Experimental Toxicology Services (ETS) Nederland BV, Frankensteeg 4, 7201KN Zutphen, The Netherlands
Accepted date: February 26, 2017
Expenditure for discovery and development of a new crop protection product is now approaching the $ 300 million mark, while at the same time underpinning information gaps in environmental safety assessment. Large information gaps also exist for the safety of a vast number of existing chemicals in commerce. Addressing safety information gaps of new and existing chemicals with increased conventional toxicity testing would skyrocket costs and animal usage and contravene the 3Rs principles to refine, reduce, and replace animal testing. There is a great need to improve predictive safety assessment with minimum animal use. Recent progress in the understanding of the molecular biology of chemical dose response relationships, which has confirmed general applicability of the Druckrey-Küpfmüller theorem for receptor-mediated toxicity, can now be employed at an early stage of product development to identify (and eliminate) hazardous chemicals using short-term chronic dose response tests in invertebrates. Time-to-effect (TTE) approaches can identify hazardous chemicals with time-cumulative toxicity. This approach would shift research and development programs to chemicals with ascertained strictly dosedependent toxicity, for which no-observable-adverse-effect levels (NOAELs) can be reliably established with short-term tests, which could make a myriad of resource-intensive animal studies obsolete.
Toxicity, Chronic, Predictive Safety Assessment, Hazardous Chemicals, 3Rs principles.
A recent survey of five leading crop protection companies  revealed that the expenditure over the last 20 years for discovery and development of a new crop protection product has nearly doubled to $ 289 million, mostly due to markedly increased expenditure for environmental chemistry studies, field trials and regulatory affairs (Table 1), which reflects a tremendous rise in environmental safety data required by regulatory bodies. The average lead time from first synthesis to commercial introduction is now more than 11 years, and the biology of an average 160,000 newly synthesized molecules is screened for the registration of one new crop protection product .
|Toxicology & Environmental Chemistry||10||7||0.70|
Table 1. Discovery and development costs (in million us dollars) of a new crop protection product, according to a survey of 5 leading crop protection companies .
Within the same time period, new products such as the neonicotinoid insecticides, which were developed according to latest regulatory standards, have become highly controversial because risks to beneficial non-target organisms were underestimated [2-4], underpinning information gaps in environmental safety assessment. There are additional large information gaps on the safety of a vast number of existing chemicals in commerce , and removing hazardous chemicals from the market and replacing them with safer alternatives has become a key objective of EU REACH chemical management regulation  and reform of the Toxic Substances Control Act in the United States . Addressing safety information gaps of new and existing chemicals with increased conventional toxicity testing would skyrocket costs and animal usage and contravene the 3Rs principles to refine, reduce, and replace animal testing . The chemical industry has a vital interest in predictive safety assessment rather than extensive and resource intensive toxicology programs and there has been a steady increase in development and use of in vitro alternatives to animal testing by industry over the last 30 years . A paradigm shift is emerging in which computational approaches, systems biology, high-throughput in vitro toxicity assays, and high-throughput exposure assessments are beginning to be applied to mechanismbased risk assessments in a time- and resource-efficient fashion . Structure-based (quantitative structure–activity relationship [QSAR]) prediction approaches were among the first developed and most commonly used cheminformatics tools for safety assessment , although there are concerns about the reliability of their predictions .
It seems possible to further improve predictive safety assessment with minimum animal use following recent progress in the understanding of the molecular biology of chemical dose response relationships , which can be employed at an early stage of product development to identify and eliminate hazardous chemicals with short-term dose response tests. This approach would shift research and development programs to chemicals with ascertained strictly dose-dependent toxicity, for which no-observable-adverse-effect levels (NOAELs) can be reliably established with short-term tests, which could make a myriad of resource-intensive animal studies obsolete.
The Renaissance physician Paracelsus (1493–1541), often regarded as the ‘Father of Toxicology’, laid the groundwork for chemical risk assessment when he coined his dictum, ‘What is there that is not poison? All things are poison and nothing is without poison. Solely the dose determines that a thing is not a poison’. Paracelsus inferred that lower doses – below a threshold – could cause otherwise poisonous substances to become harmless . Paracelsus paved the way for the threshold concept and the no-adverse effect level  for risk assessment. The notable exception is the linear non-threshold (LNT) dose-response model widely adopted for carcinogenic agents. Carcinogenic risk assessment originates from the assertion of Hermann J. Muller (in his Nobel Prize Lecture of December 12, 1946) that the dose-response relationship for radiation-induced mutations is linear  ruling out a threshold dose. Muller’s statement received first recognition by the Genetics Panel of the U.S. National Academy of Sciences’ Committee on Biological Effects of Atomic Radiation (BEAR) in 1956 , and was subsequently endorsed by leading regulatory authorities for estimates of the cancer risk from radioactive fallout  and genotoxic carcinogens [19,20]. However, the LNT doseresponse model has been challenged by several authors who hypothesized potential thresholds and protective mechanisms throughout the process from initial DNA damage induction to tumor formation [21-26]. Chemical risk assessment was compounded even further by recent observations that some nongenotoxic chemicals show dose-response relationships identical to that of nitrosamines, arguably the most potent carcinogens known [27-31]. The dose response is described in quantitative terms by what is known as the Druckrey-Küpfmüller equation:
d t n =constant (1)
where d=daily dose and t=exposure Time-to-effect, and n ≥ 1.
Adverse effects of genotoxic carcinogens and several hazardous non-carcinogens are dependent not only on exposure levels but also on exposure duration. The essence of equation (1) is that the total dose required to produce an adverse effect is much lower at low exposure levels even tough exposure times needed to produce the effect are much higher. From a mechanistic point of view, the common denominator of the dose-response relationship is irreversibility of receptor binding and irreversibility of the associated effect .
Shortly after the Second World War, two reputable German scientists, the pharmacologist and cancer researcher Hermann Druckrey, and Karl Küpfmuller, a mathematically adept electrical and communications engineer, while detained in an allied internment camp, developed groundbreaking theoretical knowledge about the relationship between drug dosage and tissue response in pharmacology and toxicology in general, and the action of carcinogenic substances in particular . They hypothesized, with theoretical approaches to reaction kinetics, that irreversible receptor binding with an associated irreversible effect would lead to reinforcement of the effect by exposure time (Table 2) . Many years later, the carcinogenicity of nitrosamines [30,31] was indeed shown to be a result of irreversible receptor binding (alkylation of DNA) associated with irreversible effects (gene mutations) [34,35]. Essentially similar receptor-mediated mechanisms of toxic action leading to time-cumulative toxicity have now been validated for a considerable number of non-carcinogens , indicating that the Druckrey-Küpfmüller theorem is generally applicable for chemical toxicity.
|Reversibility of receptor binding||Receptor binding in relation to compound concentration||Reversibility of the effect||Effect in relation to receptor binding||Effect in relation to compound concentration||Dose-response characteristics|
|TR → 0||CR ~ C||Tr → 0
Tr → ∞
|E ~ CR
E ~ ∫CR dt
|E ~ C
E ~ ∫ C dt
Ct = constant*
|TR → ∞||CR ~ ∫ Cdt||Tr→ 0
Tr → ∞
|E ~ CR
E ~ ∫CR dt
|E ~ ∫ C dt
E ~ ∫ ∫ C dt
|Ct = constant
Reinforced by time
TR is the time constant for the reversibility of receptor binding
Tr is the time constant for the reversibility of the effect
CR is the concentration of bound receptors
C is the concentration of the poison at the site of interaction
*known as Haber’s Rule (the product of concentration and time produces a constant effect). These dose response relationships may occur with irreversible receptor binding
or irreversible effect when C is practically constant over time
Table 2: Dose-response characteristics according to Druckrey and Küpfmüller .
This evidence has major implications for chemical risk assessment. The Paracelsus paradigm and threshold model cannot be upheld for chemicals with time-dependent toxicity, and is valid only for reversible effects resulting from reversible receptor binding (Table 2), when effects are indeed entirely dose-dependent. An additional implication is that chemical risk assessment requires quantitative analysis of the dose-response relationship to discriminate between strictly dose-dependent toxicity and dose- and time-dependent toxicity.
Shift of research and development to chemicals with dose-dependent toxicity
The linear non-threshold (LNT) dose-response model and the ALARA principle (“as low as reasonably achievable”), currently used in risk assessment and management of nitrosamines, may have to be extended to all chemicals with time-cumulative toxicity , because risks of long-term exposure to low concentrations of such chemicals have been seriously underestimated. In fact, equation (1) suggests a threshold dose may not exist for chemicals with time-cumulative toxicity. The neonicotinoids are a case in point. Bayer scientists demonstrated that the neonicotinoid insecticide imidacloprid blocks nicotinic acetylcholine (nACh) receptors in the central nervous system of insects , leading to irreversible neuronal damage , underpinning the Druckrey-Küpfmüller theorem [32,33]. Unlike the normal neurotransmitter acetylcholine, acetylcholinesterase cannot remove imidacloprid from the nACh receptor. Although irreversible receptor binding has recently been retracted by Bayer experts  in response to the discovery of time-cumulative toxicity of imidacloprid and thiacloprid to arthropods , it can hardly be disputed that dissociation, if it occurs at all, is bound to be very slow, and that cumulative nACh receptor binding leading to irreversible neuronal toxicity can be easily envisaged . Timecumulative toxicity to invertebrates has been demonstrated for the neonicotinoids imidacloprid, thiacloprid, thiamethoxam and clothianidin [13,27,28,41-43], which, due to their environmental properties, threatens the survival of invertebrates. They are prone to leach from soils , and have been demonstrated to contaminate surface water in Europe and North America [3,44]. In the Netherlands, surface water contamination with imidacloprid has been demonstrated to correlate with decline of macro-invertebrates  and insectivorous birds [44,46], and entomological surveys in Dutch and German nature reserves have revealed a staggering decline of ground beetles and flying insects since the introduction of imidacloprid in agriculture in the 1990s [44,47]. The assumption of thresholds of toxicity for neonicotinoids was a serious error of judgement and has resulted in disastrous insect decline with knock on effects on all insectivores, which could have been prevented if the ALARA principle would have been applied to time-cumulative toxicity.
Dose: Time-to-effect analyses to identify hazardous chemicals
The corollary of described dose response relationships is that repeated dose toxicity testing must first and foremost focus on establishing dose: Time-to-effect relationships and eliminate hazardous chemicals with time-cumulative toxicity at an early stage of development. Traditional repeated dose toxicity tests usually involve the administration of 3 dose levels of test chemicals to animals, which then are observed for adverse effects. Adverse effects are expected to be observed at high dose, while mid-dose and low-dose level are expected to provide the lowest-observed-adverse-effect level (LOAEL) and the NOAEL (no-observed adverse effect level).
This approach to toxicity testing is to consider dose (concentration): effect relationships at arbitrarily fixed exposure durations, which are supposed to reflect ‘acute’ or ‘chronic’ time scales, and measures the proportion of all exposed individuals responding with adverse effects by the end of such exposure times. This is valid when toxicity is mainly dependent on exposure concentrations (Paracelsus), but it is insufficient when toxic effects are influenced by exposure time, because the impact of low exposure concentrations may be underestimated if the duration of the experiment is shorter than the latent period for toxicity. Toxicological databases established in this way are collections of endpoint values obtained at fixed times of exposure. As such these values cannot be linked to make predictions for the wide range of exposures encountered by humans or in the environment. By contrast, Time-to-effect (TTE) approaches provide more information on the exposure concentrations and times needed to produce toxic effects on tested organisms. Indeed, TTE bioassays differ from standard chronic toxicity tests in that TTE approaches record effects of at least 5 dose levels at consecutive times during the exposure, so the data form a matrix that can be analysed to extract information about the effective concentrations (e.g. NEC, EC10, LC50, etc.) or about the Time-to-effect for a given endpoint (e.g. t50). This is an essential requirement to detect chemicals showing timedependent toxicity, and it allows prediction of toxic effects for any combination of concentration and time found in the environment.
The OECD is committed to the implementation of the 3R-principles, Replacement, Reduction and Refinement, as first laid down by Russel & Burch in 1959 , in their “The Principles of Humane Experimental Technique”. Since the adoption in 1981 of the first set of Test Guidelines, many of the short-, and long-term toxicity tests have been developed or revised to introduce aspects of the 3R-principles . A considerable number of guidelines for in vitro assays have been developed to predict diverse toxicological endpoints (Table 3) which not only provide in vitro alternatives to the notorious Lethal Dose-50% (LD50) and Draize tests that inflict severe suffering on animals , but may also identify carcinogens, mutagens, and endocrine disruptors. The industry’s interest in in vitro assays has arisen from the need to support the early identification of promising new molecules but also through legislation requiring adherence to the 3Rs .
|Toxicity endpoints||Assay||OECD or EPA Guideline(s)|
|Eye corrosion or irritation||Bovine corneal opacity and permeability test||OECD 437|
|Isolated chicken eye test||OECD 438|
|Fluorescein leakage test
Reconstructed human Cornea-like Epithelium (RhCE)
|Skin corrosion or irritation||In vitro membrane barrier test||OECD 435|
|Reconstructed human epidermis (RHE) test||OECD 431, 439|
|Transcutaneous electrical resistance test||OECD 430|
|Acute toxicity to fish||Zebrafish embryo acute toxicity test
|Skin sensitisation||Human Cell Line Activation Test (h-CLAT)
|Endocrine disruption||BG1Luc estrogen receptor transactivation test||OECD 457|
|H295R steroidogenesis assay||OECD 456|
|Amphibian metamorphosis assay
Stably Transfected Transactivation to detect Androgen Receptor Agonists and Antagonists
|Genetic toxicity||Bacterial reverse mutation test||OECD 471|
|In vitro mammalian chromosomal aberration test||OECD 473|
|In vitro mammalian cell gene mutation test||OECD 476|
|In vitro mammalian cell micronucleus test||OECD 487|
|Skin absorption||Skin absorption in vitro
|Phototoxicity||3T3 neutral red uptake phototoxicity test
|Chronic toxicity||Sediment-water chironomid toxicity (spiked sediment)||OECD 218|
|Sediment-water chironomid toxicity (spiked water)||OECD 219|
|Sediment-water chironomid life-cycle toxicity test||OECD 233|
|Daphnid chronic toxicity test||OPPTS 850.1300|
|NOAEL determination||Combined Repeated Dose Toxicity Study with the Reproduction/Developmental Toxicity Screening Test||OECD 422|
Carcinogenesis and mutagenesis are of special concern in product development, and in vitro genotoxicity studies (Table 3) are usually conducted at a very early stage. Now that the Druckrey-Küpfmüller theorem has been shown to be generally applicable, dose: Time-to-effect analyses can identify similarly hazardous non-carcinogens in early development, and improve hazard identification. Time-dependent toxicity can be conveniently investigated with short-term chronic toxicity tests in invertebrates [48,51]. Daphnia magna appears to be an excellent model because its sensitivity to lethal toxicity is very similar to mammals and humans [52,53]. Moreover, if short-term assays demonstrate absence of mutagenic potential and timedependent toxicity, the Druckrey-Küpfmüller theorem (Table 2) indicates that biological effects of a candidate product are bound to be strictly dose-dependent, which would also rule out carcinogenic potential (invariably a dose- and time-dependent process). Safe exposure levels can be extracted from short-term repeated dose experiments, such as the Combined Repeated Dose Toxicity Study with the Reproduction/Developmental Toxicity Screening Test adopted by the OECD on 22 March 1996 (Table 3) . A myriad of resource-intensive animal studies would become obsolete, simplifying product development programs.