Reviewing the thoroughness of human safety testing for succinate dehydrogenase inhibitors: fluopyram as a case study

Abstract

It has been suggested that fungicides which are succinate dehydrogenase inhibitors also inhibit this enzyme in humans, with potentially serious health consequences. It has also been suggested that regulatory studies conducted on these fungicides have missed succinate dehydrogenase inhibition and its effects. Using fluopyram as a case study, this paper addresses these suggestions by critically examining all the studies related to human safety. The mechanism causing each toxicity is considered, including the possibility that inhibition of succinate dehydrogenase might be the root cause. If fluopyram were to inhibit succinate dehydrogenase, then the consequent effects and their likelihood of detection are evaluated. Many fluopyram toxicity studies are completely untargeted, with a huge number of endpoints examined, any one of which might reveal an important toxic effect. Rat and mouse studies from a single dose to near lifetime exposure are included, whilst dog studies add biological coverage. The consistent pattern seen is that the primary toxicities caused by fluopyram are effects on the liver and thyroid, plus a kidney effect in ageing male rats. The mechanisms causing these effects have been proven in bespoke mechanistic studies—they are not relevant to humans and do not involve the inhibition of succinate dehydrogenase. If fluopyram were to inhibit succinate dehydrogenase then this would be apparent, as the tissues most likely to be affected—the nervous system, heart and muscle—were examined microscopically in numerous studies with no effects seen. The low affinity of fluopyram for mammalian succinate dehydrogenase and its rapid metabolism explain the results.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00204-025-04279-7.

Introduction

It is estimated that fungal pathogens are responsible for crop losses of up to 30% worldwide; fungicides are therefore a critical tool for farmers to control crop diseases. One fungicidal mode of action is inhibition of the enzyme succinate dehydrogenase (SDH; complex II of the electron transport chain), which plays a key role in mitochondrial respiration.

Bénit et al. (2019) have suggested that the SDH inhibiting fungicides inhibit SDH not only in fungi, but also in other species such as bees, earthworms and mammals, including humans. The same authors also propose that there may be people who are particularly sensitive to SDH inhibitors, for example because they have genetic defects in the activity of SDH, or because they are hypersensitive to oxidative insults, resulting in neurological and neurodegenerative diseases and cancer. An expert working group set up by the authorities in France for regulating pesticides reviewed this information, and concluded that it did not provide any evidence to support a health alert for humans and the environment (ANSES 2019). Nevertheless, the potential toxicity of SDH inhibitors to man is the subject of subsequent papers (Hospital et al 2023; Bouillard 2023). There are also four draft Adverse Outcome Pathways (AOP) on the OECD AOP Wiki that link the inhibition of SDH to toxic outcomes (OECD AOP #457, #474, #534 & #546—OECD 2025). In contrast, there is also interest in the potential of SDH inhibition to treat cancer (Kluckova, 2013), though the investigations have mainly been in vitro and in animals so far. It has been proposed that the human safety studies conducted on SDHi fungicides as a part of the regulatory process have missed the specific impact of these molecules on SDH and its consequences in terms of adverse effects. In other words, it is suggested that these studies lack thoroughness and are not fit for purpose. This paper will critically examine these suggestions, specifically focussing on mammalian SDH and on human health using fluopyram as a case study.

Bayer Crop Science invented fluopyram as a novel SDH inhibitor, which is effective against fungal species and nematodes. Fluopyram was chosen for development based on its biological efficacy and safety profile. After years of testing, including the conduct of almost 3000 safety studies (addressing human and environmental safety as well as metabolism), dossiers of the data and required analyses were submitted to regulatory authorities. Authorities must be satisfied that all necessary studies have been performed, and that risk assessments for humans and the environment are acceptable before registrations are granted that enable sales to start (e.g. for the EU see EC 2009 & EFSA 2021).

The assessment of human safety for crop protection products is a highly technical subject and is little known outside of this narrow domain. Using fluopyram as a case example, this paper describes the human safety studies that are routinely conducted as well as those done to address specific findings, in a way that is accessible to the general scientist and addresses the implication from the public literature that the work done lacks thoroughness, is incomplete and not fit for purpose. The results of the fluopyram human safety studies and the human risk assessment will be described. Then the following questions will be addressed:

• Could any of the toxicities seen in these studies be caused by the inhibition of SDH?

• Could toxicities caused by the inhibition of SDH have been missed?

Fluopyram human safety studies

A list of mandated studies must be carried out (Nganga et al. 2018), which are designed to identify health hazards as well as the dose levels at which they occur and to determine the amount of the chemical that can be tolerated without causing adverse effects. These studies are performed in compliance with internationally approved test guidelines (OECD 2024), most of which involve the use of animals. A list of commonly mandated test guidelines can be found in Table 1 of the Online Resource. These studies must take place in facilities which are audited by government bodies to ensure they comply with animal welfare legislation and operate to the principles of Good Laboratory Practice (GLP). GLP is an auditing and record keeping procedure which ensures that all procedures and results are accurately recorded and reported, to guarantee trustworthiness (WHO 2009). Summaries of fluopyram safety studies are available at https://www.bayer.com/en/agriculture/safety-results-crop-protection-products, whilst full study reports can be requested using the form here https://www.bayer.com/en/agriculture/safety-study-report-request-forms.

A literature search was performed to find any published papers concerning the mammalian toxicity of the fluopyram active ingredient. The few papers that were found will be referred to in the relevant section, alongside the studies performed by or on behalf of Bayer Crop Science.

General toxicity studies

There are typically eight groups of animals in a study, four groups of males and four of females: one group of each gender is an undosed control group, whilst the other three of each gender are given different dose levels of fluopyram. A large number of endpoints are examined, to see if dosing with fluopyram has an effect when compared to the undosed control. Some of these endpoints are observed during the study, but most can only be examined at the end of the study. The lowest dose level at which fluopyram has an adverse effect in a given study is called the LOAEL (lowest observed adverse effect level), and the next lowest dose level, which is important for risk assessment, is called the NOAEL (no observed adverse effect level). Any one of the large number of endpoints examined might determine the LOAEL, so in this sense the studies are untargeted. The endpoints examined in the fluopyram combined 12 month and two-year rat study are shown in Table 1 as an example. Sixty animals per sex were used for each of three dose levels plus an undosed control for part of the study with two-years of exposure, plus an additional ten animals per sex per group for the 12-month exposure, resulting in the use of 560 animals in total. The final report of this study was 4088 pages long and contained ~ 1 million raw data points, which were compiled into summary tables, figures, pathologist reports and statistical analyses. In short, it is one of the most complex and data-rich study types conducted in toxicology.

Table 1.

Endpoints measured in the fluopyram combined 12 month and two-year toxicity study in the rat^a

Endpoint category	Specific endpoints and details
Mortality	Checked twice daily (once at weekends and on public holidays)
Clinical signs	Observed for physical or behavioural abnormalities at least once daily
Physical examination	Detailed physical examination, including feeling for lumps, performed weekly
Ophthalmoscopy	Detailed eye examination before start of study and after 1 and 2 years
Food consumption	Measured twice weekly for first 6 weeks, then weekly until week 13, then every 4 weeks
Body weight	Weighed weekly for first 13 weeks then every 4 weeks
Blood properties	Measured for 10 animals/group at weeks 12/13, 25/26, 51, 78 and 103/104
- Haematology	11 blood properties measured for 10 animals/group at weeks 12/13, 25/26, 51, 78 and 103/104
- Clinical chemistry	25 blood biochemical concentration measurements (6 electrolytes, 9 enzymes and 10 other biochemicals)
Urine properties	12 properties determined for 10 animals/group at weeks 12/13, 24/25, 52/53, 77 and 104
Gross pathology	Immediately after humane termination, examined in detail externally and internally for any visible abnormalities
Organ weights	14 organs weighed
Histopathology	49 different tissue samplesb plus any gross abnormalities were taken, fixed, sectioned onto glass slides, stained and examined under a microscope for abnormalities by two pathologists

^a All endpoints were determined for every possible animal unless otherwise stated; fewer endpoints were examined for animals dying during the course of the study

^b List of the tissue samples is given in Table 2 of the Online Resource

The studies in this section are conducted over a wide range of durations, as some effects can be observed after a single dose, whilst others, such as tumours, can only be seen in longer studies. Significant consideration goes into setting dose levels. In the case of fluopyram, the traditional concept of the maximum tolerated dose (MTD) was used to help decide the highest dose level in a study. At the MTD there should be significant toxicity, so that the effects seen are a worst case, but the dose should not be so high that the animals die or have to be euthanised due to excessive toxicity. Setting the top dose is both a scientific and an animal welfare consideration. The shorter studies are conducted first, and the doses at which adverse effects are seen in those studies then inform the doses to be used for the longer studies. Because toxicity generally increases when animals are exposed to a test compound for a longer period, the doses used for the longest studies are lower than those for shortest studies.

Except for some single-dose studies, all the studies in this section include groups of both genders, to examine the possibility of gender-specific toxicities. For shorter duration studies, fewer endpoints are examined than shown in Table 1 for the two-year rat study.

The general toxicity studies conducted for fluopyram are shown in Table 2, which displays the route of exposure to the animals and the number of animals used of each gender. There are 15 general toxicity studies, with durations ranging from a single dose to 18 months and two years (mouse and rat lifetime, respectively) of daily dosing. The total number of animals used in these studies was 1540.

Table 2.

General toxicity studies for fluopyram showing species, duration, dosing route and the number of animals of each gender used

Species	Once	Duration of dosing
		28 days	90 days	1 year	Longer
Rat	oral (6♀) dermal (5♂ + 5♀) inhalation (10♂ + 10♀)	dietary (20♂ + 20♀) dermal (40♂ + 40♀)	dietary (70♂ + 70♀)	dietary (40♂ + 40♀)	2 year dietary (240♂ + 240♀)
Mouse		dietary (20♂ + 20♀)	dietary (40♂ + 40♀)	dietary (40♂ + 40♀)	18 month dietary (200♂ + 200♀)
Dog		oral gavage (8♂ + 8♀)a	dietary (16♂ + 16♀)	dietary (16♂ + 16♀)

^a This was preceded by a dose range finding study (2♂ + 2♀)

These general toxicity studies generate a huge amount of data. The minimum number of datapoints recorded for a typical two-year rat study is approximately 500 for each of 400 animals, so 200,000 in total. Statistical tests compare each dose level with the untreated control for each endpoint, resulting in approximately 1500 statistical tests. Potential effects are reviewed to distinguish those that are treatment-related and those that are chance, and to evaluate if the treatment-related effects are adverse or not adverse (Lewis et al. 2002). The minimum number of datapoints for some common longer-term human safety studies is shown in Table 3 in the Online Resource. The study report thoroughly documents the conduct and results of the studies and provides various summaries of the results to aid interpretation. Key findings at the NOAEL and LOAEL for the fluopyram general toxicity studies are shown in Table 3.

Table 3.

Key findings from fluopyram general toxicity studies with NOAELs proposed by Bayer Crop Science

Species	Duration of dosing	No observed adverse effect level (NOAEL)		Lowest observed adverse effect level (LOAEL)
		Dose levela	Treatment effectsb	Dose levela	Treatment effectsb
Rat	Once	2000 mg/kg	No effects seen at limit dose for oral and dermal exposure routes
	Once	5112 mg/m3	No significant toxicity for the inhalation exposure route
	28 d	31.0 M 36.1 F	Adaptive liver changes; hyaline droplets in kidney (males)	254 M 263 F	Reduced bodyweight gain; haematology & clinical chemistry changes; adaptive changes; thyroid weight increase & hypertrophy (males); hyaline droplets in kidney (males)
	90 d	12.5 M 14.6 F	Adaptive liver changes (males); cellular casts in urine and hyaline droplets in kidney (malesc)	60.5 M 70.1 F	Haematology changes (males); clinical chemistry changes; cellular casts in urine and hyaline droplets in kidney (males); liver weight increase; kidney weight increase (males); liver & thyroid histopathology (males & females), kidney histopathology (males)
	1 yr	1.37 M 1.88 F	None	6.9 M 9.6 F	Adaptive liver changes (males); chronic progressive nephropathy (kidney, males); thyroid hypertrophy (males)
	2 yr	1.20 M 1.68 F	None	6.0 M 8.6 F	Adaptive liver changes (males); chronic progressive nephropathy with tubular hypertrophy and dilatation (kidney, males); thyroid hypertrophy (males); thyroid colloid alteration (females); eye effects (males)
Mouse	28 d	24.7 M 31.1 F	Adaptive liver changes	162 M 197 F	Adrenal hypertrophy (females); liver hypertrophy & necrosis & weight increase (adaptive)
	90 d	26.6 M 32.0 F	Adaptive liver changes; reduced cholesterol (males)	188 M 216 F	Increased food consumption (males); clinical chemistry effects; adrenal histopathology; adrenal weight increase (males); liver hypertrophy, focal necrosis & weight increase
	1 yr	4.4 M 5.7 F	None	22.2 M 28.6 F	Increased liver weight; thyroid hyperplasia (males)
	1.5 yr	4.2 M 5.3 F	None	20.9 M 26.8 F	Liver hypertrophy and weight increase; liver single cell degeneration (males); thyroid hyperplasia (males)
Dog	28 d	150	None	750	Haematology changes (males); clinical chemistry changes; adaptive liver changes
	90 d	28.5 M 32.9 F	Reduced bilirubin (males); adaptive liver changes (males); slight thymus effect (males); no adverse effects	171 M 184 F	Slight food consumption and body weight effects; clinical chemistry changes; liver hypertrophy and other histopathology & weight increase; slight thymus effect due to lower body weight
	1 yr	13.2 M 14.4 F	None	67.6 M 66.1 F	Food consumption and body weight effect (due to palatability, esp. females); liver enzyme effects; adaptive liver changes (males)

^a units are mg of fluopyram per kg bodyweight per day unless otherwise stated; for studies with dietary dosing, achieved doses are given separately for males (M) and females (F); some studies used dose levels higher than the LOAEL, for which results are shown in Online Resource Table 4

^b effects were seen in both genders unless otherwise stated

^c cellular casts in urine were also seen in male rats at 3.06 mg/kg/day, but are not relevant to human risk

In several studies there was at least one dose level higher than the one determined to be the LOAEL. Effects at these higher doses are not relevant for risk assessment, because that is based on the NOAELs. However, some effects seen at the highest dose are considered by regulators when considering hazard assessment—i.e. the potential of a chemical to cause a toxicity irrespective of how high the dose may be. Cancer is one such toxicity, and in this context, it is worth noting that at the highest dose in the 18-month mouse study (105 & 129 mg/kg/day for males and females, respectively) fluopyram caused thyroid tumours in males, but no tumours in females, or at lower doses in either gender. In the 2-year rat study, fluopyram caused liver tumours at the top dose in females (89 mg/kg/day) but no tumours in males (29 mg/kg/day) or at lower doses in either gender. Effects at doses higher than the LOAEL are summarised in Online Resource Table 4.

The main toxicities seen are liver and thyroid effects in rats and mice of both genders, and a certain form of kidney toxicity that is specific to male rats. Liver effects were also seen in the dog. The adverse effects in liver and thyroid manifesting as tissue hypertrophy, hyperplasia and eventually tumours are common findings in rodents exposed throughout their lifetime to high dose levels of many diverse xenobiotics, including pharmaceuticals. The common factor amongst these xenobiotics is that they stimulate their own metabolism in the liver via activation of a number of nuclear receptors. The liver and thyroid effects seen in the long-term studies resulted in the subsequent conduct of a series of additional mechanistic studies, which are described in a later section. The kidney effect seen in male rats is a well characterized type of toxicity. Chronic progressive nephropathy occurs spontaneously in male rats, and is exacerbated by many chemicals, including some natural substances (Hard et al. 2009). It has been thoroughly investigated and the mechanism resulting in the effect is understood. A protein called alpha2mu-globulin is essential to this mechanism of toxicity, and this protein does not occur in humans (Swenberg et al. 1989; Borghoff et al. 1990; Lehman-McKeeman and Caudill 1992). Therefore, there is no human equivalent of male rat chronic progressive nephropathy, and this finding is recognised as not being relevant to human risk assessment (IARC 1999; Hard et al. 2009).

Table 3 includes Bayer’s original proposed NOAELs and LOAELs, but this is not the end of the process of interpreting the results of these studies. Regulatory authorities review the same studies and come to their own conclusions about the results and the NOAELs. These conclusions can differ between countries and sometimes change over time. For example, the EU authorities disagreed with the proposed NOAEL of 26.6 mg/kg/day in the 90-day mouse study and selected 5.4 mg/kg/day instead. It is the NOAELs determined by regulators which underpin the human risk assessments and regulatory decisions that enable the use of fluopyram in agriculture.

Human risk assessment for pesticides relies on studies that mimic relevant exposure routes, primarily oral (dietary) and dermal (applicator). A study by Özgöçmen and Toğay (2021) reported kidney effects in mice given fluopyram but administered it via subcutaneous injection. This route is not considered relevant for human safety assessment because it bypasses critical metabolic defences in the skin, gut, and liver that would normally occur with dermal or oral exposure. Consequently, the study’s findings are difficult to interpret and cannot be reliably applied to a human context.

Targeted toxicity studies

Unlike the general toxicity studies reviewed in the previous section, the studies in this section have a specific question to address, and so a specific focus. The observations and measurements made are targeted to reflect what needs to be examined. The general toxicity studies for fluopyram all involve dosing live animals, whilst some of the studies in this section are in vitro, i.e. studies in which cells are exposed to fluopyram.

Does fluopyram cause local toxicity at the site of application?

In single dose studies in which fluopyram was applied to the skin and eye of rabbits, it was classified to be non-irritating. In a study looking for skin sensitization in the mouse, fluopyram was found to have no potential to cause skin sensitization (i.e. an allergic skin reaction). These studies used 6 rabbits and 25 mice.

Does fluopyram damage genes?

If a chemical adversely interacts with DNA this is called genotoxicity, which makes use of the chemical as a pesticide unacceptable. A range of in vitro and short duration in vivo studies exist to evaluate these potential effects on DNA. No single test is a perfect predictor of the potential to cause tumours by a genotoxic mechanism, so a battery of studies is performed and are evaluated together to assess genotoxic potential. The studies are:

• Ames test: this test in specific strains of Salmonella bacteria evaluates the ability of a test compound to cause mutations (mutagenicity). Fluopyram did not cause mutations.

• HPRT test: this test in Chinese hamster cells cultured in vitro, also evaluates the ability of a test compound to cause mutations. Fluopyram did not cause mutations.

• Chromosome aberration test: this test in Chinese hamster cells cultured in vitro, evaluates the ability of a test compound to cause the disruption or breakage of chromosomes (clastogenicity). Fluopyram did not cause chromosome aberrations.

• Human in vitro micronucleus test: this test is an evaluation of the ability of a test compound to cause chromosome aberrations, using human lymphocytes from blood samples. Fluopyram did not cause chromosome aberrations in this test.

• Mouse in vivo micronucleus test: this test is an evaluation in vivo of the ability of a test compound to cause chromosome aberrations. Fluopyram did not cause chromosome aberrations in this test. Including a dose range finding step, this study used 37 mice.

Fluopyram showed no potential to affect DNA in any of these studies and so is not genotoxic.

Can fluopyram cause effects on fertility, the unborn child and the developing child?

The rat multigeneration study looks for effects in all parts of the reproductive cycle from the production of ova and sperm, mating behaviour, fertilization, implantation, embryogenesis, foetal development, birth, infancy, childhood, adolescence and sexual maturity. The fluopyram study exposed animals continuously over three generations. A rat and a rabbit study were also performed, focused specifically on a very detailed examination of any effects on the developing foetus. The parameters examined in these studies are shown in Tables 5a, b of the Online Resource. Study results are summarised in the Table 4.

Table 4.

Studies of effects of fluopyram on fertility, the unborn and developing offspirng

Study	Endpoint	No observed adverse effect level (NOAEL)		Lowest observed adverse effect level (LOAEL)
		Dose (mg/kg/day)	Treatment effects	Dose (mg/kg/day)	Treatment effects
Rat multi-generation	Parental toxicity	220	None	1200	Body weight effects (females); haematology changes (females); clinical chemistry changes; adaptive liver changes; kidney effects (males); lung effect (females)
	Offspring toxicity	220	None	1200	Body weight effects; slightly delayed development due to lower body weight (males); reduced spleen and thymus weight due to lower body weight
	Reproductive toxicity	1200	None
Rat development	Maternal toxicity	30	Transient lower food consumption and body weight gain (not considered adverse)	150	Reduced food consumption and body weight gain; increased liver weight with liver hypertrophy
	Foetal toxicity	150	None	450	Reduced body weight; increased incidence of two visceral variations and two skeletal variations; no malformations
Rabbit development	Maternal toxicity	25	None	75	Reduced food consumption, body weight gain and body weight
	Foetal toxicity	25	None	75	Reduced body weight and increased incidence of small foetuses

In the multigeneration study, the toxic effects seen in adults were ones also seen in other repeat-dose studies (see Table 3). The reduced body weight of adult females, as expected, resulted in body weight effects in their offspring. No specific reproductive endpoints were affected. This study started with 240 rats, with a further 2427 offspring born during the study.

In the rabbit developmental toxicity study, the dose level of 75 mg/kg/day caused reduced body weight in the pregnant females, which consequently resulted in smaller foetuses. In the rat, the pregnant females showed body weight effects from 150 mg/kg/day, and though there were no effects on the foetuses at this dose, their weight was reduced at the highest dose level (450 mg/kg/day) and the incidence of four minor developmental variations was also increased (these variations often occur spontaneously), which coincided with more pronounced maternal body weight effects and increased liver toxicity compared to 150 mg/kg/day. Fluopyram did not cause malformations in rats or rabbits at any dose level and did not cause any effects on foetuses in the absence of maternal toxicity. Including preliminary studies used to determine suitable dose levels, these studies used 124 pregnant rats, 132 pregnant rabbits and all their foetuses.

In addition to the foregoing studies using mammalian test systems, there is also a zebrafish study (Zhang et al. 2024). However, in this study the lowest concentration tested (1.5 mg/L) was higher than the lower confidence limit of the LC50 in the study (0.554 mg/L). Therefore, there was mortality at all dose levels, and no information on the effects of fluopyram at sublethal doses was obtained.

Does fluopyram affect the endocrine system?

The effects of endocrine-disrupting chemicals are expected to be detected in medium and long-term repeat dose toxicity studies, as these assess a wide range of hormone-sensitive endpoints such as impacts on reproductive organ pathology, fertility, and the growth and sexual development of offspring. These are central components of the multigeneration rat study. In addition, specific in vitro tests have been developed to focus on specific effects of concern. The in vitro endocrine disruption studies that were conducted and their results are as follows.

• Mimicking oestrogen. Oestradiol, an oestrogen hormone, is the main female sex hormone in humans, and its binding to human oestrogen receptors is critical for its effects. Assays were performed to evaluate the ability of a test substance to bind to two types of human oestrogen receptor (alpha and beta) and so potentially mimic the effects of oestradiol. The assays are conducted in insect cells which have been transformed to express the human receptors. Fluopyram did not bind to either human oestrogen receptor.

• Oestrogenic activity. Rather than looking directly for binding to an oestrogen receptor, this assay looked for any effects on the functioning of the oestrogen receptor type alpha. This uses human oestrogen receptor type alpha in a specialist cell-free system designed to enable its functioning to be evaluated. Fluopyram had no effect on the functioning to the oestrogen receptor in this system.

• Mimicking testosterone. Testosterone, an androgen hormone, is the main male sex hormone in humans, and its binding to human androgen receptors is critical for its effects. This assay evaluates the ability of a test substance to bind to the human androgen receptor and so potentially mimic the effects of testosterone. The assays are conducted in cytosol from a human cell line that is especially sensitive to androgens. Fluopyram did not bind to the human androgen receptor.

• Aromatase inhibition. Aromatase is an enzyme responsible for a key step in the synthesis of oestrogens, converting testosterone to oestradiol. This assay evaluated the ability of fluopyram to inhibit the functioning of this enzyme, and it was found that it did not have any such effect.

• Thyroperoxidase inhibition. Thyroperoxidase is a key enzyme in the production of thyroid hormones and is found in the thyroid gland. This assay evaluated the ability of fluopyram to inhibit the functioning of this enzyme, and it was found that it did not have any such effect and therefore has no direct effect on the thyroid.

• Sodium/iodide symporter (NIS) assay. This symporter mediates the uptake of iodide into thyroid follicular cells, a key step in thyroid hormone production. The assay looks for the inhibition of this symporter, and fluopyram was found to have no such effect.

Based on these study results fluopyram has been found to have no endocrine activity in vitro.

Does fluopyram affect the nervous or immune systems?

In neurotoxicity studies, specialised evaluations are made to examine potential effects on the rat nervous system, behaviour and movement, whilst post mortem the brain, eyes, nerves and muscles are examined microscopically for effects. One neurotoxicity study examines the response over two weeks to a large single dose of fluopyram, whilst the other assesses neurotoxicity during 13 weeks of continuous dietary exposure to fluopyram. Following a single oral dose of 500 mg/kg and higher to male rats and 100 mg/kg and higher to females the rats, animals were affected on the day of dosing, showing reduced motor and locomotor activity, but had fully recovered by day 7. There were no visible or microscopic changes to the nervous system or muscles at any dose up to 2000 mg/kg, and the neurotoxicity NOAEL in this study was 125 mg/kg for males and 50 mg/kg for females. During 13 weeks of dietary dosing of fluopyram and at post mortem assessment, no neurotoxicity was seen at any dose, so the neurotoxicity NOAEL in this study was 164.2 mg/kg/day for males and 197.1 mg/kg/day for females.

In addition to studies performed to established protocols, there is also a mouse study in which fluopyram was dosed by subcutaneous injection at 0.5, 1 and 2 mg/kg daily for 21 days (al Sammarraie et al. 2023); however, as mentioned earlier this route is not considered relevant for human safety assessment. A decrease in motor activity was reported at the highest fluopyram dose level, whilst histological and immunohistochemical changes were reported at all dose levels, including the occurrence of Lewy bodies, an increase in alpha-synuclein and decrease in tyrosine hydroxylase staining in the substantia nigra. No information is given about how the substantia nigra was located, bringing into question whether this structure was successfully found. The reported occurrence of Lewy bodies in the brains of Swiss albino mice is a novel finding and questionable, with no evidence being given that the objects identified as Lewy bodies were truly Lewy bodies. Overall, the quality of histology and its interpretation is questionable. The number of animals per sex per group is unclear, but based on the information given must be no more than three, which is too small for a reliable study of this nature. Overall, the quality of the study assessed using the Klimisch score is 3, Not Reliable (Klimisch et al 1997).

In an immunotoxicity study, the immune reaction to an injection of sheep red blood cells is assessed following four weeks of dietary exposure of female rats to fluopyram. There was no effect on immunological response at any dose, so the immunotoxicity NOAEL was 156.3 mg/kg/day. For the neurotoxicity and immunotoxicity studies, a total of 290 rats were used.

What is the fate of fluopyram in the body?

Studies of the absorption of fluopyram into the body from an oral dose, its distribution within the body, its breakdown to form metabolites and the excretion of fluopyram and its metabolites from the body have been conducted. These studies use fluopyram which has been labelled by the inclusion of carbon-14, which enables its fate and breakdown to be followed in detail, including identifying the structure of metabolites. The laboratory species used for these studies is the rat. The fluopyram molecule contains two rings and two sets of studies were performed, one with the carbon-14 located in one ring, and one with it located in the other ring. After dosing, the fate of fluopyram is followed in plasma, urine, faeces and exhaled air over time, whilst post mortem the fate of fluopyram is studied in a large number of organs and tissues. In some rats, the bile produced by the liver is also collected for study. This work showed that fluopyram is rapidly and almost completely absorbed following an oral dose. Fluopyram is rapidly degraded in the liver, and the majority of the breakdown products are excreted via bile into the gut, where some are further degraded by gut microbes, to then be re-absorbed into the body and excreted by the kidney into urine, or to pass out of the body in faeces. The total amount of fluopyram metabolites excreted in urine and faeces are similar. A small amount of unchanged fluopyram molecule was excreted in faeces, but not in urine. The first degradation step seen was the oxidation of fluopyram, producing metabolites that still had the two rings present as in fluopyram, then the molecule was split, separating the two rings. This degradation is one of the ways the liver protects the body from unfamiliar chemical structures, and it makes them easier to excrete. Another method to aid excretion is to attach a small naturally-occurring molecule to a chemical to form what is called a conjugate—many of these conjugates are formed in the degradation of fluopyram. In all, over 30 metabolites were structurally identified, about half of which still had the two rings present as in fluopyram, the others only having one of the rings. After seven days only a few percent of the dose of fluopyram remained in the body. The route of metabolism was the same in male and female rats, with small differences in rates of metabolism and concentration in tissues. The fate of fluopyram was similar at 5 and 250 mg/kg dose levels and was similar whether dosed once or 15 times.

Other studies of the fate of fluopyram were conducted in goats, cows and hens, which had the specific objective of understanding whether fluopyram or its metabolites might find their way into meat, milk and eggs that consumers buy and, if so, at what concentration. These showed that the fate of dosed fluopyram is similar to the rat in these species. Further studies comparing the fate of fluopyram in cell extracts from humans, rats, dogs, mice and rabbits showed that the degradation of fluopyram was as quick or quicker in the other species compared to the rat, and there were no metabolites seen in the human cell extracts that were not also seen in other species. In the toxicity studies reviewed in this paper the animals were dosed fluopyram, but their organs and cells will have been exposed to a combination of fluopyram and its many metabolites. In this sense, the toxicity of these metabolites has been tested, and forms a part of the toxicity profile of fluopyram.

The studies in this section used 82 male and 64 female rats, 84 laying hens, 2 lactating goats and 16 lactating cows.

Mechanistic studies

As described earlier, in long-term studies at the highest dose tested, fluopyram causes an increase in liver tumours in female rats, and an increase in thyroid tumours in male mice. Without further studies the default approach is to consider these tumours to be relevant to humans. However, if the tumour-causing mechanism was understood, it may be possible to determine its likelihood of occurring in humans. There is a recognized process for investigating effects seen in laboratory animals to determine their relevance to humans, called the Human Relevance Framework. This has been developed by the World Health Organization (Meek et al. 2014) and provides guidance on how to generate and assess the relevant evidence. This approach has been followed for female rat liver tumours and male mouse thyroid tumours found in the case of fluopyram by the conduct of additional mechanistic studies. These studies robustly establish the processes resulting in these tumours in rodents and have been used to investigate if these tumours were likely to occur in humans. The studies are summarised here and are more fully described in two papers (Tinwell et al. 2014; Rouquié et al. 2014).

The consistent hepatic effects seen in all general toxicity studies identify the liver as a sensitive target organ for fluopyram, forming the basis of an Adverse Outcome Pathway (AOP) for liver tumour formation that mirrors the mode of action of phenobarbital (OECD AOP #107; OECD 2023). The first two Key Events of this AOP were examined in mechanistic studies in which fluopyram was dosed to female rats, and in these studies one group was dosed phenobarbital as a reference compound. The key events are:

1. Activation of CAR and PXR nuclear receptors. Fluopyram and phenobarbital activate these nuclear receptors, as indicated by dose-related increases in hepatic mRNA levels and enzyme activities for Cyp2b and Cyp3a. The increased production and activity of these liver enzymes was seen after just a few days of dosing and is responsible for accelerating the metabolism of exogenous (and some endogenous) molecules.

2. Increased hepatocyte division. The activation of hepatic CAR and PXR nuclear receptors also induces a dose-related increase in the rate of liver cell division, which was seen after just three days of dosing, which contributed, via hepatic hyperplasia, to an increased liver weight.

3. Clusters of altered liver cells. When liver hyperplasia is maintained for a long time (however caused) then DNA replication errors (mutations) have an opportunity to accumulate, which can result in distinctive clusters of altered liver cells. These were seen in female rats at the top dose after one year of fluopyram exposure.

4. Liver tumours. Given time and especially with continued exposure, some of the clusters of altered liver cells can develop into adenomas and/or carcinomas, as was seen with fluopyram at the top dose after 2 years of dosing.

Each of these Key Events must occur, in the appropriate sequence, for fluopyram to produce liver tumours. The mode of action of fluopyram was found to closely match that of phenobarbital, as indicated by clear dose-related increases in the first two Key Events of the AOP (activation of CAR/PXR receptors and hepatocellular proliferation within the liver lobules). There was also concordance between the high dose level inducing liver tumours in females in the two-year rat study and the dose level in mechanistic studies where maximum changes were observed in the early Key Events 1 to 3. After 28 or 90 days of fluopyram dosing, Key Events 1 and 2 and associated events such as increased liver weight and hepatocellular hypertrophy were reversible and normalised during a 28-day recovery period.

Other modes of action that are known to also result in liver tumours in rodents were considered and were either excluded for fluopyram, or were considered unlikely, based on the evidence obtained in studies (Table 6 in Online Resource). Having established the mode of action by which fluopyram causes rat liver tumours, this can then be used to evaluate whether this mechanism could also cause liver tumours in humans. An in vitro study comparing the effects of fluopyram in rat and human hepatocytes showed that fluopyram and phenobarbital both substantially activated CAR and PXR receptors in rat liver cells (Key Event 1), but this effect was much weaker in human liver cells. Critically, rat liver cell division was stimulated when exposed to fluopyram or phenobarbital, but human liver cell division was unaffected. This difference in cell division response between rodent and human hepatocytes has been observed for many other chemicals and drugs (Elcombe et al 2014; Yamada et al 2021). Therefore, a critical step in the mode of action does not occur in humans, since without increased cell division and tissue hyperplasia the neoplastic lesions cannot develop, and so this mode of action lacks the potential to cause liver tumours in humans. This is further supported by the observation that long-term treatment with phenobarbital as a medicine does not cause liver tumours in humans (IARC 2001; Friedman et al 2009; La Vecchia & Negri 2014; Olsen et al 1995).

The mode of action established for the thyroid tumours in male mice was also established in bespoke mechanistic studies. The key events are:

1. Activation of CAR and PXR nuclear receptors. This is the same Key Event as for liver tumours in the female rat resulting in increased liver enzyme activity and was also demonstrated in the male mouse using similar mechanistic studies.

2. Increased thyroxine metabolism in the liver. The consequences of increased hepatic enzyme activity to metabolise fluopyram is to also increase the breakdown of endogenous hormones including thyroxine, resulting in a decrease of thyroxine in the blood. The increase in thyroxine-metabolising enzyme activity (and associated gene expression) was measured in the livers of mice dosed fluopyram, and the reduction of thyroxine in the blood could be measured after just three days of exposure. A radiolabelled dose of thyroxine was also used to demonstrate more rapid break down in fluopyram-treated compared to control mice.

3. Increased thyroid stimulating hormone (TSH). By a feedback mechanism, the reduction of thyroxine in the blood caused an increase in TSH produced in the pituitary gland, which was demonstrated by measuring pituitary TSH transcripts after 3 or 28 days of fluopyram treatment in mice.

4. Increased thyroid follicular cell division. The increase in TSH causes an increase in the rate of cell division in the thyroid. An increase in the rate of division of these thyroid cells was demonstrated in male mice after 28 days of fluopyram exposure.

5. Thyroid tumours. Given time and continued exposure, as with the liver, sustained high rates of cell division can result in the accumulation of mutations and the production of thyroid follicular cell tumours, as was seen at the highest dose in male mice in the 18-month study.

Additional mechanistic studies used mice which had been genetically modified to have no CAR and PXR nuclear receptors (referred to as CAR / PXR knockout mice), with normal mice as a comparator. Exposure of normal mice to fluopyram for 28 days produced liver toxicity, large increases in the Cyp2b and Cyp3a enzymes consequent on CAR and PXR activation (Key Event 1), increases in the liver enzymes that metabolise thyroxine (Key Event 2), increased expression in the pituitary of the gene responsible for producing TSH (Key Event 3), and an increased rate of thyroid follicular cell division (Key Event 4). In contrast, the knockout mice had no microscopic changes in the liver, small changes in the Cyp2b and Cyp3a enzymes consequent on CAR and PXR activation, no increase and in fact some decreases in the liver enzymes that metabolise thyroxine, a small decrease in expression in the pituitary gland of the gene responsible for producing TSH, and most importantly no effect on the rate of cell division in the thyroid. This clearly demonstrates that CAR/PXR activation is the molecular initiating event in the mouse thyroid tumour mode of action, without which the later Key Events cannot occur. In addition, the mechanistic studies showed that there were dose-related increases in all four Key Events preceding the formation of thyroid follicular cell adenomas. There was also concordance between the high dose level inducing thyroid tumours in male mice in the cancer bioassay and the dose level in the mechanistic studies where significant changes were observed in the first four Key Events. Reversibility of effects for the first four Key Events and associated events such as the induction of Cyp enzyme activities was clearly shown following a recovery period of 28 days after 28 days of fluopyram dosing.

Alternative modes of action based on direct toxicity of fluopyram to the thyroid were considered. The responses to fluopyram exposure seen in wild type and CAR / PXR knockout mice were inconsistent with direct thyroid toxicity, and this was also supported by in vitro assays in which fluopyram did not inhibit thyroid peroxidase or the sodium iodide symporter (Table 7 in Online Resource).

Having established the mode of action by which fluopyram causes mouse thyroid tumours, this can now be used to evaluate whether this mechanism could also cause thyroid tumours in humans. An in vitro comparative study with rat and human hepatocytes showed that fluopyram activated CAR and PXR receptors in rat and human liver cells. However, the enzyme critical for thyroxine clearance, namely UGT-T4, was activated only in the rat hepatocytes. These data as well as the known physiological differences (including buffer capacity in plasma thyroid hormone levels) between rodents and humans strongly indicate that the induction of thyroid hyperplasia and potential tumour formation via this mode of action is not effective in humans. This ineffective mode of action is also observed in the dog, where thyroid effects were not observed with fluopyram.

These mechanistic studies used 420 female rats and 1004 male mice.

Human safety risk assessment

Summary of fluopyram toxicity

The complete package of human safety studies takes years to complete and involves the use of several thousand laboratory animals and the generation of several hundred thousand datapoints. Once this has been completed, an overall assessment of the potential of fluopyram to cause adverse effects can be done. In studies in which fluopyram was included in the diet, food consumption and body weight data are used to estimate the dose the animals received in units of milligrams of fluopyram per kilogram of body weight per day. This is done separately for males and females, and the lower of the male and female NOAELs for a study is used. Figure 1 shows the proposed NOAELs for general toxicity studies plotted against exposure duration, with the toxicity of fluopyram increasing with longer exposures. This pattern is observed for many substances, reflecting that continual exposure may not allow for recovery and effects can often be more easily detected in longer studies which have more animals and so more statistical power. The toxicity of fluopyram to the mouse is similar to the rat in short-term studies, but less in longer-term studies, whilst the dog seems to be less sensitive.

Fig. 1.

Comparison of toxicity of fluopyram across species and exposure durations

Additionally, based on the studies the following statements can be made:

• fluopyram does not irritate eyes or skin or cause an allergic skin reaction.

• fluopyram does not damage genes.

• fluopyram does not affect fertility, the unborn child or the developing child.

• fluopyram does not affect the nervous or immune systems, or the endocrine system via oestrogenic, androgenic or steroidogenesis mechanisms.

• fluopyram causes liver tumours at high doses in female rats and thyroid tumours at high doses in male mice; however, the mechanisms causing these tumours have been robustly established, and it has been demonstrated that these mechanisms are rodent-specific and unlikely to occur in humans.

For use in human risk assessment, all the studies are examined and the doses which would be safe for short- medium- and longer-term exposure are identified. This is done by looking at the no adverse effect levels for the studies of appropriate duration and selecting the lowest one. This lowest no adverse effect level is then divided by at least 100 to account for both intraspecies and interspecies variability, ensuring protection for sensitive human populations (EFSA, 2013). These safe doses are the acute reference dose (ARfD) for single exposures (Solecki et al. 2005) and the acceptable daily intake (ADI, also called a chronic reference dose) for lifetime exposure. The ADI is the amount of a substance that can be consumed daily over a lifetime without a significant health risk.

Regulatory authorities frequently differ in their views on what the NOAEL is for the same toxicology study. Similarly, the acute and chronic reference doses set by regulatory authorities can differ. The human safety risk assessment presented here is not the one from the USA or EU regulatory authorities, but is the one conducted by the Joint FAO/WHO Meeting on Pesticide Residues (JMPR, 2012). The JMPR has set the acute reference dose for fluopyram at 0.5 mg/kg/day, based on a single dose NOAEL of 5 mg/kg in the acute neurotoxicity study, whilst the JMPR acceptable daily intake is 0.01 mg/kg based on a NOAEL of 1.2 mg/kg/day from the two-year rat study (JMPR, 2010).

Fluopyram exposure and risk assessment

An overview of the exposure and risk assessment conducted by JMPR will now be presented (JMPR, 2010, 2012). Studies in rats, goats, hens, grape, potato, beans and sweet peppers were used to study the metabolism of fluopyram. Based on these studies it was determined that the dominant residue in crops was parent fluopyram, whilst in animal commodities both parent fluopyram and the metabolite 2-(trifluoromethyl)benzamide were important and so were included in the exposure assessment. Across the world, fluopyram is used on many different crops including different types of fruit, nuts, vegetables, pulses and oilseed crops. Field trials were conducted for each type of crop, applying a product containing fluopyram under worst-case conditions. These worst-case conditions are defined by the label on the bottle of fluopyram-containing product, which is a legal document approved by the authorities, and which specifies the highest application rate, the maximum number of applications allowed, the minimum interval between application, and the shortest duration allowed between application and harvest. The harvested crops were analysed for the presence of fluopyram residues. For each crop type, several field trials are conducted in different conditions and in different regions, resulting in several different concentrations of fluopyram residues. For the chronic exposure assessment, a median residue value is used, whilst for acute exposure assessment the highest value is typically used. Additional studies examined the transfer of residues from harvested crops into processed commodities, e.g. from grapes into grape juice, wine and raisins. Processing studies of this kind were conducted for tomatoes, sugar beet, apples, peanuts, strawberries and rapeseed as well as for grapes.

The procedure for estimating potential consumer dietary exposure to fluopyram residues in meat, milk and eggs is more involved. In addition to being fed commodities that also go into the human diet, there are commodities only used for animal feed e.g. sugar beet tops and almond hulls—crop residue field trials also analyse these animal feed commodities for fluopyram residues. Information on the maximum fraction of the diet that can be from each animal feed commodity are available for beef cattle, dairy cattle, broiler poultry and laying poultry in North America, the EU and Australia. These are used to estimate the maximum dietary fluopyram residue that could occur in each case. Cow and hen feeding studies are then used to estimate the potential fluopyram residue in meat, liver, kidney, fat, milk and eggs.

Foods in the diet vary around the world, and JMPR use 13 different regional diets to reflect this. The fluopyram residues in raw and processed agricultural commodities and in meat products, milk and eggs were used to estimate the total potential dietary exposure for each of the 13 diets, referred to as the International Estimated Dietary Intake (IEDI), which was found to range from 0.00017 to 0.0013 mg fluopyram / kg bodyweight. These represent 1.7–13.2% of the ADI, i.e. the amount that can be consumed daily over a lifetime without a significant health risk. Expressed in a different way, the highest consumer exposure estimate is over 900 times lower than the level that caused no adverse effects in the most sensitive animal study. It is important to note that these dietary intake estimates are extremely conservative. To take apples as an example, the dietary estimates effectively assume that every apple someone eats was from an orchard treated with fluopyram on the maximum number of occasions, at the maximum allowed application rate and so on.

Acute dietary exposure assessment accounts for a worst-case possibility occurring on a single day. Taking apples again as an example, the acute exposure assessment considers someone who eats an unusually large number of apples in one day, and assumes that those apples came from an orchard where the fluopyram residue was the highest that has been measured in all apple field trials, and assumes the fluopyram residue was not the same for each apple but was exceptionally high in the first apple they ate. Using this methodology, the International Estimated Short-Term Intake (IESTI) was found to range from 0 to 0.05 mg fluopyram / kg bodyweight, which represents 0–10% of the ARfD, i.e. the amount which can be consumed in 24 h without appreciable health risk.

Risk assessments are also conducted to ensure that people applying pesticides, people working in fields to which pesticides have been applied, bystanders and people living nearby are protected (e.g. EFSA 2014). Fluopyram-containing products pass all these risk assessments.

Discussion

The process for assessing the safety of crop protection chemicals has become increasingly sophisticated over the last 50 years, with foundational principles established by the turn of the century (Tait 2001) and continuous refinement through modern scientific advancements (EFSA 2023). The human safety studies conducted for fluopyram reflect this. The number of studies conducted, the number of datapoints generated and the number of animals used may be a surprise to those not familiar with this domain. Internationally agreed study guidelines ensure best practice is followed, and GLP ensures the trustworthiness of the work. Human safety studies are distilled into human risk assessments that are conservative in nature and ensure protection. Government regulators are responsible for the assessment process, and their decisions and reasoning are publicly available, in an increasingly transparent process. However, the challenge made against SDH inhibitors is that the human safety studies conducted have no value because effects caused as a result of the inhibition of SDH have been missed. This is a serious allegation, which will now be addressed through two questions:

• Could any of the toxicities seen in these studies caused by the inhibition of SDH?

• Could toxicities caused by the inhibition of SDH have been missed?

The first question concerns whether the mechanisms causing the effects of fluopyram are understood, and whether the role of SDH inhibition in these mechanisms can be excluded. The second question concerns the toxic effects that might be expected if SDH inhibition was occurring, and whether the fluopyram human safety studies have looked thoroughly for these effects.

Are any of the toxicities seen in fluopyram studies caused by the inhibition of SDH?

In other words, is there a credible set of key events which could associate the effects of fluopyram with the inhibition of SDH? In the most sensitive species, the rat, fluopyram reduced food consumption, resulting in body weight effects. In addition, the liver and thyroid were identified as target organs for toxicity. Very similar liver and thyroid effects were observed in the mouse, whilst in the dog the liver effects were seen but no thyroid effects. In addition, a form of kidney toxicity was seen which is specific to male rats. At the highest dose tested in lifetime exposure studies, liver tumours were seen in rats and thyroid tumours in mice. Each of these effects will now be considered.

Reduced food consumption with consequent effects on body weight gain and body weight

This is probably the most common effect seen in toxicity studies for chemicals in general. The occurrence of this effect across a broad range of chemicals is an indication that it is a non-specific secondary effect, i.e. it suggests that the animals are experiencing some form of physiological distress. The medical community continues to search for medicines that reduce human food consumption, to tackle obesity, using three main approaches:

• Increase metabolic rate. The inhibition of SDH would reduce metabolic rate, not increase it, and so could not reduce food consumption by this mechanism.

• Appetite suppression. Amphetamines have been used in the past to suppress appetite but also cause restlessness and agitation (and have caused fatal pulmonary hypertension, heart valve damage and haemorrhagic stroke in humans). Such effects are not seen with fluopyram. Semaglutide is now used to suppress appetite by acting as mimic of the hormone glucagon-like peptide-1 (GLP-1). Fluopyram is structurally dissimilar to peptides, and no plausible mechanism has been proposed that connects SDH inhibition with the induction of GLP-1.

• Reducing the absorption of fat by the gut. For example, orlistats work by inhibiting an enzyme responsible for fat breakdown in the gut. No plausible mechanism that could connect such an effect to SDH inhibition has been proposed.

Another potential reason for reduced food consumption in animal studies would be if the chemical tested was causing local irritation and damage of the wall of the gastrointestinal tract—this does not occur with fluopyram because it is not an irritant. In fluopyram studies, effects on food consumption and body weight were seen at doses that also caused other effects, mainly liver toxicity. Therefore, the effects of fluopyram on these endpoints are likely to be secondary effects caused by the animals being physiologically compromised.

Liver and thyroid toxicity

A comprehensive set of mechanistic studies demonstrated that fluopyram activates CAR and PXR in the liver, and in rodents, this starts a chain of events that can result in liver and thyroid toxicity, including in some circumstances tumours in these tissues. In mice lacking CAR and PXR, fluopyram had little effect on the liver, with only very small increases in liver weight seen, without any microscopic changes to the liver. In the absence of microscopic changes, small increases in liver weight are an adaptive response, not an adverse one. The evidence is that all the liver and thyroid toxicity seen in fluopyram studies can be attributed to the activation of CAR and PXR. Might SDH inhibition cause the activation of CAR and PXR? The role of CAR and PXR is to act as a master switch, sensing foreign substances and increasing liver enzymes to accelerate their breakdown. As you might expect, many structurally diverse chemicals, including pharmaceuticals, are known to activate CAR and PXR. In contrast, the structural requirements for inhibiting SDH are very narrow. There is no evidence that SDH inhibition is a step that can result in the activation of CAR or PXR either for SDH inhibitors or for any other chemical.

Male rat kidney toxicity

The chronic progressive nephropathy that is exacerbated by fluopyram in male rats is a well characterised toxicity. It has been demonstrated that this toxicity is caused by reversible binding of substances to the protein alpha2mu-globulin (Lehman-McKeeman and Caudill 1992), and it is supposed that this binding occurs with fluopyram and/or its metabolites. It is theoretically possible that neither fluopyram nor its metabolites bind to alpha2mu-globulin, and that SDH inhibition results in the binding of some endogenous substance to this protein, but this is a much more complex explanation and seems unlikely. In any event, it would only be of academic interest and would have no implications for human risk assessment.

Other effects of fluopyram

Whilst the main toxicities associated with dosing fluopyram have been addressed above, there are other sporadic effects of fluopyram that were observed, mostly in single studies. The effects are summarised in Table 8 of the Online Resource. Many of these effects occurred only at doses higher than the LOAEL and so would not normally be considered significant. Plausible explanations exist for the causes of most of these effects, such as being secondary effects of body weight loss, liver toxicity or chronic progressive nephropathy. Reductions in body temperature are a well-documented, non-specific response to high-dose toxicant administration in rodents and are considered a general indicator of systemic toxicity (Gordon 1991). Therefore, rather than being a specific consequence of SDH inhibition, the hypothermia observed in the two acute studies can be interpreted as a sign of the animals’ overall adverse health status.

Could toxicities caused by the inhibition of SDH have been missed?

In other words, if fluopyram were to inhibit SDH in mammals, then what effects would be expected in animal safety studies, are these effects seen, and has the possibility of these effects occurring been thoroughly investigated.

Removal of the gene encoding SDH is fatal at the embryonic stage in mice. Similarly, the complete inhibition of the SDH enzyme would also be expected to be fatal. Therefore, if fluopyram was inhibiting SDH, significant acute toxicity would be expected. In fact, the acute toxicity of fluopyram dosed orally, dermally or by inhalation is low, which is not consistent with the hypothesis that fluopyram inhibits SDH in mammals. In the mouse micronucleus study, two doses were made by intraperitoneal injection (IP), separated by 24 h. IP injection is a route which is not relevant to risk, and which bypasses all natural barriers to systemic fluopyram exposure. However, even for this exposure route, lethality was only observed at a 2000 mg/kg dose level.

A range of human health effects have been reported as being a consequence of mutations in the four subunits of SDH. In general, homozygous mutations can result in degenerative brain diseases (Fullerton et al. 2020), whilst individuals who are heterozygous for a mutation have an increased risk of specific rare tumour types, including paragangliomas and pheochromocytomas, gastrointestinal stromal tumours, renal cell carcinomas and pituitary adenomas (MacFarlane et al. 2020; Gill 2018). Reduced SDH expression not associated with SDH mutations has also been reported to cause effects, including reduced rat sperm mobility and consequent male infertility (Tomar et al. 2012). The main purpose of the 2-year rat and 78-week mouse studies is to look for any carcinogenic potential. These studies cover most of the lifetime of the animals, the group sizes are large, and the number of tissues examined microscopically for abnormalities, including tumours and their precursors, covers all relevant organs. Yet in these studies fluopyram did not result in any tumours associated with SDH defects in humans. These and other studies, especially the multigeneration study, also demonstrate the lack of brain abnormalities caused by fluopyram, despite extended exposure to high dose levels. Sperm mobility and male infertility are parameters specifically examined in the multigeneration study, and no effects of fluopyram were seen.

Looking more widely, defective mitochondria cause a wide array of human diseases. The most common effects are seen in the nervous system (deafness, neurological impairment and developmental disability, stroke-like episodes), in other tissues with high energy demand (cardiomyopathy and heart arrythmias, kidney tubular dysfunction, myopathy), and elsewhere (hypothyroidism, gastrointestinal dysmotility, diabetes, diverse effects in the eye, fatigue) (Parikh et al. 2017). Damage to the nervous system would be expected to result in observable clinical signs, which are monitored in every animal study, yet no clinical signs of this nature have been seen with fluopyram. The functional observation battery (FOB) is a detailed set of clinical examinations and tests performed in several fluopyram studies. In the acute neurotoxicity study, reduced motor and locomotor activity was observed in the FOB on the day of dosing, for oral gavage doses of 500 mg/kg and higher for male rats and 100 mg/kg and higher for female rats. However, the animals quickly recovered and by day 7 there was no effect seen in the FOB. There were no visible or microscopic changes to the nervous system or muscles at any dose up to 2000 mg/kg. If the effects seen in this study were neurological in nature, then they were certainly temporary and completely reversible and were most likely due to general toxicity immediately following such a high oral dose. During 13 weeks of dietary dosing of fluopyram, including an FOB on four occasions, no neurotoxicity was seen at any dose or in detailed post mortem examinations.

Looking beyond the nervous system for signs of disease caused by mitochondrial dysfunction, heart and muscle disease would be visible in the many studies in which these tissues were examined microscopically, and yet fluopyram did not cause these effects. Regarding kidney dysfunction, fluopyram exacerbated chronic progressive nephropathy in male rats, but this is known to have no human correlate (IARC 1999; Hard et al. 2009). In the 78-week mouse study, fluopyram caused kidney toxicity in females (see Table 8 in Online Resource). However, this occurred only at the top dose of 129 mg/kg/day and is believed to be a milder manifestation of the chronic progressive nephropathy seen in aging male rats. In mice, fluopyram did reduce thyroxine (a key part of hypothyroidism), but detailed mechanistic studies proved that this was a consequence of an increased rate of breakdown of thyroxine in the liver, and so it was not caused by SDH inhibition. Effects of gastrointestinal motility might be detected in any study but especially in dog studies, yet no such effects of fluopyram were seen. Diabetes would be detected by changes in blood glucose and other measured made as a part of routine clinical chemistry tests, and in urine analysis, but effects consistent with diabetes were not seen. Eye toxicity was seen in the 2-year rat study, and the mechanism by which fluopyram caused this is unknown. However, a review of 2-year rat studies for SDH inhibitors showed that other than fluopyram only 2 out of 13 compounds caused any eye toxicity, which is not consistent with the pattern expected if the eye toxicity seen for fluopyram were a consequence of SDH inhibition.

The responses of mammalian cell cultures to fluopyram exposure are shown in Table 9 in the Online Resource. These varied studies show cells behaving as expected, with no evidence of mitochondrial toxicity. Liver is the primary target tissue for fluopyram, so the absence of effects on ATP concentrations in cultured rat and human hepatocytes exposed to up to 100 µM fluopyram, and the appropriate and uncompromised hepatocyte response (liver enzyme induction and hepatocyte proliferation) are not suggestive of the occurrence of SDH inhibition.

It is quite possible that SDH inhibition would result in effects in animals that differ from the effects seen in humans with SDH gene mutations or other forms of mitochondrial toxicity. However, it is reasonable to suppose that effects would still occur in high energy use tissues such as the nervous system, heart and other muscles, which are well covered by the tissues examined in toxicology studies (for example see Table 3 of the Online Resource). Regardless of the specific outcomes of SDH inhibition in animals, such effects would have been detected in the extensive body of detailed studies performed with fluopyram across four different mammalian species..

Why would fluopyram not inhibit SDH in vivo?

Fluopyram is a fungicide that works in the fungus by inhibiting SDH. So how is it that it doesn’t seem to inhibit SDH in mammals? As a critical component of respiration, an essential process in living organisms, SDH is highly conserved, meaning that its structure is similar in all species. Fluopyram binds to SDH at the site designed to accept the coenzyme ubiquinone (Coenzyme Q), and small changes in the structure of this binding site can have large effects on binding as shown in Table 5. The IC50s shown are the concentration of fluopyram which inhibits the activity of SDH by 50%. By this measure, the affinity of fluopyram for human SDH is thousands of times less than its affinity for SDH in fungi and nematodes. This principle is known as differential toxicity, and it forms the fundamental basis for the development and use of all modern pesticides.

Table 5.

Inhibition potency of fluopyram for succinate dehydrogenase from different species in vitro

Species	SDH IC50 (nM)	Source
Fungus (Botrytis)	200	Bénit et al. 2019
Fungus (Septoria, Botrytis)	6.3–25	Bayer Crop Science (unpublished)
Nematode (C. elegans)	1.8	Burns et al. 2015
Nematode (C. elegans)	3.2	Schleker et al. 2022
Earthworm (E. fetida)	> 1000	Schleker et al. 2022
Earthworm (L. terrestris)	30,600	Bénit et al. 2019
Housefly	> 1000	Schleker et al. 2022
Honey bee	3800	Bénit et al. 2019
Rat	> 1000	Schleker et al. 2022
Human	160,000	Bénit et al. 2019

The rapid mammalian metabolism of fluopyram might also be a contributing factor to explain the different toxicity profile between fungus and mammals.

The fluopyram human safety studies have thoroughly investigated the toxicity of fluopyram. A wide array of targeted and mechanistic studies has been conducted to address the possibility of specific toxicities. However, the most powerful set of studies in many ways are the general toxicity studies. These studies are completely untargeted, and a huge number of endpoints are examined, any one of which might reveal the critical toxic effect. Rat and mouse studies from a single dose to near lifetime exposure are included, whilst dog studies add biological span beyond rodents. The consistent pattern seen is that the primary toxicities caused by fluopyram are effects on the liver and thyroid, plus a kidney effect in ageing male rats. The mechanisms causing all these effects have been proven beyond all reasonable doubt, and they are not relevant to humans and do not involve the inhibition of SDH. If fluopyram was inhibiting SDH this would be apparent, as the tissues most likely to be affected—brain, heart and muscle—were examined microscopically in numerous studies with no effects seen. The rapid metabolism of fluopyram, combined with its low affinity for mammalian SDH, explains the results.

In summary, no signs of SDH mediated adverse effects have been identified in vivo in four mammalian species and in vitro in rat and human hepatocytes. The toxicities caused by fluopyram have been thoroughly characterised and have been used by regulators to conservatively determine human exposures at which no effects are expected. The consumer dietary risk assessments for fluopyram demonstrate ample safety margins despite their conservatism, providing protection for any subgroups who might be more sensitive than the general population. Fluopyram is a potent inhibitor of SDH in fungi and nematodes, but a weak inhibitor of SDH in mammals.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Helen Tinwell and Remi Bars had the idea for the manuscript and conceived and supervised many of the studies reviewed. Kim Travis and Remi Bars wrote the manuscript, whilst all authors critically reviewed the manuscript and approved the final version.

Funding

Helen Tinwell is, and Remi Bars was formerly employed by Bayer SAS. Kim Travis and Remi Bars are consultants with Regulatory Science Associates, who were paid by Bayer to write the manuscript.

Data availability

Summaries of fluopyram safety studies are available at https://www.bayer.com/en/agriculture/safety-results-crop-protection-products. The full study reports themselves can be requested using the form here https://www.bayer.com/en/agriculture/safety-study-report-request-forms.

Declarations

Competing and financial interests

Helen Tinwell is, and Remi Bars was formerly employed by Bayer SAS. Kim Travis and Remi Bars are consultants with Regulatory Science Associates, who were paid by Bayer to write the manuscript.

Ethical and animal welfare

The animal studies were evaluated by the ethical committee of the laboratory in order to guarantee that the animal welfare was considered on the basis of the cutting-edge knowledge of laboratory animal sciences. The care and use of animals was in accordance with the “Décret n° 2013 − 118 du 1er février 2013 relatif à la protection des animaux utilisés à des fins scientifiques”: implementation into the French law of “The Directive 2010/63/EU of the European Parliament and the Council of 22 September 2010 on the protection of animals used for scientific purposes, by the Official Journal of the European Union, L276/33–79, 2010”.

All studies involving human samples were performed in line with the principles of the Declaration of Helsinki, including obtaining informed consent. These studies are: two using human hepatocytes, one using pooled human liver microsomes and one using human lymphocytes.