Dibutyl phthalate

Endpoint summary

Key value for chemical safety assessment

Effects on fertility

Effect on fertility: via inhalation route

Endpoint conclusion:

no study available

Effect on fertility: via dermal route

Endpoint conclusion:

no study available

Effects on developmental toxicity

Effect on developmental toxicity: via oral route

Endpoint conclusion:

adverse effect observed

Dose descriptor:

LOAEL

52 mg/kg bw/day

Species:

rat

Effect on developmental toxicity: via inhalation route

Endpoint conclusion:

no study available

Effect on developmental toxicity: via dermal route

Endpoint conclusion:

no study available

Additional information

Based on the available developmental studies in mice an oral NOAEL of 100 mg/kg bw, can be derived for teratogenicity, embryotoxicity and maternal toxicity. At the next higher dose-level of 400 mg/kg bw embryotoxic and teratogenic effects were seen in the presence of maternal

toxicity.

In rats, developmental studies with exposure during gestation or during gestation and lactation, revealed preputial separation and reproductive tract malformations in male offspring at oral doses≥250 mg/kg bw. At the lowest oral dose of 100 mg/kg bw, studied in developmental studies in rats, still delayed preputial separation in male progeny was seen. Maternal toxicity was seen at oral doses≥500 mg/kg bw. From the developmental studies in rats a NOAEL of 50 mg/kg bw/d could be derived.

Concerning reproduction, fertility as well as developmental studies a NOAEL of 50 mg/kg bw can be established based on embryotoxicity in a one-generation reproduction study in rats with exposure of females only. However, a LOAEL of 52 mg/kg bw can be established based on embryotoxic effects in rats in the absence of maternal toxicity in a two-generation reproduction study with a continuous breeding protocol including improved sensitive endpoints (such as sperm parameters, estrous cycle characterisation and detailed testicular histopathology) and with

exposure of both male and female animals. The protocol of this study was supposed to adequately identify compounds with endocrine activity.

In some special in vitro assays DBP showed weak estrogenic activity. However, the estrogenic effects were not confirmed in in vivo studies. Therefore the relevance of the estrogenic effects observed in vitro for th ein vivo estrogenic activity of DBP is questionable. Moreover results of

recent developmental studies are indicative of an antiandrogenic effect rather than an estrogenic effect of DBP.

No reproduction, fertility or developmental studies with dermal exposure or exposure by inhalation to DBP are available.

The epidemiological study on possibly reproductive effects in occupationally exposed women is inadequate for assessment of possible reproductive effects caused by DBP in man in the working environment.

Based on all available studies an overall oral LOAEL of 52 mg/kg bw can be established for dibutylphthalate. This figure is derived from a two-generation reproduction study in rats with a continuous breeding protocol and based on embryotoxic effects. (1)

(1)

European Union Risk Assessment Report dibutyl phthalate, Volume 29, pp. 16 -17 (2003)

Editors: B. G. Hansen, S.J. Munn, R. A/Ianou, F. Berthault, J. de Bruin, M. Luotamo, C. Musset, S. Pakalin, G. Pellegrini, S. Scheen S. Vegro.

Office for Official Publications of the European Communities, ISBN 92—894—1276—3

Reproductive toxicity studies are organised on the basis of test procedure, mainly timing of exposure (adult, gestation or early postnatal). The effects on fertility (as adults) and development (as foetuses or early in postnatal development) are then discussed separately, to the extent possible. Fertility is tested by exposing sexually mature adults to a chemical and examining the effects on reproductive capacity. Developmental toxicity is studied by exposing pregnant dams and looking for effects on the foetuses. Chemicals like DBP that affect the developing reproductive system following prenatal exposure, also affect sexual maturation or produce functional reproductive disorders that might only be apparent at maturity. Developmental toxicity can therefore lead to effects on fertility and the two end points are clearly interdependent.

The reproductive and developmental toxicity of DBP in rodents was exhibited as perturbations in testicular structure and function, altered steroidogenesis and developmental malformations of the urogenital tract. Decreased anogenital distance (AGD) and nipple retention were also observed in male animal rodents. Overall, rats are the most sensitive to DBP toxicity, followed by mice and hamsters. Reproductive effects in mammals have not been examined extensively and limited studies with marmosets, mainly sexually mature animals, are not sufficient for robust assessments of DBP reproductive toxicity in mammals as compared to rodents.

Fertility

In adult rodents, repeated exposure to DBP doses of 500 mg/kg bw/d and higher is consistently associated with distinct testicular changes, including decreased organ weight and histopathological perturbations in the testes, indicative of testicular degeneration (Section 6.2.4 and Table 6.2) likely to lead to adverse effects on fertility.

Fertility in rodents is also adversely affected by DBP through toxicity in females, although often in the absence of overt histopathological changes in the reproductive tract (Wolf et al. 1999; Gray et al. 2006).

Continuous exposure of LE hooded female rats to DBP, starting from weaning continuing through adulthood and during crossmating with untreated males, showed clear effects on reproductive parameters (reduction of the percentage of females delivering live pups and reduction of live pups per litter) at 500 mg/kg bw/d. Ex vivo progesterone and oestradiol production by the ovarian cultures in these females was also altered at the same dose. The NOAEL for fertility in female rats in this study is 250 mg/kg bw/d and the LOAEL is 500 mg/kg bw/d, based on decreased fertility in the P0 generation (Gray et al. 2006). F1 animals were not examined in this study.

In the study by Wolf et al (1999) with LE hooded rats, fertility was monitored for both females and males (crossover mating to untreated animals). Fertility was decreased at 500 mg/kg bw/d for both sexes. The LOAEL for fertility in this multigenerational study is established at 250 mg/kg bw/d, based on reduced fertility in both sexes in the P0 generation and also decreased epididymal sperm counts in F1 observed at the same, lowest tested dose.

Fertility of males exposed to DBP only through gestation (GD13.5–21.5) and mated to untreated females was also affected (Mahood et al. 2007). In this study, primarily aimed at assessing developmental toxicity of DBP, adverse effects on fertility of the males (F1) treated during gestation was observed at 500 mg/kg bw/d. At the same dose, the testis weight in the F1 adults was also significantly decreased. However, a NOAEL for male fertility can not be established with certainty in this study because a statistically non-significant, but notable, decrease of fertility was also observed at lower doses and it correlated with significant testicular toxicity in the adult F1 animals (increase of dysgenic areas at 100 and 500 mg/kg/d).

In a continuous breeding dietary study with both male and female SD rats exposed to DBP (NTP 1995*; Wine et al. 1997), the LOAEL for fertility and embryotoxicity was at the lowest tested dose of 52–80 mg/kg bw/d (males-females), based on the decreased total number of live pups per litter following breeding of the F0 generation. This adverse effect was observed in the absence of maternal toxicity (observed only at the highest dose). In addition, no histopathological changes in the reproductive tract in F0 animals (males or females) or change in the average number of litters per pair were observed at any dose.

In this study, fertility indices (percentage of females with plug, pregnant and fertile) were significantly decreased at the highest dose (509–794 mg/kg bw/d for males–females) for the F1 but not for the F0 generation, suggesting greater sensitivity of the generation dosed from gestation, compared with F0, dosed seven days premating at adulthood (NTP 1995*; Wine et al. 1997).

Mice appear to be less sensitive than rats. The NOAEL for fertility, parental and embryotoxicity was 420 mg/kg bw/d. and the LOAEL was 1410 mg/kg bw/d in a 14-week dietary study with CD-1 mice (Lamb et al. 1987*; Morrissey et al. 1989*).

Studies in humans examining association of DBP exposure and effects on fertility are limited in number and quality, and without consistent results.

For males, the end points most examined to assess effects on fertility are related to sperm parameters such as sperm motility, morphology and velocity. Some studies involve in vitro exposure of sperm suspensions with DBP. In one such study (Fredricsson et al. 1993), a dose-dependent decrease in the mean motility and straight-line motion was observed at concentrations of 4 mM and above. However, it is unclear what the relationship of this treatment concentration is to the in vivo exposure dose, and the mechanism of action for such an effect is not addressed. In the two studies by Pant et al. (2008, 2011), there were concentration-dependent decreases in sperm motility, duration-dependent decreases in sperm viability, and positive correlations of DBP in semen with abnormal sperm or DNA fragmentation index. These results, together with in vivo studies, suggest that DBP can act directly on sperm.

There is limited evidence in humans associating MBP with effects on sperm motility. Duty et al. (2003, 2004) found that lower values for semen parameters (below the WHO reference values) were associated with higher urinary levels of MBP in a study group of 168 males. Similar results were obtained by the same group with an extended sample of 463 males (Hauser et al. 2006). In both studies, no significant association was found with metabolites of any of the other phthalates. Given that the studies were performed with subjects who presented to the collection centre for reasons of suspected infertility, the results may not be representative of the general population. However, because they are indicative of negative association between MBP and sperm quality, further studies and closer examination of the confounding factors are needed.

In contrast to the studies above, Jonsson et al. (2005) found no significant associations between highest versus lowest urinary MBP quartile values and any of the semen parameters studied in group of 234 military recruits.

Effects of DBP exposure on testosterone levels in adult males have also been examined. Pan et al. (2006) found that urinary MBP and MEHP levels (normalised to creatine) in 74 male workers in a PVC factory were significantly higher compared with 63 controls, and that serum testosterone was significantly lower in exposed workers compared with the controls.

Studies examining the relationship between DBP exposure and fertility in women focus on end points such as hormonal status, vaginal cycle disruptions, and occurrence of uterine conditions associated with decreased fertility, such as endometriosis.

One cross-sectional study, with 189 women working in conditions involving DBP exposure, concluded that DBP induces hormonal changes leading to menstrual cycle disruptions and decreased fertility. However, quantitative data were unavailable and the women were also exposed to other unknown compounds (Aldyreva et al. 1975*).

Two recent studies report significantly higher plasma levels of DBP and DEHP (Reddy et al. 2006) or urinary levels of their metabolites (Huang et al. 2010) in women diagnosed with endometriosis compared with controls. No significant association was found in women diagnosed with two other related conditions: adenomyosis and leiomyoma (Huang et al. 2010).

In contrast, a study by Itoh et al. (2009) with 137 women (50 with endometriosis and 80 controls), found no correlation between urinary levels of DBP (and DEHP) metabolites and endometriosis.

In all studies, the sample size was quite small and from a single sampling centre. In addition, occupational exposure was rarely considered and the measurements of phthalates done at the time of diagnosis may not reflect historical habitual exposure. Therefore, additional prospective studies are warranted.

In a more comprehensive cross-sectional study on 1227 women from the National Health and Nutrition Examination Survey (Wauve et al. 2010), a positive correlation was found between the urinary MBP metabolite and self-reported history of endometriosis and uterine leiomyoma. In contrast, an inverse association was observed for MEHP, MEHHP and MEOHP for both conditions. No significant associations were observed for MEP and MBzP. While more comprehensive, this study is not sufficient to establish a positive causal relationship between MBP (or by inference DBP exposure) and endometriosis or leiomyoma in the general population, for which further investigation in prospective studies is needed.

In summary, evidence in rodents constantly shows that DBP adversely affects fertility. The mechanism of toxicity involves overt effects on the reproductive tract organs in males. However, in female rodents, fertility is decreased even in the absence of obvious genital organ toxicity although the potential for progesterone and oestradiol production was shown to be altered in the ovarian cultures of the affected females in one study where this endpoint was measured. Testosterone synthesis is also affected by DBP in male rodents and this is particularly demonstrated in multigenerational and developmental studies (see below). Studies with humans are limited and often contradictory. They do not directly assess fertility but evaluate associations between indicators of DBP exposure, such as DBP or MBP levels in serum or urine, and parameters linked with (in)fertility such as sperm quality, testosterone levels and endometriosis.

Overall, the toxicity of DBP on fertility in rodents is similar to the related phthalate DEHP (NICNAS 2010), as it is mediated through similar adverse effects to the reproductive tract organs and perturbations in oestrogen and androgen synthesis, a mechanism of reproductive toxicity also relevant for humans.

Developmental toxicity

Prenatal

Numerous studies with rats exposed to DBP during gestation show toxic effects on foetal and postnatal development, particularly on male reproductive organs (see Appendix 1).

One of the lowest NOAELs for developmental toxicity following prenatal exposure is 20 mg DBP/kg bw/d, based on reduced testosterone levels correlated with significant testes dysgenesis at the LOAEL dose of 100 mg/kg bw/d (Mahood et al. 2007). In the other studies a NOAEL could not be determined as the LOAEL was at the lowest tested dose, except in the study by Mylchreest et al. 2000, where NOAEL was 50 mg/kg bw/d, based on increased seminiferous tubule atrophy and retained nipples at 100 mg/kg bw/d.

Higher doses of DBP given to rats during gestation in these studies also increase the incidence of cryptorchidism and hypospadias in the male offspring.

In a study with SD rats focused on details of histopathological effects (Mylchreest et al. 2002), exposure to 500 mg DBP /kg bw/d during gestation (GD 12–21) was associated with Leydig cell hyperplasia and an increased number of proliferating cell nuclear antigen (PCNA) positive Leydig cells in the foetal testes. The treatment was also associated with a decrease in testicular testosterone. At GD 21, testis atrophy was apparent, seminiferous cords were enlarged and contained PCNA-positive multinucleated gonocytes.

Persistent decrease in androgen concentration is consistent with the occurrence of reproductive tract malformations, as observed in a number of studies where hormone status was not examined. Sertoli cell dysfunction is also indicated by the alterations in gonocyte development (multinuclearity and hyperproliferation).

Studies that focused on examining the critical window of exposure significant for toxicity show that even a short, two-day, exposure to DBP during a critical window of development is sufficient to induce permanent developmental abnormalities. In SD rats, exposure to 500 mg DBP/kg bw/d on GD 15–16 and GD 18–19 was associated with decreased AGD, while retention of areolar nipples was observed in male offspring following exposure on GD 16–17. For increased epididymal malformations and incidences of small testes, two-day exposure on GD 17–18 was sufficient (Carruthers and Foster 2005).

In Wistar rats, DBP treatment from e13.5–e19.5, but not during late development (e19.5–21.5), critically affects development of rat testes. Toxicity exhibited as occurrences of Sertoli cells outside the normal seminiferous tubules and intermingled with Leydig and interstitial cells, and significant reduction of Sertoli cell numbers at foetal testes at e21. However the latter effect appeared reversible by postnatal day 25 in scrotal testes (Hutchison et al. 2008a, b).

Similarly, in Wistar rats, 500 mg DBP /kg bw/d treatment from e13.5 to e21.5 was associated with a significant decrease in the number of gonocytes in the early postnatal testes, with recovery by adulthood (PND 90). Short-term DBP treatment during late gestation (e19.5–e20.5) had no effect on gonocyte numbers at PND 4. However it induced MNGs at e21.5 with a frequency similar to that induced by daily DBP treatment from e13.5. (Ferrara et al. 2006).

In a more recent study, DBP exposure immediately following testis differentiation in rats (e13.5) caused a major reduction in foetal germ cell numbers. Foetal DBP exposure delayed postnatal resumption of germ cell proliferation, which led to more reductions in germ cell numbers and delayed re-expresssion of DMRT1 and germ cell migration. In contrast, late gestation effects included germ cell aggregation when germ cells are quiescent and had switched off OCT4 (Jobling et al, 2011).

Mice appear to be less sensitive to DBP than rats. In mice exposed to DBP during gestation, malformations of foetuses were observed at 100 mg/kg bw/d in the presence of maternal toxicity and exhibited as a significant increase in incidence of non-closing eyelids, encephalocoele, cleft palates and spina bifida, and increased incidence of skeletal abnormalities (Hamano et al. 1977*). In another study, the NOAEL for foetal toxicity was 350 mg/kg bw/d based on decreased pup weight at 660 mg/kg bw/d without maternal toxicity (Shiota et al. 1980).

Marmosets also appear to show low sensitivity to DBP toxicity. Exposure to 500 mg MBP/kg bw/day during gestation from week seven to 15 is not associated with adverse developmental or reproductive effects in the male offspring studied at birth (1–5 days) or in adulthood (18–21 months of age) (McKinnell et al. 2009). However, reliability of this study is limited considering that only one treatment dose was used, together with a small number of animals (maximum 6) for which significant individual variations were reported in some of the measured end points.

Postnatal and multigenerational studies

Early postnatal treatment, even a short, 7-day exposure to DBP is also associated with toxicity to the development of the reproductive system in rats. However, most studies that aim to evaluate end points relating to postnatal development include prenatal and postnatal treatment, the latter often only indirect through lactation, and can therefore be considered trans-generational studies.

Exposure of prepubertal (3-week-old) SD rats to 250 mg DBP/kg bw/d for seven days, was associated with decreased testes weight, while treatment at higher doses (500 mg/kg/bw/d and above) resulted in histopathological alterations, such as seminiferous tubule lesions, decreased tubular size, depletion of spermatogenic cells, wider tubular lumen and thin layer of seminiferous tubules (Alam et al. 2010). Thirty-day exposure at prepubertal age with 500 mg/kg bw/d, but not 200 mg/kg bw/d, was associated with decreased testostosterone levels in serum and non-reversible histomorphological alterations in testes (Xiao-feng et al. 2009).

Reproductive tract malformations were also observed in adolescent SD male rats (7-week-old) that were exposed to DBP during development (PND 1–21) and only indirectly through lactation (Zhang et al. 2004). The NOAEL for postnatal development in this study was 50 mg/kg bw/d (maternal dose) and the LOAEL at 250 mg/kg bw/d, based on decreased pup weight and male reproductive tract malformations including significant reduction of anogenital distance; increased frequency of testicular atrophy; under developed/absent epididymis; cryptorchidism; decreased epididymis weight and epididymal sperm motility; and decreased sperm heads per gram of testis. Mild degeneration of seminiferous epithelium was also observed at 250 mg/kg bw/d and was more severe at 500 mg/kg bw/d.

The lowest NOAEL for postnatal male developmental toxicity was established at 14 mg DBP/kg bw/day (NICNAS 2008b) in a dietary study with SD rats following gestational (from GD 15) and early indirect exposure (lactation) to DBP (Lee et al. 2004). This NOAEL is based on significant severity and incidence of adverse testicular effects (reduction in the spermatocyte development) at 148 mg/kg bw/d observed at PND21. However, considering that the incidence of the effect was also statistically significant at the lower doses (1.5 and 14 mg DBP/kg bw/day) even though the degree of severity was graded as minimal or slight, a lower LOAEL could be considered more appropriate. On the other hand, even the severe effects observed at the at 148 mg/kg bw/d appear to be reversible by PNW 20 (the highest dose of 1712 mg/kg bw/d was not examined at PNW 20).

In this study (Lee et al. 2004), the developmental NOAEL for females can be established at 3 mg DBP/kg bw/day, based on the significant decrease of relative pituitary weight at 29 mg/kg bw/d and above on PNW 20. However, this is also associated with significant uncertainty as no consistent histopathological changes were correlated with the decrease in weight.

Only a few studies are available where exposure continued during mating and lactation for both sexes. In only one study, with SD rats (NTP 1995*; Wine et al. 1997), treatment of F1 and F2 generations continued postweaning at the same dose level as their parents.

The NOAEL for developmental toxicity in F1 was 52 mg/kg bw/d, based on significant increase of kidney weight in F1 males at 256

mg DBP /kg bw/d. In F2 males, significant testicular atrophy and seminiferous tubule degeneration was observed at the same dose. However, no data were reported for the lowest tested dose and therefore a NOAEL could not be established for the F2 generation. (NTP 1995; Wine et al. 1997).

A similar NOAEL was established in an early study with Charles River COBS CD rats (IRDC 1984*) where both male and female rats were treated with DBP in the diet before mating (60 and 14 days, respectively), through mating, gestation and lactation. F1 weanlings were given a control diet (recovery group) or equivalent to that of their mothers, during a 7-week post-weaning period. The NOAEL for developmental toxicity was 50 mg/kg bw/d, based on slight decreases in testicular weights in weanlings in the 500 mg DBP/kg bw/d group. The decrease in weight correlated with testicular histopathological lesions in six of 10 weanlings continuously treated with the same dose, and in two out of nine weanlings in the corresponding recovery group.

Studies in humans mostly examine correlations between maternal MBP levels (in the urine or milk) and developmental parameters such as gonadotrophins and gonadal hormones, cryptorchidism or anogenital index. Behavioural and neuropsychological parameters have also been analysed. No significant association is reported for cryptorchidism but MBP levels in breast milk showed positive correlations with sex-hormone binding globulin and LH/free testosterone ratio, whereas the correlation with free testosterone was negative (Main et al. 2006). Studies that focused on measuring the anogenital distance in newborns found an inverse correlation between AGI and the maternal urinary concentrations of MBP, but not MEHP, using one statistical methodology (Swan et al. 2005). Another methodology calculated an inverse correlation for both metabolites (Swan et al. 2008).

Overall, the reliability of these studies for determining the effect of DBP in humans is limited because of the inconsistent results, most likely due to the low power of studies (small sample size, unrepresentative sample usually one study centre) and also uncertainties about the significance of the measured end points, for example AGD, as an indicator of developmental toxicity in humans. (2)

(2) Priority Existing Chemical Assessement Report No. 36 - Dibutyl phthalate; Australian Government, Department of Health; National industrial chemicals notification and assessment scheme, GPO Box 58, Sydney NSW2001, Australia; ISBN: 978 -0 -9874434 -4 -1, pp. 84-89