2-(4-tert-butylbenzyl)propionaldehyde

Endpoint summary

Key value for chemical safety assessment

Effects on fertility

Description of key information

NOAEL for testicular toxicity (rat): 25 mg/kg bw/day.

NOAEL for testicular toxicity (dog): 44.6 mg/kg bw/day.

Additional information

Concerning fertility, a variety of repeated dose studies on male rats (some of them were designed for the assessment of male reproductive organ toxicity), one-generation range finding studies and an extended one-generation reproductive toxicity study (EOGRTS) are available. Furthermore, repeated dose studies on dogs, mice, guinea pigs, rabbits and primates are available for the assessment of reproductive toxicity of lysmeral (see Chapter "Repeated dose toxicity").

 

Evidence for testicular toxicity of lysmeral after bolus oral application via gavage is available from existing repeated dose studies in rats, however, data on continuous oral application of lysmeral was lacking and has been considered in the present one-generation studies. Concerning the properties of lysmeral as fragrance material, palatability has been assessed in a preliminary feeding test in rats (BASF 2002). Lysmeral (analytical purity 99.1%) has been administered for 14 days to 3 Wistar rats per sex and dose at a dietary concentration of 0, 100 and 1000 ppm. Food consumption and body weights were monitored and animals were examined for signs of toxicity and mortality. Clinical examinations and palpations have been performed.

Mean test substance intake ranged from 10.1-10.6 mg lysmeral/ kg bw/ day in the low dose group and 97-105.3 mg lysmeral/ kg bw/ day in the high dose group. Mean food consumption was slightly decreased in males of the high dose group (-6% versus controls at day 7 and 14) and in females of the high dose group (-12% and -3% versus controls after day 7 and day 14 respectively) without gaining statistical significance. Food efficiency has been decreased significantly in high dose males after day 7 and in high dose females at both observation time points.

In male animals, mean body weights were slightly decreased in the high dose group (-5% and -7% compared to controls after day 7 and day 14 respectively) and body weight gains were slightly decreased in the low dose group and more severe in the high dose group (-2% and -20% versus controls, respectively) at both observation time points. In female animals, slightly decreased body weights were found in the low dose group (-3% versus controls) and a statistically significant decrease has been observed in the high dose group (-10% versus controls) after 14 days. Body weight gains were slightly decreased in the low dose females (-8% and -14%) and significantly decreased in the high dose females (-39% and -42%) after day 7 and day 14 respectively.

No substance related clinical signs of toxicity were observed and no deaths occurred during the observation period.

 

For further assessment of lysmeral induced effects after continuous oral application via feed, the subsequent study , i.e. a one-generation range finding study for a two-generation study, has been performed using microencapsulated lysmeral in order to exclude stability issues and palatability induced effects when applied over a longer time period.

This type of test substance administration and the chosen study types allow to:

·        verify the relevance of the adverse testicular effects observed after bolus application via gavage compared to a continuous application via feed

·        close the gap between the observed adverse changes in testes and sperm and infertility

·        to assess the no effect levels concerning fertility for a sound risk assessment

 

In the older range finding study, the test substance was administered to groups of 10 male and 10 female young Wistar rats (F0 parental generation) via the diet (30.7% lysmeral, microencapsulated with gelatin based capsules in sunflower oil; BASF SE 2006C). In the 400, 800, 1700 and 3400 ppm group, the uptake of lysmeral via the diet was accounted to 14.0/15.0, 28.0/29.4, 62.6/62.7 and 116.8/123.2 mg/kg bw/day in males/females respectively. Due to dose adjustment during gestation and lactation, the dams received 200, 400, 850 and 1700 ppm of the test substance in feed, resulting in an uptake of 12.9/10 and 25.8/18.3 mg/kg bw/day lysmeral for the two low dose groups, respectively (no assessment of the two high dose groups was performed due to the absence of offspring). About 6 weeks after the beginning of treatment, F0 animals were mated to produce a litter (F1). The female F0 animals were allowed to deliver and rear their F1 pups until weaning (postnatal day 21). The study was terminated with the sacrifice of the F1 weanlings and F0 adult animals.

Male F0 animals showed dose dependent reduced body weights and body weight gains (5-30% and 10-40% below control, respectively) and food consumption was 15% below controls in the high dose group. Increases in relative liver weights (10-20% above control) starting at the 800 ppm dose group and decreases in relative kidney weights (15% below control) were found in the high dose group. Significant changes in clinical chemistry such as increased levels of plasma alanine aminotransferase by 20-45%, alkaline phosphatase by 30-55% and a 4 to 5 fold increase of the glutamate dehydro­genase above mean controls were observed starting at 1700 ppm. Mean gamma glutamyltransferase was increased two fold in the high dose group only compared to controls.

Testicular toxicity and spermatotoxicity, i.e. effects on sperm parameters, decreases in relative testes (30-45% below control) and cauda epididymis (30-40% below control) weights, diffuse testes degeneration and aspermia of the epididymis, were observed at the 1700 ppm and 3400 ppm groups. In the high dose group, weights of additional organs were decreased (i.e. seminal vesicle (10%) and prostate (20%) below control) and hyperplasia of Leydig cells was observed.

Maternal toxicity was manifested by decreases in body weights and body weight gain (5-10% and 10-30% below control) during/after premating in the 800 ppm group and at higher doses. During gestation and lactation, mean maternal body weights and body weight gain were approx. 10% below control in the 800 ppm group and food consumption during lactation was 20% below controls. Furthermore, significant changes in clinical chemistry were seen in all dose groups observed, i.e. a 2-8 fold increase in gamma glutamyltransferase and decreases in serum cholinesterase by 50-65% compared to controls. From 800 ppm onward, glutamate dehydrogenase was found to be increased by 5-75% as well. However, no significant changes in mean relative liver or kidney weights were observed.

In the 1700 ppm group only 1 of 8 mated females became pregnant and a relationship to the adverse effects observed in male reproductive organs is indicated.

No viable offspring has been derived from animals treated with 1700 ppm and 3400 ppm microencapsulated lysmeral. In the 1700 ppm group, the only pregnant female had only 1 implant which was resorbed. In contrast, only a slight and non-significant increase in mean implantation losses have been observed for the two lower dose groups; i.e. 16% and 11% mean losses per litter in dose groups 400 ppm and 800 ppm versus 5% in controls, respectively (see Attachment 1). No corpora lutea have been determined in this study.

A slight decrease in the mean number of delivered pups per dam (7.9 in dose group 800 ppm versus 9.4 and 8.7 in controls and dose group 400 ppm) was recorded. However, no effects on gestation and the live birth indices became evident, due to the absence of any stillborn in the lysmeral treated dose groups with offspring.

Pup survival was minimally decreased for postnatal day 0 to 4 (94% in the 800 ppm dose group versus 99% in the 400 ppm dose group and controls), and no pup mortality was observed between postnatal day 4 and 21 in all dose groups with offspring. Overall the respective viability and lactation index was not considered to be affected by treatment.

For the 400 and 800 ppm dose groups, a significant reduction in birth weights (19% and 22% below controls, respectively) and pup weight at weaning (17% and 21% below controls, respectively) has been recorded for male and female pups. Accordingly the pup body weight gain was decreased in the 400 and 800 ppm dose groups (16% and 21% below controls, respectively).

The NOAEL for fertility was set at 800 ppm (approx. 28 mg/kg bw/d lysmeral for males) based on testicular toxicity and infertility observed. A LOEL for parental toxicity (females) of <= 400 ppm and a NOAEL for parental toxicity (males) was set, since the adversity of changes in the clinicochemical findings at 400 ppm is questionable (see also Chapter “Repeated dose toxicity”).

 

In the recent range finding study, performed to ensure a sound dose selection for the main EOGRTS, groups of 10 male and female Wistar rats were treated with lysmeral formulated in alginate based microcapsules via diet (17.7% lysmeral, microencapsulated in sunflower oil; BASF SE 2017B). In the 230, 750, 2300 ppm group, lysmeral uptake in males was accounted to a mean of 2.8/2.3, 9.1/7.4, 27.5/25.1 mg/kg bw/day (pre-/postmating), respectively. A dose reduction of 50% was performed for females during lactation, and the dams received doses of 3.3-3.6, 10.6-11.9, 30.6-34.7 mg/kg bw/day during premating and gestation and 3.7, 10.7 and 21.0 mg/kg bw/day in the low, mid and high dose group, respectively. About 2 weeks after the beginning of treatment, F0 animals were mated to produce a litter (F1). The female F0 animals were allowed to deliver and rear their F1 pups until weaning (postnatal day 21), and the study was terminated with the sacrifice of the F1 weanlings and F0 adult animals.

Evidence of systemic toxicity in male F0 animals represented decreases in body weights (starting at the second week of treatment) and body weight gains of 5-11% and 45-84% below control, respectively, in the high dose group. Food consumption was reduced within the first week in mid and high dose males (7% and 9% below control, respectively). Hematological changes in mid and high dose males were observed. Total protein (mid and high dose males), albumin, globulin, cholesterol, triglycerides, sodium and calcium concentrations in serum were significantly decreased and aspartate aminotransferase activity (26% above control) was increased in high dose males. Increases in absolute and relative liver weights (14-30% above control) accompanied by increased incidences of discoloration was observed in high dose males. Liver weight increases were seen to a lower extent in low and mid dose group males (< 10% versus controls).

Indications of testicular and epididymal effects were observed in the high dose males. Weights of the cauda epididymis (-29% / -19%; absolute/relative), epididymides (-16% absolute) and seminal vesicles (-19%; absolute) were reduced compared to the respective placebo control. Histopathology revealed minimal to moderate tubular degeneration in 3 of 10 males and epididymidal changes, i.e. minimal to moderate ductal atrophy (8 /10 males), slight to moderate oligospermia (6 /10 males) and slight to moderate cellular debris (2 /10 males). The mean fraction of motile sperm accounted to 25% in high dose animals (compared to 85%, 88% and 86% in placebo control, low and mid dose males and the mean fraction of abnormal sperm was 72% versus 6-7% in the other test group males. Whereas the mean spermatid head counts in the testis were not significantly affected (i.e. 115 in high dose males versus 124 mio/g in placebo controls), the mean sperm head count in the cauda epididymis was significantly reduced in high dose males (469 mio/g) compared to 674, 760 and 640 mio/g tissue in control, low and mid dose animals. The nature of the capsule material (gelatine or alginate) and the differences in lysmeral concentrations used (30.7% versus 17.7%) in the two range findings studies might impact the overall bioavailability of the test substance lysmeral and explain the difference in effective testicular and spermatotoxic dose levels observed.     

Maternal toxicity was manifested by an initial body weight loss in the first dosing week and decreases in body weight gain (32-59% below control) during premating and gestation in the high dose females. Accordingly, body weights were significantly decreased starting at day 14 of gestation (10-16% below control) in high dose females. These effects were seen less severe in mid dose females as well. During lactation, mean maternal body weights were approx. 10% below control in the mid and high dose group, respectively, and a recovery of these body weights was observed at the end of lactation.

High dose female food consumption was reduced in the first study week (14% below control) and during lactation (44-48% below controls) and mid dose food consumption was decreased less severe mainly during the first two weeks of lactation as well. Hematological parameters were affected and decreases in serum total protein, albumin, globulin, cholesterol (high dose females), triglycerides, sodium and calcium were seen in mid and high dose females. Furthermore, increase in aspartate aminotransferase (23-47% above control) and gamma glutamyltransferase (9-24 fold above control) in the mid and high dose females and additional increases of creatinine, total bilirubin, chloride and inorganic phosphate in the high dose females were observed. 

Whereas the mating index was comparable between test groups, fertility indices were affected in the high dose animals and a reduction of the mean implantations sites were observed (see Attachment 2).  A slight and non-significant increase in mean implantation losses have been observed in the high dose dams. Consequently, a decrease in the number of litters and a mean number of delivered pups per dam (4.0 in high dose dams versus 11.1 in controls) was observed. An increase in the gestation duration was seen in the high dose group (23 days versus 22.2 days in the respective placebo control), which is considered to be a secondary consequence of the clearly lower number of pups per litter and the affected offspring, leading to a lower stimulus in starting the process of parturition. No effects on live birth indices became evident, since stillborn rates in the lysmeral treated dose groups were not increased compared to the respective placebo control.

Pup survival was decreased for postnatal day 0 to 4 (86% and 75% in the mid and high dose groups versus 99% and 95% in the low dose group and placebo control group), whereas no pup mortality was observed between postnatal day 4 and 21 in all dose groups.

A significant reduction in birth weights (17% and 18% below controls, respectively) and pup weight at weaning (13-21% and 30-32% below controls, respectively) has been recorded for mid and high dose pups. Accordingly, the pup body weight gain was decreased in the mid and high dose groups (13% and 33% below controls, respectively). A higher percentage of live females compared to live males was seen in the high dose group (69% / 31% at day 0 and 67% / 33% at day 21) versus placebo control (53% / 47% at day 0 and day 21). Due to the low number of total pups delivered in this dose group (16 pups versus 111 pups in placebo control group), this finding is considered to be a chance finding, given that changes in the sex ratio has not been observed at any other study. 

 

In the EOGRTS (according to OECD test guideline No. 443 and OECD Principles of Good Laboratory Practice), encapsulated lysmeral was administered to groups of 35 male and female Wistar rats for low, mid and control groups and to 40 male and female rats for the high dose group as a homogeneous addition to the food in concentrations of 75, 230 and 750 ppm, based on the outcome of the precedent dose range finding study (BASF 2017). The concentrations of the formulated capsules corresponded to 13, 41 and 133 ppm of the active ingredient lysmeral and a targeted nominal dose of 1, 3 and 10 mg/kg body weight/day lysmeral was aimed for. The overall mean dose of lysmeral administered to the male and female Wistar rats throughout all study phase and across all cohorts was approx. 1.4, 4.5 and 15.1 mg/kg bw/d in the 75, 230 and 750 ppm dose group, respectively. Thus, the targeted lysmeral dose levels were achieved or exceeded. A negative control group given plain diet and an additional placebo-control group was dosed with capsules without lysmeral via the diet in parallel.

TheF0 animals were treated at least for 13 days prior to mating to produce a litter (F1 generation). Pups of the F1 litter were selected (F1 rearing animals) and assigned to 7 different cohorts (seeAttachment3) which were continued in the same fashion as their parents and which were subjected to specific post-weaning examinations. F1 Cohort 1B animals selected for breeding were continued in the same dose group as their parents, and the breeding program was repeated to produce a F2 litter. The study was terminated with the terminal sacrifice of the F2 weanlings and F1 Cohort 1B parental animals.

Animals were assessed for their state of health, detailed clinical signs, body weights, food consumption, clinical pathological investigations (including thyroid hormone measurements) and urinalysis. Mating and reproductive performances were investigated in P0 and P1 animals and estrous cycling or sperm parameters were assessed. The offspring was sexed, monitored for their viability, body weights and the presence of nipple/areola anlagen and their sexual maturation was followed via vaginal opening and balanopreputial separation.

Corresponding to their cohorts, F1 animals were assessed by specific (histo-) pathological examinations. Key aspects were general organ and reproductive pathology in cohorts 1 (A/B), specific pathology of central and peripheral nervous system in cohorts 2 (A/B) and specific investigations of immune response in cohort 3. Additional cohort specific investigations included T-cell dependent antibody response to sheep red blood cells or motor activity measurements, auditory startle response and functional observational battery examination. For the add-on cohorts 4 (A/B), blood serum, erythrocytes, brain and muscle tissue were investigated for acetyl cholinesterase activities in F0 parental animals, PND 4 surplus pups, PND 22 and young adult F1-offspring.

The high-dose (nominal 10 mg/kg bw/d) produced adverse systemic effects in the F0 parental rats and F1 offspring (see Attachment 4). In F0 and F1 females of Cohort 1B,food consumptionwas consistently reduced during lactation (F0 females: 5% and F1 cohort 1B females: 12% below placebo-control), whereas this parameter remained unchanged in males and females of other cohorts. Further,body weightswere consistently reduced in high-dose F0 females during gestation and the first two weeks into lactation, which was caused by a reducedbody weight gainduring different sections of premating and gestation. No such effects were observed in the high-dose F0 males. In line, the high-dose F1 females of Cohort 1B were similarly affected. The decrease of body weight persisted throughout gestation and lactation period for the F2 litters and a reduction of body weight gain was observed during pregnancy. The body weights of the high-dose males in the different F1 cohorts were below the concurrent control throughout the in-life period after weaning (up to 11%).

Effects of the red blood cell parameters were observed in high dose F0 females (higher red blood cell counts, hemoglobin and hematocrit values) and high dose F1 males and females (higher red blood cell counts and hemoglobin values and decreased mean corpuscular volumes). An isolated decrease in mean serum acetylcholinesterase activities was seen in F0 females (-16% in mid dose and -21% in high dose group animals when compared to placebo controls), which was not confirmed in males and other peripheral tissues including erythrocytes. Further, a decrease in the mean acetyl-cholinesterase activities of the Musculus gastrocnemius in the F0 high dose males (-18% compared to placebo controls) was not considered to be an adverse lysmeral related effect.

Indications for an altered metabolism of liver cells in F0 and F1 high dose females were seen by prolonged prothrombin time (i.e. reduced synthesis of coagulation factors), increasedγ-glutamyl transferase (GGT) activity and reduced albumin levels. Significant absolute and relative weight increase of the liver weight (119% and 120%, respectively) were associated with minimal to slight centrilobular hypertrophy accompanied by minimal to slight apoptosis/single cell necrosis of hepatocytes in high-dose F0 females. Furthermore, periportal vacuolation and multinucleated hepatocytes were noted in few animals. In the female F0 mid-dose group (3 mg/kg bw/d) a significant liver weight increase (absolute: 112%, relative: 110%) was within the historical control range values and occurred without a histopathological correlate. Therefore, a treatment related effect is assumed, but was regarded as non-adverse.

A comparable picture was obtained in the F1 high-dose females (i.e. Cohort 1A), showing significant absolute and relative weight increases of the liver (126% and 128%, respectively), that correlated with minimal to slight centrilobular hypertrophy and minimal to slight apoptosis/single cell necrosis of hepatocytes. In the female mid-dose group, the significant absolute and relative liver weight increases (116% and 109% respectively) were above the historical controls. They were assessed as treatment-related and non-adverse, since no histopathological correlate was noted.

There was no evidence from clinical examinations as well as gross and histopathology, thatlysmeraladversely affected thefertility or reproductive performanceof the F0 and F1 parental animals up to and including the administered nominal high-dose of 10 mg/kg bw/d (see Attachment 5). Estrous cycle data, mating behavior, conception, gestation, parturition, lactation and weaning as well as sexual organ weights and gross and histopathological findings of these organs (specifically the differential ovarian follicle count) were comparable between the rats of all groups including control and ranged within the historical control data. A slightly (non-statistically) higher mean percentage of abnormal sperms (9.8%) in the high-dose F0 males above the placebo percentage (6.3%) and the historical control range (6.0-6.6 %) was observed. However, there were no findings in sperm motility (86% versus 88% in placebo control) or sperm head counts in testes (108 mio/g versus 102 mio/g in placebo control) and epididymidis (717 mio/g versus 723 mio/g in placebo control) of these animals. Further, the corresponding F1 offspring males did not show these effects, and there were no indications whatsoever of an impairment of fertility in the F0 and F1 generations. Thus, there is no evidence in the frame of this study that this marginal non-statistical increase is of toxicological relevance.

In F0 animals, implantation was not affected and no postimplantation losses, indicative for lysmeral induced intrauterine embryo/-/fetolethality, was observed for F0 and F1 animals (Attachment5). In high dose F1 females, an observed statistically significant decrease in the mean number of implantation sites (10.5 implants/dam) was within the historical control range (9.4-13.9 implants/dam), was not observed in F0 animals and no other findings supportive of a potential effect on fertility (i.e. follicle numbers, sperm quality) exist, making this finding an incidental event. The finding of decreased mean numbers of delivered F2 pups in these animals is associated with the lower number of implants and not an independent finding. The live birth index was not affected in these and all other treated animals.   

Thepup body weight developmentof the high-dose F1 and F2 offspring was affected, as these offspring weighed about 14-16% less than control after birth and did not recover until weaning (weights about 10% below control at PND 21; see Attachment 4). These body weight effects had no evident influence onpostnatal pup survival, neither during early lactation nor later. There was a higher number of cannibalized high-dose F1 pups around PND 1 (20 vs. 1 in placebo control), however, most of these cannibalizations (16) were clustered in 2 litters, and there were no such effects evident in the high-dose F2 offspring. Thus, these were most likely incidental events. Postnatal survival after PND 4 of the offspring of all test groups until weaning remained unaffected by lysmeral administration. Furthermore, clinical and/or gross necropsy examinations of the weaned F1 pups revealed no adverse findings.

 

Lysmeral did not affect blood thyroid hormone levelsin F0 parental animals and F1 offspring. The sex ratio of F1 and F2 pups at day of birth and on PND 21 was not affected by lysmeral administration. In contrast to F1 pups, the anogenital distance of the high-dose F2 pups was statistically significantly below the concurrent placebo-control values (about 4%, respectively) and at the lower limit of the historical control range (2.99-3.15 mm for males, 1.48-1.60 mm for females). In turn,anogenital indicesof high-dose male and female F1 and F2 pups were found to be somewhat above the concurrent controls. These effects were considered as a secondary consequence to the lower pup body weights and represent not an independent treatment effect. In addition, further check of sensitive marker of potential endocrine-mediated imbalances (i.e.presence of nipples/areolas) revealed no lysmeral related effects. For theonset of puberty,no effect of lysmeral treatment was noted for vaginal opening in the offspring. A statistically significant delay in preputial separation beyond the placebo-control (41.6 days) was observed in the high-dose male F1 offspring (42.3 days), but the delay is well within the historical control range (40.5-45.2) and can in addition be attributed to a general delay in the development of high-dose male F1 offspring. It is thus not considered to be a direct test substance-related effect on male sexual maturation. Overall, the sexual development of the offspring was not impaired by lysmeral administration.

In male PND 4 pups and PND 76 females of the high dose group, lower peripheralacetylcholinesterase (AChE) activities were seenin serum, erythrocytes and diaphragm tissue (up to 50% of the means in placebo control animals). Such changes were not found in the respective other sex. No corresponding clinical signs ofdevelopmental neurotoxicitywere evident in male and female F1 offspring at any dose level. There were no compound related effects on motor activity, auditory startle habituation, and in the field observation battery following exposure to the test compound in these animals. The only notable finding in neurobehavioral testing were lower maximum amplitudes in the auditory startle response test of the high-dose F1 males. However, in comparison to corresponding negative control data (plain diet) and high-dose F1 female data the placebo-control values were rather unusually high, therefore this finding appears to be irrelevant due to high placebo control incidences. Moreover, no such findings were noted in the high-dose F1 females and no corresponding effects were recorded for startle response latency. Thus, this was not regarded as an effect caused by the treatment. In addition, regardingneuropathology, brain weight determination, brain length and width measurements as well as brain morphometry and neuropathological examination by light microscopy did not reveal any neurotoxicological treatment-related findings.

There was no evidence that lysmeral produced anydevelopmental immunotoxicity. Neither T-cell dependent anti-SRBC IgM antibody response, nor absolute and relative lymphocyte subpopulation cell counts in the spleen tissue (B-, T-lymphocytes, CD4-, CD8-T-lymphocytes and natural killer (NK) cells) displayed any treatment-related changes.

Based on the findings of this extended one-generation reproduction toxicity study theNOAELforgeneral, systemic toxicityis set at 3 (4.5) mg/kg bw/d for the F0 and F1 parental as well as adolescent animals, based on evidence for distinct liver toxicity, as well as corresponding effects on food consumption, body weights and clinical pathological parameters. TheNOAELforfertilityandreproductive performancefor the F0 and F1 parental rats is 10 (15.1) mg/kg bw/d, and theNOAELfordevelopmental toxicityin theF1 and F2 progenyis 3 (4.5) mg/kg bw/d, based on reduced pup body weights in the F1 and F2 offspring, which were observed at the LOAEL of 10 (15.1) mg/kg bw/d. As these weight reductions were only observed in the presence of maternal toxicity, including lower weight gain during pregnancy, they are not regarded as independent effect of the treatment. Although an inhibitory effect of the high dose of lysmeral on the peripheral AChE activity in pups and adolescent rats cannot entirely be excluded, there were no corresponding effects evident in the neurobehavioral or neuropathological examinations. Accordingly, theNOAELfordevelopmental neurotoxicity and developmental immunotoxicityfor the F1 progeny is 10 (15.1) mg/kg bw/d.

 

To facilitate comparison with testicular toxicity observed in repeated dose studies a summary of the various studies previously discussed in chapter "repeated dose toxicity" is provided in Attachment 6 focussing on (no-)effect dose levels for testicular toxicity.

An in vitro study on chorionic gonadotrophin stimulated testosterone secretion of primary rat Leydig cells after treatment with lysmeral did not indicate any specific inhibition at the level of testosterone secretion. It was therefore assumed, that the atrophic changes of the testes in vivo might not be attributed to a specific and direct action of lysmeral on the main function of Leydig-cells which is the secretion of testosterone (Roche 1994).

In vitro tests using estrogen responsive MCF7 human breast cancer cell line and human recombinant ER alpha and ER beta gave indications for an estrogenic activity of lysmeral (Charles 2009). However a much lower potency of lysmeral concentrations have been observed in these test systems when compared to estradiol activity. In competitive ER binding assays, a 3 000 000 molar excess of lysmeral was needed to archive a moderate inhibition of estradiol binding by 15-50% and the responses to lysmeral were found to be lower than respective estradiol responses even at 10 000 fold higher lysmeral concentrations. Furthermore, lysmeral is very rapidly metabolized in vivo (and a quantitatively relevant systemic interaction of lysmeral with the estrogen receptor in vivo is questionable. In addition, an extensive in vivo database on lysmeral, which includes repeated dose toxicity studies in various laboratory species, a developmental toxicity study, two one-generation range finding studies and a key extended one-generation reproductive toxicity study do not confirm the in vitro estrogenic activity findings presented in this publication. Overall, the moderate estrogen receptor mediated effects observed after incubation of high lysmeral concentrations in vitro are of questionable relevance for the in vivo situation.

 

A number of studies with pharmacological agents have demonstrated the pivotal nature of RARαsignalling in maintaining functionality in the testis and ablation of the RARαresults in testicular toxicity and spermatotoxicityin rodents similar to lysmeral related findings (Chung et al, 2016, 2013). In order to assess the RAR pathway as a putative mode of action for lysmeral and TBBA induced testicular toxicity, these substances have been tested in a reporter gene assay to detect potential agonistic or antagonistic activities directed against the human RARα, RARβand RARγ(BASF 2017C). The CHO cells used, express a receptor hybrid in which the native N-terminal DNA binding domain (DBD) has been replaced with a yeast Gal4 DBD and the reporter vector comprises the firefly luciferase gene functionally linked to the Gal4 upstream activation sequence. Neither incubation with lysmeral nor TBBA at a concentration range of 0.00128 - 100 µM resulted in a relevant induction of luciferase activity for any of the nuclear receptor investigated. Furthermore, lysmeral and TBBA did not influence the reporter gene expression in the presence of suitable receptor agonists (i.e. 9-cis and all-trans retinoic acid) at non-cytotoxic concentrations, demonstrating the absence of a potential antagonistic effect towards human RARα, RARβand RARγ. Therefore, these results indicate, that lysmeral and TBBA mediated testicular and spermatotoxicity are not mediated via the RARα, which is highly conserved between species.

 

Testicular toxicity induced by p-tert-butylbenzoic acid (TBBA).

Clear evidence of adverse testicular and spermatotoxic effects, being identical in quality to lysmeral induced testicular toxicity, have been observed in repeated dose studies, testicular toxicity screening studies and fertility studies with p-tert-butylbenzoic acid (TBBA). For comparison, a summary is provided in Attachment7focussing on (no-)effect dose levels for testicular toxicity.

 

Testicular toxicity induced by p-tert-butyl-benzaldehyde (TBB) and p-tert-butyltoluene (TBT).

As described in Chapter "Toxicokinetics", oral administration of two other substances, namely p-tert-butyl-benzaldehyde (TBB) and p-tert-butyltoluene (TBT), resulted in the formation of systemic TBBA. Therefore, both compounds and lysmeral share TBBA as a common metabolite. As outlined below, oral TBB and TBT administration to rats resulted in adverse testicular effects, being identical to the testicular findings observed for TBBA and lysmeral. In line with findings for lysmeral, the rat represents the most sensitive species and evidences for testicular effects in dogs exist, whereas other species, i.e. mouse or guinea pigs, show low susceptibility for testicular toxicity.

For comparison, a summary is provided in Attachment 8 focussing on (no-)effect dose levels for testicular toxicity.

Effects on developmental toxicity

Description of key information

Developmental toxicity, oral (OECD 414, GLP): NOAEL = 5 (4.1) mg/kg bw/day for maternal and prenatal developmental toxicity.

Additional information

For the assessment of developmental toxicity of lysmeral (2-(4-tert-butylbenzyl) propionaldehyde), a study in Wistar rats was performed in accordance with the OECD test guideline No. 414 in line with the OECD Principles of Good Laboratory Practice (BASF SE 2004). Lysmeral (analytical purity 98.1%) was administered via gavage at nominal doses of 5, 15 and 45 mg/kg bw/day from day 6 through day 20 post coitum (p .c.). The effective dose levels amounted to 4.1; 12.7 and 40.7 mg/kg body weight/day.

Clear signs of maternal toxicity were observed starting at the mid dose level. The high dose animals showed transient salivation. Slight but statistically significant reduction of mean food consumption (18% below controls) was observed in the high dose group on day 6-8 p.c. By study termination, food consumption was comparable to control animals (see Attachment 9). Although no evident decrease in food consumption was detectable in mid dose animals, mean maternal weight gains significantly decreased on day 6-8 p.c of about 56 % below control which recovered during the study period.

In high dose animals, a statistically significant mean body weight loss was observed on day 6-8 p.c and the mean body weight gain over the entire treatment phase was found to be about 25% below controls. Furthermore, a statistically significant reduction of mean body weights on day 13 - 20 p.c. (about 7% below controls at study termination) was found. In line, the corrected body weight gain was statistically significantly lower (about 32% below control), representing a direct, substance-related sign of maternal toxicity.

Concerning clinical chemistry, increases in mean alanine aminotransferase levels (20-30% above control) and decreases in serum cholinesterase levels (20-45% below control) were found, starting from the mid dose. In the high dose group, mean glutamate dehydrogenase levels were found to be 79% above controls.

Increases in absolute and relative liver weights (10% and 10-20% above controls, respectively) were found at all dose levels, however, due to the lack of changes in respective clinical parameters, only the liver weight changes in the mid and high dose group were considered as adverse. In high dose animals, reduced mean uterus weights (20% below controls) were observed.

Gestational parameters such as number of corpora lutea, implantation sites and preimplantation loss were not influenced by the test substance at any dose level (see Attachment 10). However, mean postimplantation losses (mainly early resorptions) were found to be increased significantly in the high dose group. In animals receiving 45 mg/kg bw per day, mean resorptions accounted to 15.1% per dam compared to 4.4%; 4.7% or 4.9% resorptions at 0, 5 or 15 mg/kg body weight/day, being outside the historical control range. Subsequently, a decrease in the mean number of fetuses and live fetuses per dam became evident in the high dose group, i.e. 7.4, when compared to controls or lower dose groups (8.1; 8.2 and 8.8 at 0; 5 and 15 mg/kg body weight/day). These high dose findings were slightly below the historical controls of the mean number of fetuses per dam. Sex distribution and placental weights were not influenced by the test substance. No dead fetuses, abortions or premature births have been observed in control and all dose groups of this study.

 

Sporadic malformations were observed in 3 out of 170 or 1.8% of all high dose group fetuses. Three out of 23 or 13% of the litters were affected in this dose group. Findings were reported as anasarca with a small spleen, polydactyly due to a supernumerary phalanx and cervical hemivertebra. The mean percentages of affected fetuses per litter with total malformations amounted to 0; 0; 0 and 2.4% at 0; 5; 15; or 45 mg/kg bw/day. These findings are not regarded as sufficient evidence for a selective teratogenic effect of lysmeral, since the observed malformations lacked a consistent pattern and were not found in any other dose group. Furthermore, they occurred in very few of the large number of examined fetuses and its low incidence is to be found within the respective control range of the given testing laboratory; i.e. affected fetuses (0-2.7%), affected litters (0-25%) and affected fetuses/litter (0-2.79%).

 

External variations were not observed and soft tissue variations (dilated renal pelvis, ureters and/or cerebral ventricles) occurred in a dose independent manner in all test groups including control animals. The mean percentages of affected fetuses per litter with total soft tissue variations amounted to 7.8%, 7.7%, 3.6% and 5.3%, in controls, low, mid and high dose animals, respectively.

 

Skeletal variations were seen in all tested dose groups and litters including controls. Every litter was affected, resulting in 100% litter incidence for all groups assessed. Although within the historical control range and lacking a relation to dosing, a statistically significant increase in mean percentages of affected fetuses per litter were found in mid and high dose animals (99.1% and 98.3% in mid and high dose groups versus 89.1%, 92.%, in controls and the low dose group, respectively). The fetal incidence of skeletal variations in total was increased accordingly.

As shown in Attachment 9, these skeletal variations mainly represent delays and minor disturbances in ossification of the skull, sternebrae and pubic girdle. Non or incompletely ossified structures were statistically significantly increased in the mid and/or high dose group compared to the concurrent control, and incidences of single findings were above the study related and/or current historical control range, i.e. incomplete ossification of supraoccipital, sternebra, pubis or unossified sternebra. As described in detail for skeletal structural variations further down, these findings coincide with decreased mean fetal body weights. These were dose dependently lowered in the mid (10% below controls) and high dose groups (20% below controls). Approximately 85 % of the fetuses in the high dose group and 50 % in the mid dose group showed body weights below one standard deviation of the control group mean body weight (< 3.3 g). Decreases in mean maternal body weight gains or even body weight losses in combination with decreased food consumption occurred at these dose levels as well. 

 

Although significantly increased, structural skeletal variations such as supernumerary (14^th) ribs were found in control and dosed animals at high incidences within historical control ranges, whereas incidences for a supernumerary thoracic vertebra (14^th) or a misshapen sacral vertebra (1^stsacral arch; right or left side) were increased in the high dose group fetuses above historical control ranges.

One fetus of one litter in the control group showed a supernumerary thoracic vertebra which places the study specific control group parameters into the lower part of the updated current historical control range. High dose group incidences are well above the historical control range whereas mid dose group values are within the updated current historical data range. Both dose groups contained single litters with multiple fetuses showing a supernumerary thoracic vertebra. Findings from these litters are the main driver for the increased fetal incidences and affected fetuses/ litter given in Attachment9. Out of 23 litters, a single mid dose litter with 4 affected fetuses and 2 mid dose litters with 2 affected fetuses each were observed. Similarly, a single litter contained 4 affected fetuses and 3 litters contained 2 affected fetuses each in the high dose group. For each of these litters, a decreased mean litter weight has been observed, i.e. ≤3.3 g in the mid dose group and ≤3.0 g in the high dose group. The respective dams showed a decreased body weight gain in the mid dose and a body weight loss in the high dose group on day 6-8 p.c. Furthermore, increases in absolute and/or relative liver weights and changes in clinical chemistry were noted. Three additional litters contained 1 fetus with such a variation in the mid and high dose group, being within the range of incidence per litter also observed in the control group. One additional vertebra in the thoracolumbar region is generally considered to be a variation in the rat and occurs quite frequently in the rabbit (Solecki 2001).

 

A misshapen sacral vertebra was found in 1 fetus of 2 control group litters each (fetal incidence of 2.1% and litter incidence of 8.7%). Incidences in the high dose group exceeded the historical control data. Two of 23 litters contained 2 fetuses with the named variation. In addition, 7 litters contained 1 affected fetus each, leading to an increase in the litter incidence. For all these litters, mean fetal body weights were decreased (≤3.0 g/litter) and respective dams experienced a body weight loss and decreased food consumption on d6-8 p.c., changes in clinicochemical parameters and increases in absolute and/or relative liver weights. Furthermore, the observed structural changes in the morphology of the sacral vertebra were minor and provide, together with the high control animal incidence, the rationale for classifying them as variations.

As described above, the observed skeletal variations are well correlated to the statistically significantly decreased mean fetal body weights and maternal adverse effects in the respective dose groups. Such a delay in fetal body weight development and subsequent increase in skeletal variations is considered to be caused by the evident maternal toxicity.

Significantly increased mean percentages of affected fetuses per litter with incomplete ossification of parietal (29.8% versus 11.7% in controls) or interparietal (36.0% versus 20.7% in controls) with unchanged cartilage was observed only in the low dose group and considered to be spontaneous in nature, since no dose dependency was observed. Discoloration of fetal livers was evident in some mid and high dose animals with a mean percentage of affected fetuses per litter of 1.7% and 15.5%, respectively, being in line with the liver changes of the respective dams.

Based on these findings, the NOAEL is set at 5 (4.1) mg/kg bw/day for maternal and prenatal developmental toxicity.

Further information on developmental toxicity can be deduced from the respective endpoints of the one-generation range-finding studies and the EOGRTS. In the older range-finding study, no viable offspring has been derived from animals treated with 1700 ppm (63 mg/kg bw/d) and 3400 ppm (120 mg/kg bw/d) microencapsulated lysmeral (BASF SE 2006C; for further details see “Additional information” in the section “Fertility”). In the 1700 ppm group, the only pregnant female had only 1 implant which was resorbed. In contrast, only a slight and non-significant increase in mean implantation losses have been observed for the two lower dose groups; i.e. 16% and 11% mean losses per litter in dose groups 400 ppm (10-15 mg/kg bw/d) and 800 ppm (18-29 mg/kg bw/d) versus 5% in controls, respectively (seeAttachment1). A slight decrease in the mean number of delivered pups per dam (7.9 in dose group 800 ppm versus 9.4 and 8.7 in controls and dose group 400 ppm) was recorded. However, no effects on gestation and the live birth indices became evident, due to the absence of any stillborn in the lysmeral treated dose groups with offspring. Pup survival was minimally decreased for postnatal day 0 to 4 (94% in the 800 ppm dose group versus 99% in the 400 ppm dose group and controls), and no pup mortality was observed between postnatal day 4 and 21 in all dose groups with offspring. Overall the respective viability and lactation index was not considered to be affected by treatment.

No effects in sex ratios have been observed and pup necropsy revealed only sporadic and non-dose related findings, including post mortem autolysis, situs inversus, hemorrhagic thymus, dilated renal pelvis and a small kidney. The overall pup incidence for these observations was 6.4%, 2.6%, and 1.3% in controls and dose groups 400 ppm and 800 ppm respectively.

For the 400 and 800 ppm dose groups, a significant reduction in birth weights (19% and 22% below controls, respectively) and pup weight at weaning (17% and 21% below controls, respectively) has been recorded for male and female pups. Accordingly the pup body weight gain was decreased in the 400 and 800 ppm dose groups (16% and 21% below controls, respectively).

In the recent range finding study (BASF SE 2017B), a slight and non-significant increase in mean postimplantation loss have been observed in the high dose dams (2300 ppm, 25-35 mg/kg bw/d; (see Attachment 2). No effects on live birth indices were observed, since stillborn rates in the lysmeral treated dose groups were not increased compared to the respective placebo control. An observed decrease in the number of delivered pups per dam (4.0 in high dose dams versus 11.1 in controls) can predominantly be attributed to lower numbers in implantation sites and affected fertility indices.

Pup survival was decreased for postnatal day 0 to 4 (86% and 75% in the mid and high dose groups versus 99% and 95% in the low dose group and placebo control group), whereas no pup mortality was observed between postnatal day 4 and 21 in all dose groups.

A significant reduction in birth weights (17% and 18% below controls, respectively) and pup weight at weaning (13-21% and 30-32% below controls, respectively) has been recorded for mid (750 ppm, 7-12 mg/kg bw/d) and high dose pups. Accordingly the pup body weight gain was decreased in the mid and high dose groups (13% and 33% below controls, respectively).

Taken together, significant lysmeral related developmental toxicity in the one-generation rangefinding studies have been observed predominantly in terms of decreased pup weights and indications for effects on early pup survival exist. These dose levels were associated with impaired maternal body weight development and food consumption during premating phase, gestation and lactation, and resulted in changes of clinical chemistry and hematological parameters (for further details see “Additional information” in the section “Fertility”).

In the EOGRTS, nominal doses of 1, 3 and 10 mg/kg body weight/d lysmeral (approx. 1.4, 4.5 and 15.1 mean mg/kg bw/d throughout all study phase and across all cohorts) were administered to groups of 35-40 male and female Wistar rats as a homogeneous addition to the food (BASF 2017).

No postimplantation losses, indicative for lysmeral induced intrauterine embryo/-/fetolethality, was observed for F0 and F1 generation animals up to the highest dose level tested (Attachment 5). A decrease in the mean number of implantation sites (10.5 implants/dam) in high dose F1 females was within the historical control range, was not observed in respective F0 animals and no other findings supportive of a potential effect on fertility (i.e. follicle numbers, sperm quality) exist, making this finding an incidental event. The finding of decreased mean numbers of delivered F2 pups in these animals is associated with the lower number of implants and not an independent finding. The live birth index was not affected in these and all other treated animals.   

 

Thepup body weight developmentwas affected in high-dose F1 and F2 offspring (about 14-16% less than control after birth and no recovery until weaning; see Attachment 4). Organ weight changes (brain, thymus and spleen) were observed at this dose and were considered to be secondary to the changes in body weight, rather than independent findings.

 

No evident influence onpostnatal pup survivalduring early lactation nor later was observed. A higher number of cannibalized high-dose F1 pups around PND 1 were mostly clustered in 2 litters and no such effects evident in the high-dose F2 offspring. Thus, these were most likely incidental events.

 

Blood thyroid hormone levelsin parental animals offspring were not influenced by lysmeral administration and the sex ratio of F1 and F2 pups was not affected. A decrease in the anogenital distance of the high-dose F2 pups (not observed in F1 pups), and a slight increase of theanogenital indicesof high-dose male and female F1 and F2 pups were considered as a secondary consequence to the lower pup body weights and represent not an independent treatment effect. Further marker of potential endocrine-mediated imbalances (i.e.presence of nipples/areolas) revealed no lysmeral related effects. A delay in preputial separation in the high-dose male F1 offspring, was within the historical control range and can be attributed to a general delay in the development of high-dose male F1 offspring. No effect of lysmeral treatment for vaginal opening was noted. Overall, the sexual development of the offspring was not impaired by lysmeral administration.

 

During the course of development, i.e. in high dose PND 4 male pups and PND 76 females, lower peripheralacetylcholinesterase (AChE) activities were observedin serum, erythrocytes and diaphragm tissue. However, no corresponding signs of developmental neurotoxicity were observed. There was no evidence that lysmeral produced any developmental immunotoxicity.

 

Overall, developmental toxicity observed in the EOGRTS represent only reductions in pup body weights in the high dose F1 and F2 offspring (10 mg/ kg bw/d nominal; approx. 15 mg/kg bw/d ingested). This dose level resulted in adverse maternal liver effects, effects on food consumption, body weights and clinical pathological parameters (for further details see “Additional information” in the section “Fertility”).

1. Solecki et al. (2001); Harmonisation of rat fetal skeletal terminology and classification. Report of the third workshop on the terminology in developmental toxicology Berlin, 14-16 September 2000. Reproductive Toxicology 15: 713-721.

Toxicity to reproduction: other studies

Additional information

Formation of para-alkyl benzoyl-CoA by lysmeral and lysmeral-like materials and correlation to rat male reproductive toxicity.

Lysmeral and its metabolite TBBA is transformed to TBBA-CoA conjugates in hepatocytes (Givaudan 2017, 2020). Additional testing of a wide variety of aldehydes, benzoic acids and other chemicals potentially transformed to benzoic acid metabolites in rat hepatocytes revealed, that chemicals with a para-substituent at the benzyl ring accumulate alkyl-benzoyl-CoA conjugates in a similar fashion as lysmeral and TBBA. All chemicals with this metabolic outcome (lysmerylic acid, lysmerol, BHCA, PMHCA, PHCA, iBMHCA, TBT and p-isopropyl benzoic acid) were reported to cause testicular and spermatotoxic effects in the rat.

In contrast, meta substituted substances such as meta-lysmeral and m-iP2MHCA and other structurally related substances (Floralozone, Tropional, Fennaldehyde, Jasmorange, Nympheal, benzoic acid, p-hydroxy benzoic acid and ethylparaben) showed no accumulation of corresponding alkyl-benzoyl-CoA conjugates especially at the 22-hour time point and did also not affect male reproductive organs such as the testis. Thus, a very strong correlation exists between the metabolic formation of benzoyl-CoA at a sustained elevated level in hepatocytes and spermatoxic and testicular effects in the rat.

Lysmeral and TBBA exposure disrupts CoA dependent intracellular processes which leads to the inhibition of hepatic lipogenesis and gluconeogenesis. Metabolome studies in vivo in rats with lysmeral and lysmerol revealed a common decrease of lipids, fatty acids and fatty acid related metabolites, whereas the non-testicular toxicant meta-lysmeral led to a divergent metabolite pattern especially for complex lipids such as sphingolipids, ceramides and phosphatidylcholines. Complex lipids such as very-long-chain-polyunsaturated fatty acid (VLCPUFA) or ceramides, sphingolipids and phosphatidylcholines containing these PUFAs are present in high amounts in mammalian sperm and play an important role for spermatogenesis. These lipids are considered to stabilize cellular membranes with high curvature such as the rims of the sperm head and provide membrane flexibility needed for efficient sperm formation. Defects in enzymes relevant for synthesis of such lipids are associated with testes toxicity and male infertility. This has been shown in transgenic mice with disrupted enzymes such as FADS2 desaturase, ELOVL2 elongase or LPAAT3 acyltransferase (Stroud 2009; Zadravec 2011; Iizuka-Hishikawa 2017). Disruption of these enzymes are associated with decreases in n-6 and n-3 polyunsaturated fatty acids (PUFAs) such as arachidonic acid, docosapentanoic acid and docosahexanoic acid and result in disturbances of polyunsaturated ceramide, sphingolipid and phosphatidylcholin formation. As seen for the Elovl2-/-mice, both Sertoli cells and the Leydig cells appeared normal in the interstitium, whereas primary spermatocytes degenerated, formed multinucleated giant cells and a complete arrest of spermatogenesis was observed. The data imply, that ELOVL2 synthetized VLCPUFAs are essential membrane components for normal completion of spermatocyte cytokinesis, accumulating in sphingolipids with these VLCPUFAs. For the other ko. mice, normal spermatozoa formation was disturbed and these animals proved to produce no offspring.

Overall, complex lipids are essential and the disturbance in formation of these lead to disruption in spermatogenesis which results in male infertility. The formation of these lipids is dependent on CoA and disturbances by CoA-TBBA complex formation represent a mode of action of lysmeral and TBBA induced testicular toxicity. This is further supported by the fact, that lysmeral induced liver toxicity - which is also caused by disruption of CoA dependent intracellular processes - occurs at doses leading to testicular/spermatotoxic effects.

Para-benzoyl-CoA conjugates were formed in human and rabbit hepatocytes qualitatively in vitro (Givaudan 2020). However, lower levels were initially formed for Lysmeral especially in human hepatocytes, and a strong clearance over time was noted for both Lysmeral and PMHCA in rabbit and human cells. Thus, lower levels of the p-benzoyl-CoA conjugates were detected after 22 h in both rabbit and human hepatocytes, while sustained accumulation at higher levels was found in rat hepatocytes. This species difference observed reflects the species difference for effects on sperm formation and function in vivo between rats and rabbits. Furthermore, the difference in the kinetic profile between the different species for the para-substituted analogues is very similar to the difference in the kinetic profile in rat hepatocytes between reprotoxic and non-reprotoxic derivatives. Whereas Lysmeral reduced the endogenous C8-CoA formation in rat hepatocytes, no effects were observed in human and rabbit hepatocytes indicating specific interference with lipid metabolism in rat hepatocytes only. Due to the observed effects on endogenous C8-CoA formation in vitro, interference with lipid metabolism seems to be the likely cause of toxicity. Taken together, there is strong correlative evidence that sustained accumulation of p-alkyl-BA-CoA after 22 h in hepatocytes is predictive for the toxic outcome for a given derivative and in a given species, and that the rat is most susceptible to adverse reprotoxic effects.

 

Further toxicological data on position isomers of tert-butyl benzoic acid (TBBA) have been generated in three ex vivo study using a 3D cell culture with primary seminiferous tubules from juvenile Sprague Dawley rats (Kallistem 2018-2020; 1501-03, 1501-08, 1501-10) or a from a transgender male (Kallistem 2019; 1501-09), who underwent castration (Bio-Alter®). Cultures were incubated with para/p- or meta/m-TBBA and cytotoxicity, blood-testis barrier functionality (for 2/3 rat Bio-Alter® studies only) and cell numbers of different somatic and germ cell populations have been quantified. In addition, the content of TBBA conjugated with CoA has been assessed in cell culture lysates of the rat (2 studies) and human cultures (1 study). For the rat Bio-Alter® test setup additional endogenous metabolite profiling has been performed in the cell lysates (1 study).

A clear difference was observed between p- and m-TBBA concerning the endpoints for meiotic processes, i.e. decreased numbers of total germ cells, middle to late pachytene spermatocytes, secondary spermatocytes and round spermatids. These results were confirmed in 3 comparable follow-up experiments, using the same testing conditions. In line, the respective Coenzyme A conjugate (p-TBBA-CoA) was detectable, whereas no levels of CoA conjugated with the meta-substituted benzoic acid derivative (m-TBBA-CoA) was found. In the endogenous metabolite profiling, principal component analyses show, that p-TBBA treatments are clearly separated from controls and m-TBBA treatments and a clear metabolic difference of p-TBBA is evident irrespective of dose and time.

In contrast, no clear-cut differences in germ cell populations were observed between p-TBBA and m-TBBA in the human Bio-Alter® model. Low proportions of germ cell populations present and dose- or time-independent changes question the relevance of findings of single test concentrations. However, neither p-TBBA nor m-TBBA decreased the number of germ cells and spermatogonia and no clear dose related decreases in the numbers of pachytene spermatocytes, secondary spermatocytes and round spermatids were observed for both test substances. In all culture samples dosed with p-TBBA, p-TBBA-CoA was below the limit of quantification and no m-TBBA-CoA was detectable.

A related p-alkyl-BA-CoA conjugate was detected in the testes and liver of PMHCA treated rats in vivo, which further links these in vitro observations to the in vivo situation and confirms the in vitro findings as no in vitro artefact (Givaudan 2020).    

In conclusion, the present data provide strong indications, that the formation of para-alkyl-benzoyl-CoA conjugates is an essential step for the observed male reprotoxic rat effects in this close structural chemical group due to the strong correlation of elevated and stable para-alkyl-benzoyl-CoA levels in plated rat hepatocytes and testicular toxicity/spermatotoxicity. Furthermore, liver toxicity in vivo correlates with male reproductive toxicity at comparable doses as well. The liver toxicity is likely to be caused by the disruption of CoA dependent metabolic processes due to conjugation with p-alkyl-benzoic acids. This conjugation further represents a critical hallmark of the reproductive outcome and serves as a mode of action for lysmeral/TBBA induced testes toxicity and spermatotoxicity. Besides the differences in the formation of TBBA levels, the kinetics of TBBA-CoA conjugation fundamentally differs in human (and rabbit) hepatocytes in terms of lower concentrations and a transient and rapid decrease compared to the rat, which strongly indicates, that the observed metabolic fate of lysmeral is rat specific and the testicular toxicity/spermatotoxicity is a species-specific effect with little relevance for humans.

In support to the in vivo data with PMHCA treated rats,  the rat Bio-Alter® studies confirmed, that TBBA-CoA conjugates are formed in reproductive tissues and later stage germ cell populations decrease in the presence of p-TBBA but not with m-TBBA. The levels of TBBA-CoA conjugates detected are approx.100 fold lower when compared to in vitro dosed liver hepatocytes, however the same differences between the position isomers (para vs. meta) observed in hepatocytes in vitro have been detected in this ex vivo system on reproductive tissues as well. In contrast to meta-TBBA, concentrations of para-TBBA affecting spermatogenic processes lead to detectable CoA-conjugate levels. The clear difference in the potency of para-TBBA and meta-TBBA to affect spermatogenesis in the Bio-Alter® system supports the findings from available in vivo data (i.e. para-TBBA and para-Lysmeral (metabolized to para-TBBA) vs. meta-Lysmeral concerning testes/sperm effects). A generic assessment of the metabolome indicated an impact on lipid biosynthesis and energy production, which further supports the hypothesis that the initiating event for the toxicity of p-TBBA is the conjugation of CoA, reducing its availability for cellular metabolism. Neither TBBA-CoA nor a clear dose related decrease of later stage germ cell populations were observed human Bio-Alter® model, further supporting the species specificity and low human relevance of Lysmeral related testicular toxicity.

Overall, the formation of systemic TBBA is a key metabolic event for Lysmeral-induced testicular toxicity. A strong correlation between the formation of TBBA-CoA conjugates in rat hepatocytes, disruption of lipid synthesis and testicular toxicity has been found, which is further supported by the data of the ex vivo study on primary rat seminiferous tubule cultures. Accordingly, these data strengthen our conclusion, that the conjugation of TBBA with CoA in an extended manner represents the species specific mode of action for Lysmeral-induced testes toxicity and spermatotoxicity.

1. Iizuka-Hishikawa et al., 2017; Lysophosphatidic acid acyltransferase 3 tunes the membrane status of germ cells by incorporating docosahexaenoic acid during spermatogenesis. J. Biol. Chem.; Vol 292(29), pp. 12065–12076.


2. Stroud et al., 2009; Disruption of FADS2 gene in mice impairs male reproduction and causes dermal and intestinal ulceration. Journal of Lipid Research; Vol 50, pp. 1870-1880.


3. Zadravec et al., 2011; ELOVL2 controls the level of n-6 28:5 and 30:5 fatty acids in testis, a prerequisite for male fertility and sperm maturation in mice. Journal of Lipid Research; Vol 52, pp. 245-255.

 

Reproductive toxicity study on the impurity 3-(m-tert-butylphenyl)-2-methyl-propionaldehyde (CAS 62518-65-4).

An enhanced one-generation reproduction toxicity study with the impurity 3-(m-tert-butylphenyl)-2-methylpropionaldehyde (CAS 62518-65-4) has been performed. This impurity is found to be present at a concentration above 0.1% in the technical grade Lysmeral Technisch, which is only used as intermediate under strictly controlled conditions. There is no wide dispersive use of Lysmeral Technisch and no consumer exposure.

 

In this enhanced one-generation reproduction toxicity study according to OED 415 and GLP) 3-(m-tert-butylphenyl)-2-methylpropionaldehyde has been administered orally via gavage to Wistar rats at doses of 0, 50, 150 and 450 mg/kg bw/d in olive oil (BASF 2011; 77R0875/08070). At least 74 days after the beginning of treatment, F0 animals were mated to produce a litter (F1 animals). Mating pairs were from the same dose group and F1 animals were continued in the same dose group as their parents. Groups of 25 males and 25 females, selected from F1 pups to be brought up until sexual maturity, were treated with 3-(m-tert-butylphenyl)-2-methylpropionaldehy at dosages of 0, 50, and 150 mg/kg bw/d post weaning. The high-dose group was treated with 450 mg/kg bw/d but consisted only of 10 males and 10 females, because high pup mortality precluded to have the regular group size in this test group.

The only relevant clinical observation was temporary salivation during a short period after dosing, which is considered to be test substance-induced. From the temporary, short appearance immediately after dosing it is likely, that this finding was induced by a bad taste of the test substance or local affection of the upper digestive tract. It is, however, not considered to be an adverse toxicologically relevant finding.

In the high-dose (450 mg/kg bw/d) F0 generation females food consumption was significantly increased during premating/gestation on the one hand and significantly decreased during lactation on the other hand. The effect on body weights / body weight gain, however, was rather limited and remained non-significant throughout the almost entire study. High-dose F0 parental males had statistically significantly lower body weights during several study segments, which led to a statistically significant reduction of the mean terminal body weight resulting in several secondary weight changes of organs.

Regarding clinical pathology rats at 450 mg/kg bw/d developed an anemia shown by reduced hemoglobin and hematocrit values as well as RBC counts. At least in females the anemia seemed to be regenerative. The higher WBC and lymphocyte counts in the mentioned rats were most probably due to stress. Concurrently, to the high lymphocyte counts, the relative neutrophil and eosinophil counts were lower.

The main target organ regarding changes in clinical chemistry parameters was the liver: The alterations in protein levels (increased albumin levels in rats of both sexes in the high dose group, decreased total protein and globulin values in dosed males) as well as increased urea levels in males of the mid- and high-dose groups without a concurrent increase of creatinine levels indicate a changed protein metabolism. Decreased cholesterol values in males of the mid- and high-dose groups as well as increased triglyceride levels in high-dosed females might be a hint of an altered metabolism of lipids.

An increased oxidation of long-chained fatty acids was measured by a higher palmitoyl CoA oxidation rate indicating a greater activity of peroxisome enzymes.

Most probably, the decrease of total bilirubin levels in females of the mid- and high-dose groups was due to an increased conjugation followed by a higher excretion rate. The changed metabolism led to a liver cell swelling with increased ALP activities in rats of both sexes of the mid- and high-dose groups and additionally in males of the low-dose group. The higher ALT activities indicated a liver cell membrane degradation in high-dosed rats of both sexes as well as in females of the mid-dose group.

Increased inorganic phosphate levels in high-dosed rats of both sexes as well as increased urea levels in males of the mid- and high-dose groups may be a hint of a beginning kidney dysfunction in these animals.

Regarding pathology, the terminal body weight decrease after treatment of males with 150 and 450 mg/kg/bw is considered a treatment-related effect. The increase of liver weights in all dose groups and in both sexes and the centrilobular hepatocellular hypertrophy in 450 mg/kg/bw animals correlate with alterations of clinical chemistry parameters indicative of changed protein and lipid metabolism.

No effects were found regarding the gonads and accessory sex organs after light microscopy of the respective organs. Therefore, the significant decrease of growing and primordial follicles after treatment with 450 mg/kg/bw remains unclear and may be of no toxicological relevance, as any other effects on the reproductive organs were missing. Thus, there were no indications from clinical examinations as well as gross and

histopathology, that 3-(m-tert-butylphenyl)-2-methylpropionaldehyde adversely affected the fertility of the F0 parental animals up to and including a dose of 450 mg/kg bw/day. Estrous cycle data, mating behavior, conception, as well as sperm parameters, gross and histopathology of sexual organs were comparable between the rats of all test groups and ranged within the historical control data. Therefore, the significance of the statistical decrease of growing and primordial follicles after treatment with 450 mg/kg/bw remains unclear and may not be of any toxicological relevance, as neither functional effects on fertility nor corroborative morphological changes in the reproductive organs were identified.

However, 3-(m-tert-butylphenyl)-2-methylpropionaldehyde adversely affected the reproductive performance of the F0 parental animals at a dose of 450 mg/kg bw/day. For this dose level, this was indicated by a lower number of implants, an increased intrauterine mortality, an increased number of stillborns and complete litter losses as well as a decreased average litter size. A marginal effect was still present at a dose of 150 mg/kg bw/d as a slightly higher number of stillborn and a lower number of liveborn pups was noted in the respective litters. No adverse effect on reproductive performance was observed at the 50 mg/kg bw/d dose level. Gestation, parturition, lactation and weaning remained unaffected. 

For liveborn pups of the F0 parents, test substance-induced signs of developmental toxicity were noted at dose levels of 150 mg/kg bw/d and above. Postnatal survival of offspring in the 450 mg/kg bw/d dose group was significantly reduced as almost half of the liveborn offspring died or were cannibalized within 4 days after birth and another 8% of the survivors died between PND 5 and weaning. Six high-dose dams lost their complete litters. Three high dose dams did not nurse their pups properly, this may be a contributing factor to some of the pup mortality but cannot explain the whole extent of the effect. Pup mortality during early lactation was still observed at mid-dose level but the extent was rather minor.

The F1 offspring in the mid- (males only) and high-dose groups had significantly reduced body weights and gained also less body weight than the control offspring prior to weaning. The reductions were related to the dose. This reduced weight gain is, together with pup mortality, regarded as a treatment-related adverse effect on postnatal development.

Commencement of sexual maturity was slightly but significantly earlier in high-dose female offspring. The earlier high-dose female sexual maturation is in line with a significantly lower body weight at commencement of vaginal patency. As demonstrated by a mostly unaffected body weight gain there was no significant adverse effect on general development in these females. In contrast, male puberty was slightly delayed at the 450 mg/kg bw/d dose level. The slight delay corresponds to the slightly, non-significantly, lower high-dose body weight. No such effects were noted for the 50 and 150 mg/kg bw/d dose levels.

Gross necropsy examinations of the F1 offspring revealed no test substance-related adverse findings.

 

Thus, under the conditions of the present 1-generation reproduction toxicity study the NOAEL (no observed adverse effect level) for general, systemic toxicity is below 50 mg/kg bw/day for the F0 parental males and is 50 mg/kg bw/day for the F0 parental females, based on increased liver weights and alterations of clinical chemistry parameters indicative of changed protein and lipid metabolism at next higher dosages.

The NOAEL for fertility for the F0 parental rats is 450 mg/kg bw/day, the highest dose tested.

The NOAEL for reproductive performance and developmental toxicity in the F1 dams and progeny is 50 mg/kg bw/day, based on intrauterine mortality as well as pup mortality and decreased pre-weaning pup body weights/pup weight gain, at the LOAEL of 150 mg/kg bw/day. Developmental effects do not occur in the absence of parental toxicity.

 

Although a structural similarity exists, this study has no implication for the assessment of the reproductive toxicity potential of 2-(4-tertbutyl)propionaldehyde (CAS 80-54-6), due to the differences in their toxicological profiles, namely testicular and sperm toxicity versus evident developmental toxicity.

 

The outcome of this study on the impurity 3-(m-tert-butylphenyl)-2-methylpropionaldehyde triggers a classification for reproductive toxicity for the technical grade Lysmeral Technisch as Repr. 1B; H360D (May damage the unborn child). In contrast, the alternative grade Lysmeral Extra is to be classified as Repr. 2; H361f (Suspected of damaging fertility), due to its intrinsic hazard properties and the absence of relevant levels of the impurity 3-(m-tert-butylphenyl)-2-methylpropionaldehyde.

 

Justification for classification or non-classification

Summary of Fertility

Repeated dose studies in male rats, partly with focus on male reproductive organs, provide evidence for adverse effects on male reproductive organs in rats after oral lysmeral (2-(4-tert-butylbenzyl) propionaldehyde) administration. These effects were observed concomitantly with signs of general toxicity and adverse effects on the liver. The subchronic repeated dose toxicity study provides a NOAEL for testicular toxicity effects after oral administration at 25 mg/kg bw/day (Givaudan 1986A), and according to the findings from further repeated dose and reproductive toxicity studies, these effects can be expected to occur at doses above this NOAEL. This effect level was found to be independent from treatment duration. Adverse testicular findings were observed even after a single oral administration. These data support the conclusion for a clear dose threshold for the induction of testicular toxicity in rats independent of dose duration.

Accordingly, impairment of male fertility combined with signs of general toxicity and changes in clinical parameters of the liver was observed in the one-generation range finding studies in the rat after oral administration of lysmeral. Due to the obvious testicular and spermatotoxic effects of lysmeral, the relation between the observed lack of pregnancies, lack of delivered offspring and impairment of male fertility is clearly indicated. These findings were obtained at comparable dose levels also used in repeated dose studies. In contrast, dermal administration on rats led to no testicular toxicity except for dose levels above the limit dose. In the EOGRTS, oral administration of lysmeral via feed did not affect male or female fertility and reproductive performance of parents and offspring at doses up to10 mg/ kg bw/d nominal (approx. 15 mg/kg bw/d ingested). In dogs, general adverse effects together with liver and testicular toxicity were observed after oral administration, however, adverse testes effects occurred at higher dose levels than in the rat. Considering the findings from the available studies in dogs, a NOAEL for testicular toxicity is set at 44.6 mg/kg bw/day. No testicular toxicity was observed in the mouse, guinea pig, rabbit and primates.

Identical adverse testicular effects and species specificity has been observed after oral administration of p-tert-benzaldehyde (TBB) and p-tert-butyltoluene (TBT); (see Annex 3). The rat has been found to be the most sensitive species for TBB and TBT induced testicular toxicity. In analogy to lysmeral, systemic formation of p-tert-butylbenzoic acid (TBBA) has been observed after oral administration of TBB and TBT. Clear evidence of adverse testicular and spermatotoxic effects - identical in quality to lysmeral - have been observed for the metabolite TBBA as well (see Annex 2). Based on the lowest adverse effect level for testicular toxicity, TBBA application in rats revealed the highest potency and is included in Annex VI of the CLP regulation with aclassification asRepr. 1B (H360F; Index No. 607-698-00-1). TBBand TBT showed lower potencies in exerting comparable testes effects. Lysmeral showed the lowest potency in testes toxicity when compared to TBB, TBT and especially to TBBA. Testes toxicity potencies correlated well with systemically formed urinary TBBA amounts. Therefore TBB, TBT and lysmeral share TBBA as common metabolite andthe formation of the systemic TBBA intermediaterepresents a metabolic key eventfor lysmeral induced testicular toxicity.

A strong correlation between the formation of TBBA-CoA conjugates in rat hepatocytes, disruption of lipid synthesis and testicular toxicity has been found. Complex lipids are present in high amounts in mammalian sperm and play an important role for spermatogenesis. Their synthesis depends on intracellular process, that requires a sufficient pool of available CoA. Lysmeral treatment was found to disrupt fatty acid/ lipid synthesis and induced testes toxicity is always observed in the presence of liver toxicity. Other chemicals potentially transformed to benzoic acid metabolites show a strong correlation between sustained formation of benzoyl-CoA complexes in hepatocytes and spermatoxic/testicular toxicity in rats.

Taken together, the comparable pattern of testicular effects, the species dependencies and the observed differences in potencies substantiate, that the formation of systemic TBBA is a metabolic key event for lysmeral and TBB/TBT induced testicular toxicity. Furthermore, the conjugation of TBBA with CoA represents the mode of action for lysmeral induced testes toxicity and spermatotoxicity.

 

Summary of Developmental Toxicity

Developmental toxicity of lysmeral has been assessed by oral (gavage) administration of lysmeral to pregnant rats in a developmental toxicity study according to OECD test guideline No. 414.

High dose dams (41 mg/kg bw/d) showed clinical signs (transient salivation), transient reduction of mean food consumption and body weight loss on day 6-8 p.c. Mean body weight gain was decreased over the entire treatment phase resulting in lower mean body weights on day 13 - 20 p.c. and net body weight gain compared to controls. Increased levels of alanine aminotransferase and glutamate dehydrogenase, decreases serum cholinesterase levels and organ weight changes (increased liver weights, reduced uterus weights) were noted.

In mid dose dams (13 mg/kg bw/d) body weight gains were transiently decreased on day 6-8 p.c. Furthermore, alanine aminotransferase levels were increased, serum cholinesterase levels were decreased and increased liver weights were found.

These findings reflect a lysmeral induced general systemic and liver toxicity for high dose and less pronounced for mid dose dams.

The number of mainly early resorptions was increased due to postimplantation losses in the high dose group whereas gestational parameters were not significantly influenced in lower dose groups (5, 15 mg/kg bw/d). Subsequently, the number of fetuses and live fetuses per dam was found to be slightly below the respective historical control range in the high dose group.

Concomitantly, prenatal developmental toxicity in terms of reduced fetal body weights was observed in the mid and high dose groups. These findings coincided with significant maternal toxicity at the same dose levels. 

Sporadic malformations were observed, which lacked a consistent pattern, occurred in very few of the large number of examined fetuses and there incidences were found within the respective historical control ranges. External variations were not observed and soft tissue variations occurred in a dose independent manner in all test groups including control animals.

In contrast, an overall incidence of skeletal variations was statistically significantly increased in mid and high dose animals. These variations represented mainly delays and minor disturbances in ossification processes of the skull, sternebrae and pubic girdle. Supernumerary (14th) ribs were found in control and dosed animals at high incidences, and structural variations like a supernumerary thoracic vertebra (14th) or a misshapen sacral vertebra (1^stsacral arch) were found to be increased evidently in the high dose group fetuses. The observed skeletal variations are well correlated to statistically significantly decreases in mean fetal body weights and evident maternal toxicityin the respective dose groups. Clustering of incidences for a supernumerary or misshapen vertebra in single litters was observed, and a maternal predisposition which affects the respective offspring in situations of maternal stress conditions could be hypothesized here.

Supernumary ribs and delays of ossification in rodent offspring are among the common endpoints related to chemical exposure stress (ECETOC, 2004). Delays in ossification are by definition transitory, occur in conjunction with decreased fetal weights and represent an indicator for adverse effects on fetal maturation rather than a teratogenic potential (Daston, 2007).

Overall, the increased numbers of fetuses with common skeletal variations are considered an embryo-/fetotoxic effect due to fetal growth retardations, representing a manifestation of a non-specific stress on the dams and not a teratogenic effect of lysmeral. Increased early resorptions and the subsequent decrease in number of fetuses are further manifestations of the non–specific maternal stress induced by lysmeral administration.

The findings of the one-generation rangefinder studies are largely consistent with the effects observed in the present key teratogenicity study. Slight, non-significant and dose independent increases in postimplantation losses were found in dose groups having offspring. A slight reduction in the number of delivered pups has been observed at doses not affecting fertility indices. Furthermore, a significant reduction in birth weights, pup weights at weaning and pup weight gain has been seen when compared to controls. These findings coincided with adverse systemic effects to the dams. No effects on the gestation and live birth indices were observed due to the absence of any stillborn in the dosed animals. Whereas effects on early pup survival occurred, lactation indices were not significantly affected and no test substance related findings in pup necropsy have been found.

Furthermore, the highest lysmeral dose tested in the EOGRTS (in the range of the LOAEL of the developmental toxicity study) resulted in pup body weight reductions of the F1 and F2 offspring and was associated with adverse maternal liver and general systemic effects. Lysmeral did not have a consistent impact on the number of postimplantation losses, delivered pups and pup survival up to this dose. Further developmental toxicity endpoints including developmental neurotoxicity and immunotoxicity were not affected by treatment with lysmeral.  

Taken together, developmental toxicity has been observed at doses leading to evident maternal toxicity and is considered to be a secondary non-specific consequence of general systemic toxicity in the dams. Therefore, these findings do not warrant a classification with respect to developmental toxicity.

 

Discussion

Adverse effects of lysmeral on the male reproductive system have been observed in various oral repeated dose toxicity studies in rats and were confirmed in feeding one-generation range-finding studies, whereas no evidence for testicular toxicity was observed in the mouse and guinea pig. Considering non-rodent species, the dog has been shown to be susceptible towards testicular toxicity as well, however at higher dose levels than the rat. Testicular toxicity in rats and dogs after oral lysmeral application was only observed at dose levels showing also general signs of toxicity in these animals. In studies with more detailed observations, the liver was found to be the main affected organ upon treatment with doses also inducing testicular toxicity. In contrast, short-term oral exposure to rabbits did not indicate a potential of lysmeral to induce testicular toxicity. Furthermore, in rhesus monkeys, no indication of testicular toxicity, at doses causing testicular toxicity in the rats, was observed. Therefore, the rat represents the most susceptible species, and it appears, that a single oral exposure to lysmeral above a clearly defined threshold dose seems sufficient to cause testicular toxicity.

 

Testicular effects and species dependencies identical to lysmeral have been observed after application of p-tert-benzaldehyde (TBB) and p-tert-butyltoluene (TBT). Para-tert-butylbenzoic acid (TBBA) is formed as metabolite after administration of TBT, TBB or lysmeral. Therefore TBB, TBT and lysmeral all share the same metabolite, namely TBBA. TBBA application in rats revealed the highest testicular toxicity potency based on effect levels and is included in Annex VI of the CLP regulation with aclassification as Repr. 1B (H360F).In line, TBB and TBT showed lower potencies in exerting comparable testes effects. Lysmeral showed the lowest potency in testes toxicity when compared to TBB, TBT and especially to TBBA. Testes toxicity potencies correlated well with systemically formed urinary TBBA amounts. Overall, these findings substantiate, that the formation of systemic TBBA represents a metabolic key event for lysmeral and TBB/TBT induced testicular toxicity.

Based on the clear evidences from animal studies, it is considered appropriate to classify lysmeral for reproductive toxicity, i.e. adverse effects on fertility. In determining the respective hazard category, the assessment of the relevance of the hazard to humans is to be considered. The adverse effects on male reproductive organs are considered to underlie species specific mechanisms and the present data indicate, that primates are considerably less or even not susceptible towards the testicular toxicity observed in the dog and more effectively in the rat. In accordance, quantitative differences in the formation of metabolites such as TBBA exist and the urinary excretion of glycine conjugated TBBA differs between the rat and the other rodent species investigated, i.e. mouse and guinea pig.

Based ona qualitative and quantitative evaluation of metabolic profiles for different species in anin vitro metabolism study, a predominant formation of TBBA concentrations in rat hepatocytes was found when compared to other rodent, non-rodent animal or human hepatocytes. The TBBAconcentrations found in the model using human hepatocytes were approx. 4 fold lower compared to rat hepatocytes at corresponding incubation concentrations of lysmeral (10, 50 100 µM). The lysmeral concentrations used reflect plasma levels obtained after oral administration of lysmeral doses below and above the lowest adverse testicular effect level.Furthermore, the TBBA levels formed in human hepatocytes after incubation of lysmeral concentrations related to adverse testicular effect doses (50 and 100 µM) werecomparable to TBBA levels found in the rabbit, a species not sensitive totesticular toxicity. Although, the metabolite TBBA is classified as Repr. 1B (H360F), its endogenous formation after lysmeral exposure is species dependent and the formation in humans is found to be comparable to species, showing no lysmeral induced testicular toxicity.

A strong correlation has been established between the formation of TBBA-CoA conjugates in rat hepatocytes, disruption of lipid synthesis and testicular toxicity. It is concluded, that the conjugation of TBBA with CoA represents the mode of action for lysmeral induced testes toxicity and spermatotoxicity. The kinetics of TBBA-CoA conjugation fundamentally differs in human hepatocytes when compared to the rat. Lower and transient concentrations due to a rapid decrease of TBBA-CoA conjugates have been observed in human hepatocytes, which strongly indicates, that the observed metabolic fate of lysmeral is rat specific and the testicular toxicity/spermatotoxicity is a species-specific effect with little relevance for humans.

Based on differences in the endogenous formation and the mode of action between rat and humans, the harmonized classification of the metabolite TBBA as Repr. 1B (H360F) has no implication for the classification proposal provided for lysmeral.

 

The adverse testicular effects in the rat and dog were observed after administration of lysmeral via the oral route. Because of its properties as fragrance material, lysmeral administration needed to be performed via gavage or encapsulation. Besides test substance stability issues, palatability was a major obstacle due to the unpleasant smell of concentrated lysmeral in order to attain study relevant doses. Dermal application of lysmeral represents the most appropriate route of administration, having regard to the likely route of human exposure as a fragrance material (for further details see Annex 1). In contrast, the administration via gavage or encapsulation represents an unrealistic and non-relevant form of application. Compared to oral studies, dermal administration of lysmeral in rats led to testicular toxicity only at an excessive dose level, clearly above the limit dose, whereas at 1000 mg/kg body weight, no adverse testicular effects were observed. When compared todoses leading to rat testicular toxicity,a prolonged human uptake of lysmeral doses inducing systemic toxicity (testes toxicity or spermatotoxic effects) is highly unlikely.

Overall, studies on species variability provide clear evidence, that higher order mammalian species including humans are less or not susceptible than rats. This is due to the observed lack of testicular toxicity in these studies, differences in metabolic profiles (including TBBA formation) and fundamental differences concerning the mode of action identified (sustained TBBA-CoA conjugation). In support,testicular toxicity in susceptible species has a clear threshold and it is highly unlikely that lysmeral levels taken up by humans would lead to the formation of relevant systemic levels of TBBA.Taken together, the relevance of the species-specific testes toxicity observed in test animals for humans is doubtful.

 

Comparison with criteria

Lysmeral(2-(4-tert-butylbenzyl)propionaldehyde)has been found to induce testicular toxicity and spermatotoxicity when administered orally to rats and at higher dose levels to dogs.Infertility in rats dueadverse effects of orally administered lysmeral on the male reproductive system has been confirmed in feeding one-generation range-finding studies.Based on clear evidences fromexperimental animals, it is considered appropriate to classify lysmeral for reproductive toxicity, i.e. adverse effects on fertility.

The CLP regulation criteria for classification as reproductive toxicants are as follows:

The classification in Category 1A (Known human reproductive toxicant) “is largely based on evidence from humans”.

The classification of a substance in Category 1B (Presumed human reproductive toxicant) “is largely based on data from animal studies. Such data shall provide clear evidence of an adverse effect on sexual function and fertility or on development in the absence of other toxic effects, or if occurring together with other toxic effects the adverse effect on reproduction is considered not to be a secondary non-specific consequence of other toxic effects.”

Besides the reference toclear evidences fromexperimental animals, the CLP regulation further states in its criteria for classification in Category 1B:

“However, when there is mechanistic information that raises doubt about the relevance of the effect for humans, classification in Category 2 may be more appropriate.”

 

Further, substances are classified in Category 2 (Suspected human reproductive toxicant), “when there is some evidence from humans or experimental animals, possibly supplemented with other information, of an adverse effect on sexual function and fertility, or on development, and where the evidence is not sufficiently convincing to place the substance in Category 1”.

In determining the appropriate hazard category for the adverse effects on fertility, the assessment of the relevance of the given hazard to humans needs to be taken into account.

 

• Species specificity for lysmeral induced testicular toxicity has been observed.Adverse effects of lysmeral on the male reproductive systemat aclearly defined threshold dose have been found in rats whereas no evidence for testicular toxicity was observed in the mouse and guinea pig. Considering non-rodent species, the dog has been shown to be susceptible towards lysmeral induced testicular toxicity. In contrast, short-term oral exposure to rabbits did not indicate a potential of lysmeral to induce testicular toxicity. Furthermore, in rhesus monkeys, no indication of testicular toxicity, at doses causing testicular toxicity in the rats, was observed.


• Based on the accordance in the testicular toxicity profile of lysmeral,p-tert-benzaldehyde (TBB),p-tert-butyltoluene (TBT)and the shared metabolitepara-tert-butylbenzoic acid (TBBA), the formation of the systemic TBBA intermediaterepresents a metabolic key eventfor lysmeral induced testicular toxicity.


• Species specificity for lysmeral induced testicular toxicity is reflected by species dependent differences in the conversion of lysmeral to TBBA in hepatocytes. TBBA formation in human hepatocytes is of low magnitude compared to rats and iscomparable to concentrations found in the rabbitat toxicologically relevant doses, a species not sensitive to Lysmeral inducedtesticular toxicity.


•  A strong correlation has been established between the formation of TBBA-CoA conjugates in rat hepatocytes, disruption of lipid synthesis and testicular toxicity. The conjugation of TBBA with CoA represents the mode of action for lysmeral induced testes toxicity and spermatotoxicity. In human hepatocytes, lower and transient concentrations due to a rapid decrease of TBBA-CoA conjugates strongly indicate, that testicular toxicity/spermatotoxicity is a species-specific effect with little relevance for humans.

• Because of the properties of lysmeral as fragrance material leading to palatability issues, substance administration needed to be performed via gavage or encapsulation. However, dermal application of lysmeral represents the most appropriate route of administration, having regard to the likely route of human exposure as a fragrance material. Dermal studies in rats showed no testicular toxicity up to the limit dose.

Although the human exposure considerations provided in Annex 1 do not represent a main argument for the classification for lysmeral, they are included to support the justification on classification. Based on these considerations, a prolonged human uptake of lysmeral doses inducing systemic toxicity (testes toxicity or spermatotoxic effects) is highly unlikely.

A clear evidence for a species specificity and, if at all, a low human susceptibility concerning lysmeral induced testicular toxicity raises doubt about the relevance of the effect for humans. Furthermore,evident reproductive toxicity has been observed after substance administration via gavage or encapsulation, representing a non-relevant form of application. In support, a prolonged human uptake of lysmeral doses inducing systemic toxicity (testes toxicity or spermatotoxic effects) is highly unlikely, when compared todoses leading to rat testicular toxicity.

As outlined in the CLP criteria listed above, a classificationin Category 2 (Suspected human reproductive toxicant;Repr. 2 - H361f; CLP regulation EC/1272/2008) is appropriate, whereasclassification in Category 1A or 1B (Known or presumed human reproductive toxicant) is not justified.

 

Concerning developmental toxicity, the CLP regulation states as a basis of classification:

“…Classification as a reproductive toxicant is intended to be used for substances which have an intrinsic, specific property to produce an adverse effect on reproduction and substances shall not be so classified if such an effect is produced solely as a non-specific secondary consequence of other toxic effects.”

Developmental toxicity has been observed at doses leading to evident maternal toxicity and is considered a secondary non-specific consequence of general systemic toxicity in the dams. Therefore, based on the present data, no classification concerning developmental toxicity is warranted.

Classification for effects on or via lactation is intended to indicate when a substance may cause harm due to its effects on or via lactation, and it is independent of consideration of the reproductive toxicity of the substance. According to Table 3.7.1 (b) of the CLP-regulation, classification for effects on or via lactation can be assigned on the:

a) human evidence indicating a hazard to babies during the lactation period; and/or

b) results of one or two generation studies in animals which provide clear evidence of adverse effect in the offspring due to transfer in the milk or adverse effect on the quality of the milk; and/or

c) absorption, metabolism, distribution and excretion studies that indicate the likelihood that the substance is present in potentially toxic levels in breast milk.

The present data do not allow to specifically assess the effects of lysmeral on or via lactation. No human evidence indicating a hazard to babies during the lactation and no information on presence and concentration of lysmeral or its metabolites in milk is available and the reproductive toxicity studies did not provide clear evidence of adverse effect in the offspring due to milk transfer or effects on the milk quality. Based on currently available data, classification for effects on or via lactation is therefore not warranted.

 

Conclusion on classification

Lysmeral(2-(4-tert-butylbenzyl)propionaldehyde)has been identified to induce testicular toxicity when administered orally to rats and at higher dose levels to dogs.Infertility in rats dueadverse effects of orally administered lysmeral on the male reproductive system has been confirmed in feeding one-generation range-finding studies. In an EOGRTS, lysmeral doses below the identified LOAEL for testicular toxicity did not affect male or femalefertility and reproductive performance of parents and offspring.Based on clear evidences from animal studies, it is considered appropriate to classify lysmeral for reproductive toxicity, i.e. adverse effects on fertility.

Based on the accordance in the testicular toxicity profile of lysmeral,p-tert-benzaldehyde(TBB),p-tert-butyltoluene (TBT) and the shared metabolitepara-tert-butylbenzoic acid (TBBA), the formation of the systemic TBBA intermediaterepresents a metabolic key eventfor lysmeral induced testicular toxicity. On the basis of the effective doses determined, lysmeral possesses evidently a lower potency for testicular toxicity than TBBA.

Furthermore,a strong correlation has been established between the formation of TBBA-CoA conjugates in rat hepatocytes, disruption of lipid synthesis and testicular toxicity. The conjugation of TBBA with CoA was found to be the mode of action for lysmeral induced testes toxicity and spermatotoxicity.

In determining the respective hazard category, the assessment of the relevance of the given hazard to humans needs to be taken into account.Species specificity for lysmeral induced testicular toxicity has been observed.Adverse effects of lysmeral on the male reproductive systemat aclearly defined threshold dose have been found in rats but not in the mouse and guinea pig. Considering non-rodent species, the dog has been shown to be susceptible towards lysmeral induced testicular toxicity. In contrast, short-term oral exposure to rabbits did not indicate a potential of lysmeral to induce testicular toxicity. Furthermore, in rhesus monkeys, no indication of testicular toxicity, at doses causing testicular toxicity in the rats, was observed

Species specificity for lysmeral induced testicular toxicity is reflected by species dependent differences in the conversion of lysmeral to TBBA in hepatocytes. TBBA formation in human hepatocytes is of low magnitude compared to rats. In fact, TBBA formation in humans is comparable to levels produced in species that did not demonstrate testicular toxicity (i.e. rabbits) at biologically relevant doses.In addition, the kinetics of TBBA-CoA conjugation, i.e. the identified mode of action, fundamentally differs in human hepatocytes when compared to the rat. Lower and transient concentrations due to a rapid decrease of TBBA-CoA conjugates have been observed in human hepatocytes, which strongly indicates, that the observed metabolic fate of lysmeral is rat specific.Overall, clear evidences for a low susceptibility of humans regarding lysmeral induced testicular toxicity further support the absence of itshuman relevance.

Because of the properties of lysmeral as fragrance material leading to palatability issues, substance administration needed to be performed via gavage or encapsulation. This represents a non-relevant form of application.Studies via the relevant route to humans (i.e. dermal) in rats showed no testicular toxicity up to the limit dose (1000mg/kg bw/day).As a supportive argument, a prolonged human uptake of lysmeral doses inducing systemic toxicity (testes toxicity or spermatotoxic effects) is highly unlikelyand the relevance of the observed testicular effects for humans is doubtful.

Overall, clear evidence for a species specificity and, if at all, a low human susceptibility concerning lysmeral induced testicular toxicity question a relevance for humans.

Therefore, a classification as substance to be suspected of damaging fertility, i.e. Repr. 2 (H361f; regulation 1272/2008), is warranted.

Developmental toxicity has been observed at doses leading to evident maternal toxicity and is considered a secondary non-specific consequence of general systemic toxicity in the dams. Therefore, based on the present data, no classification concerning developmental toxicity is warranted.

Based on currently available data, classification for effects on or via lactation is not warranted.

 

The assessment of the registrant was not confirmed by ECHAs Committee for Risk Assessment (RAC) as laid out in their Opinion proposing harmonised classification and labelling at EU level of 2-(4-tert-butylbenzyl)propionaldehyde, 2019.

According to the RAC, there is evidence that species differences exist, but these differences are considered as quantitative rather than qualitative. The RAC considered data from other species (mice, guinea pigs, rhesus monkeys) not sufficient to deem them “non-responders”, since the doses used may have been too low, and exposure periods (5 days) too short, to induce testicular effects. According to the RAC, a direct effect of Lysmeral on testes tissue cannot be ruled out, since it appeared not clear, how relevant mechanistic findings from in vitro tests in hepatocytes were for the effects seen on testes tissue. The RAC found, that the testicular toxicity and spermatotoxicity shown in two species (rats and dogs) is relevant for humans despite a quantitatively different metabolism of the compound in different species. Furthermore, the RAC considered the proposed MoA, although plausible, not sufficient to preclude relevance for humans. To dismiss effects, and downgrade the classification to Category 2, the RAC would have needed stronger (mechanistic) evidence. Thus, the RAC considered classification of Lysmeral as Repr. 1B; H360F warranted.

Concerning developmental toxicity, the RAC considered the skeletal variations observed as likely secondary to the marked maternal toxicity and decreased foetal weights. Malformations lacked a consistent pattern and occurred at a very low incidence inside the historical control range. Hence, the RAC does not consider these effects as relevant for classification.

However, the RAC emphasized that, the effects on post-implantation loss were seen consistently in several studies, albeit with a high variation. These effects were associated with doses also leading to clinical chemistry changes indicative of maternal liver toxicity. However, only in some cases these were accompanied by changes in liver weight and liver histopathology, or by markedly reduced maternal body weights or food consumption. According to the RAC, the effects on pup body weights were also consistently observed, starting from doses without marked maternal toxicity.

Thus, the effects on post-implantation loss and pup body weights are considered to warrant classification, as the RAC considers these not unequivocally attributable to the maternal toxicity seen in the studies. Therefore, the RAC considers classification as Repr. 2; H361d warranted.

1. Daston et al. (2007);Skeletal Malformations and Variations in Developmental Toxicity Studies: Interpretation Issues for Human Risk Assessment.Birth Defects Research (Part B) 80:421-424.


2. ECETOC (2004);Influence of Maternal Toxicity in Studies on Developmental Toxicity.Workshop Report No.4.