Triboron trimethyl hexaoxide

Endpoint summary

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

key study

Study period:

No data

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: Comparable to a guideline study.

Objective of study:

excretion

Qualifier:

according to guideline

Guideline:

other: No data

Deviations:

not specified

Principles of method if other than guideline:

Groups of 10 Sprague-Dawley (Charles River) pregnant or non-pregnant rats, were maintained on a specially prepared low boron diet (approximately equivalent to 0.3 boric mg/kg/day) for 7 days prior and switched to a lower boron diet 24 h prior to treatment. Doses of 0.3 (0.052), 3.0 (0.52) or 30 (5.2) mg boric acid (mg B)/kg in water was administered by gavage. Plasma boron was determined at 3 h and 15 h after administration of boric acid and urine was collected for 12 h after the first blood sample was taken. A separate experiment was performed to estimate the plasma half-life. Six pregnant and six non-pregnant rats were treated in the same way as for the renal clearance study. They received a single dose of 30 (5.2) mg boric acid (mg B)/kg (on GD 16) for the pregnant rats. Blood samples (0.25 mL) were taken by periorbital puncture after light anaesthesia at 3, 5, 7, 9,12 and 15 hours post dosing.

GLP compliance:

not specified

Radiolabelling:

not specified

Species:

rat

Strain:

Sprague-Dawley

Sex:

female

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Age at study initiation: 12-weeks old
- Weight at study initiation: 200 - 250 g
In addition 37 timed-pregnant rats of simillar weight range were used.
- Diet: Ad libitum
- Water: Ad libitum

ENVIRONMENTAL CONDITIONS
- Temperature (°C): 22 ± 2 °C
- Humidity (%): 50 - 60 %
- Photoperiod (hrs dark / hrs light): 12h dark/light
- Other: Animals were maintained in stainless steel cages and potential soures of boron were minimised.

Route of administration:

oral: gavage

Vehicle:

water

Details on exposure:

All pregnant and non-pregnant rats were maintained on a casein-based rat diet. To ensure an adequate amount of dietary boron a "low boron" diet (1.4 mg BA/kg diet, or 0.25 mg B/kg diet) was used and supplemented with boric acid (3.5 mg BA/kg diet, or 0.64 mg B/kg diet) and the rats were given the supplemented diet on days 1 through 7 of the study (prior to gavage adminsitration of boric acid). Pregnant rats were placed on the supplemented diet beginning on day 9 of gestation for the first 7 days of the study. This diet aimed to acheive steady state conditions for rats given a diet comparable to that ingested by humans in terms of boron intake. The supplemented diet contained about 15 - 25 times less than boron in Purina rat chow and was designed to deliver a dose of approximately 0.3 mg/kg/day of boric acid (equivalent to 0.05 mg B/kg/day).
On the afternoon of the 7th day (GD 15 for pregnant rats) through the evening of the 8th day of the study all pregnant and non-pregnant rats were switched to low-boron caein diets without the boric acid supplementation to minimise any cross-contaminiation of the urine with boron in the diet.

Duration and frequency of treatment / exposure:

Single administration

Remarks:

Doses / Concentrations:
Renal clearance study: 0.3, 3.0 or 30 mg boric acid/kg bw; 0.052, 0.52 and 5.2 mg boron /kg respectively by gavage.
Plasma half-life study: 30 mg boric acid/kg.

No. of animals per sex per dose / concentration:

Renal clearance study: 10 rats per group, sex not specified
Plasma half-life study: 6 pregnant and 6 non-pregnant females.

Control animals:

yes, concurrent vehicle

Positive control reference chemical:

No data

Details on study design:

Renal clearance study:
The lowest dose was comparable to the high end of the normal range of human dietary intake of boron. The highest dose was approximately half of the NOAEL.

Plasma half-life study:
The dose was equivalent to the highest dose in the clearance study and was selected because if any dose level exhibits a non-linear plasma curve it is most likely to be the high dose. It is assumed that if the high dose is linear then the lower doses would also be linear.

Details on dosing and sampling:

Renal clearance study:
Two blood samples were drawn from each rat, the first after approximately 3 h after administration; the second approximately 12 h after the first.
A 12 h urine sample was collected from each rat the clearance study during the period between the first and second blood samples being taken.

Plasma half-life study:
Six blood samples were drawn from each animal during a 12 h period starting 3 h after dosing on Day 8 of the study.

Statistics:

Reanal clearance was expressed as mean ± standard deviation. Two way analysis of variance, multiple range test (Student-Newman-Keuts Method) was used as appropriate. For all statistical analyses p values < 0.05 were considered statistically significant.

Type:

excretion

Results:

Renal clearance: 3.1 mL/min/kg for non-pregnant rats, 3.2 mL/min/kg for pregnant rats. Clearance independent of dose up to 30 mg boric acid/kg bw. (5.24 mgB/k).

Details on absorption:

Plasma half-life evaluation:
Gavage administration resulted in plasma levels of 1.82 ± 0.32 and 1.78 ± 0.32 μg B/mL among non-pregnant and pregnant rats in the first blood sample which was taken 3 h after dosing. This was followed by a monophasic decline in plasma boron concentrations in both pregnant and non-pregnant rats; the plasm levels were consistent with a compartmental model. There was no evidence of saturation kinetics. The estimated half-lives of boric acid in non-pregnant and pregnant rats were 2.93 ± 0.24 and 3.23 ± 0.28 h respectively. This difference was not statistically significant.

Details on distribution in tissues:

No data

Details on excretion:

The urinary concentration of boron was significantly higher in pregnant compared to non-pregnant rats at the high dose, but not at the mid or low dose. The concentration of boron in the urine during the 12 h collection period in the urine of non-pregnant rats was 1.67 ± 0.62, 10.12 ± 8.16 and 66.82 ± 47.00 μg B/mL at the low, mid and high doses respectively. In pregnant rats the corresponding urine boron concentrations were 1.62 ± 0.49, 12.30 ± 5.12 and 121.45 ± 47.09 μg B/mL, respectively. The amount of boron excreted in the urine increased proportionately with increasing dose. The percentage of the administered dose recovered in the urine was significantly higher in the low dose group compared to the mid and high dose groups. No significant dose-related differences in boric acid clearance were observed in either non-pregnant or pregnant rats.

Toxicokinetic parameters:

half-life 1st: The plasma half-life of boric acid in non-pregnant and pregnant rats given boric acid by gavage was 2.93 ± 0.24 and 3.23 ± 0.28 hours, respectively.

Metabolites identified:

no

Details on metabolites:

Boric acid is not metabolised.

Conclusions:

Interpretation of results (migrated information): no data
Gavage administration resulted in plasma levels of 1.82 ± 0.32 and 1.78 ± 0.32 μg B/mL among non-pregnant and pregnant rats in the first blood sample which was taken 3 h after dosing. This was followed by a monophasic decline in plasma boron concentrations in both pregnant and non-pregnant rats; the plasma levels were consistent with a compartmental model. There wsa no evidence of saturation kinetics. The estimated half-lives of boric acid in non-pregnant and pregnant rats were 2.93 ± 0.24 and 3.23 ± 0.28 h respectively. This difference was not statistically significant.
The urinary concentration of boron was significantly higher in pregnant compared to non-pregnant rats at the high dose, but not at the mid or low dose. The concentration of boron in the urine during the 12 h collection period in the urine of non-pregnant rats was 1.67 ± 0.62, 10.12 ± 8.16 and 66.82 ± 47.00 μg B/mL at the low, mid and high doses respectively. In pregnant rats the corresponding urine boron concentrations were 1.62 ± 0.49, 12.30 ± 5.12 and 121.45 ± 47.09 μg B/mL, respectively. The amount of boron excreted in the urine increased proportionately with increasing dose. The percentage of the administered dose recovered in the urine was significantly higher in the low dose group compared to the mid and high dose groups. No significant dose-related differences in boric acid clearance were observed in either non-pregnant or pregnant rats.

Reference 2

Endpoint:

basic toxicokinetics

Type of information:

migrated information: read-across from supporting substance (structural analogue or surrogate)

Adequacy of study:

supporting study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: see 'Remark'

Remarks:

well-documented publication which meet basic scientific principles Read-across is justified on the following basis: The family of zinc borates that include Zinc Borate 500, Zinc Borate 2335 and Zinc Borate 415 (also known as Zinc Borate 411). Zinc borate 500 is anhydrous Zinc Borate 2335 and Zinc Borate 415 has different zinc to boron ratio. Zinc borate 2335 (in common with other zinc borates such as Zinc borate 415 and 500) breaks down to Zinc Hydroxide (via Zinc oxide) and Boric Acid, therefore the family of zinc borates shares the same toxicological properties. Zinc borates are sparingly soluble salts. Hydrolysis under high dilution conditions leads to zinc hydroxide via zinc oxide and boric acid formation. Zinc hydroxide and zinc oxide solubility is low under neutral and basic conditions. This leads to a situation where zinc borate hydrolyses to zinc hydroxide, zinc oxide and boric acid at neutral pH quicker than it solubilises. Therefore, it can be assumed that at physiological conditions and neutral and lower pH zinc borate will be hydrolysed to boric acid, zinc oxide and zinc hydroxide. Hydrolysis and the rate of hydrolysis depend on the initial loading and time. At a loading of 5% (5g/100ml) zinc borate hydrolysis equilibrium may take 1-2 months, while at 1 g/l hydrolysis is complete after 5 days. At 50 mg/l hydrolysis and solubility is complete (Schubert et al., 2003). At pH 4 hydrolysis is complete. Zinc Borate 2335 breaks down as follows: 2ZnO • 3B2O3 •3.5H2O + 3.5H2O + 4H+ ↔ 6H3BO3 + 2Zn2+ 2Zn2+ + 4OH- ↔ 2Zn(OH)2 ____________________________________________________________ Overall equation 2ZnO • 3B2O3 •3.5H2O + 7.5H2O ↔ 2Zn(OH)2 + 6H3BO3 The relative zinc oxide and boric oxide % are as follows: Zinc borate 2335:zinc oxide = 37.45% (30.09% Zn) B2O3 = 48.05% (14.94% B) Water 14.5% Zinc borate 415: zinc oxide = 78.79%; (63.31% Zn) B2O3 = 16.85% (5.23% B) Water 4.36% Zinc borate, anhydrous: Zinc oxide = 45 % B2O3= 55% (17.1 % B)

Objective of study:

other: Re-evaluation of Toxicokinetics for Data-Derived Uncertainty Factors

Qualifier:

no guideline followed

Principles of method if other than guideline:

A critical analysis of the existing data on boron toxicokinetics was conducted to clarify the appropriateness of replacing default uncertainty factors (10-fold for interspecies differences and 10-fold for intraspecies differences) with data-derived values.

GLP compliance:

not specified

Radiolabelling:

no

Species:

other: humans and rats

Details on study design:

Boron, which is ubiquitous in the environment, causes developmental and reproductive effects in experimental animals. This observation has led to efforts to establish a Tolerable Intake value for boron. Although risk assessors agree on the use of foetal weight decreases observed in rats as an appropriate critical effect, consensus on the adequacy of toxicokinetic data as a basis for replacement of default uncertainty factors remains to be reached. A critical analysis of the existing data on boron toxicokinetics was conducted to clarify the appropriateness of replacing default uncertainty factors (10-fold for interspecies differences and 10-fold for intraspecies differences) with data-derived values.
The default uncertainty factor for variability in response from animals to humans of 10-fold (default values of 4-fold for kinetics and 2.5-fold for dynamics) was recommended, since clearance of boron is 3- to 4-fold higher in rats than in humans and data on dynamic differences—in order to modify the default value—are unavailable. A data-derived adjustment of 6-fold (1.8 for kinetics and 3.1 for dynamics) rather than the default uncertainty factor of 10-fold was considered appropriate for intrahuman variability, based on variability in glomerular filtration rate during pregnancy in humans and the lack of available data on dynamic differences. Additional studies to investigate the toxicokinetics of boron in rats would be useful to provide a stronger basis for replacement of default uncertainty factors for interspecies variation.

Type:

absorption

Results:

Absorption of Boron varied between 81 and 98 % in humans

Type:

excretion

Results:

Excretion and elimination, respectively was found to be from 81 to 99 %.

Type:

other: assessment factor

Results:

A data-derived AF of 6-fold (1.8 for kinetics and 3.1 for dynamics) was considered appropriate for intrahuman variability, based on variability in glomerular filtration rate during pregnancy in humans and the lack of available data on dynamic differences.

Details on absorption:

Absorption of Boron varied between 81 and 98 % in humans (based on the studies: Hunt et al., 1997, Nielsen, 1996, Schou et al., 1984, Job, 1973, Kent and McCance, 1941 and Vanderpool et al., 1994).
refer to Table 2.

Details on excretion:

Excretion and elimination, respectively was found to be from 81 to 99 % (based on the studies: Hunt et al., 1997, Nielsen, 1996, Schou et al., 1984, Job, 1973, Kent and McCance, 1941 and Vanderpool et al., 1994)
refer to Table 2.

The 19 toxicokinetic studies that were evaluated are listed in Table 1.

Of these studies, six human studies and five animal studies were considered to contribute meaningfully to this database, although confidence in the data was considered low for some of these. Lack of confidence in these data related primarily to limited information about time of sampling or lack of kinetic analysis.

Eight studies were excluded completely. In general, human studies were excluded based on lack of quantified intake or reports of acute boron overdoses, which precluded adequate analysis of boron kinetics and prediction of steady-state conditions. Three of the animal studies were excluded from further analysis owing to difficulties in interpretation of the data.

Table 1
B Toxicokinetic Studies
Author	Year	Comments*
Human Studies
Barr et al.	1993	included
Gordon et al.	1973	clear overdose; calculated clearance implausible
Hunt et al.	1997	included
Jansen et al.	1984	included
Job	1973	included
Linden et al.	1986	clear overdose; no relevant data for clearance rate determination
Lilovitz et al.	1988	no individual data on intake or blood concentrations
Nielsen	1996	included
O'Sullivan and Taylor	1983	included
Wiley	1906	inadequate methodology
Wong et al.	1964	intake not quantified; time between intake and blood sampling unknown, calculated clearance rates for 3 subjects were not comparable
Animal Studies
Bai and Hunt	1996	included
Chapin et al.	1997	included
Ku et al.	1993	included
Ku et al.	1991	included
Nielsen et al.	1992	blood concentrations were the same with and without supplementation, animals fasted overnight before sampling
Magour et al.	1982	profile of boron concentrations in selected tissues inconsistent with other studies
Price el al.	1997	included
Vanderpool et al.	1994	no relevant data for clearance rate determination
* Comments refer to whether or not clearance was estimated from data in the study.

The ad hoc joint panel agreed that in view of the lack of metabolism of boron in experimental animals and humans and the similarity in absorption and elimination, interspecies variation in kinetics relates principally to renal clearance rates. Table 2 provides an evaluation of the literature on boron absorption and elimination. Over a wide range of doses, the percentage of the boron dose absorbed ranged from 64% to 98% and elimination ranged from 67% to 98% in the human studies. Similar results were reported in the rat. These data indicate that boron is nearly completely absorbed and does not tend to accumulate in humans and animals. Table 3 provides the data on blood boron levels as a function of administered dose.

Table 2
B Absorption and Elimination
Species	Route	Dose mg/kg-day	% Absorption	% Elimination	Reference
human	diet	0.0054	81-92	81-92	Hunt et al. 1997
human	diet	0.049	87-89	89-92	Hunt et al., 1997
human	i.v.	1.4-1.5	NA	99	Jansen et al., 1984
human	drinking water	1.4	>91	>91	Job, 1973
human	diet	ND	83-98	83-98	Kent and McCance, 1941
human	diet	0 05	84	84	Nielsen, 1996
human	drinking water	1.9	94	94	Schou et al., 1984
human	oral	2.5	92	92	Schou et al.. 1984
human	oral	-26	83	86	Wiley, 1906
human	oral	-17	83	86	Wiley, 1906
human	oral	-23	64	67	Wiley, 1906
human	oral	-17	83	86	Wiley, 1906
human	oral	-22	75	78	Wiley, 1906
human	oral	ND	84	87	Wiley, 1906
rat	oral	0.02	95	99	Vanderpool et al., 1994

Table 3
Blood B Levels as a Function of Dose in Humans and Rats
Dose mg/kg-day	Human ng/mL	Rat ng/mL	Route	Reference
0 01	22 (2)"		drinking water	Barr et al., 1993
0 02	34 (10) plasma		diet	Nielsen, 1996
0 02	68 (34)		drinking water	Barr et al., 1993
0 04	52(15)		drinking water	Barr et al., 1993
(1 049	95 (56) plasma		diet	Hunt et al., 1997
0 06	53 plasma		diet	Nielsen, 1996
0,08	347(163)		drinking water	Barr et al., 1993
0 2		200	diet	Chapin et al., 1997
0.3	585(166)		drinking water	Barr et al , 1993
0 35		229 (143)	diet	Price et al., 1997
0 4	450(87)		drinking water	Barr et al., 1993
0 5	659 (337)		drinking water	Barr et al., 1993
0 5		110	diet	Bai et al., 1996
0 65		300	diet & gavage	Bai et al., 1996
1 4	3000(633)		drinking water	Job, 1973
1 7		800	diet	Chapin et al., 1997
3 3		564 (211)	diet	Price et al., 1997
6 3		975 (261)	diet	Price et al., 1997
8 4		2400	diet	Chapin et al., 1997
9 6		1270 (298)	diet	Price et al., 1997
13		1530 (546)	diet	Price et al., 1997
25		2820 (987)	diet	Price et al., 1997
26		6700(1000) scrum	diet	Ku et al., 1993
38		10300(600) serum	diet	Ku et al., 1993
52		13300(700)scram	diet	Ku et al., 1993
68		17300 (2200) serum	diet	Ku et al., 1993
95		16000 (710) plasma	diet	Ku et al., 1993
"Values in parentheses represent the standard deviation.

Based on this information, boron clearance was calculated [clearance (mL/kg/h) = dose (mg/kg/h)/blood concentration (mg/mL)] and the resulting analysis is provided in Table 4. The mean clearance rate for the rat studies was 163 mL/kg/h and the mean clearance rate from human studies was 41 mL/kg/h, therefore, rats have an approximately 4-fold higher boron clearance rate than humans. This 4-fold difference is similar to what one would suspect based on analysis of other chemicals.

Table 4
Blood B Clearance as a Function of Dose in Humans and Rats
Dose mg/kg-day	Clearance ml/kg/hour			Reference
	Human	Rat	Pregnant ral
0.01	18			Barr et al., 1993
0.02	23			Nielsen, 1996
0.021	13			Barr et al., 1993
0.042	33			Barr et al., 1993
0.042	92			Nielsen, 1996
0.049	21			Hunt et al., 1997
0.15 (peak)		94		Bai and Hunt, 1996
0.28	20			Barr et al., 1993
0.317		240		Nielsen et al., 1992
0.35	32			Barr et al., 1993
0.45	28			Barr et al., 1993
0.5 (peak)		189		Bai and Hunt, 1996
0.65 (peak)		152		Bai and Hunt, 1996
1.4	51			Job, 1973
1.5 (peak)	54			Jansen et al., 1984
1.7		92		Chapin et al., 1997
3.3			417	Price et al., 1997
6.3			350	Price et al., 1997
8.4		140		Chapin et al., 1997
9.6			385	Price et al., 1997
13.3			426	Price et al., 1997
20	97			O'Sullivan and Taylor, 1983
25			409	Price et al., 1997
26		162		Ku et al., 1993
38		154		Ku et al., 1993
52		163		Ku et al., 1993
68		164		Ku et al., 1993
95		247		Ku et al., 1991
Mean	40 (28)"	163 (49)	397 (31)
"Values in parentheses represent the standard deviation.

The ad hoc joint panel also recommended a change in the uncertainty factor for intraspecies variation.

Figure 1 provides some indication that blood boron levels may vary with pregnancy based on the seeming difference between pregnant and nonpregnant rats; however, the difference in kinetics of boron during pregnancy is likely due to an increase in the glomerular filtration rate (GFR). This observation is of particular interest, since the critical effect used to derive the tolerable intake (TI) is decreased fetal weight. Data describing clearance of boron in pregnant humans are not available, but an increase in GFR is a recognized physiological adaptation in pregnancy. Available data from studies of GFR in healthy pregnant females were pooled; the mean GFR was 144 (32 mL/min (± the standard deviation) during late pregnancy was determined (28-30). In order to estimate the degree of intraspecies variation in this factor, the ratio of the mean GFR (i.e., 144) and the mean GFR minus two standard deviations (i.e., 2 x 32) from the mean (i.e., 144 - 64 = 80) was calculated: 144 ml/min + 80 ml/min = 1.8. Thus, this factor was used directly as the adjustment for intra-human variability in kinetics.

DISCUSSION

The estimate of a Tolerable Intake value requires determination of a critical effect level and application of uncertainty factors to ensure this critical effect level captures the variability in dose-response for the human population. In this article, the available toxicokinetic data for boron were analyzed in detail to determine if they were adequate to replace default values with data-derived values for inter- and intraspecies variation. The ad hoc joint panel analysis supports the idea that differences in blood boron levels between rats and humans at equivalent doses reflects differences in clearance of this compound and that metabolism, absorption, and overall elimination are nearly the same among species. Calculation of the mean clearance rates for boron indicate that clearance is approximately 4-fold higher in rats than humans. This is the same as the default value for the toxicokinetic component of interspecies variation proposed by Renwick and adopted by IPCS, which is not unexpected since species differences in renal function contributed to the selection of the default factor of 4. When a subset of the data with greater confidence was used, a 3-fold difference was estimated. Since the lower clearance rate observed in humans would tend to increase boron body burden relative to rats, it appears premature to modify the default UF for toxicokinetics from animals to humans. As no data were available to modify the default UF of 2.5 for animal to human toxicodynamics, a total UF of 10 is recommended for animal to human variability.

The ad hoc joint panel also analyzed the data for intraspecies susceptibility. A potential difference in clearance of boron between pregnant and nonpregnant rats is indicated. Variations in the toxicokinetics of boron are principally a function of differences in renal clearance, which increases during pregnancy, as reflected in the pregnant versus nonpregnant animals. Available data are inadequate; however, to make any comparison between pregnant rats and pregnant humans. Although the clearance of boron in pregnant humans has not been studied, data are available for GFR in this population. Analysis of the available literature indicates that a factor of 1.8 accounts for the difference in GFR for the average to susceptible individual. Use of this value (instead of the default value of 3.1) to replace the default toxicokinetic component of the intra-human variability uncertainty factor coupled with the default value for the toxicodynamic portion yields a total UF of 6-fold.

The original review of the boron database had identified reports of boron toxicity in infants and children. A review of the available case reports indicated that these reports provide poor characterization of exposure and patient outcome. Effects were reported at relatively low doses, 9 to 33 mg/kg/d, but these doses were still at least an order of magnitude greater than the recommended TI, which was based on effects in neonatal or young animals.

In summary, based on the ad hoc joint panel analysis, it was recommended that the default factor for interspecies variation of 10-fold be retained for derivation of the guideline for boron in drinking water. A toxicokinetic study in rats would be useful to compare to the available human study of Jansen et al. and would provide a better basis for a data-derived uncertainty factor. It was also recommended that the default value for the uncertainty factor for human variability be replaced with one derived on the basis of the available information on the variability in GFR for the population relevant to the critical effect, pregnant females. Based on the available data, a factor of 6 was considered appropriate. Thus, the overall factor was 60.

The use of a data-derived uncertainty factor for a boron TI illustrates the power of utilizing the types of toxicokinetic information of a complete database, when available. Uncertainty factors for intraspecies and interspecies variability are both comprised of toxicokinetic and toxicodynamic components. Being able to define mathematically the contribution of each component of the UF and applying it to the calculation of the TI more accurately defines the TI and allows the risk assessor to estimate the safe level with more precision. Use of a reduced UF is consistent with the methodology of several regulatory bodies, such as the IPCS or other groups such as Health Canada or the US EPA. For example, reduction of UFs used in the derivation of inhalation Reference Concentrations is performed routinely, particularly the reduction of interspecies variability when using animal data that has been dosimetrically modeled. The reduction of the intraspecies UF for boron is consistent with the practice of these groups. Such reductions should be undertaken when the appropriate data are available. Boron provides an example for reduction of the intraspecies UF for oral exposure.

Conclusions:

Interpretation of results (migrated information): no bioaccumulation potential based on study results
An assessment factor of 6 (1.8 for kinetics and 3.1 for dynamics) was considered appropriate for intrahuman variability, based on variability in glomerular filtration rate during pregnancy in humans and the lack of available data on dynamic differences.

Executive summary:

A critical analysis of the existing data on boron toxicokinetics was conducted to clarify the appropriateness of replacing default uncertainty factors (10-fold for interspecies differences and 10-fold for intraspecies differences) with data-derived values (Dourson et al., 1998). The default uncertainty factor for variability in response from animals to humans of 10-fold (default values of 4-fold for kinetics and 2.5-fold for dynamics) was recommended, since clearance of boron is 3- to 4-fold higher in rats than in humans and data on dynamic differences—in order to modify the default value—are unavailable. A data-derived adjustment of 6-fold (1.8 for kinetics and 3.1 for dynamics) rather than the default uncertainty factor of 10-fold was considered appropriate for intrahuman variability, based on variability in glomerular filtration rate during pregnancy in humans and the lack of available data on dynamic differences. Moreover, information on absorption andelimination of boron from the human body was given. In detail, absorption of boron varied between 81 and 98 % in humans and Excretion and elimination, respectively was found to be from 81 to 99 % (based on the studies: Hunt et al., 1997, Nielsen, 1996, Schou et al., 1984, Job, 1973, Kent and McCance, 1941 and Vanderpool et al., 1994)

Reference 3

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Test procedures in accordance with accepted standard methods, well documented.

Objective of study:

toxicokinetics

Principles of method if other than guideline:

In the toxicokinetics part of the study, blood kinetics for methanol as well as methanol blood concentrations, formate plasma concentrations and serum folate concentrations were determined after methanol inhalation exposure in female monkeys during the prebreeding period and during pregnancy. Changes in blood clearance and/or distribution kinetics of methanol related to repeated exposure and to pregnancy were assessed.

GLP compliance:

not specified

Radiolabelling:

no

Species:

monkey

Strain:

Macaca fascicularis

Sex:

female

Details on test animals or test system and environmental conditions:

TEST ANIMALS
Cohort 1
- Source: all feral born
- Age at assignment to project: 5.5-11 years old (estimated on the basis of dental records)
- Weight at assignment to project: 2.3-3.7 kg
Cohort 2
- Source: feral born (n=15), colony born (n=9, Texas Primate Center, Charles River Primates, CV Primates or Johns Hopkins University)
- Age at assignment to project: 5-13 years old
- Weight at assignment to project: 2.2-5.7 kg
Both cohorts
- Fasting period before study:
- Housing: Individual (social contact through wire mesh)
- Diet: Purina Laboratory Fiber-Plus® Monkey Diet, once per day in the afternoon
- Water: ad libitum
- Acclimation period: The females were transferred to and from the laboratory (inhalation chamber) in a transfer cage on a daily basis.

The four adult males were feral-born with age estimates between 10 and 12 years. The males weighed between 5 and 7.6 kg, and each had sired an offspring during the project.

Route of administration:

inhalation: vapour

Vehicle:

unchanged (no vehicle)

Details on exposure:

TYPE OF INHALATION EXPOSURE: whole body

GENERATION OF TEST ATMOSPHERE / CHAMPER DESCRIPTION
- Exposure apparatus: Inhalation chamber housing 1 animal in a cage (47, 61, 80 cm (w, h, d))
- Source and rate of air: Dayton Model 5K901C blower (Dayton Corporation, Moraine, OH), 420 L/min

TEST ATMOSPHERE
Methanol vapour was generated by passing compressed air through gas dispersion bottles filled with methanol. The methanol was heated by placing the bottles in a water bath set at a temperature of approximately 36 °C. The methanol vapour was delivered to the chamber via insulated polypropylene tubing that ran from the bottles to the vapour inlet port of each chamber.

ANALYSIS OF METHANOL AND CARBON DIOXIDE CONCENTRATIONS
Methanol, carbon dioxide, and dew point were measured by withdrawal of an air sample through a polypropylene tube located in the chamber at a level 5 cm above the monkey cage. The air sample was drawn at a rate of approximately 1.5 L/minute. Methanol and carbon dioxide were measured by a General Analysis Corporation (Norwalk, CT) infrared analyzer. Dew point was measured by a General Eastern Instruments (Woburn, MA) hygrometer, and chamber temperature was measured via resistance temperature detectors placed in each chamber. Dew point and temperature were used to calculate the relative humidity (RH) of each chamber.

Duration and frequency of treatment / exposure:

Daily for 2.5 hours (7 days/week) before breeding, during breeding and during pregnancy.
At the end of each 2-hour exposure, the animals remained in the chamber for another 30 minutes while the methanol dissipated.
During prebreeding: approximately 120 days; during breeding: approximately 70 days; during pregnancy: approximately 165 days.

There was a baseline period (prior to the prebreeding period) which was approximately 4 months long in order to characterize at least 3 menstrual cycles for each female, as well as to provide data regarding background (endogenous) blood methanol and formate concentrations and behavioural responses.

Remarks:

Doses / Concentrations:
0, 0.27, 0.8, 2.39 mg/L (corresponding to 0, 200, 600, 1800 ppm)

No. of animals per sex per dose / concentration:

11-12 females

Control animals:

yes

Details on study design:

- Dose selection rationale: The target air concentrations were chosen to provide a range of blood methanol concentrations from just above background to just below that reported to cause nonlinear clearance kinetics in primates (Horton et al., 1992).

The two-cohort study design utilized 48 adult females (24 females/cohort), 4 adult males (2 males/cohort), and their offspring. This design minimized the number of subjects tested simultaneously, yet achieved a sufficient sample size to detect subtle changes. For each cohort, adult females were initially separated into 6 groups, with 4 animals per group based on known or estimated age, size, and colony parity. Females from each of the 6 groups were then randomly assigned to one of four methanol-exposure groups.


Reference:
Horton VL, Higuchi MA, Rickert DE (1992). Physiologically based pharmacokinetic model for methanol in rats, monkeys, and humans. Toxicol Appl Pharmacol 117: 26–36.

Details on dosing and sampling:

TOXICOKINETIC STUDY
- Tissues and body fluids sampled: blood/plasma
- Time and frequency of sampling:
Biweekly sampling throughout the study (monitoring).
Sampling for full toxicokinetic profile determination (Studies 1-4) was as follows.
After initial methanol exposure (Study 1)
After approximately 3 months of exposure (Study 2)
During pregnancy: 66-72 days (Study 3)
During pregnancy: 126-132 days (study 4)

Throughout the study, blood was collected from all females every other week for methanol and formate analyses. The blood draws occurred approximately 10 minutes after the females were removed from the inhalation chambers, that is, 30 minutes after the end of methanol flow to the chambers. Approximately 3 mL of blood were collected by venipunctures of the saphenous veins of the unanesthetized females. In addition to biweekly monitoring of blood methanol and formate, four full toxicokinetic profile studies were performed (Studies 1-4).
For each toxicokinetic profile study, 2-mL blood samples were collected by venipunctures of the saphenous veins of the unanesthetized females at 7 time points: 30 minutes prior to exposure, and 30, 60, 120, 180, 240, and 360 minutes following termination of exposure.
Blood was also collected from all females for serum folate analyses. Blood was collected at five time points during the study: once prior to methanol exposure, twice after initial exposure but prior to breeding, and twice during pregnancy (51-57 days and 111-117 days).

Statistics:

Statistical analyses were performed using Systat, SAS, or Splus.Basic hypotheses were developed for the toxicokinetics part of the study:
1. There will be no significant differences in methanol toxicokinetics
(a) across the methanol-exposure groups following the initial methanol exposure,
(b) between initial exposure and following repeated exposures (90 days) to methanol,
(c) between preconception and mid- and late-stage pregnancy.
2. There will be no significant differences across the methanol-exposure groups in plasma formate and serum folate concentrations during pregnancy.

In addition to these hypotheses, statistical analyses of maternal characteristics (age, weight, crown–rump length, gravidity, parity) at the outset of the study were performed to examine the results of random assignment of adult females to the 4 exposure groups. The general approach to testing the hypotheses was to first assess whether an exposure effect existed, both globally and specifically. A global F test (or equivalent) was used for assessing whether there were detectable differences among the 4 exposure groups. Because this test has less power than specific alternatives, a no exposure–effect hypothesis was also examined that compared the control group with the combination of all methanol-exposure groups. The control group was also compared with each methanol-exposure group using pairwise comparisons. Finally, the impact of controlling for cohort was assessed in mean models.

To test Hypotheses 1a through 1c, three separate measures of methanol disposition kinetics were used. Standard ANOVA models were used to examine hypothesis 1a, whereas repeated measures ANOVA models were used to examine hypotheses 1b and 1c. To test hypothesis 2, plasma formate concentrations obtained biweekly and serum folate concentrations obtained twice (prior to and during pregnancy) were used. Repeated measures ANOVA models were used to examine plasma formate and serum folate concentrations during pregnancy.

Test no.:

#1

Toxicokinetic parameters:

half-life 1st: 64.9 ± 2.8 min

Test no.:

#1

Toxicokinetic parameters:

Cmax: 30-40 µg/mL (range at first reading after exposure)

Test no.:

#1

Toxicokinetic parameters:

other: Km: 63.0 ± 11.3 µg/mL (Michaelis constant)

Test no.:

#1

Toxicokinetic parameters:

other: Vmax: 2855 ± 403 µg/min (maximum rate of metabolism)

Test no.:

#1

Toxicokinetic parameters:

other: Vd: 1.25 ± 0.07 L/kg bw (volume of distribution)

Test no.:

#1

Toxicokinetic parameters:

other: Cl: 14.3 ± 1.1 mL/min/kg bw (first order clearance)

Test no.:

#1

Toxicokinetic parameters:

other: all values represent mean ± SE [at 1800 ppm, saturable Michaelis-Menten kinetics (one-compartment model), 32/44 data sets]

Methanol exposures, temperature, humidity:

Average chamber concentrations obtained for the 11 samples taken from 14 minutes after onset of methanol flow until 6 minutes prior to offset of methanol flow (total, 100 minutes) were all within 5 % of the target concentrations. For the prebreeding and breeding periods, the average daily chamber methanol concentrations were near the target values of 0, 200, 600, or 1800 ppm. During these periods, the mean chamber temperatures remained between 22 and 24 °C. The relative humidity of the chambers varied greatly.

TOXICOKINETIC STUDIES:

Mean blood methanol concentrations in the control group remained reasonably stable over time within each toxicokinetic study (Studies 1-4) and did not appear to differ among studies. Across the 4 studies at 200 ppm, blood methanol concentrations barely rose above pre-exposure concentrations (roughly 2-fold, 4.3 to 5.5 μg/mL) at 30 minutes following exposure, and declined to near control concentrations by 120 minutes. At 600 ppm, blood methanol concentrations at 30 minutes after exposure ranged from 9.5 to 12.1 μg/mL across studies, which were 3- to 4-fold higher than pre-exposure or control values. Blood methanol concentrations declined to near control levels by 240 minutes. At 1800 ppm, blood methanol at 30 minutes after exposure rose to concentrations that were about 13- to 16-fold higher (33.2 to 40.4 µg/mL) than in the control group across the 4 toxicokinetic studies. Mean blood methanol concentrations remained slightly (2 times) above background or control concentrations at 360 minutes.

For all exposure groups, mean plasma formate concentrations did not show consistent rises following methanol exposure. Given the absence of a measurable elevation in plasma formate concentration following methanol exposure, toxicokinetic analysis was focused on the blood methanol data.

The net rise in blood methanol concentration (Cnet) was estimated by subtracting the pre-exposure blood methanol concentration from each observed blood methanol concentration. Since a full kinetic analysis of blood methanol data at 200 ppm was not feasible, only the data from the 600 and 1800 ppm–exposure groups were used for toxicokinetic modeling.

At corresponding times during each study, the mean net blood methanol concentrations at 1800 ppm were consistently more than 3 -fold higher than those at 600 ppm (disproportionate relation). The mean exposure, dose-normalized, net blood methanol concentrations were significantly higher at 1800 ppm than at 600 ppm. These observations collectively pointed to the presence of nonlinear elimination kinetics at 1800 ppm.

One compartment models:

On the basis of these findings, net blood methanol concentration–time data were fitted to a one-compartment model featuring either first-order or saturable (Michaelis-Menten) metabolic kinetics.

The data at 600 ppm were well described by the linear model, which was used to determine elimination half-life, blood clearance, and distribution for Studies 1-4. At 1800 ppm, better fits were observed for 32 of the 44 data sets with the Michaelis-Menten model than with the linear model. Data from the remaining 12 studies were fitted with the linear model.

Summarizing over all 4 toxicokinetic studies (Michaelis-Menten model fits (32/44 data sets)), the Km estimates ranged from 32.7 to 107.7 μg/mL (mean ± SE = 63.0 ± 11.3 μg/mL). The apparent Vmax estimates ranged from 1502 to 4672 μg/min (mean ± SE = 2855 ± 403 μg/min). The observed blood methanol concentrations at the earliest sampling time of 30 minutes were generally around 30 to 40 μg/mL. Hence, peak blood methanol concentrations at 1800 ppm were just below or within range of the apparent Km of methanol. The mean apparent volume of distribution ranged from 1.12 to 1.44 L/kg bw (mean ± SE = 1.25 ± 0.07 L/kg bw). Elimination half-life for the terminal first-order decline in blood methanol concentration ranged from 56.6 to 77.6 minutes (mean ± SE = 64.9 ± 2.8 minutes). The first-order clearance estimate ranged from 11.7 to 18.2 mL/min/kg bw (mean ± SE = 14.3 ± 1.1 mL/min/kg bw).

Assessment of dose effects of initial methanol exposure (Hypothesis 1a):

The estimates for elimination half-life, blood clearance, and distribution volume of methanol obtained from Study 1 were analyzed by ANOVA to examine whether these 3 toxicokinetic parameters differed between the 600 and 1800 ppm groups after the first exposure. The results did not indicate a significant dose effect for any of the 3 toxicokinetic parameters (p > 0.32). The mean elimination half-life, blood clearance, and distribution volume of methanol at 600 ppm were comparable with those at 1800 ppm.

Assessment of effects of repeated methanol exposures (Hypothesis 1b):

ANOVA models were also fitted to parameter estimates for the toxicokinetic study following the initial methanol exposure and for the study following approximately 3 months of exposure (Study 1 versus Study 2). The results from the ANOVA models indicated a significant change in elimination half-life (p = 0.03), a significant change in blood clearance of methanol (p = 0.02), and no change for the distribution volume of methanol (p = 0.46) following repeated exposures. These effects were not dose dependent; that is, there was no significant interaction effect between study and dose (p > 0.18). The mean elimination half-life of methanol decreased by 20 and 7 % after 90 days of exposure at 600 and 1800 ppm, respectively. The percentage increase in mean blood clearance was the same for the two exposure groups (~30 %).

Assessment of pregnancy effects (Hypothesis 1c):

For monkeys at 600 and 1800 ppm that conceived, repeated measures ANOVA models were used to compare the values of the 3 toxicokinetic parameters from studies conducted prior to conception with values from the studies conducted during early and late pregnancy. The results of the ANOVA models indicated a significant change due to pregnancy in the mean distribution volume of methanol (p = 0.01); however, no changes in the mean elimination half-life and blood clearance of methanol were indicated (p > 0.17). These effects were not dose dependent (no study-by-dose interaction). The respective mean distribution volume of methanol decreased by approximately 22 and 17 % at 600 and 1800 ppm during pregnancy. The results also indicated a significant main effect of dose on the elimination half-life of methanol, reflecting a consistently shorter mean elimination half-life over all 3 studies at 600 ppm (63 versus 79 minutes at 600 and 1800 ppm, respectively).

Summary (Hypotheses 1a-c):

In summary, postexposure blood methanol concentrations were elevated by more than 3-fold as exposure level increased from 600 to 1800 ppm. Moreover, saturable metabolic kinetics were observed at 1800 ppm. A significant decrease in the elimination half-life and blood clearance of methanol was observed. There was a significant decrease in distribution volume during pregnancy, but no consistent changes in elimination half-life or clearance of methanol were observed. The mean elimination half-lives were consistently shorter at 600 ppm than at 1800 ppm across the 2 toxicokinetic studies performed before pregnancy and those performed during mid and late stages of pregnancy.

MONITORING BLOOD METHANOL, PLASMA FORMATE, AND SERUM FOLATE CONCENTRATIONS

Monitoring concentrations were assessed 30 minutes postexposure. During the baseline period, the mean endogenous methanol concentrations in blood were at or below 3 μg/mL (0.9 to 11.0 µg/mL) for all exposure groups. During the prebreeding period, mean blood methanol concentrations of approximately 5 μg/mL (2.0 to 8.3 µg/mL) were reached at 200 ppm. The mean blood methanol concentrations were approximately 10 μg/mL (5.5 to 22.4 μg/mL) at 600 ppm and 35 μg/mL (18.6 to 77.5 µg/mL) at 1800 ppm. For the females that conceived, there was little change in blood methanol concentrations during pregnancy. The mean blood methanol concentrations during pregnancy were approximately 3 μg/mL for the controls, 5 μg/mL for the 200 ppm–exposure group, 11 μg/mL for the 600 ppm–exposure group, and 35 μg/mL for the 1800 ppm–exposure group.

Methanol exposure effects on plasma formate and serum folate concentrations (Hypothesis 2):

Plasma formate concentrations remained low during the entire study with mean values ranging between 0.14 and 0.24 mM. Different ANOVA models were used for comparisons. The results of the ANOVA models did not indicate a significant difference in the plasma formate concentrations across the 4 methanol exposure groups during the baseline period (p = 0.60). Significant changes in plasma formate concentrations were observed, when the baseline formate concentrations were compared with those obtained after approximately 3 months of methanol exposure (p = 0.005) and when formate concentrations obtained after 3 months of methanol exposure were compared with those obtained during pregnancy (p = 0.0001). The changes observed following initial methanol exposure were nearly dose dependent (dose-by-period interaction, p = 0.06), reflecting a slightly greater rise in plasma formate concentrations for the methanol-exposed monkeys than for the controls. The changes observed during pregnancy were not dose dependent, reflecting a small overall increase in plasma formate concentrations for all exposure groups.

Serum folate concentrations were typically within the normal range of values for macaques. The mean serum folate concentrations during baseline and prepregnancy were 12 to 15 ng/mL (individual value range: 5.5 to 20 ng/mL) for the 4 methanol exposure groups,

For the females that conceived, mean folate concentrations during pregnancy remained between 13 and 17 ng/mL. The results of the ANOVA models did not indicate a significant difference in the serum folate concentrations across the methanol exposure groups during the baseline period (p = 0.47). Significant changes in serum folate concentrations were observed, however, when the baseline folate concentrations were compared with those obtained after methanol exposure but prior to pregnancy (p = 0.02) and when folate concentrations obtained prior to pregnancy were compared with those obtained during pregnancy (p = 0.007). The observed changes in serum folate concentrations were not dose dependent.

___

No significant differences in the blood concentrations between the 4 phases (measured 0.5 h post-exposure, n = 11 or 12 (non-pregnant); n = 9 or 10 (pregnancy) (Fig 7, Tab. 11, Fig. 14, Part I):

Baseline range approx. 2 µg/mL [approx. 0.06 mM] 200 ppm: range approx. 5 µg/mL [approx. 0.15 mM] 600 ppm: range approx. 10 µg/mL [approx. 0.30 mM] 1800 ppm: range approx. 35 -40 µg/mL [1mM-range].

At 1800 ppm, after 5 h elimination, the residual methanol level was near baseline (max. 2-fold higher). The mean estimated elimination half-lives (for 600 and 1800 ppm) ranged between about 60 to 90 min.

Formate: Irrespective of concentration levels or exposure intervals, there was no evidence of a significant increase above the background range of about 0.15 - 0.30 mM (approx. 7 to 14 µg/mL) (Fig. 8, Fig. 15, Tab. 12, Part I).

Folate in serum: There was no significant shift in the folate levels during pregnancy, independent of exposure (Tab. 13, Part I).

Reference 4

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

supporting study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Study well documented, meets generally accepted scientific principles, acceptable for assessment.

Reason / purpose for cross-reference:

reference to same study

Objective of study:

toxicokinetics

Principles of method if other than guideline:

This study determined the pharmacokinetics of methanol and formate in normal and folate-deficient monkeys after methanol inhalation in the animal blood.

GLP compliance:

not specified

Radiolabelling:

yes

Remarks:

14C-radiolabelled methanol

Species:

monkey

Strain:

Macaca fascicularis

Sex:

female

Route of administration:

inhalation: vapour

Vehicle:

unchanged (no vehicle)

Duration and frequency of treatment / exposure:

2 hours

Remarks:

Doses / Concentrations:
0.01; 0.06; 0.27 and 1.2 mg/L (corresponding to 10; 45; 200 and 900 ppm)

No. of animals per sex per dose / concentration:

1

Control animals:

no

Details on metabolites:

The results suggest that low-dose short-term exposure to methanol would not result in toxic elevated methanol or formate levels in blood even under folate deficiency, because, even under folate deficiency, the methanol-related formate levels remained at a factor of 10 to 100 below endogenous blood concentrations and are several orders of magnitude lower than levels of formate known to be toxic (>7 mM).

The methanol blood concentrations increased during exposure and declined rapidely after completion of the expsoure and was no longer detectable between 8 and 10.5 hours after exposure. No significant difference in end-of-exposure blood methanol concentration was found between the 1.2 mg/L (normal) and the 1.2 mg/L (folate-deficient) exposures.

The methanol AUC was linearly related to the inhaled methanol dose (Dorman et al., 1994). The half-life (t1/2) for elimination of methanol was unaffected by the concentration of the inhaled methanol (see Table) and ranged from mean t1/2 of 0.56, to 0.95 h, respectively, to 1.31 h under folate deficiency showing a positive increasing trend, although not significant (Dorman et al., 1994; Medinsky et al., 1997).

Average blood formate concentrations during exposure were approximately 10- to 40 -fold lower than methanol concentrations. Exposure of folate-normal monkeys resulted in blood formate concentrations that increased during the 2 hour exposure and rapidly declined after cessation of methanol exposure.

The blood formate concentration following a 2 hour 1.2 mg/L exposure was significantly higher in folate deficient monkeys when compared to 2 hour 1.2 mg/L exposures conducted under normal folate conditions.

Methanol and formate levels in blood and the half-life (t1/2) for elimination of methanol following a 2 hour exposure.

Methanol concentration [mg/L]	Methanol blood levels [µM]	Formate levels blood [µM]	Methanol t1/2 [hours]
0.01 (folate normal)	0.65 ± 0.3	0.07 ± 0.02	0.56
0.06  (folate normal)	3.0 ± 0.8	0.25 ± 0.09	0.88
0.27  (folate normal)	21 ± 16	2.3 ± 2.9	0.62
1.2  (folate normal)	106 ± 84	2.8 ± 1.7	0.95
1.2 (folate-deficient)	211 ± 71	9.5 ± 4.7	1.31

The endogenous formate blood level was not measured, but assumed to be 0.1 to 0.2 mM (taken from other sources). The range of the author´s group is given as 0.29 to 0.56 mM.

It is unclear why the endogenous animal-specific total formate blood levels were not determined in the same test animals. This restriction is not considered to invalidate the overall result and conclusion.

Reference 5

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Study well documented, meets generally accepted scientific principles, acceptable for assessment.

Reason / purpose for cross-reference:

reference to same study

Objective of study:

toxicokinetics

Principles of method if other than guideline:

The study was designed to clarify methanol and formate kinetics in blood at low and high exposure concentrations to methanol in monkeys (and rats with extrapolation to humans [Horton et al., 1992]).

GLP compliance:

not specified

Radiolabelling:

no

Species:

monkey

Strain:

other: Macaca mulatta

Sex:

male

Route of administration:

inhalation

Vehicle:

unchanged (no vehicle)

Duration and frequency of treatment / exposure:

single exposure for 6 hours;
in 3 animals also repeated inhalation: 2.66 mg/L, 6 hours/day, 5 days/week, 14 days

Remarks:

Doses / Concentrations:
0.06; 0.27; 1.60 and 2.66 mg/L (corresponding to 50; 200; 1200 and 2000 ppm)

No. of animals per sex per dose / concentration:

3

Control animals:

yes

Details on metabolites:

Comparison of monkey blood-methanol elimination rate constant with those derived for humans (HEI, 1987) and rats shows an interspecies similarity in the overall elimination of methanol after low doses (Horton et al., 1992).
The dose of methanol required to saturate the folate-dependent metabolism of formate, calculated from an equation derived by the Health Effects Institute (HEI, 1987), is about 250 mg/kg. If one assumes a 6-kg monkey breathes 118.8 l/h and 100% absorption of methanol occurs, a dose equivalent to 308 mg/kg is absorbed during a 6-h exposure to 2.66 mg/L methanol (corresponding to 2000 ppm). While this amount of methanol could overwhelm the formate metabolism pathway if given as a bolus, the saturation of metabolism with this amount divided over 6 h is unlikely (Horton et al., 1992).

Kinetic key data:

Generally, the elimination rate from blood decreased with increasing concentrations. No significant increases in blood values above background were evident at 0.06 mg/L methanol. The end-of-exposure blood concentrations were directly proportional to the atmospheric vapour concentration between 1.6 and 2.66 mg/L

inhalation exposure [mg/L]	t(1/2) [h]	methanol blood levels [µg/mL]
0.27	1.1 ± 0.2	3.9 ± 1.0
1.60	3.2 ± 1.2	37.6 ± 8.5
2.66	2.9 ± 0.6	64.4 ± 10.7

Comparison with other species:

At the lowest concentration, 0.27 mg/L the elimination was only slightly lower in monkeys than in rats (approx. -30 %), but was distinctly slower at the higher burdens.

Formate: The peak blood levels in methanol-exposed rats and monkeys ranged between 5.4 and 13.2 µg/mL (data not shown in Horton et al., 1992, but Medinsky and Dorman (1995) present data in Fig. 1). There were no statistically significant differecnes betweeen control and methanol-exposed animals.

Blood acidosis: no signs of acidosis which is in line with the formate blood levels in normal range.

Prediction of in-vivo time-course data by the PBPK model:

Below 1.6 mg/L (corresponding to 1200 ppm), all three species (rat, monkey, human) would exhibit similar post-exposure blood levels of methanol which would be proportional to atmospheric concentrations. Above 1.6 mg/L, increase in blood would become non-linear for rat and monkey, but remain linear for human.

Prediction of in-vivo time-course data by the PBPK model:

Below 1.6 mg/L (corresponding to 1200 ppm), all three species (rat, monkey, human) would exhibit similar post-exposure blood levels of methanol which would be proportional to atmospheric concentrations. Above 1.6 mg/L, increase in blood would become non-linear for rat and monkey, but remain linear for human.

Reference 6

Endpoint:

basic toxicokinetics in vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Study well documented, meets generally accepted scientific principles, acceptable for assessment.

Objective of study:

metabolism

toxicokinetics

Principles of method if other than guideline:

The metabolism of formate and methanol was studied in monkeys after i.v infusion of radiolabeled formate or gavage of radiolabeled methanol via a nasogastric tube. Additionally, the influence of folate-deficiency on their metabolism was investigated.

GLP compliance:

no

Radiolabelling:

yes

Species:

monkey

Strain:

Macaca fascicularis

Sex:

male/female

Route of administration:

other: methanol via nasogastric tube, formate via intravenous infusion

Vehicle:

not specified

Duration and frequency of treatment / exposure:

single administrationof either formate or ethanol

Remarks:

Doses / Concentrations:
formate: 1, 2.5, 5, 10 mmol/kg; methanol: 3000 mg/kg
folate-deficiency: formate: 2.5 mmol/kg; methanol: 500 mg/kg
Clay et al.: 50, 72, 200, 255, 470 mg/kg formate i.v.

No. of animals per sex per dose / concentration:

3-4 per dose in total, sex not further specified

Control animals:

not specified

Details on absorption:

not determined

Details on distribution in tissues:

not determined

Details on excretion:

Only excretion in form of CO2 was addressed.

Details on metabolites:

The oral administration of 3000 mg/kg methanol produced a syndrome in cynomolgus similar to that observed in other primates, with a latent period of 8 to 12 hours, followed by metabolic acidosis, coma and death. Formic acid accumulates in the blood to levels as high as those in other species (12 mEq/L), and arterial blood bicarbonate decreases correlate with increases of formate ions.
The rate of metabolism to CO2 increased with increasing dose of formate, the maximal rate occurred at 10 mmol/kg, but higher rates were not employed as this dose already induced severe vomiting in a number of animals. At the higher dose levels of formate, the rate of formate metabolism to CO2 in the monkey was less than half that observed in the rat. At lower doses the rate of oxidation to CO2 in the monkey was about 25 % of that in the rat. Elimination half-lifes for formate from blood ranged from 31 min to 51 min in the monkey after administration of sodium formate as single i.v. injections of doses ranging from 50 to 470 mg/kg. When formate elimination was measured in the rat, doses up to 100 mg/kg yielded a half-life of 12 min, and 670 mg/kg of 23 min (Clay et al., 1975). The lower rate of formate metabolism to CO2 in the monkey must be considered as an important factor in explaining formate accumulation and acidosis in the methanol-poisoned monkey but not in the rat.
In order to investigate the role of catalase in formate oxidation in the monkey, the animals were treated with the catalase inhibitor AT prior to [14C]formate. The half-life of formate elimination from the blood of monkeys treated with AT was 38 ± 5 min, whereas in control monkeys it was 37 ± 5 min at 2.5 mmol/kg formate, demonstrating that catalase is not important for formate metabolism in the monkey.
Folate depletion reduced the rate of formate oxidation to CO2 (9.6±0.4 mg/kg/h vs. 17.3±2.8 mg/kg/h) and the rate of formate elimination from the blood (half-life: 74 min vs. 39 min) to about half that observed in control monkeys. Administration of AT to folate-deficient monkeys did not further decrease the rate of formate oxidation to CO2, indicating that catalase does not contibute to formate metabolism in monkeys, even in monkeys with folate-deficiency. In contrast, exogenously administered folate increased the rate of formate oxidation to about 50% above the rate observed in control animals. Furthermore, folate-deficiency rendered monkeys more susceptible to the effects of methanol, administration of a low dose of methanol (500 mg/kg) induced a striking increase in blood formate in the folate-deficient monkeys compared to the control animals (approx. 5 vs. 2 mEq/L blood).

Description of key information

The toxicokinetic profile of TMBX has been based upon the properties of the two hydrolysis products , methanol and boric acid. Dissociation into the two species is expected due to exposure to physiological fluids following oral or inhalation exposure. 100% absorption rates are therefore assumed for methanol and boric acid by these routes. Dermal exposure results in moderate skin irritation, manifesting as erythema and is indicative of dermal absorption. It is anticipated that TMBX hydrolyses as it enters the stratum corneal barrier, and that the barrier function of the stratum corneum to these species is maintained. Therefore it is assumed that the dermal absorption of methanol and boric acid are 100% and 0.5% respectively. These values have been used in the derivation of the DNEL values based on the most appropriate points of Departure for Methanol and Boric Acid.

Methanol is readily absorbed after inhalation, ingestion and dermal contact and distributed rapidly throughout the body. The clearance from the body is mainly due to metabolism (up to 98%), with more than 90% of the administered dose exhaled as carbon dioxide. Renal and pulmonary excretion rates contribute to only about 2 – 3%. The metabolism and toxicokinetics of methanol varies by species and dose. In humans, the half-life time is approximately 2.5 – 3 hours at doses lower than 100 mg/kg bw. At higher doses, the half life can be 24 hours or more (IPCS/WHO, 1977; Kavet and Nauss, 1990).

The mammalian metabolism of methanol occurs mainly in the liver, where methanol is initially converted to formaldehyde, which is in turn converted to formate. Formate is converted to carbon dioxide and water. In humans and monkeys, the oxidation to formaldehyde is mediated by alcohol dehydrogenases and basically limited to the capacity of those enzymes. In rodents, the oxidation to formaldehyde predominantly employes the catalase-peroxidase pathway which is of less capacity than the ADH-pathway in humans but on the other hand produces oxygen radicals which may be involved into the developmental effects in rodents which - in contrast to humans - tolerate high methanol levels without signs of CNS or retinal toxicity. The last oxidation step, conversion of formate to carbon dioxide employes formyl-tetrahydrofolate synthetase a co-enzyme, is of comparably low capacity in primates which leads to a low clearance of formate, possibly also from sensitive target tissues (such as CNS and the retina) (DFG 1999; IPCS/WHO, 1997; Dorman et al., 1994; Medinsky et al., 1997, Medinsky and Dorman, 1995; Mc Martin et al., 1977).

In humans, when exposed to methanol via inhalation up to an air concentration 65 mg/m3, no increase of blood methanol is expected. Up to 260 mg/m3 (single or repeated exposure) the methanol blood level is likely to increase only 2- to 4- fold above the endogenous methanol concentration in humans, but still remains significantly below 10 mg/L (Lee et al., 1992; NTP, 2003). Up to air concentrations of 1600 mg/m3 the blood methanol levels increase to a similar extent in rats, monkeys, and humans. However, above this concentration rats show a steep exponential increase which apparently reflects the saturation of the catalase-dependent pathway. A smaller exponential increase was observed in monkeys, whereas in humans there appears to be a linear relationship between air concentrations and blood methanol levels.

Baseline levels of formate in blood are about 3 to 19 mg/L (0.07 – 0.4 mM) in humans. Toxic blood formate concentrations are reported to be 220 mg/L and higher (> 5 mM formate). Inhalation of about 1200 mg methanol/m3 for 2.5 hours contributed only insignificantly to the internal formate pool in monkeys (in the μM-range). This also hold true for folate-deficient conditions. After repeated inhalation of 2600 mg/m3 for 6 hours/day, 5 days/week, for 1 or 2 weeks, monkeys showed no discernible increase in formate concentrations in blood (estimated body burden 200 to 300 mg/kg bw/d). Formate accumulation, however, has been observed in primates upon bolus administration of more than 500 mg Methanol/kg bw (Horton et al., 1992; Medinsky and Dorman, 1995). The critical methanol dose that saturates the folate pathway in humans is estimated to be ≥ 200 mg/kg bw. Based on data produced in monkeys, metabolic saturation in humans is also less likely to happen upon inhalation where the dose is distributed over several hours (DFG 1999; IPCS/WHO, 1997; Burbacher et al., 1999).

There is a strong link between saturation (zero-order) kinetics and the onset of acute toxic effects. Exposure levels in humans above 5000 ppm (750 mg/kg bw in the course of 8 hrs) are prone to a zeroorder kinetic and a strong accumulation of methanol in the blood. Transient blindness has been reported for exposure levels between 1000 and 5000 ppm. (This saturation point could be reached after oral uptake at lower dose levels.) 10.000 ppm are still tolerated in rodents but would be highly detrimental in humans.

There is little difference between animals and humans in absorption, distribution, and metabolism of Boric Acid. A difference in renal clearance is the major determinant in the differences between animals and humans, with the renal clearance in rats approximately 3 times faster than in humans.

Boric acid is not metabolised in either animals or humans, owing to the high energy level required (523 kJ/mol) to break the B - O bond (Emsley, 1989). Other inorganic borates convert to boric acid at physiological pH in the aqueous layer overlying the mucosal surfaces prior to absorption.Most of the simple inorganic borates exist predominantly as undissociated boric acid in dilute aqueous solution at physiological and environmental pH, leading to the conclusion that the main species in the plasma of mammals is un-dissociated boric acid. Since other borates dissociate to form boric acid in aqueous solutions, they too can be considered to exist as un-dissociated boric acid under the same conditions. Additional support for this derives from studies in which more than 90 % of administered doses of inorganic borates are excreted in the urine as boric acid. Absorption of borates via the oral route is nearly 100 %. For the inhalation route also 100 % absorption is assumed as worst case scenario. Dermal absorption through intact skin is very low with a percent dose absorbed of 0.226 ± 0.125 in humans. Using the % dose absorbed plus standard deviation (SD) for boric acid, a dermal absorption for borates of 0.5 % (rounded from 0.45 %) can be assumed as a worse case estimate.

In the blood boric acid is the main species present and is not further metabolised. Boric acid is distributed rapidly and evenly through the body, with concentrations in bone 2 - 3 higher than in other tissues. Boric acid is excreted rapidly, with elimination half-lives of 1 h in the mouse, 3 h in the rat and < 27.8 h in humans, and has low potential for accumulation. Boric acid is mainly excreted in the urine.

Interspecies differences in toxicokinetics based on data for boron clearance rates in rats versus humans and intraspecies differences in human toxicokinetics based on data on human variability in glomerular filtration rates (GFR) are critical determinates in evaluating human toxicity of boric acid. GFR was identified as the primary determinant of boron clearance rates. A toxicokinetic adjustment factor for boron for human variability is based on the variability in GFR during pregnancy (Dunlop, 1981; Krutzén et al., 1992; Sturgiss et al., 1996) ensuring adequate coverage of the sensitive subpopulation of preeclamptic women (US. EPA 2004; Dourson et al. 1998; Maier et al. 2014).

Key value for chemical safety assessment

Bioaccumulation potential:

no bioaccumulation potential

Absorption rate - oral (%):

100

Absorption rate - dermal (%):

100

Absorption rate - inhalation (%):

100

Additional information

Further relevant studies in Section 7.10 Exposure related observations in humans:

Boric Acid

Hui et al., 1996 and Hartway et al. 1997

Pahl and Culver, 2000

Dunlop 1981, Krutzen et al. 1992, Sturgiss et al. 1996.

Maier et al., 2014

Methanol:

IPCS/WHO, 1977;

Kavet and Nauss, 1990

Lee et al., 1992;

NTP, 2003

IPCS/WHO, 1997