Registration Dossier

Administrative data

Key value for chemical safety assessment

Effects on fertility

Effect on fertility: via oral route
Dose descriptor:
NOAEL
94.6 mg/kg bw/day
Additional information

Effects on male fertility have been investigated in detail. A dose related effect on the testis was observed in rats, mice and deer mice, with confirmation from limited studies in dogs. Effects in rats start with reversible inhibition of spermiation after 14 days (at 39 mg B/kg bw/day) and 28 days (at 26 mg B/kg bw/day). At doses equal to and above 26 mg B/kg bw/day testicular atrophy, degeneration of seminiferous tubules and reduced sperm counts were observed. Male fertility was further investigated in two serial mating studies of treated male rats with untreated female rats. Infertility of treated males correlated well with germinal aplasia. Similar effects on male fertility were described in deer mice (Peromyscus maniculatus) after treatment with boric acid. Fertility studies in rats (two three-generation study with for boric acid and disodium tetraborate decahydrate) and mice (a continuous breeding study with boric acid) further support effects on testes as the underlying cause for reduced male fertility.

Diminished sperm production may be due to testicular effects on germ cell, Sertoli cell, or Leydig cell function or act via an alteration of the pituitary-hypothalamic axis. There is an indication that LH and FSH are elevated under boric acid treatment (Lee et al., 1978) and that serum testosterone may be decreased in CD-1 mice and F344 rats (Grizzle et al., 1989; reviewed in Fail et al., 1991; Treinen & Chapin, 1991). The decrease in prostate weight at 111.3mg B/kg bw/day observed by Fail et al. (1991) might be caused by reduced testosterone levels.

A NOAEL of 17.5 mg B/kg bw/day for effects on female fertility was derived in the Transitional Annex XV dossier (TD 2008) based on Weir (1966c-d) and Fail et al. 1991. However, the TD failed to adequately distinguish between effects on female fertility and effects on development. Fertility is generally defined in males as the ability to produce sperm which are capable of producing fertilisation of an ovum leading to conception.  In females, it is defined as the ability to produce and release ova which can be fertilised leading to conception.  To test fertility in animals males and females are pretreated to cover the period of development of the sperm and eggs, then mate and treat until the time of implantation, around Day 6 following mating, and then stop treatment in the females.   To test for effects on development pregnant females are treated from Day 6 till the end of pregnancy. Neither the Weir and Fisher multigeneration study nor the Fail RACB studies were performed with this division of treatments. They both treated animals continuously before and during pregnancy and also after delivery.

In a three generation study in rats groups of 8 males and 16 females were treated with boric acid or disodium tetraborate decahydrate equivalent to 0, 5.9, 17.5 and 58.8 mg B/kg bw/day (Weir 1966c, d). An attempt was made to study the fertility of the P1 females at the top dose level by mating them with untreated males but only one litter of 16 pairs was produced. This highest dose level was clearly clinically toxic to the females after 2-3 weeks of dosing, with rough fur, scaly tails, inflamed eyelids and staining of the fur on the face and abdomen. The mating procedure to test the fertility of the females was not a satisfactory one. To avoid treatment of the males used for pairing, food was withdrawn from the cages of the females for 8 hours per day during the pairing process, and this is known to be very stressful to laboratory rats. There was no evidence on whether mating actually occurred for any of the rats, and no vaginal examinations for the presence of sperm were carried out. The females of the top dose P1 generation were sacrificed after 45 weeks of treatment and histopathological examination of the ovaries and uterus carried out. In the ovaries the presence of corpora lutea was regarded as a major indication of cyclic function, and these were found in 7 of 15 females, with reduced or absent function in the remaining 8 animals. The changes in the ovaries were not clearly different from those of controls.  No treatment related changes were found in the uterus. No changes were found that could account for the reduced litter production, and no conclusions could be drawn about fertility in the top dose females.  Comparable results were found in the Weir and Fisher multigeneration study on borax, with clear testicular atrophy at the top dose levels in males, and no clear explanation of the reduced number of litters in the top dose females, using the same unsatisfactory mating technique.  The authors of the study concluded that testis atrophy was clearly produced in males at the top dose level, but that the evidence of the decreased ovulation in females did not account for the reduced number of litters in the cross mating study in females.  Thus the Weir and Fisher studies produced clear evidence of adverse effects on male fertility, but did not produce clear evidence for an adverse effect on female fertility.

In a continuous breeding study of boric acid in Swiss mice (NTP, 1990; Fail et al., 1991), the three administered doses were 1000 ppm (26,6 mg B/kg bw/day), 4500 ppm (111,3 mg B/kg bw/day) and 9000 ppm (220,9 mg B/kg bw/day). A dose-related effect on the testis (testicular atrophy and effects on sperm motility, morphology and concentration) was noted; fertility was partially reduced at 111 mg B/kg bw/day, and absent at 221 mg B/kg bw/day.

For cross over mating only the mid dose group (111.3 mg B/kg bw/day) could be mated with control animals, since the high dose produced no litter. Indices of fertility for mid dose males with control females, control males with mid dose females and control males with control females were 5 %, 65 % and 74 %, respectively. The according indices of mating (incidence of copulatory plugs) were 30 %, 70 % and 79 %. This indicates that the primary effect was seen in males, however, slight effects were also noted in females. Live pup weight (adjusted for litter size) was significantly reduced compared to control litters, the average dam weight was significantly lower on postnatal day 0 compared to control dams and the average gestational period of the mid dose females was 1 day longer than in control females. The latter finding has also been observed in the developmental toxicity study by Price et al. (1996, see section 5.9.2).

In task 4 of this continuous breeding study control animals and low-dose F1 animals were mated because in the 9000 ppm groups no litters and in the 4500 ppm group only 3 litters were produced. While mating, fertility and reproductive competence were un-altered compared to control, the adjusted pup-weight (F2) was slightly but significantly decreased. F1 females had significantly increased kidney/adrenal and uterus weights and the oestrus cycle was significantly shorter compared to control females. A crossover mating study of controls and 4500 ppm groups confirmed the males as the affected sex. Necropsy at 27 weeks confirmed reduced testes weight, seminiferous tubule degeneration, decreased sperm count and motility and increase in abnormal sperm. In females at 27 weeks, 4500 ppm boric acid was toxic with decreased liver, kidney and adrenal weights, but no effect on oestrous cycles, mating, number of litters and number of pups. In F1 males a reduction in sperm concentration was observed, but no other sperm parameters were influenced.

While in this study the NOAEL for females of the F0-generation is 1000 ppm this is a LOAEL for males of the F0-generation (motility of epididymal sperms was significantly reduced: 78% ± 3 in controls vs. 69% ± 5 at 1000 ppm). For the F1-generation 1000ppm can be identified as a LOAEL, based on the 25% reduction of sperm concentration in males at this dose. Further, though normal in number, the F2-pups had reduced adjusted bodyweights at 1000 ppm, which is therefore also a LOAEL for F2-generation.

The authors concluded that the male is the most sensitive sex and that the testis is the primary target organ for boron. The NOAEL for testicular pathology in the present mouse study is probably 1000 ppm (26mg B/kg bodyweight). While males are more sensitive to boron induced toxicity, data also suggest an effect of boron on the female reproductive system. A reduced number of pups per litter and number of pups born alive at high dose levels are in agreement with earlier reports and could result from an effect of boron to alter implantation or to disrupt cell division in the embryo. This is supported by results of developmental toxicity studies in rats and mice in which higher dose levels can reduce the number of implants. Although F1 females had significantly increased kidney/adrenal and uterus weights and the oestrus cycle was significantly shorter compared to control female, similar effects were not observed in the 4500 ppm dose group, therefore the NOAEL in females was the dose level in diet of 4500 ppm, 846 mg/kg bw of boric acid or equivalent to 148 mg B/kg bodyweight.

In conclusion, the effects described in the Fail study on fertility show that 4500 ppm (111.3mgB/kg bw) is a NOAEL for the females, and that other small effects in females are the result of developmental toxicity for which a NOAEL of <1000ppm (26.6mg B/kg bw) may be valid.

No further studies on the effects of boron on female fertility were reported by the National Toxicology Program team who published several other studies on the mechanism of action of boron on male fertility and on spermatogenesis. No effects on steroidogenic function were found in Leydig cells, and no clear mechanism of action to cause testis atrophy was identified by Ku and Chapin (1994).

Although boron has been shown to adversely affect male reproduction in laboratory animals, male reproductive effects attributable to boron have not been demonstrated in studies of highly exposed workers. For further information on epidemiologic studies with workers exposed to high concentrations of boron, please refer to chapter 7.10.2 of this dossier and the respective endpoint summary.

Read Across

A number of these studies were conducted on an analogue substance. Read-across is justified on the following basis:

In aqueous solutions at physiological and acidic pH, low concentrations of simple inorganic borates such as boric acid B(OH)3, potassium pentaborate (K2B10O16.8H2O), potassium tetraborate (K2B4O7.4H2O), disodium tetraborate decahydrate (Na2B4O7.10H2O; borax), disodium tetraborate pentahydrate (Na2B4O7.5H2O; borax pentahydrate), boric oxide (B2O3) and disodium octaborate tetrahydrate (Na2B8O13.4H2O) will predominantly exist as undissociated boric acid. Above pH 9 the metaborate anion (B(OH)4-) becomes the main species in solution (WHO, 1998). This leads to the conclusion that the main species in the plasma of mammals and in the environment is undissociated boric acid. Since other borates dissociate to form boric acid in aqueous solutions, they too can be considered to exist as undissociated boric acid under the same conditions.

For comparative purposes, exposures to borates are often expressed in terms of boron (B) equivalents based on the fraction of boron in the source substance on a molecular weight basis. Some studies express dose in terms of B, whereas other studies express the dose in units of boric acid. Since the systemic effects and some of the local effects can be traced back to boric acid, results from one substance can be transferred to also evaluate the another substance on the basis of boron equivalents. Therefore data obtained from studies with these borates can be read across in the human health assessment for each individual substance. Conversion factors are given in the table under CSR section 5.1.3, which corresponds to IUCLID section 7.1 (toxicokinetics, metabolism and distribution endpoint summary).

References:

WHO. Guidelines for drinking-water quality, Addendum to Volume 1, 1998


Short description of key information:
A multigeneration study in the rat (Weir, 1966) gave a NOAEL for fertility in males of 17.5 mg B/kg/day.
No multigeneration studies with dipotassium tetraborate were available however since all the borates will exist as undissociated boric acid under physiological and environmental conditions, the toxicology of all these simple borates is similar on an equivalent boric acid basis or boron basis. Therefore the data for boric acid and disodium tetraborate decahydrate can be read across to the other borates for toxicological effects.
The following oral data were obtained (NOAEL):
Dipotassium tetraborate (anhydrous): 94.6 mg/kg bw/day
Dipotassium tetraborate (tetrahydrate): 123.7 mg/kg bw/day

Effects on developmental toxicity

Description of key information
No studies with dipotassium tetraborate were available however since all the borates will exist as undissociated boric acid under physiological and environmental conditions, the toxicology of all these simple borates is similar on an equivalent boric acid basis or boron basis. Therefore the data for boric acid and disodium tetraborate decahydrate can be read across to the other borates for toxicological effects.
A benchmark dose of 10.3 mg B/kg bw/day for developmental toxicity developed by Allen et al. (1996) was based on the studies of Heindel et al. (1992), Price, Marr & Myers (1994) and Price et al. (1996).
Effect on developmental toxicity: via oral route
Dose descriptor:
BMDL05
55.7 mg/kg bw/day
Additional information

Developmental effects have been observed in three species, rats, mice and rabbits. The most sensitive species being the rat with a NOAEL of 9.6 mg B/kg bw/day. This is based on a reduction in mean foetal body weight/litter, increase in wavy ribs and an increased incidence in short rib XIII at 13.3 mg B/kg bw/day. The reduction in foetal body weight and skeletal malformations had reversed, with the exception of short rib XIII, by 21 days post natal. At maternally toxic doses, visceral malformations observed included enlarged lateral ventricles and cardiovascular effects.

The NOAEL for this endpoint is 9.6 mg B/kg bw/day corresponding to 55 mg boric acid/kg bw/day; 85 mg disodium tetraborate decahydrate/kg, 65 mg disodium tetraborate pentahydrate/kg and 44.7 mg disodium tetraborate anhydrous/kg.

The critical effect is considered to be decreased fetal body weight in rats, for which the NOAEL was 9.6 mg/kg body weight per day. A benchmark dose developed by Allen et al. (1996) was based on the studies of Heindel et al. (1992), Price, Marr & Myers (1994) and Price et al. (1996). The benchmark dose is defined as the 95% lower bound on the dose corresponding to a 5% decrease in the mean fetal weight (BMDL05). The BMDL05of 10.3 mg/kg body weight per day as boron is close to the Price et al. (1996) NOAEL of 9.6 mg/kg body weight per day.

There is no evidence of developmental effects in humans attributable to boron in studies of populations with high exposures to boron (Tuccar et al 1998; Col et al. 2000; Chang et al. 2006).

BDML05:

Potassium pentaborate anhydrous: 55.7 mg/kg bw/day

Potassium pentaborate tetrahydrate: 72.8 mg/kg bw/day

Read Across

A number of these studies were conducted on an analogue substance. Read-across is justified on the following basis:

In aqueous solutions at physiological and acidic pH, low concentrations of simple inorganic borates such as boric acid B(OH)3, potassium pentaborate (K2B10O16.8H2O), potassium tetraborate (K2B4O7.4H2O), disodium tetraborate decahydrate (Na2B4O7.10H2O; borax), disodium tetraborate pentahydrate (Na2B4O7.5H2O; borax pentahydrate), boric oxide (B2O3) and disodium octaborate tetrahydrate (Na2B8O13.4H2O) will predominantly exist as undissociated boric acid. Above pH 9 the metaborate anion (B(OH)4-) becomes the main species in solution (WHO, 1998). This leads to the conclusion that the main species in the plasma of mammals and in the environment is undissociated boric acid. Since other borates dissociate to form boric acid in aqueous solutions, they too can be considered to exist as undissociated boric acid under the same conditions.

For comparative purposes, exposures to borates are often expressed in terms of boron (B) equivalents based on the fraction of boron in the source substance on a molecular weight basis. Some studies express dose in terms of B, whereas other studies express the dose in units of boric acid. Since the systemic effects and some of the local effects can be traced back to boric acid, results from one substance can be transferred to also evaluate the another substance on the basis of boron equivalents. Therefore data obtained from studies with these borates can be read across in the human health assessment for each individual substance. Conversion factors are given in the table under CSR section 5.1.3, which corresponds to IUCLID section 7.1 (toxicokinetics, metabolism and distribution endpoint summary).

References: WHO. Guidelines for drinking-water quality, Addendum to Volume 1, 1998

Justification for classification or non-classification

Boric acid and disodium tetraborate are classified under the 1stATP to CLP as Repr. 1B; H360FD.

However, text of the 30th ATP as published in the EU Official Journal, 15 September 2008 stated that “The classification and labelling of the substances listed in this Directive should be reviewed if new scientific knowledge becomes available. In this respect, considering recent preliminary, partial and not peer-reviewed information submitted by industry, special attention should be paid to further results of epidemiological studies on the Borates concerned by this Directive including the ongoing study conducted in…”

While boron has been shown to adversely affect male reproduction in laboratory animals, there was no clear evidence of male reproductive effects attributable to boron in studies of highly exposed workers (Whorton et al. 1994; Sayli 1998, 2001; Robbins et al. 2010; Scialli et al. 2010). Not only are these the most exposed workers, but the Chinese worker study is themost sensitive study that has been carried out as semen analysis was performed, a very sensitive detection system for testicular damage. There is no evidence of developmental effects in humans attributable to boron in studies of populations with high exposures to boron (Tuccar et al 1998; Col et al. 2000; Chang et al. 2006).

Comparison of Blood, Semen and Testes Boron Levels in Human and Rat

A comparison of blood, semen and target organ boron levels in studies of laboratory animals and human studies shows that boron industry worker exposures are lower than untreated control rats. Background boron levels in standard rat chow are high (10-20 ppm), as a result control rats in toxicity studies receive 45 times more boron than background exposure in humans. Blood boron levels in female control rats is about 0.23 µg B/g (Price et al. 1997), approximately equal to the blood levels in boron industry workers in China, Turkey and U.S. of 0.25, 0.22 and 0.26 µg B/g, respectively (Scialli et al. 2010; Culver et al. 1994; Duydu et al. 2011). Plasma and seminal vesicle fluid (the major component of semen) boron levels in untreated male control rats were 1.94 and 2.05 µg B/g, respectively, while boron levels in testes in rats dosed at the rat fertility LOAEL (26 mg B/kg) was 5.6 µg B/g (Ku et al. 1991,1993). Values in male control rats were higher than corresponding boron levels in the highest exposed Chinese boron industry workers with blood boron levels of 1.56 µg B/g and 1.84 µg B/g in semen (Scialli et al. 2010). Blood and semen boron levels in highly exposed Turkish boron workers were also lower than control rats with levels of 0.22 and 1.88 µg B/g, respectively (Duydu et al. 2011). Boron levels in testes of rats dosed at the rat fertility LOAEL was over 3x the blood boron levels in highest exposure group of Chinese boron industry workers. The blood level at the lowest animal LOAEL (13 mg B/kg) was 1.53 µg B/g, about 6 times greater than typical boron industry workers (Price et al. 1997). No adverse effects on sperm were seen in Turkish boron industry workers or in the most highly exposed subgroup of Chinese boron industry workers drinking boron contaminated water (mean blood level 1.52 µg B/g, the human NOAEL). Only under extreme conditions do human levels reach those of the animal LOAEL: the subgroup of Chinese boron workers who also drank contaminated water. Since no boron accumulation occurs in soft tissues (testes) over plasma levels biological monitoring in humans provide direct comparison to test animal target organ boron levels. Workers in boron mining and processing industries represent the maximum possible human exposure however their blood and semen boron levels are less than levels in untreated control rats. This provides an explanation whystudies of highly exposed boron industry workers have shown no adverse effects anddemonstrates that maximal possible exposures in humans are insufficient to cause reproductive toxicity effects.Graphs comparing the rodent and human exposure, blood, semen and tissue boron levels are presented in Appendix C.

A weight of evidence approach was used in evaluating numerous independent studies on the determination of the hazard of boric acid to humans. Information that was considered together included results of in vitro tests, animal data, occupational exposure data, epidemiological studies and mechanistic data.

Extensive evaluations of sperm parameters in highly exposed workers in Turkey and China have demonstrated no effects on male fertility. No evidence of developmental effects in humans attributable to boron (B) has been observed in studies of populations with high exposures to boron. Although the epidemiological studies have methodological deficiencies, collectively these studies consistently show an absence of effects in highly exposed populations.

Workers in boron mining and processing industries represent the maximum possible human exposure. However a comparison of blood, semen and target organ boron levels in studies of laboratory animals and human studies shows that boron industry worker exposures are lower than untreated control rats.

Mechanistic data provide possible explanations for the absence of developmental and reproductive effects in humans exposed to high levels of boron. Recent studies provide evidence that boric acid may act by similar mechanisms in causing developmental effects in mice as sodium salycilate (the natural deacetylated form of aspirin and a rodent teratogen) including effects on Hox gene expression and inhibition of embryonic histone deacetylases. Although aspirin is known to cause developmental effects in laboratory animals, controlled human studies have not demonstrated developmental effects in humans. Similar mechanisms of action of boric acid and aspirin, and the absence of developmental effects in humans ingesting aspirin suggest that boric acid related developmental effects in humans are unlikely. 

Additionally, zinc levels in soft tissue in humans is over 2 times greater than in comparative tissues in rats (King et al. 2000; Yamaguchi et al. 1996), which explain in part the absence of fertility and developmental effects in humans. Zinc has been shown to protect against testicular toxicity of cobalt and cadmium (Anderson et al. 1993), and the developmental effects of cadmium (Fernandez et al. 2003). There is evidence that zinc interacts with boric acid in the body reducing the toxicity of boric acid. The interaction of zinc and boric acid is evident by the low acute toxicity of zinc borate (absorbed as boric acid and zinc) with a LD50 value greater than 10,000 mg/kg-body weight in rats (Daniels 1969) compared to disodium tetraborate pentahydrate (similar % boron composition as zinc borate) with a LD50 value of 3300 mg/kg-body weight. Furthermore, no toxic effects were observed in the testes of males (a target organ of boric acid) administered 1000 mg zinc borate/kg/day in a 28-day repeated dose oral gavage toxicity study, equivalent dose of boron of 50 mg B/kg bodyweight (Wragg et al. 1996). The LOAEL for testicular effects is 26 mg B/kg body weight. 

Based on the total weight of evidence, the data show that it is improbable that boric acid or potassium pentaborate will cause reproductive or developmental effects in humans. Therefore, based on a total weight of evidence, Category 2 H361d: suspected human reproductive toxicant, suspected of damaging the unborn child is considered the appropriate classification.