m-tolylidene diisocyanate

Endpoint summary

Key value for chemical safety assessment

Genetic toxicity in vitro

Description of key information

In an in vitro bacterial reverse mutation assay according to OECD Guideline 471 (AMES test) (Seel, 1999), three out of five tested strains (TA98, TA 100 and TA1537) showed a biological relevant increase of mutant colonies after metabolic activation (S9 mix).

In an in vitro mammalian chromosome aberration test according to OECD Guideline 473 (Ministry of Health Labour Welfare Japan, 2001), Chinese hamster CHL/IU cells showed increased number of structural chromosomal aberrations.

Link to relevant study records

Reference

Reference 1

Endpoint:

in vitro gene mutation study in bacteria

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

guideline study with acceptable restrictions

Qualifier:

equivalent or similar to guideline

Guideline:

OECD Guideline 471 (Bacterial Reverse Mutation Assay)

Deviations:

yes

Remarks:

Only four strains. E.coli or TA102 missing.

GLP compliance:

not specified

Type of assay:

bacterial reverse mutation assay

Species / strain / cell type:

S. typhimurium TA 1535, TA 1537, TA 98 and TA 100

Metabolic activation:

with and without

Metabolic activation system:

S9 mix

Test concentrations with justification for top dose:

125.0, 250.0, 500.0, 1000.0, and 2000.0 µg/plate

Vehicle / solvent:

- Vehicles/solvents used: Ethyleneglycol dimethylether (EDGE), dimethylsulphoxide (DMSO)

Positive controls:

yes

Positive control substance:

other: 2-aminoanthracene

Details on test system and experimental conditions:

Ames test

Key result

Species / strain:

S. typhimurium TA 1535

Metabolic activation:

with and without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

no cytotoxicity, but tested up to precipitating concentrations

Positive controls validity:

valid

Key result

Species / strain:

S. typhimurium TA 1537

Metabolic activation:

without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

no cytotoxicity, but tested up to precipitating concentrations

Positive controls validity:

valid

Key result

Species / strain:

S. typhimurium TA 98

Metabolic activation:

without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

no cytotoxicity, but tested up to precipitating concentrations

Positive controls validity:

valid

Key result

Species / strain:

S. typhimurium TA 100

Metabolic activation:

without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

no cytotoxicity, but tested up to precipitating concentrations

Positive controls validity:

valid

Key result

Species / strain:

S. typhimurium TA 1537

Metabolic activation:

with

Genotoxicity:

positive

Cytotoxicity / choice of top concentrations:

no cytotoxicity, but tested up to precipitating concentrations

Positive controls validity:

valid

Key result

Species / strain:

S. typhimurium TA 98

Metabolic activation:

with

Genotoxicity:

positive

Cytotoxicity / choice of top concentrations:

no cytotoxicity, but tested up to precipitating concentrations

Positive controls validity:

valid

Key result

Species / strain:

S. typhimurium TA 100

Metabolic activation:

with

Genotoxicity:

positive

Cytotoxicity / choice of top concentrations:

no cytotoxicity, but tested up to precipitating concentrations

Positive controls validity:

valid

Remarks on result:

other: TDI 80, solvent: EDGE

The HPLC analyses of the dissolved aromatic diisocyanates indicate that their degradation is considerably accelerated if DMSO is the solvent and may be completed before the Salmonella microsome test has even begun.

2,4-TDI, 2,6-TDI, and TDI 80, all of which were dissolved in EGDE, showed a consistently negative response in the absence of S9 mix. Furthermore, no mutagenicity was observed in any of the TA 1535 strain experiments. Clearly positive results were obtained in the other strains after metabolic activation of 2,4-TDI, 2,6-TDI, and TDI 80.

2,4-TDI, 2,6-TDI and TDI 80, all of which were dissolved in EGDE, showed a consistently negative response in the absence of S9 mix. Furthermore, no mutagenicity was observed in any of the TA 1535 strain experiments. With respect to these findings, the results of the TDIs were in complete agreement with those obtained for the MDIs. In contrast to the MDIs, however, clearly positive results were obtained in the other strains after metabolic activation of 2,4-TDI, 2,6-TDI and TDI 80. The results are summarised in Tables 1-3. Consistently positive results were obtained in strain TA 98 with all types of TDI tested. In TA 1537, clearly positive results were obtained for 2,4-TDI (Table 1) and weak effects were observed for TDI 80 whereas no effects were found for 2,6-TDI (Tables 2 and 3). Weak positive results were also obtained for TDI 80 with TA 100 (Table 3). S9 mixtures with varying amounts of the S9 fraction (10 and 30 %) were used for 2,4- and 2,6-TDI. The results demonstrate (Table 1 and 2) that effects are slightly reduced in the presence of the S9 mix containing 30 % of the S9 fraction. This reduction of effects may be examined with a selective or at least preferred reaction of the test samples with the proteins of the S9 fraction. [See remarks on results below for Tables 1-3)

See Additional information on results for Tables 1 -3.

Table 1: Results with 2,4-TDI (dissolved in EGDE) and metabolic activation

Strain S9, µg/plate	TA 1537		TA 98
	10 % +S9	30 % +S9	10 % +S9	30 % +S9
0	19	12	54	39
50	18	16	64	52
100	23	16	104*	76*
200	44*b	25	117*	76*
400	52*b	22b	153*b	87*b
600	51*b	45*b	172*b	93*b
800	49*bp	41*b	125*bp	128*bp
1000	41*bp	42*bp	117*bp	58*bp
AA 3	233*	80*bp	1245*	509*

AA = 2-aminoanthracene * = mutagenic effect, b = background growth reduced, p = precipitation

Table 2: Results with 2,6-TDI (dissolved in EGDE) and metabolic activation

Strain S9, µg/plate	TA 1537		TA 98
	10 % +S9	30 % +S9	10 % +S9	30 % +S9
0	12	14	47	45
150	13	21	67	57
300	13	16	87*	61
600	15	15p	130*	73*
1200	7p	10p	166*p	97*p
2400	7p	9p	128*p	82*p
4800	p	p	p	p
AA 3	360*	66*	1538*	618*

AA = 2-aminoanthracene * = mutagenic effect, b = background growth reduced, p = precipitation

Table 3: Results with TDI 80 (dissolved in EGDE) and 10 % S9 mix

Strain +S9, µg/plate	TA 100	TA 1537	TA 98
0	71	9	36
125	143	12	86*
250	188*	15	101*
500	211*p	21p	123*p
1000	57p	28*bp	88*p
2000	21bp	2bp	23bp
AA 3	869*	306*	888*

AA = 2-aminoanthracene * = mutagenic effect, b = background growth reduced, p = precipitation

Results

The stability of TDI solutions prior to salmonella/microsome tests was investigated (table 4a). The N=C=O content of 50 - 500 mg TDI, dissolved in 100 mL relatively 'dry' DMSO (0.02 - 0.03 % water), dropped to 60 % or less within the first 15 min of the test. Theoretically, the hydrolysis of 174 mg (1.0 mM) of TDI could consume 18 mg (1.0 mM) of water to form highly reactive intermediates, the aminoisocyanatotoluenes (TDAIs), and carbon dioxide. A homogeneous solution containing 0.02 % (1.11 mM) water, as was the case for the 500 mg (2.86 mM) 2,4-TDI sample, would therefore convert about 40 % of the TDI into reactive intermediates on a purely stoichiometric basis. These aminoisocyanates would then be available for further reactions with remaining 2,4-TDI or with themselves to produce a number of monomeric, oligomeric and polymeric ureas, which may be terminated by N=C=O and / or NH2 groups. At this point it is important to recall that a residual N=C=O content by no means indicates the presence of unmodified 2,4-TDI. The IR spectrum provides only insufficient information on the location of the N=C=O groups. The observed decline of isocyanate absorptions is, however, proof of the fact that chemical reactions have occurred.

In an additional experiment, 500 mg (2.86 mM) 2,4-TDI was dissolved in DMSO with an increased water content of 0.1 % (5.56 mM), a level perfectly conceivable in practice. This amount of water led to an accelerated reduction of the N=C=O absorptions, so that after 15 min, only 43 %, and after 4 h, 14 % of the isocyanate groups could be detected.

The fate of aromatic diisocyanates in Salmonella/microsome tests were also investigated (Table 4b). When a sample of 500 mg 2,6-TDI / 100 mL EGDE solution was mixed with the test ingredients or with water (0.1:2.6 mL), up to 6.6 % of the TDI was converted into 2,6-TDA within 45 s (Table 5). With respect to the amount of diamine produced, no difference was seen between the distilled water and the aqueous system of the Salmonella/microsome test. Concerning the decline of diisocyanate content, however, the transfer to a different environment became apparent. In water, the 2,6-TDI concentration dropped to ~60 %, which is basically in agreement with the results of the 10-fold approaches shown in Table 4 for the different types of TDI. When the ingredients of the Salmonella/microsome test replaced water, more than 90 % of the initial 2,6-TDI disappeared within 15 - 45 s. In this case, not 500 µg, but less than 50 µg 2,6-TDI, enriched with around 25 µg 2,6-TDA and further unquantified products will be poured onto the plate.

Table 4a: Stability TDI (in EGDE) during the first minute of a simulated test (a); HPLC determination of residual TDI and its reaction products. [This table has been separated by TDI type : 2,4-TDI)

Diisocyanate	2,4-TDI
TDI in 100 mL EGDE	5000 mg		500 mg
Dose / plate	5000 µg		500 µg
Analysed products (b)	2,4-TDI [%]	2,4-TDA [%]	2,4-TDI [%]	2,4-TDA [%]
Start	100	nd	100	nd
After 15 s	95	0.35	57	3.9
After 30 s	83	0.65	48	5.7
After 45 s	93	0.60	41	6.6
After 60 s	93	0.65	35	7.4

nd: Not detectable, detection limit: 0.1 %, e.g., 0.5 µg for the 500 mg/100 mL concentration. na: Not available. (a) Simulating the mixing of dissolved TDI with the test ingredients (1 mL:26 mL mix with water). (b) Ureas and (insoluble) polyureas were not quantified.

Table 4b: Stability TDI (in EGDE) during the first minute of a simulated test (a); HPLC determination of residual TDI and its reaction products. [This table has been separated by TDI type : 2,6-TDI)

Diisocyanate	2,6-TDI
TDI in 100 ml EGDE	5000 mg		500 mg
Dose / plate	5000 µg		500 µg
Analysed products (b)	2,6-TDI [%]	2,6-TDA [%]	2,6-TDI [%]	2,6-TDA [%]
Start	100	nd	100	nd
After 15 s	91	0.27	67	3.5
After 30 s	92	0.56	50	6.6
After 45 s	92	0.81	37	9.5
After 60 s	90	0.98	40	10.7

nd: Not detectable, detection limit: 0.1 %, e.g., 0.5 µg for the 500 mg/100 mL concentration. na: Not available. (a) Simulating the mixing of dissolved TDI with the test ingredients (1 mL:26 mL mix with water). (b) Ureas and (insoluble) polyureas were not quantified.

Table 4c: Stability TDI (in EGDE) during the first minute of a simulated test (a); HPLC determination of residual TDI and its reaction products. [This table has been separated by TDI type : TDI 80)

Diisocyanate	TDI 80
TDI in 100 mL EGDE	5000 mg		500 mg
Dose / plate	5000 µg		500 µg
Analysed products (b)	2,4-TDI [%]	2,6-TDI [%]	2,4-TDA [%]	2,6-TDA [%]
Start	100	100	nd	nd
After 15 s	61	68	3.1	0.7
After 30 s	52	61	4.4	1.0
After 45 s	36	47	5.7	1.6
After 60 s	35	45	6.0	1.9

nd: Not detectable, detection limit: 0.1 %, e.g., 0.5 µg for the 500 mg/100 mL concentration. na: Not available. (a) Simulating the mixing of dissolved TDI with the test ingredients (1 mL:26 mL mix with water). (b) Ureas and (insoluble) polyureas were not quantified.

Table 5. Stability of 2,6-TDI during the first minute of the mutagenicity test (a): HPLC determination of residual 2,6-TDI and its reaction products

2,6-TDI in 100 ml solvent	500 mg EGDE		500 mg/EGDE		500 mg/DMSO
Reaction medium	Dist. water		Test ingredients(b)		Test ingredients(b)
2,6-TDI/plate	500 µg		500 µg		500 µg
Analysed products(c)	2,6-TDI [%]	2,6-TDA [%]	2,6-TDI [%]	2,6-TDA [%]	2,6-TDI [%]	2,6-TDA [%]
Start	100	nd	99.5	0.5	12.3	9.1
After 5 s	77.8	1.6	23.1	1.6	2.3	6.4
After 15 s	70.0	3.4	8.4	4.7	3.0	8.4
After 30 s	60.7	5.3	5.6	5.8	2.6	9.1
After 45 s	61.9	6.6	8.1	5.6	2.5	8.3

nd: Not detectable; detection limit: 0.1 %, e.g. 0.5 µg. (a) Mixing 0.1 mL dissolved 2,6 -TDI with 2.6 mL water or 2.6 mL test ingredients. (b) 2.0 mL agar + 0.5 mL S9 mix + 0.1 mL nutrient broth. (c) Ureas and (insoluble) polyureas were not quantified.

Reference 2

Endpoint:

in vitro cytogenicity / chromosome aberration study in mammalian cells

Type of information:

experimental study

Adequacy of study:

supporting study

Reliability:

3 (not reliable)

Rationale for reliability incl. deficiencies:

significant methodological deficiencies

Remarks:

, less than full details available. DMSO as solvent invalidates test

Qualifier:

according to guideline

Guideline:

OECD Guideline 473 (In Vitro Mammalian Chromosome Aberration Test)

GLP compliance:

yes

Type of assay:

other: in vitro mammalian chromosome aberration test

Species / strain / cell type:

other: Chinese hamster CHL/IU cells

Metabolic activation:

with and without

Metabolic activation system:

S9 mix

Test concentrations with justification for top dose:

0, 78.1, 156, 313, 625 µg/mL (without S9)
0, 625, 1250, 2500, 5000 µg/mL (with S9)

Positive controls:

yes

Positive control substance:

benzo(a)pyrene

mitomycin C

Details on test system and experimental conditions:

DURATION
- Preincubation period:
- Exposure duration: 6 h

NUMBER OF REPLICATIONS: 2 plates/dose

Key result

Species / strain:

other: Chinese hamster CHL/IU cells

Metabolic activation:

with and without

Genotoxicity:

positive

Additional information on results:

Structural chromosomal aberrations were induced at 313 and 625 µg/mL after 6 h short term treatment without S9 mix (10.0 and 13.5 % respectively). Polyploidy was not induced in any treatment group. Chromosomal aberrations were not induced in any treatment group in the presence of S9 metabolic activation.

There are various other in vitro genotoxicity studies reported. In a poorly reported chromosomal aberration study in which the cell type was not stated (JETOC, 1996), 2,4-TDI was negative with and without metabolic activation. 2,6-TDI appears to have been slightly positive but neither untreated control data nor test concentrations were provided. DMSO was used as the solvent. This and the other in vitro genotoxicity studies are summarized below.

In a cytogenetics assay, human whole-blood lymphocytes were exposed to TDI 80/20 dissolved in acetone (Mäki-Paakkanen and Norppa, 1987). TDI slightly increased chromosome aberrations at the high two doses tested (0.075 and 0.15 µL/mL) in the absence of S9 and at the second highest dose (0.038 µL/mL) in the presence of S9. There was no dose-response. These authors also studied the formation of SCE in these cells but could not detect an increase. The quality of the study is questionable due to the formation of polymer-like fibers in the medium when TDI was added to the culture that made analysis of metaphase chromosomes for sister-chromatid exchange impossible. Acetone, a polar solvent like DMSO, is likely to hydrolyze TDI to TDA and ureas.

Gulati et al. (1989) studied the formation of chromosome aberrations and sister chromatid exchanges in Chinese Hamster Ovary cells after treatment with 2,4-TDI and 2,6-TDI (in DMSO) in the absence and presence of S9. 2,4-TDI did not induce chromosome damage at doses of 300 - 1000 µg/mL either with or without S9. SCE data were judged to be equivocal because, although there was no evidence of SCE in the presence of S9, 2/3 studies in the presence of S9 showed a significant incidence of SCE, without there being a consistent dose response. 2,6-TDI induced chromosome aberrations and SCE without S9 but the test substance inhibited cell growth and the cells were incubated for an extended time before assessing. Under these abnormal experimental circumstances, treatment at 600-1000 µg/mL produced a significant, dose related increase in aberrations with a pronounced increase occurring at the highest dose at which severe toxicity was evident. Induction of SCE occurred within a concentration range 50-300 µg/mL but the level of response did not always correlate with dose, possibly because of precipitation of the test material and the resulting variable decreases in cell culture exposures. Taking into account the instability of TDI in the diluent used and the evident toxicity of TDI in this test, the significance of these findings is questionable.

Endpoint conclusion

Endpoint conclusion:

adverse effect observed (positive)

Genetic toxicity in vivo

Description of key information

In an in vivo mammalian erythrocyte micronucleus test similar to OECD guideline 474 (Mackay, 1992), mice were inhalatively exposed to TDI. No genotoxic effects were observed.

In an unscheduled DNA synthesis (UDS) assay (Benford, 1988), TDI did not induce UDS at any of the administered doses in lung cultures from any of the treated animals.

In an in vivo mammalian erythrocyte micronucleus test (Loeser, 1983), mice or rats were inhalatively exposed to TDI. No dose- or treatment-related percentage increase of micronucleated erythrocytes in animals exposed to TDI was observed.

Link to relevant study records

Reference

Reference 1

Endpoint:

in vivo mammalian somatic cell study: cytogenicity / erythrocyte micronucleus

Remarks:

Type of genotoxicity: chromosome aberration

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

comparable to guideline study

Qualifier:

equivalent or similar to guideline

Guideline:

OECD Guideline 474 (Mammalian Erythrocyte Micronucleus Test)

GLP compliance:

yes

Type of assay:

micronucleus assay

Species:

mouse

Sex:

male/female

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: Barriered Animal Breeding Unit, ICI Pharmaceuticals, Alderley Park, Macclesfield, Cheshire, UK.
- Age at study initiation:
Phase 1: 6 - 12 weeks
Phase 2: 8 - 12 weeks
Phase 3: 8 - 9 weeks
- Housing: housed by sex with 5 - 10 animals per cage on single sided wire mesh mouse cage racks or in the long-term inhalation chambers.
- Diet: ad libitum, Porton Combined Diet [PCD] (supplied by Special Diets Services Limited, Stepf ield, Witham, Essex, UK)
- Water: ad libitum, filtered tap water
- Assigned to test groups randomly: yes, using a Latin square method until each group contained the appropriate number of mice.

ENVIRONMENTAL CONDITIONS
- Temperature (°C): ~21
- Humidity (%): 45 - 55 The relative humidity (RH) was measured using a Kew Pattern wet and dry bulb hygrometer. Excursions outside RH range were noted throughout the study but this was considered not to affect the integrity of the study.
- Air changes (per hr): 20 - 30 (positive pressure)
- Photoperiod (hrs dark / hrs light): 12/12

Route of administration:

inhalation

Details on exposure:

Clinical observations were performed in ~30 minute intervals during the exposure period and at least once daily, following exposure.

TYPE OF INHALATION EXPOSURE: whole body
GENERATION OF TEST ATMOSPHERE / CHAMBER DESCRIPTION
- Exposure apparatus: long-term inhalation chambers
- Source of air: clean dry air from the laboratory air supply
- System of generating particulates/aerosols:
Phase 1: passing clean dry air from the laboratory air supply via a flow controller and a flow meter to a jacketed bubbler containing the test material. The bubbler was heated by a flow of warm air at a temperature controlled by a thermocirculator. The generation air passed through the bubbler picking up vapourised test material and was split into two streams, each stream being passed to one of the 2 exposure chambers being used. Two streams of dilution air were each passed through flow controllers and flow meters. Each of the dilution streams then joined a generation stream prior to entry to the exposure chamber. The diluted test material stream was then passed through the exposure chamber and was subsequently vented into a fume cupboard. Air flows were monitored continuously using flowmeters (KDG Flowmeters, Burgess Hi 11, Sussex, UK) and were recorded at approximately 30 minute intervals during the exposure periods.
Phase 2 and 3: The atmospheres for Phases 2 and 3 of the study were generated using the system described above except that flow controllers were not fitted into the system and as only single chambers were supplied from each generation system, only one dilution stream was used.
Vinyl chloride: Vinyl chloride was extracted from the cylinder and mixed with compressed air at a flow rate to allow generation of 50000 ppm. Both the air and vinyl chloride flow rates were monitored using in-line flow meters with needle valves.

Duration of treatment / exposure:

6 hours

Dose / conc.:

5.9 ppm (nominal)

Remarks:

♂

Dose / conc.:

11.8 ppm (nominal)

Remarks:

♂

Dose / conc.:

18.9 ppm (nominal)

Remarks:

♂

Dose / conc.:

3.7 ppm (nominal)

Remarks:

♀

Dose / conc.:

7.5 ppm (nominal)

Remarks:

♀

Dose / conc.:

11.9 ppm (nominal)

Remarks:

♀

No. of animals per sex per dose:

5

Positive control(s):

Vinyl chloride

Tissues and cell types examined:

Bone marrow, erythrocytes

Details of tissue and slide preparation:

CRITERIA FOR DOSE SELECTION:
Phase 1 involved the determination of the median lethal concentration (MLC) calculated on the deaths observed over a 4-day observation period using a single 6-hour inhalation exposure.

TREATMENT AND SAMPLING TIMES:
Bone marrow smears were prepared 24, 48, and 72 h after the end of the exposure periods in Phase 2 and 24 h after the end of the exposure period in Phase 3. The preparations were stained with polychrome methylene blue and eosin to visualise the various cell types. One thousand polychromatic erythrocytes per slide were originally evaluated for the presence of micronuclei. An additional 2000 polychromatic erythrocytes were also evaluated for the presence of micronuclei from all slides from male animals exposed to the air control or TDI 24 h after exposure and female animals exposed to the air control or TDI 24 and 48 h after exposure in Phase 2. In addition, 1000 erythrocytes were counted to determine the percentage of polychromatic erythrocytes in the total erythrocyte population. This provides an indication of any cytotoxicity in the bone marrow.

Statistics:

The incidence of micronucleated polychromatic erythrocytes and percentage polychromatic erythrocytes in the erythrocyte sample was considered by analysis of variance, regarding each combination of sampling time, concentration and sex as a separate group. The results were examined to determine whether any differences between air control and TDI treated groups were consistent between sexes and across sampling times. The data from the extended counts were similarly analysed as an independent database and also after combination with the original counts. All analyses were carried out after calculating the average number of micronuclei per 1000 polychromatic erythrocytes. The values for micronucleated polychromatic erythrocytes were transformed using a natural logarithmic transformation, to stabilise the variance, before analysis. All analyses were carried out using the GLM procedure in SAS (1985). Unbiased estimates of the group means were provided by the least square means (LSMEANS option in SAS) but for simplicity standard means are presented. Each treatment group mean was compared with the air control group mean at the corresponding sampling time using a one-sided Student's t-test based on the error mean square in the analysis.

Key result

Sex:

male/female

Genotoxicity:

negative

Toxicity:

yes

Positive controls validity:

valid

Additional information on results:

Atmosphere Analysis:
Examination of peak areas of the gas chromatograph confirms that the ratio of 2,4 to 2,6 TDI in the exposure chambers approximated closely to the expected ratio of 80:20. There was no TDI detected in control group atmospheres and no TDI or vinyl chloride were detected in the room air samples.

Phase 1 - MLC Determination
Groups of 5 male and 5 female mice were exposed to TDI at target concentrations of 7, 10, 15, 20, and 30 ppm. From the resulting mortalities the MLC over a four day observation period was calculated by logistic regression as 14.1 ppm for females and by linear log interpolation as 19.0 ppm for males. Atmosphere concentrations of 11.8 and 18.9 ppm for males and 7.5 and 11.9 ppm for females were used in Phase 2 of the study. In both cases these concentrations were selected to represent 50 and 80 %, respectively of the median lethal concentration (MLC). Due to an error in the original calculation of the MLC values the target concentrations used actually represented 62 and 99 % of the MLC in males and 53 and 84 % of the MLC in females.

Phase 2 and Phase 3 - Micronucleus Test
In Phase 2 of the study clinical signs were recorded for mice exposed to TDI as follows: male mice exposed to TDI at the target concentration of 11.8 ppm had a reduced response to stimulus throughout the exposure period, subdued nature, increased breathing depth, reduced breathing rate and piloerection, whereas male mice exposed to TDI at 18.9 ppm had no visible response to stimulus, very subdued nature, hunched posture, reduced breathing rate and increased breathing depth. The males exposed to TDI at 18.9 ppm were also noted to be subdued the day following exposure. In addition, one male exposed to TDI at 18.9 ppm was found dead in its cage approximately 48 h after exposure. Female mice exposed to TDI at the target concentration of 7.5 and 11.9 ppm exhibited reduced response to stimulus, reduced breathing rate and increased breathing depth during exposure. In addition the females exposed to the target concentration of 11.9 ppm exhibited hunched posture and piloerection during exposure. After exposure females exposed to TDI at 7.5 ppm exhibited hunched posture, subdued nature and piloerection, whereas those exposed at 11.9 ppm exhibited clinical signs including subdued nature, hunched posture, piloerection and reduced temperature. In addition, one female exposed to TDI at 11.9 ppm was found dead in its cage 24 h after the end of the exposure period.
Males exposed to vinyl chloride were noted to have a slightly subdued nature, a reduced response to stimulus and hunched posture during exposure, and one male exhibited a subdued nature the day following exposure. Females exposed to vinyl chloride were noted to be exhibiting a subdued nature, hunched posture, pi loerection, reduced response to stimulus, reduced breathing rate and increased breathing depth during the exposure period and hunched posture, subdued nature and piloerection after exposure.
Small but statistically significant increases in the incidence of micronucleated polychromatic erythrocytes over the air control values were observed in females 24 h after being exposed at the target concentration of 7.5 ppm TDI and 24 and 48 h after being exposed at the target concentration of 11.9 ppm. These increases were small and not concentration-related.
Extended analysis of a further 2000 polychromatic erythrocytes from these animals and the female air control animals at the 24 and 48 h time points was conducted. No statistically or biologically significant increases in the incidence of micronucleated polychromatic erythrocytes were observed in these extended counts. However, when the original and extended analyses were combined prior to statistical analysis small but statistically significant increases were observed in females 24 h after being exposed at both target concentrations.
Small but statistically significant increases in the incidence of micronucleated polychromatic erythrocytes, over the air control values, were observed in males 24 h after being exposed at the target concentrations of 11.8 and 18.9 ppm but there was no clear concentration-response relationship. Extended analysis of a further 2000 polychromatic erythrocytes from these animals and the male air control animals at the 24 h time point was conducted. A small but statistically significant increase in the incidence of micronucleated polychromatic erythrocytes was observed only at the lower target concentration (11.8 ppm) in these extended counts and when the original and extended analyses were pooled prior to statistical analysis.
In order to further investigate the increases observed in both males and females exposed to TDI and the lack of concentrat ion-response relationships observed a second assay was conducted. Groups of 5 male mice were exposed to TDI for a 6 h period by the inhalation route at target concentrations of 5.9, 11.8, and 18.9 ppm and groups of 5 female mice were similarly exposed to TDI at target concentrations of 3.7, 7.5, and 11.9 ppm. In both cases these concentrations were selected to represent the concentrations used in the first study with an additional lower concentration to investigate any concentration-response relationships. Due to error in the original calculation of the MLC values, the target concentrations used actually represented approximately 31, 62, and 99 % of the MLC in males and 26, 53, and 84 % of the MLC in females. Bone marrow samples were taken 24 h after the end of the exposure period for all concentrations.
Adverse reactions to treatment was recorded for mice exposed to TDI. Clinical signs recorded for male mice exposed to TDI at 5.9, 11.8 and 18.9 ppm were reduced response to stimulus, subdued nature and decreased breathing rate during exposure, although due to misting of the inside of the exposure chamber difficulty was experienced in carrying out the clinical observations on the 11.8 and 18.9 ppm concentration groups. In addition, 4 males exposed to TDI at 11.8 ppm were found dead in their cages and the remaining male was killed in extremis. One male exposed to TDI at 18.9 ppm was killed in extremis. Clinical signs recorded for female mice exposed to TDI at 3.7, 7.5, and 11.9 ppm included reduced response to stimulus and reduced breathing rate. In addition, females exposed to TDI at 7.5 and 11.9 ppm exhibited hunched posture and little movement although difficulty was experienced in carrying out the clinical observations due to misting of the exposure chambers. In addition, six females exposed to TDI at 11.9 ppm were found dead in their cages following exposure to TDI.
Males exposed to vinyl chloride were noted to have a reduced response to stimulus, piloerection and hunched posture whereas females exposed to vinyl chloride exhibited a hunched posture and reduced response to stimulus.
In this second study high levels of lethality were observed at the 11.8 ppm (62 % MLC) concentration in males and the 11.9 ppm (84 % MLC) concentration in females and therefore the slides from the males exposed at the 5.9 ppm target concentration and the females exposed at the 3.7 and 7.5 ppm target concentrations only were analysed. The maximum concentration in each case is considered to represent a maximum tolerated concentration (MT.C .) in this second study.
No statistically or biologically significant increases in the incidence of micronucleated polychromatic erythrocytes, compared to the air control values, were observed in the males exposed to TDI. Small but statistically significant increases in the incidence of micronucleated polychromatic erythrocytes, over the air control values, were observed in females 24 h after being exposed at both the 3.7 and 7.5 ppm target concentrations. These increases were small and the values fell within the range of female air control values reported in this study.
Statistically significant decreases in the percentage of polychromatic erythrocytes were observed in both males (24 and 48 h; 18.9 ppm; and females (24 h; 11.9 ppm) on the first study and in females exposed at the 7.5 ppm target concentration in the second study. The test system positive control, vinyl chloride, induced statistically and biologically significant increases in the incidence of micronucleated polychromatic erythrocytes in both male and female animals at the 24 h sampling time on both studies.

DISCUSSION
The criteria for a valid test system as laid down by OECD Guideline 474 (1983) for the conduction of micronucleus studies, are that the positive control substance should induce a significant elevation in micronucleated polychromatic erythrocytes compared to the vehicle control values and that the test compound should be tested at a level that causes a decrease in the percentage of polychromatic erythrocytes (indicating a cytotoxic effect on the bone marrow) or at the maximum tolerated dose level. The study satisfies these criteria in that TDI was tested in excess of 80 % of a median - lethal concentration (MLC), a concentration which also induced adverse reactions to treatment. Consideration of the percentage of polychromatic erythrocytes showed statistically significant decreases, compared to the air control values, in both males and females in the first study and in females in the second study. These decreases may indicate that TDI or a metabolite has induced a cytotoxic response in the bone marrow resulting in a depression of cell proliferation. The positive control substance, vinyl chloride, gave a statistically significant and biologically meaningful increase in micronucleated polychromatic erythrocytes, compared to air control values, in both male and female mice in both studies.
Small but statistically significant increases in the incidence of micronucleated polychromatic erythrocytes, over the air control values, were observed in both males and females exposed to TDI in the first study. The increases at the 24 h time point were confirmed by extended analysis of the slides and a second assay was conducted to clarify the increases observed.
No statistically or biologically significant increases were observed in the males in the second study, and although small but statistically significant increases were observed in the females the values fell within the range of female air control values reported in this study. It is therefore considered that the increases observed in the second study are due to a low control value rather than to any effect of TDI. The increases are therefore considered not to be biologically significant.
In summary, although increases in the incidence of micronucleated polychromatic erythrocytes were observed in both males and females exposed to TDI these increases were small, not concentration-related and were not reproducible at concentrations limited by lethality in a repeat study. It is therefore considered that the increases observed are of no biological significance and do not indicate any clastogenic activity of TDI in the mouse bone marrow micronucleus assay.

For individual animal data see attached full study report in section "attachments".

Exposure levels of 11.8 ppm in males and 11.9 ppm in females were lethal.  Therefore bone marrows assessed only at 5.9  ppm in males and 3.7 and 7.5 ppm in females. No effect on male bone marrow at 5.9 ppm was observed. 

Small statistically significant increase in micronucleated polychromatic erythrocytes (MPE) in females at 3.7 and 7.5 ppm were observed. However, the values were within control range, therefore changes are not considered to be biologically significant.

Positive control, vinyl chloride, induced statistically and  biologically significant increases in MPE, demonstrating the sensitivity of test system.

 

Mean Incidence of MPE/1000 PE ±SD        

Conc.	24 h	48 h	72 h	Extended Counts	Combined + Original
Males	.	.	.	(24 h)	(24 h)
Control	2.0±1.2	1.4±0.9	2.0±1.6	2.8±1.1	2.5±1.2
11.8 ppm	7.4±4.5**	.	5.9±2.1**	6.4±3.0**	.
18.9 ppm	4.4±2.0*	1.8±2.9	1.8±0.8	1.9±1.1	2.7±1.8
Repeat study	.	.	.	.	.
Control	1.2±0.5	.	.	.	.
5.9 ppm	2.0±1.9	.	.	.	.
.	.	.	.	.	.
Conc.	24 h	48 h	72 h	Extended Counts	Combined + Original
Females	.	.	.	(24 h)	(24 h)
Control	0.4±0.9	0.6±0.6	1.4±0.6	1.4±1.1	1.1±1.4
.	.	.	.	1.3±1.1	1.1±1.0(48h)
7.5 ppm	4.0±1.4**	.	.	2.3±2.1	2.9±2.0**
11.9 ppm	1.8±1.5*	2.0±1.4*	0.5±0.6	2.0±1.4	1.9±1.4*
.	.	.	.	(48 h)	(48 h)
Repeat study	.	.	.	.	.
Control	0.2±0.5	.	.	.	.
3.7 ppm	1.2±0.8*	.	.	.	.
7.5 ppm	1.4±0.9*	.	.	.	.

**Sig. at p = 0.01
*Sig. at p = 0.05

 

Mean % Polychromatic Erythrocytes ±SD   

Conc.	24 h	48 h	72 h
Males	.	.	.
Control	38.3±3.9	40.7±2.1	41.1±4.4
11.8 ppm	38.8±7.9	.	.
18.9 ppm	27.1±7.4*	29.8±13.5**	35.3±6.0
Repeat study	.	.	.
Control	48.4±3.3	.	.
5.9 ppm	46.1±4.1	.	.
.	.	.	.
Conc.	24 h	48 h	72 h
Females	.	.	.
Control	41.7±5.8	34.7±7.1	37.1±7
7.5 ppm	39.1±5.0	.	.
11.9 ppm	34.0±3.1*	29.5±4.3	33.2±7.1
.	.	.	.
Repeat study	.	.	.
Control	47.3±2.8	.	.
3.7 ppm	43.9±1.0	.	.
7.5 ppm	40.1±8.0*	.	.
7.5 ppm	40.1±8.0*	.	.

**Sig. at p 0.01
*Sig. at p 0.05

Reference 2

Endpoint:

in vivo mammalian cell study: DNA damage and/or repair

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

study well documented, meets generally accepted scientific principles, acceptable for assessment

Principles of method if other than guideline:

Ashby et al. (1985) Method for UDS adapted for a single exposure by inhalation.

GLP compliance:

yes

Type of assay:

unscheduled DNA synthesis

Species:

rat

Strain:

Fischer 344

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
All animals were checked for signs of illness or poor condition on arrival and during the pre-exposure period.

Route of administration:

inhalation

Details on exposure:

Rats were exposed via inhalation for 4 h. Animals were observed during the exposure period for clinical signs. Immediately after exposure, animals were removed from the chambers and allowed to recover overnight. Autopsies were then performed on 3 rats/group.

Duration of treatment / exposure:

4 hours

Frequency of treatment:

single exposure

Post exposure period:

over night

Dose / conc.:

0.077 ppm (nominal)

Dose / conc.:

0.4 ppm (nominal)

Dose / conc.:

1.49 ppm (nominal)

No. of animals per sex per dose:

4

Control animals:

yes

Details of tissue and slide preparation:

DETAILS OF SLIDE PREPARATION:
Sections of the livers and lungs were removed and hepatocyte and lung explant cultures were then prepared immediately. Sections of lung, trachea, nasal turbinates, liver and kidney were removed and fixed in 10 % neutral buffered formalin. The hepatocyte cultures were then incubated for 2 h and then medium removed and labeled thymidine in serum free medium was added to each plate and dishes incubated for 3 - 4 h. The labeled medium was then washed out with fresh unlabeled medium and the cultures were incubated overnight. Cells were then fixed and slides prepared. The results of the test were quantified by determining the production of net nuclear grains. The nuclear grain counts and the highest grain count from an equivalent area of cytoplasm were scored on each of 50 randomly selected cells. For the lung explant cultures, 4 pieces of lung from each animal (3 mm³) were cut and placed in petri dishes. Medium containing labeled thymidine was added and cultures were incubated in a CO2 chamber for 4 h. Cultures were then washed, transferred to 10 % formalin and then embedded into paraffin and cut into sections for slide preparation. All slides were autoradiographed and then stained with hematoxylin and eosin. Only explants with normal histology were analyzed. The net number of grains per nucleus was determined in 50 cells in healthy alveolar regions of each slide. Evidence of unscheduled DNA synthesis was examined in cultured hepatocytes and lung explant samples.

Statistics:

Mean net nuclear counts ± S.E.M. were determined for each of the triplicate slides per animal and the mean ± S.D. of net nuclear counts and percentage of cells in repair for each rat was then calculated. From these values, the mean ± S.D. for each dose group was determined. The test compound is considered positive if the mean net nuclear grain count of the treated animals is statistically greater than that of controls and equal to or greater than 2 grains/nucleus (the upper limit ofcontrol values).

Key result

Sex:

male

Genotoxicity:

negative

Toxicity:

yes

Remarks:

Severe upper respiratory tract irritation observed, but no toxicity in the lung or liver

TDI did not induce UDS at any of the administered doses in lung cultures from any of the treated animals.

Histopathology of the preserved upper respiratory tract sections from each animal, evidence of lung infection (pneumonitis or bronchopneumonia) was observed in controls and consisted of inflammatory cell infiltrate in the interstitial space and alveoli in the lungs and nasal region. Clear dose-related effects were observed in the upper respiratory tract in rats exposed to TDI by inhalation. At the lowest dose, only 1 animal was affected by hyperplasia in the epithelium covering the turbinates and nasal septum. Hyperplasia and metaplasia in these areas increased with increase in dose and at the highest dose, one animal had minute but clearly definable ulceration. The trachea and main bronchi were affected in 1 animal at this dose level. There were no lesions in liver or kidneys from any animal exposed to TDI by inhalation.

Reference 3

Endpoint:

in vivo mammalian somatic cell study: cytogenicity / erythrocyte micronucleus

Remarks:

Type of genotoxicity: chromosome aberration

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

study well documented, meets generally accepted scientific principles, acceptable for assessment

Principles of method if other than guideline:

Method: other

GLP compliance:

yes

Type of assay:

micronucleus assay

Species:

rat

Strain:

Sprague-Dawley

Sex:

male/female

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Age at study initiation: 6 - 9 weeks

Route of administration:

inhalation

Duration of treatment / exposure:

6 h / day

Frequency of treatment:

5 days per week, for 4 weeks.

Dose / conc.:

0.05 ppm (nominal)

Dose / conc.:

0.15 ppm (nominal)

No. of animals per sex per dose:

5

Details of tissue and slide preparation:

DETAILS OF SLIDE PREPARATION:
Bone marrow from femur was aspirated with 10 % foetal calf serum, the suspension was centrifuged and a film of sedimented cells was prepared. The film was stained and the number of micronuclei-containing cells per 1000 polychromatic erythrocytes was estimated.

Key result

Sex:

male/female

Genotoxicity:

negative

Toxicity:

no effects

Reference 4

Endpoint:

in vivo mammalian somatic cell study: cytogenicity / erythrocyte micronucleus

Remarks:

Type of genotoxicity: chromosome aberration

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

study well documented, meets generally accepted scientific principles, acceptable for assessment

Principles of method if other than guideline:

Method: other

GLP compliance:

yes

Type of assay:

micronucleus assay

Species:

mouse

Strain:

CD-1

Sex:

male/female

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Age at study initiation: 5 - 6 weeks

Route of administration:

inhalation

Duration of treatment / exposure:

6 h / day

Frequency of treatment:

5 days per week, for 4 weeks.

Dose / conc.:

0.05 ppm (nominal)

Dose / conc.:

0.15 ppm (nominal)

No. of animals per sex per dose:

5

Details of tissue and slide preparation:

DETAILS OF SLIDE PREPARATION:
Bone marrow from femur was aspirated with 10 % foetal calf serum, the suspension was centrifuged and a film of sedimented cells was prepared. The film was stained and the number of micronuclei-containing cells per 1000 polychromatic erythrocytes was counted.

Key result

Sex:

male/female

Genotoxicity:

negative

Toxicity:

no effects

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed (negative)

Additional information

Aromatic diisocyanates are virtually insoluble in water. Therefore, an organic solvent is required to ensure homogeneous dispersion in in vitro genotoxicity assays. Dimethylsulphoxide (DMSO) has been used routinely as the vehicle of choice for such assays. The validity of using DMSO as a solvent was queried by Gahlman et al. (1993)* when it was found that there was a chemical conversion of TDI to TDA in the solvent which could explain a number of positive responses recorded in some in vitro genotoxicity assays. A detailed evaluation of the stability of TDI in dimethylsulphoxide (DMSO) by Seel et al. (1999) showed there is a rapid breakdown of TDI in DMSO with less than 60 % of the initial amount remaining after 15 minutes. A HPLC examination of the breakdown products showed TDA was first detected at 15 minutes rising to 8 % after 30 minutes. The authors concluded that in traditional bacterial mutation assays with Salmonella typhimurium using DMSO, the solvent conversion of TDI to ureas, polyureas and TDA would be complete within minutes and the TDI would not have been tested. To determine if the positive results seen in in vitro genotoxicity assays when TDI was dissolved in DMSO was in fact a consequence of the chemical break down of TDI to TDA. Seel et al. (1999) undertook a series of mutagenic investigations using dry ethylene glycol dimethylether (EGDE) as the organic solvent as investigations indicated TDI was stable in this solvent with 98 to 99 % of original TDI remaining after 1 hour and more than 85 % after 4 hours with no detectable formation of TDA. The studies with Salmonella typhimurium showed quite clearly the absence of any mutagenic response when TDI was dissolved in EGDE. Based on such evaluation the authors concluded that positive results seen in vitro genotoxicity studies undertaken using solvents such as DMSO must be treated with caution as such effects are very well be an artifact of the testing conditions caused by the breakdown of TDI to TDA which is known to produce mutations in Salmonella typhimurium. Based on these observations the use of results from in vitro tests in aqueous cell systems are problematic because of interaction with the test system components. These studies are considered to be invalid, and not useful for determining the genotoxic potential of TDI. For this reason mammalian cell gene mutation assays in vitro are not feasible and assessment relies on the in vivo studies.

A number of in vivo genotoxicity studies have been carried out with TDI. A slight increase in numbers of micronucleated erythrocytes was measured in a non-GLP micronucleus assay in rats exposed to TDI via inhalation (Owen, 1980, Loeser 1983). As the increase was not significant, occurred at only one dose level and because of the probably hyperthermia caused by the treatment the result was not considered to be biologically significant. Negative results were obtained with mice in the same study using similar exposures. Negative results have also been seen in a well conducted micronucleus assay in mice using inhalation route of exposure (Mackay, 1992) and an unscheduled DNA synthesis assay examining effects in liver and lungs in rats after acute and sub-acute inhalation exposures to TDI (Benford and Riley, 1988). Commercial grade TDI was also inactive in inducing sister chromatid exchanges and micronuclei in lung cells after intratracheal instillation in rats (Whong et al, 1991, cited in Zeiger 2005). Studies examining DNA adduct formation have produced mixed results and are inconclusive as to their relevance to human exposures.

Overall the data on genotoxicity show:

•Weight of scientific evidence supports the conclusion that TDI is not mutagenic or genotoxic.

•As TDI is unstable in solvents such as DMSO and rapidly degrades to TDA results from the majority of in vitro genotoxicity test results are unsuitable for assessing the genotoxic potential of TDI.

•Inhalation of TDI does not induce micronuclei formation or DNA damage as measured by unscheduled DNA synthesis.

•Supplemental investigations of DNA binding have proven inconclusive, as data were, in the main, obtained with non-validated methodologies and the results are difficult to interpret.

* Gahlmann R, Herbold A, Ruckes A, Seel K. Tests on the stability of aromatic diisocyanates in dimethylsulphoxide (DMSO): toluene diisocyanate (TDI) and diphenylmethane diisocyanate (MDI) in the Ames test, Zbl Arbeitsmed 43 (1993), 34 -38.

Justification for classification or non-classification

The available experimental test data with m-tolylidene diisocyanate are reliable and suitable for classification purposes under Regulation (EC) No 1272/2008. In vitro results with m-tolylidene diisocyanate were ambiguous/weakly positive. The explanation and reliability for the in vitro results is given above. In vivo results with m-tolylidene diisocyanate were negative. As a result the test substance is not considered to be classified for genetic toxicity under Regulation (EC) No 1272/2008, as amended for the tenth time in Regulation (EU) No 2017/776.