (ADI) [46 -49].
Salmonella or Ames mutagenicity test defines as a short -term bacterial assay depending upon reverse gene mutation (damage) via various mechanisms that lead to change growth requirements in single base (or few bases) modification, insertion, deleting, or rearrangement, beside chromosome loss, rearrangement, or breakage [46].
Qualitative Structure Activity Relationship related to Ames (QSAR-Ames) prediction is an advanced study required for safety depending electrophilic and other chemical-physical characters. Also, QSAR -Ames prediction may be compared with other in vitro, in vivo, and / or other in Silico toxicological studies to get broad qualified -quantified point of view [47,50].
In QSAR related to toxicity subject, "the highest dose of the test agent given during the chronic study that can be predicted not to alter the animal" represent the maximum tolerated dose. This character is important in diagnosis and treatment of numerous diseases used stimulants, antipsychotics, antidepressant or others in early or late stages of chronic health problems like carcinogenicity or long -term treatment of liver, heart, or kidney [51].
Maximum tolerated dose of the tested herbicides was ranged from (-0.25 log mg/kg/day or 0.599 mg/kg/day belong to P8) to (1.677 log mg/kg/day or 47.6 mg/kg/day belong to T4) ( Figure 8) and (Tables 2 & 3). hERG (ether-a-go-go related gene, KCNH2, Kv11.1.2) is a human gene considered as a highly selective voltage -gated channel for potassium, expresses in brain, thymus, adrenal gland, retina, heart, and smooth muscle. ERG 2 and ERG3 appear in nervous system while ERG1 channel in the human heart controlling potential repolarization (cardiac rhythm regulation) in both activation and inactivation kinetic state [52].
Chemical(s) in early drug development phase or candidate drug related to cardio-toxicity issue can be predicated via inhibition of affinity to hERG by animal, ex vivo, or in Silico studies. In Silico method is a favorite method because it is low costly, high throughput, computerized output, with absence of experimental lab requirements providing faster cardio-toxicity estimation [53]. Here, online estimation machine model by http://biosig.unimelb.edu.au/pkcsm/ website specified unexpected results where all 30 herbicides under testing showed NO hERG I inhibition while only P1 had hERG II inhibition (Tables 2 & 3) and ( Figure 8).

EPA (U.S Environmental Protection Agency) and GHS (Globally Harmonized System of Classification and
Labelling of Chemicals) classified toxicity in term of medium lethal dose (LD50) and toxicity class where LD50 gives the required quantity of substance that kills half or the target animal (rat). This toxicity character expresses material from toxic (less than 50 mg/Kg) to safe (more than 2000 mg/Kg) [51]. Toxicity characterization may involve Oral Rat Acute Toxicity (LD50, mol/kg) and Oral Rat Chronic Toxicity (LOAEL, log mg/kg body weight/day) that tested in this paper by http://biosig.unimelb.edu.au/pkcsm/ (Tables 2 & 3) and ( Figure 2).
Lowest Observed Adverse Effect Levels (LOAEL) represents toxicity and chronic exposure with duration (52-104) weeks as long -term or (13 weeks) as short -term. In herbicide studies, there is a crucial need for in Silico foundation that may be found in food, water, soil, plant, animal and others. It is known that in Silico studies avoid legal consequences of using laboratory animals that defined by 3R (Replace, Reduce, and Refine) [54,55]. Here, online computational toxicological risk assessment was done and showed that LD50 range was (1.581(S9)-3.301(T3)) mol./Kg while LOAEL was (0.141 (T6) -3.005 (S9)) log mg/kg body weight/day (Tables 2-3, Figure 8) that highly influenced by difference in structures (Table 1) and related physical-chemical properties.
Hepatotoxicity is another important predictor of drug (or chemical) safety related to critical organ in human that is liver. This largest organ in human body support many functions such as digestion, metabolism, immunity, detoxification, storage and other functions by high efficient transporter that is dual blood supplement from both portal vein and hepatic artery. Bio-transformation (or detoxification) of any compound is depending upon formation of active site by addition of oxygen (Phase I) and conjugation with a functional group that initiated solubility in water (Phase II) by sulfonation, glucoronidation, and methylation. By combination of Phases (I &  II), fat -soluble compound (lipophilic) is converted to corresponding water -soluble (lipophobic) compound so easier to be excreted by body through urine after transportation steps (Phase III) from hepatocyte, kidney, intestine, and any other cell. Liver may be exposed to chemical (drug or any toxic material) and this dysfunction may be depending on dose or independent. With depending mode, drug -induced liver injury is predictable causing by direct toxicity at short time with reproducibility influence, inhibition of biliary efflux, production of reactive toxic metabolics, protein modification, T & B cell cytotoxicity, and mitochondrial impairment [56].
Exogenous or endogenous, hormone, or any molecule considered as foreign compound reabsorbed by intestinal or renal tubular forward excretion after counted metabolism steps that converted target molecule to its metabolic(s) with lower toxicological activity. With these processes: metabolism, detoxification, and excretion, excess accumulation of any toxin is controlled to prevent multistep induced liver injury (acute or chronic liver failure) or liver cancer in advanced stages.
Human susceptibility to hepatotoxic material is depending upon: gender, age, nutrition, chemical composition, concentration, obesity, life style, genetic base, duration, and disease(s) especially immunodeficiency, hepatitis C, and others. Selectivity -toxicity relationship of chemical is mainly caused by presence of special group(s) that may interact with nuclear receptor(s) by enzymatic step. For example, acetaminophen (known as Panadol, acetamidic phenol) and diclofenac (known as Voltaren, sodium carboxylate salt of secondary aromatic amine) have acute hepatocellular toxicity [57,58].
Skin as an integumentary organ can be defined as ''protective barrier or guard against external environmental effect(s) to protect internal bones, muscles, and organs composed of multi-layers (epidermis and dermis)''. These external environmental effects such as temperature, moister, natural or synthesized chemicals varied in their influence on skin which is skin sensations [59].
Chemicals under research, development, maintaining subjects target healthy skin and whole human body and avoiding anyone of all four types of allergy response (or hypersensitivity). Hypersensitivity type classified as type I: cutaneous test reaction, type II: antibody and related cytotoxicity, type III: antigen-antibody complex, and type IV: delayed response. So, skin sensitization is an allergic contact dermatitis of type IV occurred after more than 48 hours [60].
Sensitization symptoms may be appeared as rash, blister, and / or swollen that may be in few hours to lifetime or threat human life. Chemical reactions between any foreign material attached skin or air exposure activation, bacterial degradation, photo-activation are covalent formation with skin protein, forming adduct antigen, or / and free radical with / without electrophile -nucleophile interaction. Covalent bond formation with skin protein can be happened by nucleophilic thiol-or amine-amino acid in protein structure [61].
Changing in cell surface is a direct method in testing chemicals especially commercial (natural or synthesized) products such as pharmaceuticals and cosmeceuticals. These changing may be also tested through animals and non-animal (in Silico) models towards final steps: human testing.
In Silico skin sensitization has a great attention after Europe banning of animal testing since 2013 beside carcinogenicity, reproductive toxicity, and related kinetic toxicological studies.
These prediction models are online data source and commercial packed models subjected to computer automated structure evaluation (CASE) that depend on structural subunits having activity termed as biophore(s). These models are not limited to chemicals with direct sensitization, they also may have evaluated products of direct and enzymatic metabolic transformation including reduction, oxidation, hydrolysis… etc [62]. In this online in Silico testing model, V, T3, T6, S2, S4, P3, P5, and P9 showed skin sensitization with Yes response where most of them have amide (or thioamide) and chloro-groups ( Figures (1, 6, & 9) and (Tables (1, 2

, & 3).
Toxicity of chemicals whether used as herbicides or not used is an essential issue in safety classification and monitoring their effects on Food series (killing, inhibition, management, security, production, designing, availability, and development).
So, safety -efficacy balance is a main concern in herbicide science, health, technology, and environment. Time, cost, environmental limiting, and biological -agricultural testing of this balance as well as selectivity, action, and classification of any chemical (or herbicide) were shorten by computational models (QSAR) [63].
Recent scientific articles evaluated safety -efficacy of huge number of chemicals including crops, honey, and herbicidal activity incorporation with chemical -physical properties in high accurate identification. Online machine learning represents fast, easy, accurate prediction methods and provides scientists and drugdevelopers ADMET information in time of several minutes or less [64][65][66][67][68].
In general, Yes/ No response of herbicides towards Herbicidal activity with honey bee and avian toxicity that predicted by [13] website were summarized as below (Table 4):  Herbicidal activity -No response: T3, T7, and P10.  Honey bee toxicity -Yes response: V and S2.  Avian toxicity -Yes response: only T2.
So, most of the tested herbicides have herbicidal activity with minimum toxic effects on bees and avian. For example, Vernolate (V) is safe to avian and can be used as an herbicide but toxic to honey bees.
Cancer is considered as a big challenge to scientific and medicinal regimes around the world. It needs effective and safe chemicals for treatment and diagnosis that required multiple steps of synthesis (or extraction), characterization, and practical in vitro, in vivo, exo vivo methods as well as time -cost consuming. QSAR strategy is a primary gate of exploring and development candidate drug and may short these multiple steps besides giving a primary image of toxicity, resistance to chemotherapy, or/ and side effects with reducing time and cost [69,70]. Many published articles presented machine learning throughput screen studies of anticancer materials at high level of accuracy. One of these QSAR computational approaches is online screen websites targeting pharmacokinetic or pharmacodynamics properties of cancer cell lines based upon structural features [71,72,73,74]. In this study, cancer cell line activity of CNS, Leukaemia, ovarian, and prostate were evaluated by [14] (Tables 5-7, Figure 4, Figures (10-12)) where summary of these cell lines were as below:  Active response of all tested cell lines: T1, T3, P1, P8, and P10.  CNS: all were less than 5. These herbicides under prediction testing were characterized as (Table 8), [75]:  Harmful if swallowed: V, T2, T3, T4, T6, T7, T8, T9, S5, P1, and P1.  Causes damage to organs through prolonged or repeated exposure: T4, T6, and S3.  May cause an allergic skin reaction: T4, S9, P8, and P10.  Toxic to aquatic life with long lasting effects: V.  Moderately toxic, classified by WHO as unlikely to be hazard as in Class U toxic, causing irritation to eye, skin, and respiratory tract: S4.  Very toxic to aquatic life with long lasting effects: T3, T4, S3, S5, and P1.  Causes serious eye irritation: T4, S9, P8, and P10.  May irritate eyes, nose, throat, and skin: S7.  May cause an allergic skin reaction: T1, T4, T6, and S3.  May cause skin and eye irritation: S6.  May also cause liver and kidney damage, coma or convulsions: S6.  Did not induce unscheduled DNA synthesis in human cells: T6.  Epidemiological study did not find significant changes in neurological function after high exposure: T6.  Suggestive Evidence of Carcinogenicity, but Not Sufficient to Assess Human Carcinogenic Potential: T7.  Likely to be Carcinogenic to Humans: T7 and P5.  Not likely to be Carcinogenic to Humans: P4.  Evidence in humans: No adequate data. Evidence in animals: Sufficient evidence. Overall summary evaluation of carcinogenic risk to humans is Group 2B: The agent is possibly carcinogenic to humans: S2.  Cancer Classification: Group C Possible Human Carcinogen: P11.  Group D Not Classifiable as to Human Carcinogenicity: T9 and P7.  Group E Evidence of Non-Carcinogenicity for Humans: T8, S1, and P6.  Not classifiable as to its carcinogenicity to humans (Group 3): T1.
Here, online prediction of Tridiphane (T3) showed extraordinary activity response to Leukaemia, Ovarian, and Prostate cell lines with more than 5 while its activity towards CNS lines was less than 5 ( Figures (10 -12), Tables (5 -7)). Scientific literature foundations confirmed that human swallowing of T3 gives a harmful effect and it is very toxic to aquatic life lasting for long time (Table 8) [75].

Conclusion
Thirty herbicides were chosen and subjected to online website activity prediction belong to pkCSM -University of Melbourne -Australia as a QSAR prediction against honey bee, avian, and multiple human Leukemia, CNS, Ovarian, Prostate Cancer cell lines. Another calculated predictor of ADMET were Caco-2 permeability, human Intestinal absorption, Skin Permeability, P-glycoprotein substrate and inhibitor, BBB permeability, CNS permeability, Cytochromes P450 inhibition, Total Clearance, Renal OCT2 substrate, Ames, Human Maximum Tolerated Dose, hERG inhibition, Oral Rat Acute Toxicity, Oral Rat Chronic Toxicity, Hepatotoxicity, and Skin Sensitisation. These phytotoxins varied in their response to each calculated predictor. Sulfentrazone (S1) is a known herbicide with evidence of human non-carcinogenic character (Group E). Its prediction showed hepatotoxicity with Leukemia, CNS, Ovarian, Prostate Cancer cell lines activities less than 5. Also, it works as CYP1A2 inhibitor with negative inhibition response to p-glycoprotein, Ames test, skin sensitization, renal OCT2, and hERG. Also, S1 had the low intestinal absorption (89.659%) and lowest BBB compared to the other herbicides under testing. So, Sulfentrazone (S1) is more structurally safe, however, its active -multiple toxicological -cellular interactions must be under considerations. In comparison, Simetryn (S5) and Simazine (S6) have the same core structure except (-SCH3) group replaced with chloro group (or chemical formula: C8H15N5S and C7H12ClN5 respectively). Both S5 and S6 gave semi identical results in Caco-2, intestinal absorption, skin permeability, CNS, max. tolerated dose, oral rat acute (LD50) with the same response to herbicide activity, honey bee and avian, p-glycoprotein, cytochrome P450, renal OCT2, Ames, hERG I and II, hepatotoxicity, skin sensitization, and general anticancer activity as an inactive material. In reverse, BBB, total clearance, and oral rat chronic (LOAEL) of S5 and S6 were in different values especially LOAEL. These results briefly confirm chemical structure influences upon prediction outcomes.