NTP Center for the Evaluation of Risks to Human Reproduction: phthalates expert panel report on the...

Reproductive Toxicology 16 (2002) 489–527

NTP Center for the Evaluation of Risks to Human Reproduction:phthalates expert panel report on the reproductive and developmental

toxicity of di-n-butyl phthalate�

Robert Kavlocka, Kim Boekelheideb, Robert Chapinc, Michael Cunninghamc,Elaine Faustmand, Paul Fostere, Mari Golubf, Rogene Hendersong, Irwin Hinbergh,

Ruth Littlec, Jennifer Seedi, Katherine Sheaj, Sonia Tabacovak, Rochelle Tyll,Paige Williamsm, Timothy Zacharewskin

a National Health and Environmental Effects Research Laboratory, USEPA, Research Triangle Park, NC, USAb Brown University, Providence, RI, USA

c NIEHS, Research Triangle Park, NC, USAd University of Washington, Seattle, WA, USA

e Chemical Industry Institute of Toxicology, Research Triangle Park, NC, USAf California Environmental Protection Agency, Sacramento, CA, USA

g Lovelace Respiratory Research Institute, Albuquerque, NM, USAh Health Canada, Ottawa, Ont., Canada

i Office of Toxic Substances, USEPA, Washington, DC, USAj Duke University, Durham, NC, USA

k Food and Drug Administration, Rockville, MD, USAl Research Triangle Institute, Research Triangle Park, NC, USA

m Harvard University, Boston, MA, USAn Michigan State University, East Lansing, MI, USA

Keywords:Di-n-butyl phthalate; Reproductive toxicity; Developmental toxicity; Review; Exposure; Systemic toxicity; Toxicokinetics; Risk evaluation

1. Chemistry, usage, and exposure

1.1. Chemistry

Di-n-butyl phthalate (DBP) (Fig. 1) (CAS RN 84-74-2)is produced through the reaction ofn-butanol with phthalicanhydride[1] (Table 1).

1.2. Exposure and usage

1.2.1. OverviewAccording to the American Chemistry Council (ACC, for-

merly CMA) [1], DBP is used mainly as a coalescing aid inlatex adhesives. DBP is also used as a plasticizer in celluloseplastics and as a solvent for dyes. Although there was lim-ited use of DBP in polyvinyl chloride (PVC) plastics duringthe 1970s and 1980s, it is not currently used as a plasticizer

� Correspondence to John Moore. Tel.:+1-703-838-9440;fax: +1-703-684-2223.E-mail address:[email protected].

in PVC. Release of DBP to the environment can occur dur-ing its production and also during the incorporation of thephthalate into plastics, adhesives, or dyes. Because DBP isnot bound to the final product, it can be released during theuse or disposal of the product. Phthalates that are releasedto the environment can be deposited on or taken up by cropsintended for consumption by humans or livestock and canthus enter the food supply.

1.2.2. General population exposureExposure of the general population to DBP has been esti-

mated by at least four authoritative sources: the InternationalProgram on Chemical Safety[3], the UK Ministry of Agri-culture, Fisheries, and Food (MAFF)[4,5], Health Canada[6], and the US Agency of Toxic Substances and DiseaseRegistry[7]. Levels of DBP in exposure media, assumptionsused in exposure calculations, and estimated exposure levelsare detailed inTables 2–4.

As noted in exposure estimates by the IPCS, HealthCanada, and ATSDR, the largest source of DBP exposureto the general population is food. Sources of DBP in food

0890-6238/02/$ – see front matter. Published by Elsevier Science Inc.PII: S0890-6238(02)00033-3

490 R. Kavlock et al. / Reproductive Toxicology 16 (2002) 489–527

Fig. 1. Chemical structure of di-n-butyl phthalate.

Table 1Physicochemical Properties of DBP[2]

Property Value

Chemical formula C16H22O4

Molecular weight 278.35Vapor pressure 2.7× 10−5 mmHg at 25◦CMelting point −35◦CBoiling point 340◦CSpecific gravity 1.042Solubility in water Slight: 11.2 mg/LLogKow 4.45

Table 2IPCS exposure estimates for adults[3]

Ambient air Indoor air Drinking water Food

DBP concentrationin media

0.0045–0.0062�g/m3 0.420�g/m3<1.0�g/L Various levels in a Canadian market

basket survey (see text)Assumptions 22 m3 inhaled/day;

64 kg bw; 4/24 h outdoors22 m3 inhaled/day;64 kg bw; 20/24 h indoors

1.4 L/day intake;64 kg bw

Various intake rates for differentfood types; 64 kg bw

Estimated doses(�g/kg bw/day)

0.00026–0.00036 0.120 <0.02 7

Table 3ATSDR exposure estimates for adults[7]

Ambient air Drinking water Fish

DBP concentration in media 0.003–0.006�g/m3 0.2�g/L 78–200�g/kgAssumed intake rate 20 m3/day/70 kg adult 2 L/day/70 kg adult 6.5 g/day/70 kg adultAssumed absorption fraction 0.5 0.9 0.9Estimated dose (�g/kg bw/day) 0.0005–0.0009 0.005 0.007–0.02

Table 4Health Canada DBP exposure estimates[6]

Substrate/medium Estimated intake DBP (�g/kg bw/day)

0.0–0.5 years old 0.5–4 years old 5–11 years old 12–19 years old 20–70 years old

Ambient aira 0.00030 0.00040 0.00041 0.00038 0.00034Indoor air 0.68 0.91 0.1 0.87 0.78Drinking water 0.11 0.062 0.033 0.022 0.021Food 1.6 4.1 3.2 1.4 1.1Soila 0.0070 0.0054 0.0018 0.00049 0.00040Total estimated intake 2.4 5.0 4.3 2.3 1.9

a Value represents the upper range of the estimates.

include environmental uptake during crop cultivation or mi-gration from processing equipment or packaging materials.IPCS [3] and Health Canada[6] conducted more compre-hensive exposure estimates. Both exposure estimates werebased on a 1986 Canadian market-basket survey of 1998 dif-ferent food types. Foods reported to contain DBP includedbutter (1.5 mg/kg), margarine (0.64 mg/kg), freshwater fish(0.5 mg/kg), cereal products (0–0.62 mg/kg), baked potatoes(0.63 mg/kg), bananas (0.12 mg/kg), coleslaw (0.11 mg/kg),gelatin (0.09 mg/kg), and white sugar (0.2 mg/kg). DBPexposure through food intake in adults was estimated at7�g/kg bw/day by IPCS[3] and at 1.9�g/kg bw/day byHealth Canada[16]. DBP exposures in children were alsoestimated by Health Canada by applying appropriate as-sumptions such as intake rates of different food typesper age group. Estimated DBP exposure levels from foodranged from 2.3�g/kg bw/day in children aged 12–19 yearsto 5.0�g/kg bw/day in children aged 6 months to 4 years.

MAFF [4] estimated adult DBP exposure through dietaryintake based on a 1993 survey of fatty foods in the UK.DBP was detected in carcass meat (0.09 mg/kg), poultry(0.2 mg/kg), eggs (0.1 mg/kg), and milk (0.003 mg/kg). Incalculating dietary food exposures, MAFF assumed thatthese types of food likely account for 85% of dietary

R. Kavlock et al. / Reproductive Toxicology 16 (2002) 489–527 491

phthalate intake. Food intake levels were obtained from theDietary and Nutritional Study of British Adults, but the val-ues were not reported by MAFF. Mean and high level DBPintakes were estimated at 13�g DBP/person/day and 31�gDBP/person/day, respectively. Specific details describingthe calculations and assumptions used were not provided.Using the IPCS-assumed[3] adult body weight of 64 kg, theexposure values were converted to 0.20–0.48�g/kg bw/day.

MAFF also addressed DBP exposure in infants resultingfrom the consumption of infant formula. A survey pub-lished in 1996 reported DBP levels of 0.08–0.4 mg/kg ininfant formulas purchased in the UK, while a later surveyreported DBP levels of<0.05–0.09 mg/kg[5,8]. It is spec-ulated that the drop in DBP concentration occurred becauseinfant formula manufacturers were urged to reduce phtha-late levels after MAFF published the results of the 1996survey. Exposure levels were estimated for infants basedon the results from the 1998 survey using assumed bodyweights of 2.5–3.5 kg at birth and 7.5 kg at 6 months of age.Formula intake rates were determined from manufacturerinstructions. Exposure levels for infants were estimated at2.4�g/kg bw/day at birth and 1.4�g/kg bw/day at 6 monthsof age. Infants in the US are likely exposed to lower levelsof DBP through formula than are infants in the UK. In asurvey of infant formulas conducted in 1996, DBP levelsin the US were approximately 10-fold lower than concen-trations measured in the UK and ranged from<5 to 11 ppb(<0.005–0.011 mg/kg)[9]. DBP has also been reported inbaby food and breast milk samples collected from Germanyand Japan; average values were within ranges reportedby MAFF. DBP was measured in seven German babyfood samples (average 0.033 mg/kg), eight baby formulas(<0.2–0.9 mg/kg; average 0.042 mg/kg), and in the breastmilk of five mothers from Germany (average 0.035 mg/kg)and three from Japan (0.02–0.08 mg/kg). The time periodwhen these samples were collected was not specified[1].

In their estimates of dietary exposure, ATSDR[7] onlyconsidered fish intake because at that time it was the onlyfood source for which reliable data were available. The di-etary estimate of 0.007–0.02�g/kg bw/day was based onDBP levels of 78–200�g/kg that were reported for fish instudies published between 1973 and 1987.

Levels of DBP in drinking water were estimated to beminimal. DBP exposure to adults through drinking waterwas estimated at 0.02�g/kg bw/day by IPCS[3] and HealthCanada[6] based upon a survey of drinking water sup-plies in Ontario, Canada. Health Canada also estimated DBPexposures through drinking water intake in children andthose values ranged from 0.022�g/kg bw/day in childrenaged 12–19 years to 0.11�g/kg bw/day in infants aged 0–6months. Adult DBP exposure through drinking water wasestimated by ATSDR[7] at 0.005�g/kg bw/day. The valuewas based on a survey of drinking water in 10 unspecifiedcities prior to 1986.

Mouthing of toys is another potential source of oral ph-thalate exposure in children. However, use of DBP in toys

appears to be rare. In an analysis of 17 plastic toys, DBPwas only detected in one polyvinyl chloride doll’s head at0.01% by weight[10].

Although off-gassing from building materials has beenreported as a potential source of DBP exposure throughinhalation, exposure has been postulated to be minimal be-cause of the low vapor pressure of DBP. The available data,though minimal, support this view. IPCS[3] estimated thatadults are exposed to 0.120�g/kg bw/day through inhala-tion of indoor air. The estimate was based on the mean airconcentration of DBP measured within 125 homes in Cali-fornia in 1990. Health Canada also estimated indoor inhala-tion exposure to DBP based on a survey of DBP air levelsin nine homes in Montreal (reported in 1985). Exposure toadults was estimated at 0.78�g/kg bw/day and exposures inchildren ranged from 0.68�g/kg bw/day in 0–6-month-oldinfants to 1.1�g/kg bw/day in 5–11-year-old children. Ex-posures to DBP through ambient air was also estimated byIPCS [3] and Health Canada[6]; the values were roughly2–3-order of magnitude lower than the indoor-air estimates.

Dermal contact with products containing DBP is possible,but absorption through skin is most likely to be minimal.Studies in rats have demonstrated that absorption of DBPthrough skin is fairly slow[11]. An in vitro study conductedwith rat and human skin has demonstrated that human skinis much less permeable to DBP than is rat skin[12].

Caution is required to interpret exposure data for the gen-eral population. IPCS has emphasized that dietary intakecan vary widely depending on the types of food eaten andthe types of material in which the foods are packaged. Inaddition, the majority of data used to estimate exposurelevels was collected 15–20 years ago and may not reflectcurrent exposure levels. Lastly, exposures in children maybe higher due to non-dietary intake through mouthing ofDBP-containing objects.

1.2.3. Medical exposureAccording to IPCS[3], a DBP level of 5 mg/g was mea-

sured in plastic tubing used for oral/nasal feeding. There areno other known uses of DBP in medical equipment.

1.2.4. Occupational exposureExposure in occupational settings can occur through skin

contact and by inhalation of vapors and dust. Phthalates aremanufactured within closed systems, but workers can be ex-posed during filtering or loading/unloading of tank cars[1].Higher exposures to phthalates can occur during the incorpo-ration of the phthalate into the final product if the process isrun at a higher temperature. In a limited number of surveys,DBP levels in US plants have ranged from concentrations be-low the detection limit (0.01–0.02 mg/m3) to 0.08 mg/m3 [3].OSHA established a permissible exposure limit of 5 mg/m3

for DBP. Following a review of six studies, the ACC hasestimated exposure to DBP in the workplace based upon anassumed level of 1 mg/m3 during the production of phtha-lates[1]. Exposure levels during the incorporation of DBP


Table 5Comparison of DBP dietary estimates

Agency Exposure in infants (0–6months) (�g/kg bw/day)

Exposure in adults(�g/kg bw/day)

IPCS [3] N/A 7MAFF [4,5,8] 1.4–2.4 0.2–0.48ATSDR [7] N/A 0.007–0.02Health Canada[6] 1.6 1.1

into plastics are not known. An exposure level was esti-mated by using assumptions of a 10 m3/day inhalation rateand a 70-kg body weight. The resulting exposure estimatewas 143�g/kg bw/workday for workers employed in phtha-late manufacturing. The maximum exposure, by regulation,would be five-fold greater. As stated in theSection 1.2.2,absorption of DBP through skin is expected to be minimal.

1.2.5. ConclusionExposure estimates varied between authoritative bodies.

However, in all cases it was evident that food was the primarysource of exposure to DBP. ATSDR only considered fishintake, and their exposure estimate therefore provides noinformation on total dietary exposure. The dietary exposureestimate by MAFF is approximately one order of magnitudelower than estimates by IPCS and Health Canada. The basisfor discrepancies in dietary exposure estimates is difficultto determine for several reasons, including: use of differentfood types in calculations (e.g. fatty foods versus a variety offoods); use of different assumptions in calculations; varyingDBP levels in foods from different countries; and changingDBP levels in food over time.Table 5lists the dietary DBPestimates calculated by the different agencies for infants andadults.

The summary forSection 1is located inSection 5.1.1.

2. General toxicological and biological parameters

2.1. General toxicity

2.1.1. Human dataThere were no human data located for expert panel review.

2.1.2. Experimental animal dataMultiple evaluations are available for assessing the ef-

fects of oral exposure to DBP. A few inhalation and dermalevaluations have also been conducted; these studies areprimarily in rats with a few assessments in mice, rabbits,hamsters, and guinea pigs.

2.1.2.1. Acute studies.The oral LD50 for DBP appearsto be between 8000 and 20,000 mg/kg bw in rats[3] andthe 90-day dermal LD50 is 4200 mg/kg bw in rabbits. Slightirritation was observed in rabbit dermal occlusion studies at520 mg/kg bw.

2.1.2.2. Repeat-dose studies.In a 3-month sub-chronicstudy, 6-week-old Wistar rats, 10 of each sex per dose, werefed a diet containing 0, 400, 2000 or 10,000 ppm DBP[13](Table 6). In addition to developing a toxicological profileof DBP, a stated purpose of the study was to evaluate possi-ble neurological or testicular toxicity. A battery of standardhematological and clinical chemistry parameters (includingthyroid function) was evaluated at points approximatelyhalfway through and at the end of the study. Cyanide insen-sitive palmityl-CoA oxidation (PCoA) was also determinedas a measure of peroxisome proliferation. Urinalyses wereperformed at the midpoint and at the end of the study. Neu-rological function, using the EPA functional observationbattery, was assessed prior to DBP administration, and ondays 34, 59, and 90 of the study.

Dietary consumption was not a factor in the study;nominal daily doses were calculated to be 27 (M) and33 (F) mg/kg bw/day, 142 (M) and 162 (F) mg/kg bw/day,and 688 (M) and 816 (F) mg/kg bw/day for the three-dosegroups. Effects were observed only in the high-dose group,688 (M), and 816 (F) mg/kg bw/day. Statistically signif-icant increases in liver and kidney to body weight ratioswere observed in the absence of body weight changesin females. Histologically, a decrease in lipid depositionwas noted in hepatocytes; this effect was possibly dueto peroxisome-related enzyme increases in the liver. Anincrease in PCoA activity was confirmed. Serum triglyc-erides and triiodothyronine were both decreased. RBC,hemoglobin, and hematocrit were transiently decreasedin males. No histological effects on testes appropriatelypreserved in Bouin’s fixative were observed.

Neurological function was assessed at three time pointsduring the study and no effects were observed. A LOAELwas observed at 688 (M) and 816 (F) mg/kg bw/day basedon multiple impacts and a NOAEL was determined at 142(M) and 162 (F) mg/kg bw/day.

Marsman[14] reported two 13-week, sub-chronic NTPstudies using male and female F344 rats. One of the studieswas of traditional design; 5–6-week-old rats were exposedto either control or one of four test diets. In the secondstudy, rats placed in a standard sub-chronic design wereborn and reared by mothers exposed to 10,000 ppm DBPduring pregnancy and nursing; at weaning, they were furtherexposed to a 10,000 ppm DBP diet until 8 weeks of age.

In the standard study, 10 F344 rats/sex were exposed toDBP in their diet for 13 weeks starting at 5–6 weeks of age(Table 7). The dietary levels were 0, 2500, 5000, 10,000,20,000, and 40,000 ppm (M: 0, 176, 359, 720, 1540, and2964 mg/kg bw/day; F: 0, 177, 356, 712, 1413, and 2943mg/kg bw/day). At the end of the study the rats were killedand necropsied with extensive tissue examination (testespreserved in 10% neutral buffered formalin), hematologyand clinical chemistry, sperm morphology, and vaginal cy-tology parameters were evaluated. Zinc and testosteronelevels were measured in sera and testes of all males. An in-crease in serum albumin was observed in exposed males at


176 mg/kg bw/day, the lowest dose tested. No other effectswere seen in either sex at this dose. Adverse effects in malesseen at the next highest dose (359 mg/kg bw/day) were evi-denced by a decrease in hemoglobin and erythrocyte counts.Severity of the hematological effects, seen only in males,progressed in a dose-response manner at all other doses.Platelets and serum albumin were increased, as were liverand kidney organ to body weight ratios. An increase in PCoAactivity was seen in both sexes, and an increase in bile acidwas seen in females. Decreases in body weight occurred inmales at the 720 mg/kg bw/day dose, the third-highest out offive treatment levels. Males exposed to 359 mg/kg bw/dayand males and females exposed to 712–720 mg/kg bw/dayhad increased liver and kidney organ to body weight ra-tios. Hepatic lesions in males and females and testicularlesions were first noted at 712–720 mg/kg bw/day. Testic-ular lesions consisted of focal seminiferous tubule atrophyin 4/10 males. The chemistry changes noted at the nextlower dose (356–359 mg/kg bw/day) continued at this dose(712–720 mg/kg bw/day) with the addition of increases inalkaline phosphatase activity. The histologic hepatic lesionspersisted and testicular lesions increased in severity at thehigher doses with all males of that dose group affected. Hy-pospermia of the epididymis was observed at the two highestdoses. Decreases in testicular organ weight ratios, testicularzinc, and testosterone were not observed until the 1540 (M)mg/kg bw/day exposure level. Peroxisome proliferation wasnoted histologically at the highest dose tested (2964 [M]and 2963 [F] mg/kg bw/day). Good dose-response data wasavailable for almost all parameters in this study. A NOAELof 176 mg/kg bw/day was identified by the expert panel.

In the second NTP sub-chronic study, F344/N rats wereborn and reared by mothers exposed to 10,000 ppm DBPin diet throughout prenatal development and lactation;the weaned rats were then fed a 10,000 ppm diet until 8weeks of age[14] (Table 8). At that time, the male rats,10 per sex per group, were placed on one of five dietsfor an additional 13 weeks that contained 0, 2500, 5000,10,000, 20,000, or 40,000 ppm DBP (M: 0, 138, 279, 571,1262, or 2495 mg/kg bw/day; F: 0, 147, 294, 593, 1182, or2445 mg/kg bw/day)[14]. The sub-chronic exposure dosesand the protocols for histopathology, hematology, and chem-istry were the same as the NTP sub-chronic study discussedabove. The authors concluded that developmental exposureto DBP resulted in neither increased sensitivity nor resis-tance to DBP exposure during adulthood (compare resultsin Tables 7 and 8). The expert panel notes that there weresignificant increases in organ to body weight ratios for kid-ney and liver in females and in testes at the lowest exposuregroup, 138 (M) and 147 (F) mg/kg bw/day. Such findingswere not observed at this dose level in the other sub-chronicstudy.

The NTP also conducted a sub-chronic study in6-week-old B6C3F1 mice where 10 mice per sex were fedDBP in the diet for 13 weeks at levels of 0, 1250, 2500,5000, 10,000, and 20,000 ppm (M: 0, 163, 353, 812, 1601,

and 3689 mg/kg bw/day; F: 0, 238, 486, 971, 2137, and4278 mg/kg bw/day[14] (Table 9). Experimental design inthis study was similar to the 13-week sub-chronic study inrats. There were no clinical signs related to exposure andall mice survived until the end of the study. Decreases inbody weight gain were observed in both sexes fed levels of812 mg/kg bw/day or higher. Increases in absolute and rela-tive kidney weight were seen in all treated female groups, butabsolute kidney weight was decreased in high-dose males.There was no report of histological change in the kidney nordid weights increase with increasing dose. The liver was theonly organ identified as a site of DBP toxicity by the studyauthors. Relative liver weights were increased at doses of812 mg/kg bw/day and higher. Cytoplasmic alterations con-sisting of fine eosinophilic granules, more intensely stainingcytoplasm, and increased lipofuschin were observed at thetwo highest doses in males (1601 and 3869 mg/kg bw/day)and at the highest dose in females (4278 mg/kg bw/day).A reduced hematocrit level was observed in high-dose fe-males. Based on decreased body weight gain, the NOAEL is353 mg/kg bw/day in males. A LOAEL based on increasedkidney weight in females is 238 mg/kg bw/day, the lowestdose tested according to the expert panel.

In a series of three identical experiments, Walseth andNilsen [15] examined lung and liver effects in groups offive male Sprague–Dawley rats. The rats were exposed for6 h/day for 5 days to DBP vapors at 0, 0.5, 2.5, or 7.0 ppm (0,5.7, 28.4, and 79.5 mg/m3 as calculated by authors). Therewere no effects on lung or liver weights. In the lung, therewere dose-related decreases in microsomal cytochromeP-450 and cytochromec-reductase levels in the two highestdose groups. There were no dose-related changes in livercytochrome levels. A significant decrease in serum levels ofalanine aminotranferase (ALAT) and significant increasesin serum aspartate aminotranferase and albumin levels wereobserved, but the authors indicated that there was no evi-dence of liver cell damage. The authors concluded that thelung is the main target organ following inhalation exposureto DBP.

2.2. Toxicokinetics

2.2.1. Phthalate moiety

2.2.1.1. Absorption.

2.2.1.1.1. Humans: dermal.In an in vitro study, humanskin absorption rate was reported as 0.07�g/cm2/h [12]which was considered ‘slow’.

2.2.1.1.2. Humans: oral. DBP was detected in bloodfrom humans following ingestion of foodstuffs containingDBP [3]. Background levels of DBP in human blood weremuch higher following exposure. Unfortunately, the authorsmeasured only the parent compound so there is no estimateof total DBP equivalents absorbed in this study. Similarly,


levels of DBP in human adipose tissue were studied[16];again total DBP equivalents were not calculated.

2.2.1.1.3. Rodents: dermal.Dermal absorption of DBPwas studied in Fischer 344 rats by applying 30–40 mg/kgradiolabeled DBP to the skin (administration site occluded)and measuring the radioactivity in urine[11]. Approximately10–12% of the dose was excreted in urine per day withapproximately 60% of the dose excreted within 1 week.Thirty-three percent of the dose was present at the applica-tion site 1 week following treatment.

2.2.1.1.4. Rodents: oral.The extent of intestinal absorp-tion of phthalate esters has been estimated by monitoringurinary excretion of the parent compounds or their metabo-lites after orally administering a known amount of theradiolabeled compound. Greater than 90% of radioactivityfollowing an oral dose of DBP in rats is recovered in theurine within 2 days, indicating nearly complete intestinalabsorption of this compound over a range of administereddoses[17]. This is consistent with the general observationthat dialkyl phthalate esters are well absorbed followingoral dosing. It is generally accepted that orally ingestedphthalate diesters are quantitatively hydrolyzed by gut li-pases and absorbed almost entirely as the correspondingmonoester.

2.2.1.2. Biotransformation.

2.2.1.2.1. Humans. In a study comparing the relative ratesof monohydrolysis of DBP by rat, baboon, and human gutpreparations, Lake et al.[18] demonstrated that these speciespossess similar intrinsic lipase activity. The observed ratesin human intestinal preparations were similar enough to theother species to expect that human intestinal metabolism ofDBP would result in absorption of the monoester similarlyto rats. The activity of pancreatic lipase was not assessed, sothe quantitative relationships of this study to in vivo exposurecannot be accurately determined[18].

2.2.1.2.2. Rodents.Dialkyl phthalates including DBPwere found to be metabolized to the monoesters by enzymespresent in many tissues. It is generally accepted that orallyingested phthalate diesters are quantitatively hydrolyzedby lipases in the wall of the small intestine and pancreaticlipases and not by gut flora. Absorption occurs almost en-tirely as the corresponding monoester[19]. Metabolites ofDBP include monobutylphthalate, monobutylphthalate glu-curonide,o-phthalic acid and oxidized monobutylphthalateglucuronide metabolites[17].

2.2.1.3. Distribution.

2.2.1.3.1. Humans. No human data were located for expertpanel review.

2.2.1.3.2. Rodents.DBP is rapidly cleared following oralor intravenous (i.v.) administration. There is little or nobioaccumulation observed. Radioactivity associated withDBP administration can be found in the GI tract and ex-cretory organs of the liver and kidney, and in fat. Liver,kidney, and the GI tract probably accumulate the phtha-late esters as a mechanism of excretion and not as depots[20]. One week following dermal treatment of Fischer 344rats with 30–40 mg/kg radiolabeled DBP, no tissues ex-amined contained more than 2% of the administered dose[11].

2.2.1.3.3. Pregnant rodents.Saillenfait et al.[21] studiedmetabolism and placental transfer of14C-DBP, administeredby gavage on gestation day (gd) 14 at 500 or 1500 mg/kgto Sprague–Dawley rats. Radioactivity peaked followed bya rapid decline in all tissues within 1–2 h of administration.Maternal plasma had the highest peak concentration; alltissue levels were less than 7% of peak concentrations by24 h. Fifty-five percent and 29% of a 500 mg/kg14C dosewere detected in urine and feces, respectively, in 24-h sam-ples; there was a slight increase to about 60% in urine at48 h, whereas fecal values did not change. Radioactivity inplacenta, embryo, and amniotic fluid were 0.3, 0.15, and0.2% of the administered dose, respectively. Concentra-tions in placenta and embryo did not exceed 30 and 21%of maternal plasma levels. The 1500 mg/kg dose indicatedslower absorption from the gastrointestinal tract; total fecalradioactivity was not affected, although there was lowerexcretion in urine over 48 h. In maternal plasma, placen-tal, and embryonic tissues, monobutyl phthalate (MBuP)and its glucuronide represented most of the DBP-derivedactivity. MBuP levels ranged from 50 to 95%, dependentupon the time after administration when samples weretaken. In contrast, unchanged DBP accounted for less than1%. The authors speculate that the lower levels of MBuPglucuronide in embryonic tissues compared to those inmaternal plasma could reflect limited placental transfer orlimited ability to conjugate this substrate. Levels of ra-dioactivity in placenta and embryos associated with DBPadministration were approximately 65% of the levels foundin maternal serum and there was no bioaccumulation of ra-dioactivity observed in the embryonic tissues. DBP, MBuP,and MBuP-glucuronide were present in embryonic tissuesat levels lower than were found in maternal plasma. MBuPaccounted for most of the radioactivity recovered in mater-nal plasma, placenta, and embryos, which is consistent withthe hypothesis that MBuP is the ultimate teratogenic speciesin vivo.

Distribution following i.v. exposure produces a differentdistribution pattern than that observed following oral ad-ministration. Since DBP is not in direct contact with gutesterases, metabolism to the monoester is slowed. Thisproduces more DBP-associated radioactivity to distributeto lungs and blood in addition to liver and kidney. Ra-dioactivity was detectable in adipose tissue 7 days after


i.v. exposure[22]. The difference between the oral and i.v.distribution probably reflects a higher concentration of par-ent DBP reaching adipose tissue following i.v. exposure,which would be expected to distribute to lipophilic tissuessuch as adipose tissue.

2.2.1.4. Excretion.

2.2.1.4.1. Humans. No human data were located for expertpanel review.

2.2.1.4.2. Rodents.The primary route of MBuP, the ma-jor DBP metabolite, elimination in rodents and humansis urinary excretion. The monobutylphthalate glucuronideappears to be the primary metabolite identified in rat urine[23]. MBuP is excreted into the bile (about 45%), but onlyabout 5% is eliminated in the feces, indicating that efficiententerohepatic recirculation occurs[17]. Biliary metabolitesof DBP include monobutylphthalate, monobutylphthalateglucuronide, and oxidized monobutylphthalate glucuronidemetabolites[17]. Following dermal exposure of rats to DBP,urine was the primary route of excretion with the excre-tion rate remaining nearly constant at 10–12% of the doseexcreted per day[11].

Mice are known to excrete higher amounts of glu-curonidated phthalate ester metabolites than rats and pri-mates excrete higher levels of glucuronidated phthalateester metabolites than mice[24]. There appears to be littleretention of DBP or MBuP in tissues of rats treated withDBP for 12 weeks[20].

2.2.2. ModelsA physiologically based pharmacokinetic (PBPK) model

of the tissue distribution of DBP and its monoester metabo-lite, MBuP, in rats administered DBP by various routes hasbeen developed by Keys et al.[25]. The model is based onan earlier model developed by the same group for DEHPand its metabolite, MEHP[26]. It includes a combinedperfusion-limited and pH trapping mechanism for uptake ofMBuP into tissues, and it provides a valuable tool for ex-trapolations of tissue doses among various routes and ratesof exposure. With modification, the model can be used toextrapolate doses to target tissues among various speciesand ages and between genders and gravid versus non-gravidfemales. The model allows estimation of the internal doseto specific target tissues for the evaluation of risk, ratherthan using total exposure or total internal dose as a riskestimate.

2.2.3. Side-chain-associated toxicokinetics (butanol)Butanol, a metabolite of DBP, is a primary alcohol that

is easily oxidized to butyric acid (n-butanoic acid) by al-cohol dehydrogenase and aldehyde dehydrogenase. Furthermetabolism by oxidation pathways converts butyric acid intoacetyl-CoA conjugates in intermediary metabolism path-ways with no toxicological importance[27].

2.3. Genetic toxicity

DBP has tested negative or marginally positive in genemutation and chromosomal aberration studies. The ASTDR[7] concluded that DBP may be weakly mutagenic. Thesignificance of these findings is not known because in vivogenotoxicity studies have not been conducted. The Wood-ward et al.[28] review concluded that the evidence indicatesthat DBP is not directly genotoxic, but noted it increasesthe sister chromatid exchanges and small increases in theincidence of gaps and breaks. However, the effect does notappear to be dose-related[29]. IPCS[3] reviewed a numberof mutagenic and related endpoints for DBP and concludedthat the weight of the evidence indicated that DBP is notgenotoxic. DBP was positive in the L5178Y mouse lym-phoma assay in the presence, but not in the absence, of anaroclor-induced rat liver activation system (S9)[30]. Theauthors conclude that the positive activity was likely theresult of in vitro metabolism of the DBP to an aldehyde,and therefore, that the results may not represent any realpotential for in vivo genotoxicity. DBP is not mutagenic inthe Salmonella/mammalian microsome mutagenicity assay[31], and was negative in the Balb/3T3 cell transformationassay[30].

The summary forSection 2, including general toxicity,toxicokinetics, and genetic toxicity, is located inSection5.1.2.

3. Developmental toxicity data

3.1. Human data

There were no human data located for expert panel review.

3.2. Experimental animal toxicity

A number of studies have evaluated DBP for both prenataland postnatal developmental toxicity; the vast majority ofstudies have been performed in the rat using the oral routeof exposure. In most cases, the doses were high (>0.5% indiet; >500 mg/kg bw/day), and the number of animals perdose group was small (10–15).

3.2.1. Prenatal development

3.2.1.1. DBP. Results from a set of investigations in micehave been reported by Shiota et al.[32] and Shiota andNishimura[33] (Table 10). They evaluated the effects of oralexposure to DBP in concentrations of 0, 0.05, 0.1, 0.2, 0.4,and 1.0% in the diet. On the day a cervical plug was observed(gd 0), female ICR-JCL mice commenced eating the DBPdiet until they were killed on gd 18. Using food consumptiondata, the authors calculated mean daily intake of DBP to be 0,80, 180, 350, 660, and 2100 mg/kg bw/day. Six to nine litterswere examined per dose group, except that 15 litters were


Table 10DBP developmental toxicity, rats

Species, strain,and source

Experimental regimen Numbera Dose(mg/kg bw/day)

Maternal effects Fetal effects

ICR-JCLmice [32,33]

Prenatal developmental toxicity study. Micewere fed diets with 0, 0.05, 0.1, 0.2, 0.4, or1% DBP from gd 0–18. Body weights weremeasured on gd 0–18. Dams were sacrificedon gd 18. Corpora lutea were counted andpups were examined for skeletal and softtissue malformations

8 07 80 NE LOAELb

Delayed ossification8 180 NE Delayed ossification6 350 NE Delayed ossification9 660 NOAEL ↓Fetal weight

Delayed ossification15 2100 ↓Body weight

gain↑Resorptions(98.4% vs. 5%)↓Fetal weightDelayed ossification↑Neural tube defects(2/3 fetuses)c

↑, Statistically significant increase;↓, statistically significant decrease; NE, no effects.a Number of pregnant females at sacrifice.b Differs from author’s selection of effect level.c Effect not statistically significant.

examined from the highest dose group. Food intake levelswere not affected in pregnant dams. Maternal weight gainwas significantly reduced at the high dose (2100 mg/kg bw/day), but the effect may have been secondary to increasedfetal loss. Resorptions (prenatal mortality) were signifi-cantly increased (98.4%) in the high-dose group. At thisdose, malformations in 2/3 surviving fetuses (increase notstatistically significant) were limited to neural tube defects(exencephaly and spina bifida, to which murine species arepredisposed). Delayed ossification was observed at all doselevels as indicated by a reduction in the number of ossi-fied coccygia in treated fetuses (n = 9.4, 5.1, 4.5, 6.0, and2.6 in the control to 660 mg/kg bw/day groups). Reducedfetal body weight was observed at the two highest doses.Because ossification was delayed at all dose levels, a de-

Table 11DBP developmental toxicity, rats


Experimental regimen Numbera Dose(mg/kg bw/day)

Maternal effects Fetal effects

Wistar rats[34] Prenatal developmental toxicity study. Ratswere gavaged with DBP from gd 7–15.Body weights and food intake weremeasured daily. Dams were sacrificed ongd 20. Implantation sites were examined.Pups were sexed, weighed, and evaluatedfor external malformations. Two-thirds offetuses were examined for skeletalmalformations and one-third for visceralmalformations

11 (11) 011 (11) 500 NOAEL NOAEL12 (12) 630 ↓Weight gain Complete resorption in

2/12 litters↓Live fetuses/litter (43%)↓Fetal weight (9–10%)

12 (12) 750 ↓Adjusted weightgain (38%)

Complete resorption in10/12 litters↓Live fetuses/litter (93%)↓Fetal weight (14–18%)↑External malformations(cleft palate) in 6/10fetuses (two litters) vs.0/118 fetuses in control

11 (9) 1000 ↓Adjusted weightgain (71%)

Complete resorption in9/9 litters

↑, Statistically significant increase;↓, statistically significant decrease.a Number of pregnant rats (number of litters evaluated).

velopmental NOAEL could not be identified for this studyand, therefore, a LOAEL of 80 mg/kg bw/day was selectedby the expert panel. However, the authors stated that ‘themaximum non-embryotoxic dose’ was 370 mg/kg bw/day.The maternal NOAEL and LOAEL were identified as 660and 2100 mg/kg bw/day, respectively.

Ema et al.[34–36] used Wistar rats to evaluate the de-velopmental toxicity of DBP by exposure through gavageand feed. In all studies, dams were sacrificed on gd 20–21and examined for implantation sites. Fetuses were weighedand examined for external, skeletal, and visceral malfor-mations. In Ema et al.[34], 12 dams/group were gavagedwith 0, 500, 630, 750, or 1000 mg/kg bw/day (0, 1.80, 2.27,2.70, or 3.60 mmol/kg bw/day) on gd 7–15 (Table 11).Gestational weight gain was reduced in dams of the


630 mg/kg bw/day group and adjusted weight gain (damweight not including gravid uterus) was reduced in damsexposed to 750 mg/kg bw/day and higher. Complete resorp-tions occurred in 2/12, 10/12, and 9/9 litters of the 630,750, and 1000 mg/kg bw/day dose groups, respectively,thus resulting in decreased live fetuses/litter. Fetal weightwas reduced in groups exposed to 630 mg/kg bw/day andhigher. External malformations, consisting entirely of cleftpalate, were increased in the 750 mg/kg bw/day group. Ma-ternal and developmental NOAELs and LOAELs of 500and 630 mg/kg bw/day, respectively, were identified.

Another study conducted by Ema et al.[36] is of particu-lar interest because it examines additional endpoints includ-ing anogenital distance and testicular descent (Table 12).In this study, 11 dams/group were fed diets containing 0,0.5, 1.0, or 2.0% DBP on gd 11–21. Authors estimateddaily intake rates of 0, 331, 555, and 661 mg/kg bw/dayfor the control to high-dose groups, respectively. Mater-nal gestational and corrected weight gain were reduced indams exposed to 555 mg/kg bw/day and higher and wereaccompanied by a reduction in food intake. Fetal weightwas reduced and the incidence of external malformations(cleft palate) and skeletal malformations (fused sternebrae)were increased in the 661 mg/kg bw/day dose group. Re-duced anogenital distance and increased incidence of un-descended testes were observed in male fetuses exposedto 555 and 661 mg/kg bw/day. The maternal and develop-mental NOAEL and LOAEL of 331 and 555 mg/kg bw/day,respectively, were identified for this study.

The two remaining studies by Ema et al.[35,37] focusedon the time- and dose-dependency of DBP developmentaltoxicity. In the studies, groups of 10–13 pregnant rats weregavaged with 0, 750, 1000, 1250, or 1500 mg/kg bw/day ongd 7–9, 10–12, or 13–15. Resorptions were increased inall dose groups at all time points. All dams treated with1500 mg/kg bw/day experienced complete litter resorptions.However, the types and frequencies of malformations var-ied according to the exposure time course. Treatment ongd 10–12 did not result in an increased malformation rate.Treatment with doses of 750 mg/kg bw/day and higher ongd 7–9 resulted in increased skeletal malformations (fusionor absence of vertebral arches and ribs). Administration of750 mg/kg bw/day and higher on gd 13–15 resulted in thegreatest incidence of teratogenicity, including increased ex-ternal malformations (cleft palate) and skeletal malforma-tions (fusion of sternebrae).

Saillenfait et al.[21] exposed Sprague–Dawley rats (27per group) to a single administration of DBP by gav-age on gd 14 at 0, 500, 1000, 1500, or 2000 mg/kg bodyweight. Increased resorptions at 1500 and 2000 mg/kg andreduced fetal body weights at 2000 mg/kg were observed.Skeletal variations were also increased at these doses. Keyaspects of the paper were studies on metabolism and pla-cental transfer of14C-DBP, administered by gavage ongd 14 at 500 or 1500 mg/kg. The toxicokinetic data arepresented inSection 2.2. The authors concluded that their

data support the view that MBuP may be the proximatetoxicant.

Developmental effects were also noted in reproduc-tive toxicity studies, which are discussed in detail underSection 4. In a continuous-breeding study, two generationsof Sprague–Dawley rats were exposed to 0, 80, 385, or794 mg/kg bw/day through diet during a 98-day matingperiod [38]. Maternal effects were only observed in thehigh-dose group and included a decrease in body weightfor both generations and increased liver and kidney weightsin F0 dams. Developmental effects included a reduction inlitter size in all dose groups and in live pup weight in thetwo highest doses of F1 rats. F2 pups in all treatment groupsexperienced a reduction in body weight. A developmentalLOAEL of 80 mg/kg bw/day and a maternal NOAEL of385 mg/kg bw/day were identified.

A similar continuous-breeding study was conducted inone generation of CD-1 mice treated with 0, 53, 525, and1750 mg/kg bw/day in diet[39,40]. Fetal effects that wereobserved only at the highest dose included reductions in lit-ter size, live pups/litter, and pup weight. The developmentalNOAEL was identified as 525 mg/kg bw/day, but a mater-nal NOAEL could not be identified because necropsies wereonly conducted in the high-dose group. In a multigenera-tion reproductive study, Long Evans (LE) Hooded rats weretreated with 0, 250, or 500 mg/kg bw/day DBP by gavagefrom the time they were weanlings through the time thatthey nursed their own litters[41]. Maternal toxicity was notreported. Developmental effects included malformations inreproductive organs, kidneys, and eyes in F1 rats and reduc-tions in F2 litter size in all dose groups. The developmentalLOAEL was identified as 250 mg/kg bw/day.

3.2.1.2. MBuP. The prenatal developmental effects of ad-ministering MBuP by gavage in the Wistar rat were reported[42,43]. The expert panel noted that some of the doses usedin these studies were equimolar equivalents to doses usedin earlier studies with DBP (described above). Ema et al.[42] studied doses of 0, 250, 500, and 625 mg/kg bw/day (0,1.13, 2.25, or 2.80 mmol/kg bw/day) on gd 7–15. They ob-served maternal toxicity at the two highest doses expressedas reduced weight gain and feed consumption. Also, at thesedoses there were significant increases in post-implantationloss/litter and decreases in live fetuses/litter and fetal bodyweight/litter. Fetal malformations were increased, with cleftpalate, deformed vertebral column, and dilated renal pelvisthe predominant findings. A maternal and developmentalNOAEL and LOAEL of 250 and 500 mg MBuP/kg bw/day,respectively, were identified for this study.

Ema et al.[43] then followed up with evaluation of stagespecificity studies by administering MBuP at doses of 0,500, 625, or 750 mg/kg bw/day on gd 7–9, 10–12, or 13–15.Embryolethality was increased at all doses for all dosing in-tervals. No teratogenicity was observed from the gd 10–12dosing interval. Increased incidences of fetal external mal-formations were present at the 500 and 750 mg/kg bw/day


doses on gd 7–9 and 13–15. Increased skeletal malforma-tions were observed at 500, 625, and 750 mg/kg bw/dayon gd 7–9 and at 625 and 750 mg/kg bw/day on gd 13–15(deformed cervical vertebrae were predominant on gd 7–9).Cleft palate and fused sternebrae were observed on gd13–15. These results are consistent with the findings forDBP and imply that MBuP (and/or subsequent metabolites)may account for the developmental toxicity (embryolethal-ity and malformations) for DBP.

3.2.2. Postnatal development

3.2.2.1. DBP. Marsman et al.[14] exposed F344/N ratsand B6C3F1 mice to high dietary concentrations of DBP dur-ing gestation and lactation. Both species were exposed to 0,1250, 2500, 5000, 7500, 10,000, and 20,000 ppm. Dosages inmg/kg bw/day were estimated by using average values from2 NTP studies that included a food intake rate of 14.8 g/dayand a body weight of 203.71 g for rats and a food intake rateof 7.18 g/day and body weight of 39.63 g for mice[44–46].The dosages are listed inTables 13 and 14. After weaningon pnd 21, up to 10 F1 pups/group were fed a diet with aDBP concentration identical to that fed to their dams for anadditional 4 weeks. Author-calculated doses for pups were:143, 284, 579, 879, and 1165 mg/kg bw/day for male rats;133, 275, 500, 836, and 1104 mg/kg bw/day for female rats;199, 437, 750, 1286, and 3804 mg/kg bw/day for male mice;and 170, 399, 714, and 1060 mg/kg bw/day for female mice.Complete necropsies were performed on one rat and onemouse pup of each sex per litter at weaning and on all pupsat the end of the 4-week post-weaning dietary exposure. Or-gan weights were obtained on major organs, including testis.Histopathological examination was performed on a broadarray of tissues from all animals in the control and highestexposure group. In addition, the epididymis of rats from the2500, 5000 and 7500 ppm groups were studied.

For the rats (Table 13), gestational index was reduced(fewer live litters) at 5000 and 20,000 ppm, and gestationallength was reduced at 5000 ppm. Litter size and postnatalsurvival were reduced at 10,000 and 20,000 ppm. All F1pups died by pnd 1 in the 20,000-ppm group. Male pupbody weights were reduced during lactation in dose groupsreceiving 7500 ppm and higher. In the post-weaning pe-riod, relative liver and kidney weights were increased infemale offspring exposed to≥2500 and≥5000 ppm (275and 500 mg/kg bw/day), respectively. Increased liver andkidney to body weight ratios were observed in males of alldose groups. Reduced relative testis weights were observedat the highest dose. Mild-to-marked hypospermia was seenin all males at the 879 and 1165 mg/kg bw/day doses andin 4/10 males of the 579 mg/kg bw/day dose group. Therewere no histopathological lesions observed in liver or kid-ney. Acquisition of vaginal patency and preputial separationwere not assessed. Based on increased liver and kidney tobody weight ratios in all treated males, no NOAEL wasidentified.

For B6C3F1 mice (Table 14), length of gestation wasincreased at 2500 ppm and higher with 75 and 95% oflitters lost at 10,000 and 20,000 ppm. Decreases were ob-served in litter size and pup body weights at 2500, 7500,and 10,000 ppm. In the F1 post-weanling phase, males ex-hibited increased relative liver weights (one surviving malepup at 10,000 ppm exhibited hepatic lesions), and femalesexhibited increased relative kidney weights at 1250 ppm(170–199 mg/kg bw/day) and higher. Except for liver lesionsin the male at 10,000 ppm, no histpathological changes wereobserved, including in the testis. No NOAEL was identified.

Taking note of Wine et al.[38] continuous-breeding studyresults (seeSection 4), Mylchreest et al.[47] followed upthe study using comparable dose levels (Table 15). However,three important changes in experimental design were intro-duced: (1) shortening the exposure period to include onlygestation and lactation; (2) using gavage (with corn oil) tocontrol exposure more closely; and (3) including more sensi-tive endpoints of reproductive development, such as markersof sexual maturation. Thus, pregnant CD rats (10 per group)were administered DBP by gavage at 0, 250, 500, or 750 mg/kg bw/day from gd 3 until pnd 20. At birth, pups werecounted, sexed, weighed, and examined for signs of toxicity.Sexual maturity was assessed by observing age of vaginalopening and preputial separation in females and males, re-spectively. Estrous cycles were assessed in females for 2weeks. The F1 rats were sacrificed at 100–105 days of age.Necropsies were conducted on all males and up to three fe-males per litter. A histological examination of sex organs wasconducted on all rats with lesions and up to two unaffectedrats per litter. Testes were preserved in Bouin’s fixative.

Maternal body weight gain was comparable to controlsthroughout the dosing period. At 750 mg/kg bw/day, thenumber of live pups per litter at birth was decreased and ma-ternal effects on pregnancy and post-implantation loss arelikely to have occurred. Anogenital distance was decreasedat birth in the male offspring at 500 and 750 mg/kg bw/day.The epididymis was absent or underdeveloped in 0, 9, 50,and 71% of adult offspring (100 days old) at 0, 250, 500,and 750 mg/kg bw/day, respectively, and was associatedwith testicular atrophy and widespread testicular germ cellloss. Hypospadia occurred in 0, 3, 21, and 43% of males,and ectopic or absent testes in 0, 3, 6, and 29% of males at0, 250, 500, and 750 mg/kg bw/day, respectively. Absenceof prostate gland and seminal vesicles as well as small testesand seminal vesicles were noted at low incidence in the 500and 750 mg/kg bw/day dose groups. Dilated renal pelvis,frequently involving the right kidney, were observed in allDBP dose groups. Vaginal opening and estrous cycles werenot affected in the female offspring, although low incidencesof reproductive tract malformations, mainly involving de-velopment of the uterus, were observed in two and one ratat the 500 and 750 mg/kg bw/day doses, respectively.

In Mylchreest et al.[47], all exposed groups showed ad-verse effects on male reproductive tract structure and in-dices of puberty. Based on this, the LOAEL in this study is


250 mg/kg bw/day/day. Based on the relationship betweentestis weight/histopathology and sperm production, the rela-tionships between sperm numbers and fertility[48], and thenumber of major malformations of the reproductive tract,it is expected that at least the high- and mid-dose animalswould be sub-fertile. The panel’s confidence in the qualityof the study is high.

In a subsequent study, Mylchreest[49] reduced DBP ex-posure to just late gestation (gd 12–21) and compared theeffects of DBP to the pharmacological androgen receptorantagonist, flutamide (Table 16). Pregnant CD rats receivedDBP at 0, 100, 250, or 500 mg/kg bw/day by gavage withcorn oil (n = 10) or flutamide at 100 mg/kg bw/day (n = 5)on gd 12–21. Males were killed at approximately 100 daysof age and females at 25–30 days of age. In F1 males,DBP (500 mg/kg bw/day) and flutamide caused hypospadia,cryptorchidism, agenesis of the prostate, epididymis, andvas deferens, degeneration of the seminiferous epithelium,and interstitial cell hyperplasia of the testis. Agenesis of theepididymis was also observed at 250 mg/kg bw/day. Flu-tamide and DBP (250 and 500 mg/kg bw/day) also causedretained thoracic nipples and decreased anogenital distance.Interstitial cell adenoma occurred at 500 mg/kg bw/day intwo males from the same litter. The only effect seen at100 mg/kg bw/day was delayed preputial separation. Thelow incidence of DBP-induced intra-abdominal testes con-trasted with the high incidence of inguinal testes seenwith flutamide. Thus, the prenatal period is sensitive forthe reproductive toxicity of DBP. Uterine and vaginal de-velopment in female offspring was not affected by DBPtreatment. There were no signs of maternal toxicity with theexception of a 16% body weight loss at the time of birth andcomplete fetal mortality in 1 dam of the 500 mg/kg bw/daygroup. In addition, testicular focal interstitial cell hyperpla-sia and an adenoma (in one male) were observed in malesat 500 mg/kg bw/day at 3 months of age. A LOAEL of100 mg/kg bw/day was established in this study, based ondelay in preputial separation at all dose levels. A NOAELwas not established.

To identify a NOAEL for DBP-induced develop-mental toxicity, Mylchreest et al.[50] gavaged 19–20Sprague–Dawley CD rats/group with 0, 0.5, 5, 50, or100 mg/kg bw/day and 11 Sprague–Dawley CD rats with500 mg/kg bw/day in corn oil on gd 12–21 (Table 17).Dams delivered and pups were weighed and examined atbirth. After the pups were weaned, dams were killed andimplantation sites and organ weights were evaluated. Pupswere weighed weekly and examined for sexual maturation.When pups reached puberty they were killed and organweights were determined. The testes and epididymis werepreserved in Bouin’s solution and examined histologically.

There was no evidence of maternal toxicity at any dose. Inmale pups, the incidence of retained aereola or nipples wasincreased at the 100 and 500 mg/kg doses (31% of rats in16/20 litters and 90% of rats in 11/11 litters, respectively).Malformations observed in the highest dose group included:

hypospadia (9% of rats in 4/11 litters); and agenesis of theepididymis (36% of rats in 9/11 litters), vas deferens (28% ofrats in 9/11 litters), and prostate (1/58 rats). Reduced testis,epididymis, prostate, and levator muscle weight and reducedanogenital distance in males were also observed at the highdose. Histological effects in high-dose males included inter-stitial cell hyperplasia (35% of rats in 3/5 litters), adenoma(1/23 rats), and seminiferous tubule degeneration (56% ofrats in 3/5 litters). The single case of seminiferous tubule de-generation in the 100 mg/kg bw/day group was consideredequivocal because the lesion does occur spontaneously in asmall number of Sprague–Dawley rats. In female offspring,the age of vaginal opening and reproductive organ weightand histology were unaffected. A developmental NOAELand LOAEL of 50 and 100 mg/kg bw/day, respectively, anda maternal NOAEL of 500 mg/kg bw/day, were identified forthis study.

The qualitative findings of Mylchreest et al.[47,49,50]were confirmed by Gray et al.[41] who gavaged 8–10Sprague–Dawley rats/group from gd 14 to lactation day3 with corn oil vehicle or DBP at 500 mg/kg bw/day, andgroups of 4–6 Long Evans Hooded rats with 0 or 500 mg/kg bw/day on gd 16–19.

Gray et al. [41] also compared the effects of DBPat 500 mg/kg bw/day and an equimolar concentration of750 mg/kg bw/day DEHP administered by gavage to 8–10Sprague–Dawley rats/group from gd 14 to lactation day 3(Table 18). The male F1 pups were evaluated for sexual mat-uration and were then killed and necropsied at 5 months ofage. Organ weights were measured and a histological exami-nation of reproductive organs (preserved in Bouin’s solution)was conducted. The presence or absence of maternal toxi-city was not described. Effects in F1 males are summarizedin Table 19and included reduced anogenital distance, andincreases in percent areolas and nipples at birth, numbers ofareolas and nipples at birth and adulthood, hypospadia, andtesticular and epididymal atrophy or agenesis. A decrease inweight for prostates, epididymis, testes, penis, and the leva-tor animal muscle was also observed in the treated rats. Noneof the control pups were found to have nipple development,malformations, or testicular degeneration. DEHP and DBPexposure resulted in effects that were qualitatively similar.Several males from DEHP-treated dams also had hemor-rhagic testes. The authors stated that DEHP was consider-ably more toxic to the male reproductive system than DBP.

In an abstract, DBP was reported to have been evaluatedfor developmental toxicity in amphibian and non-rodentmammalian test systems[51]. Xenopus laeris(Africanclawed toad) tadpoles were exposed to 0 (n = 14) or 10(n = 52) ppm DBP beginning at 2 weeks of age (stage 52)through complete metamorphosis (stage 66), with mortalityand time to complete metamorphosis monitored weekly.Mortality at 10 ppm was 85% in week 1 (0% in controls)and 92% in week 16 (28% in controls). Seventy-five per-cent of the controls were metamorphosed by week 12 with100% by week 14; none of the treated tadpoles completed


Table 19Comparison of reproductive effects following in utero exposure to equimolar concentrations of DEHP (750 mg/kg bw) and DBP (500 mg/kg bw) inSprague–Dawley rats[41]

Effect Control Chemical

DEHP DBP

Anogenital distance (mm) 3.7 ± 0.09 2.45± 0.11a 2.79 ± 0.09a

Areolas at birth (%) 0 88 ± 12 55± 14Number of areolas at birth 2.7 ± 0.75 8.4± 15 2.7± 0.75Retained nipples at birth 0 8.1 ± 1.4a 2.2 ± 0.8a

Number of nipples at necropsy 0 8.1 ± 1.4a 2.2 ± 0.8a

Hypospadia (%) 0 67 ± 14 6.2± 6.2Vaginal pouch (%) 0 45 ± 17 0Ventral prostate agenesis (%) 0 14 ± 14 0Testicular and epididymal atrophy or agenesis (%) 0 90± 10 45.8± 12

a Statistically significant.

metamorphosis until week 16. The authors concluded thatDBP or its metabolite(s) may disrupt thyroid hormone cas-cade, since metamorphosis, a thyroid hormone-dependentevent, is affected at 10 ppm. The same group administeredDBP in corn syrup at 0 or 400 ppm/kg body weight topregnant Dutch belted rabbits, 6 does/group, on gd 15–30.Does were allowed to litter and male pups were monitoreduntil 12 weeks of age. At 12 weeks of age, body, testes,and epididymis weights were unaffected, but accessorygland weights and anogenital distance were lower in treatedmale offspring. In addition, analogously to male rats ef-fects, one treated rabbit had undescended testes, ambiguousexternal genitalia, hypospadia, and was missing (agene-sis of) the prostate and bulbourethral glands. The authorsconcluded that DBP disrupts androgen-dependent develop-mental events and is consistent with anti-androgenic effectsof DBP observed in rodents after perinatal exposure.

3.2.2.2. MBuP. Imajima et al. [52] gavaged pregnantWistar–King A (WKA) rats with MBuP in sesame oil at 0or 300 mg/day on gd 15–18 (equivalent to approximately1000 mg/kg bw/day based on actual rat body weights)(Table 20). Male offspring were evaluated on gd 20 and onpnd 30–40 to determine the position of the testes. In controlmales, all the testes were located in the lower abdomenon gd 20 (19 pups, three litters) and had descended intothe scrotum on pnd 30–40 (15 pups, three litters). In starkcontrast, in males exposed in utero to MBuP, all testes werelocated high in the abdominal cavity (15 pups, three litters)with significantly higher testes ascent on gd 20. On pnd30–40, MBuP-exposed males exhibited cryptorchidism (22of 26 pups, five litters) with uni- or bilateral undescendedtestes; 87% of the undescended testes were in the abdominalcavity, the remaining 13% were located at the external in-guinal ring. Testis descent is under androgenic control; theauthors suggest that phthalate esters may interfere with FSHstimulation of cAMP accumulation in Sertoli cells, resultingin the reduced secretion of Mullerian inhibiting substance, aputative mediator in trans-abdominal migration of the testis.


4. Reproductive toxicity

4.1. Human data

The relationship between either human sperm densityor total number of sperm and DBP concentration in thecellular fraction of ejaculates was studied in a group ofunselected college students[53]. A negative correlationbetween DBP concentration and the studied sperm indiceswas found. The authors point out that there was no rea-son to believe that any of the students examined had beenexposed to phthalate esters other than at ambient levels inthe environment. However, the use of this study to supporta causal relationship to DBP exposure is limited becausesubjects’ characteristics and other potential risk factors thatcould confound or modify the observed association werenot taken into account by the authors.

4.2. Experimental animal toxicity

Approximately 20 studies were reviewed in the evalua-tion of the reproductive toxicity of DBP. Collectively, thesestudies predominantly used rodents, and built on the orig-inal observation that DBP produced testicular atrophy ina sub-acute toxicity study[54]. The literature contains nu-merous redundant studies, usually at high doses (e.g. 2 g/kg,usually in rats), all of which show similar effects on thetestis. For example, Gray et al.[55] reported on the testic-ular effects of DBP in the adult rat, mouse, guinea pig, andhamster. In these studies, DBP was administered by gavagefor 7 or 9 days at doses of 2000 or 3000 mg/kg bw/day. Se-vere effects were seen on testis weight with histopatholog-ical damage (reduction in spermatids and spermatogonia)affecting almost all tubules. Mouse testis was less severelyaffected and no effects were observed in hamsters. Themonoester of DBP was also essentially without effect inthe hamster. As discussed inSection 2.1.2, sub-chronic oralexposure of adult F344 rats resulted in testicular lesions atdoses of 720 mg/kg bw/day and higher[14]. A second study[14] demonstrated that exposure to DBP during gestation


and lactation did not increase sensitivity in rats exposed toDBP for 3 months during adulthood. Sub-chronic studiesin B6C3F1 mice at doses up to 3689 mg/kg bw/day did notcause histological or organ weight changes in the testes.

A number of more specific studies in the rat have at-tempted to investigate the mode of action of DBP using invivo and in vitro protocols. The papers summarized hereillustrate important facets of DBP-induced reproductive ef-fects.

The key study for the quantitative assessment of the re-productive toxicity of DBP is reported by Wine et al.[38](Table 21). CD Sprague–Dawley rats, 10 weeks old at thestart of exposure, were used for continuous-breeding phaseand crossover mating studies. There were 20 breeding pairsin each treated dose group, and 40 pairs in the controlgroup. DBP was mixed with feed to levels of 0, 0.1, 0.5and 1.0% (w/w); this yielded calculated doses of 0, 52,256, and 509 mg/kg bw/day for males and 0, 80, 385, and794 mg/kg bw/day for females. Following a 7-day prematingperiod, the rats were housed as breeding pairs for 14 weeks.Litters were removed immediately after birth. Endpointsin life included clinical signs, parental body weight andfood consumption, fertility (numbers of pairs producing alitter/total number of breeding pairs), number of litters/pair,number of live pups/litter, proportion of pups born alive,sex ratio, and pup body weights within 24 h of birth.

In the F0 generation there was no effect on the overallfertility of the breeding pairs (i.e. the ability to produce lit-ters with at least one live pup); all produced approximatelyfive litters. There was clear indication that DBP, when ad-ministered in the diet, affected total number of live pups perlitter in all treated groups (reduced by∼8–17%) and livepup weights in the 256–385 and 509–794 mg/kg bw/daygroups by 6–12%.

A crossover mating study was conducted between thehigh-dose treatment group and the controls. The percent ofpairs mating, becoming pregnant, and delivering a litter wasunaffected, as was litter size, although adjusted live pupweight was reduced in litters from treated females. At F0necropsy, there were no gross or histopathological effects inthe reproductive tracts of treated animals. Epididymal spermcount, testicular spermatid number, and estrous cycle lengthwere not affected by DBP treatment in the F0 animals. Sys-temic effects in the F0 rats included decreased body weightin females and increased liver and kidney to body weight ra-tios in both sexes of the high-dose group. The final F1 littersfollowing the continuous F0 breeding phase were weanedand raised to sexual maturity (pnd 88) and received the samedose in feed as their parents. Upon reaching sexual matu-rity, 20 non-sibling F1 males and females within the sametreatment group were housed in pairs for 1 week and thenhoused individually until delivery of an F2 litter.

F1 pup weight was significantly reduced in the high-dosegroup on pnd 0, 14, and 21. During rearing, three high-dosemales were found to have small and malformed prepucesand/or penises and were without palpable testes. Mating,

pregnancy, and fertility were significantly lower in thehigh-dose F1 group with only 1 of 20 pairings resultingin a litter. While litter size was unaffected, F2 pup weightwas reduced in all treatment groups. All dose groups werekilled and necropsied, at which point the body weights ofthe high-dose animals were 8–14% lower than controls, butunchanged at other dose levels. For males only, kidney tobody weight ratio increased at the 256–509 mg/kg bw/daylevels and liver to body weight ratio was increased at thehighest level. The relative weights of the ventral prostateand seminal vesicles and the absolute weight of the righttestis were decreased in the F1 males from the high-dosegroup. There were no effects on the ovary of F1 females.Epididymal sperm count and testicular spermatid count wassignificantly reduced in the high-dose F1 males. Histologi-cal analysis was only performed on selected males (n = 10)from the control, mid-, and high-dose groups (the solutionused to preserve testes is not clear). Widespread seminifer-ous tubular degeneration was noted in 1/10 controls, 3/10 inthe mid-dose group, and 8/10 in the high-dose group. Thehigh-dose group also exhibited interstitial cell hyperplasia.Five of 10 high-dose males also had underdeveloped ordefective epididymis. No ovarian or uterine lesions werenoted in F1 females and there was no effect on ante-mortemestrous cyclicity.

In Wine et al.[38], the F1 high-dose group had a highrate of infertility, the middle dose had fewer (F0 mating)and lighter pups (F0 and F1 matings), while the low-doseanimals had fewer pups (F0 mating) and lighter pups (F1mating). Thus, a NOAEL was not established. The LOAELwas 52–80 mg/kg bw/day based on reductions in F0 littersize and F2 pup weight. The expert panel’s confidence inthe quality of the study is high, and our confidence is alsohigh that these doses correctly represent the LOAEL.

A multigeneration reproductive study was conducted toassess effects of DBP exposure in Long Evans Hoodedrats [41] (Table 22). Weanling male and female rats of theparental (F0) generation (10–12/sex/group) were gavageddaily with DBP in corn oil through puberty, adulthood,mating, gestation, and lactation. Females received 0, 250,or 500 mg/kg bw/day; male rats received 0, 250, 500, or1000 mg/kg bw/day. Sexual maturation and estrous cyclesof the F0 were evaluated. Treated rats were mated withuntreated controls. When the F1 litters were weaned, theparental rats were killed and necropsied. Implantation sites,serum hormone levels, organ weights, and testicular histol-ogy were evaluated.

A delay in puberty was observed in all treated F0 malesbased on the age of preputial separation (42.6, 43.4, and44.4 days from low to high-dose group versus 39.6 daysin control group). Fertility was reduced in F0 males andfemales in the 500 mg/kg bw/day group. Infertility in F0males was apparently due to testicular atrophy and re-duced sperm counts. F0 females in the 500 mg/kg bw/daygroup cycled and mated successfully, but experienced anincreased incidence of mid-term abortion. Malformations


were significantly increased in F1 pups from the 250 and500 mg/kg bw/day groups. Types of malformations includedlow numbers of hypospadia, abdominal testes, anoph-thalmia, uterus unicornous, and renal agenesis.

The F1 pups were not treated with DBP after weaning.Four to eighteen pairs of F1 pups from treated dams wereselected for continuous mating within dose groups for 11cycles and the remaining F1 pups were necropsied. TheF2 pups born during the continuous breeding phase werecounted and discarded. Fecundity was reduced in F1 ratsfrom treated dams and the number of F2 pups born was re-duced in breeding pairs from the 250 and 500 mg/kg bw/daygroups. At necropsy, a non-significant reduction in caudalsperm counts (19%) and a significant reduction in caudalsperm levels (34%) were noted in F1 males from the 250 and500 mg/kg bw/day groups, respectively. The study by Grayet al.[41] is somewhat limited because many endpoints anddetails of their experimental methods were not reported.

In Lamb et al. [39] and Reel et al.[40] (Table 23),DBP was one of four phthalate esters compared usingthe continuous-breeding protocol in CD-1 mice; the samebasic protocol as reported in Wine et al.[38]. Male andfemale CD-1 mice, 20 pairs/treatment group and 40 pairsin control, were fed a diet with DBP at 0, 300, 3000, or10,000 ppm (doses of 53, 525, and 1750 mg/kg bw/day asreported by Reel et al.[40] for 7 days prior to and duringa 98-day cohabitation period. Litters were removed im-mediately after birth. Reproductive function was evaluatedduring the cohabitation period by measuring the numbersof litters per pair and of live pups per litter, pup weight, andoffspring survival. Testes were fixed in Bouin’s solution forhistological evaluation. DBP exposure reduced litter size,numbers of litters per pair, number of fertile pairs, live pups

Table 23DBP reproductive toxicity, mice


Experimental regimen Numbera Doseb

(mg/kg bw/day)Effects

CD-1 mice[39,40]c Fertility assessment through a continuousbreeding study. DBP administered in feed.Breeding pairs housed together for 98days; body weight was measured on 7days, clinical signs, and food intake wererecorded; litters were counted, sexed,weighed, and removed following birth. In acrossover breeding study, high dose malesand females were mated with control mice.Breeding pairs were housed together for 7days or until a copulatory plug wasobserved. Necropsy and a histopathologicalexamination were conducted.

39 020 52.5 NE18 525 NOAEL20 1750 ↓Number of fertile pairs

↓Number of litters delivered/pair↓Litter size↓Live pups↑Percentage of male pups↓Pup weight↓Uterus to body weight ratio in F0females↓Body weight in F0 males↑Liver to body weight ratio in F0males and femalesNo effects on estrous cycles, spermmorphology, or sex organs in F0 mice

NE, no effect;↑, statistically significant increase;↓, statistically significant decrease.a Number of male and female pairs.b Author-calculated doses.c This study is also addressed in[14].

per litter, and proportion of pups born alive in the high-dosegroup. These effects were not seen at lower dose levels.A crossover mating trial demonstrated that female, but notmale mice were affected by DBP, as shown by significantdecreases in the percentage of fertile pairs, the number oflive pups per litter, the proportion of pups born alive, and livepup weight. Only the control and high-dose F0 DBP groupswere necropsied. There were no effects on sperm param-eters in the males, although body weight was significantlydecreased (8%) and liver to body weight ratio significantlyincreased (11%). For females, liver to body weight ratio wassignificantly increased (19%) and relative uterine weightsignificantly decreased (28%), but there was no effect onestrous cycles. No treatment-related gross or histologicallesions were noted. A second generation was not evaluated.

In Lamb et al.[39], the high-dose group was sub-fertileand the middle-dose and the low-dose groups were func-tionally unaffected. Thus, the NOAEL was calculated at525 mg/kg bw/day, based on reductions in litter size and inproportions of pairs having litters. The mid- and low-dosegroups were not necropsied or evaluated for reproductivedevelopment or performance. For these reasons, the expertpanel has moderate-to-low confidence that these doses cor-rectly represent the LOAEL and NOAEL. Confidence in thequality of the data reported is high.

4.3. Mode of action

The expert panel believes that data from studies withDEHP are relevant to a consideration of the mechanism bywhich DBP causes adverse effects. It is well understoodthat DEHP produces a range of hepatic effects (induc-tion of peroxisomes; increased Cyp4A1; PCoA) including


hepatic tumors in rat and mouse studies. The inductionof these effects in rats is believed due to activation ofPPAR-alpha. In PPAR-knockout mice, administration ofDEHP does not result in the induction of hepatic effectsobserved in the wild-type control animals. In humans,PPAR-alpha is activated upstream of different enzymesfrom those noted in the rat. Recently, an IARC[56] reanal-ysis of the cancer data led them to conclude that DEHPis not classifiable as to its carcinogenicity in humans.IARC further concluded that “. . . the mechanism by whichdi(2-ethylhexyl) phthalate increases the incidence of hepato-cellular tumors in rats and mice is not relevant to humans.”

In studies with DEHP, a genetically modified strain ofmouse (the PPAR-alpha knockout mouse) cannot activatePPAR-alpha, but is susceptible to phthalate-induced devel-opmental toxicity and testicular toxicity. This mouse doesexpress PPAR-gamma in the testis which can be activatedby MEHP[57]. PPAR-gamma may conceivably play a rolein the reproductive toxicity of phthalates. PPAR-gamma hasbeen found in human hematopoetic cell populations andother organ systems including placenta and fetal tissue[58].Other members of the PPAR family (beta and gamma) havenot been extensively studied with regard to activation by ph-thalates. Finally, the guinea pig, a non-responding species tothe peroxisomal proliferating effects of DBP, is susceptibleto the testicular effects of this phthalate.

Gray et al.[55] investigated the reason for the lack oftesticular lesions in hamsters orally administered DBP andMBuP at doses exceeding those that produced testicular le-sions in rats. Using14C-labeled DBP and monobutyl ester(MBuP), it was determined that intestinal esterase activitieswere similar in the two species and that the principal metabo-lite in the rat and hamster was MBuP glucuronide[23]. How-ever, the levels of unconjugated MBuP in urine were threeto four-fold higher in the rat. Finding that the activity of tes-ticular beta-glucuronidase was significantly higher in the ratthan the hamster, the authors speculated that the testiculardamage might be associated with greater concentrations ofunconjugated MBuP, the putative toxicant.

All phthalates that cause testicular toxicity produce acommon lesion characterized by alterations in Sertoli cellultrastructure and function[59–61]. It is known that someSertoli cell functions are mediated by follicle stimulatinghormone (FSH) interaction with membrane bound receptors.Lloyd and Foster[62] demonstrated that MEHP disturbsFSH interaction with the FSH receptor. Further studies withMEHP using primary rat Sertoli cell cultures revealed thatthe monoester of DEHP inhibited FSH-stimulated cAMPaccumulation. The MEHP-induced inhibition was specificfor FSH[63].

Factors affecting increased sensitivity to phthalate-inducedtesticular toxicity in young animals were studied for DBP,DEHP, di-n-hexyl phthalate (DnHP), and dipentyl phthalate.The monoester derivatives of DBP and DEHP have beenshown to cause similar testicular effects. Sjoberg et al.[64]demonstrated that gavage treatment with DEHP resulted in

greater absorption of MEHP, and hence, a greater systemicdose to young versus mature rats. Further, in vitro studiesdid not find that FSH-stimulated cAMP accumulation andlactate secretion were age related[65]. Lloyd and Foster[62]noted that initiation of spermatogenesis was dependent onFSH interaction with the Sertoli cell in young rats, but wasnot necessary for maintenance of spermatogenesis in adults.Their experiment in Sertoli cell cultures demonstrated thatMEHP interferes with FSH interaction at the receptor leveland provided a hypothesis for increased sensitivity to tes-ticular toxicity in young animals. The panel was not ableto reach agreement that interfering with FSH signalingfunction was the accepted mode or mechanism of action.

Several studies have examined the ability of selectedphthalate esters to compete with labeled estradiol (E2) forbinding to the estrogen receptor (ER). Sources of ER pro-tein included rat uterine[66], rainbow trout hepatic cytosol[67], recombinant human ERs (rhER) overexpressed inSF9 insect cells using the baculovirus system[68,69] andrainbow trout ERs expressed in yeast. Triated E2 was usedin the tissue cytosol binding assays while a high affinityfluorescent E2 derivative was used in the rhER binding as-says. DBP exhibited no or weak activity in in vitro assaysthat measured binding of phthalates to estrogen receptors[66,67,70]. The assays did not include the addition ofesterases or lipases to metabolize DBP to its monoester.

Selected phthalate esters have been examined in a numberof in vitro gene expression assays systems. The assays haveused stably transfected cells[66], transiently transfectedcells [66,67], yeast based assays[66,70–72] and vitel-logenin induction in rainbow trout hepatocyte cultures[70].DBP was weakly active in an assay of estrogen-inducedgene expression, but its metabolite MBuP was inactive[72].There was no synergism in estrogenic response with DBPand other phthalates[72,73].

In vivo assays demonstrated that DBP does not increaseuterine wet weight or vaginal epithelial cell cornification inimmature or mature ovariectomized rats[66] and prepuber-tal mice[71]. Uterine permeability was not affected follow-ing the subcutaneous injection of DBP[73]. Malformationsin reproductive organs and effects on androgen-relatedendpoints of male rats exposed to DBP or MBuP duringprenatal development suggest antiandrogenic activity byDBP and MbuP[41,49,50,52].


5. Data summary and integration

5.1. Summary

5.1.1. Human exposureThe major use of DBP is as a coalescing aid in latex

adhesive. It is also used as a plasticizer for cellulose plasticsand as a solvent for dyes. DBP is not used as a plasticizerfor PVC plastics[1].


Several authoritative estimates of human exposure, de-scribed inSection 1, have been published since 1990. Allestimates place total DBP exposure in the general popu-lation at less than 10�g/kg bw/day and were consistent inidentifying food as the major exposure source. In additionto food, general human exposure occurs primarily throughindoor air followed by drinking water, soil, and ambient air.Infants and young children may have higher exposures thanadults, primarily because of dietary differences and possiblemouthing of DBP-containing household articles (not lim-ited to toys). Using reasonable assumptions and data fromsurveillance and food surveys, Health Canada[6] estimatedtotal exposures of 2.4, 5.0, 4.3, 2.3, and 1.9�g/kg bw/dayfor humans aged 0–0.5, 0.5–4, 5–11, 12–19, and 20–70years, respectively. Discrepancies in food exposure esti-mates may be due to inherent variability of food eaten byindividuals based on age, sex, ethnicity, time of sampling,and geographical locations.

DBP was found in infant formula, but amounts vary inter-nationally and seem to be falling[5,9]. The most recent esti-mate of DBP exposure from infant formula to a newborn inthe UK is 2.4�g/kg bw/day[5] and is the same as the HealthCanada total exposure estimate. DBP has been found in 1 of17 European children’s toys at a very low level (0.01% byweight) [10]. Use of DBP in plastic nasogastric tubing hasalso been reported[3]. Occupational exposure during phtha-late manufacture is estimated at 143�g/kg bw/day. Expo-sures in other occupational settings have not been estimated.

5.1.1.1. Utility of data to the CERHR evaluation.DBPexposures resulting from contact with various media (e.g.food, drinking water, and air) have been estimated by severalauthoritative sources. Limitations in the dataset include thefact that most of the data used in calculations were 15–20years old and may not reflect current exposure. Further, themajority of data was collected in Europe and Canada andmay not accurately reflect US patterns. Data from HealthCanada were selected for use since they provide age-basedexposure estimates.

5.1.2. General biological and toxicological data

5.1.2.1. Toxicity. The expert panel had to rely on animaltoxicity data in its evaluation of general biology and toxicity.DBP is not acutely toxic to rodents with the oral LD50 givenin grams per kilogram (g/kg) quantities. There are sufficientdata to establish that DBP in the diet is toxic to adult ratsand mice at repeated daily doses of∼350 mg/kg bw/day andhigher [14]. The liver and testes are consistently found tobe target organs with the hematopoetic system also affectedin some strains of rats and at higher doses in mice. Testicu-lar lesions were observed at doses of 720 mg/kg bw/day andhigher in adult rats[14]. DBP increases liver to body weightand kidney to body weight ratios. These effects are consis-tent with effects seen with other phthalates. Indications ofperoxisome proliferation, such as elevated levels of PCoA

oxidation, were consistently observed. The lowest repeateddose NOAEL in rats was observed in males exposed throughdiet to 142 mg/kg bw/day[13]. The corresponding NOAELin male mice was 353 mg/kg bw/day[14] (Table 24). Chroniccarcinogenicity studies were not available for review.

5.1.2.2. Toxicokinetics.There are no conclusive in vivotoxicokinetic data in humans. Orally administered DBP inrodents is rapidly hydrolyzed to the monoester, MBuP, bypancreatic lipases secreted into the small intestine. The mo-noester is rapidly absorbed from the gut, widely distributedin tissues, and is rapidly excreted in urine, mainly as a glu-curonide. No studies are available on the absorption of orallyadministered DBP in primates. Thus, it is not known whetherDBP is more poorly hydrolyzed and absorbed in the gut ofprimates compared to rats, as has been observed with otherphthalates. Rodent studies indicate there is no bioaccumula-tion of absorbed DBP or its metabolites (including testes andprostate tissue). In vitro human and rat skin were comparedfor their absorption of DBP; and human skin was foundto be much less permeable than rat skin[12]. In rats, der-mal absorption of DBP as identified by urinary excretion ofmetabolites is 10–12% of the 30–40 mg/kg dose per day[11].

Rats treated with14C-DBP on gd 14 showed concen-trations of radioactivity in placenta and fetuses that wereapproximately 65% of the levels in maternal serum. MBuPwas the major metabolite found in both maternal and em-bryonic tissues[21]. A PBPK model of the tissue distri-bution of DBP and MBuP in rats has been developed byKeys et al.[25]; the model includes diffusion limitationsand pH trapping as mechanisms of uptake of MBuP intotissue. A model has been derived to extrapolate rodent datato predicted values in humans. The model does not containparameters for estimating fetal or pediatric values.

5.1.2.3. Genetic toxicity. IPCS[3] reviewed a number ofmutagenicity and related endpoints for DBP and concludedthat the weight of the evidence indicated that DBP is notgenotoxic.

5.1.2.4. Utility of general biological and toxicological datafor a CERHR evaluation. The oral sub-chronic studies inrats and mice are adequate for the evaluation of generaltoxicity induced by DBP. Some studies were conducted ac-cording to GLP standards and relevant exposure routes wereutilized. Small group numbers, used in some studies, are oflimited concern considering the reproducibility of effects be-tween studies. Adult rodents were tested for DBP-inducedtesticular lesions.Sections 3 and 4address the studies wherethe male rodent reproductive tract was exposed to DBP dur-ing prenatal and postnatal development. The examination ofhepatic effects was adequate and included an evaluation ofperoxisome proliferation in rodents.

There are acceptable toxicokinetic data for DBP, consist-ing of absorption, distribution, metabolism, and excretion,following oral and dermal exposure in the rat. The human


Table 24Summaries of NOAELs and LOAELs and major effects in general toxicity studies

Protocol and DBP doses (mg/kg bw/day) NOAEL(mg/kg bw/day)

LOAEL (mg/kg bw/day)and effects

Major effects at higher doses

3-Month repeat-dose dietary study in Wistar rats M: 142 M: 688; F: 816 No higher doses in study6-Week-old at start of study, 10 rats/sex/group F: 162 ↑Liver and kidney weight (F)Doses – M: 0, 27, 142, 688; F: 0, 33, 162, 816[13] Peroxisomal proliferation

Histological liver changes↓Thyroid hormoneAnemia (M)No testicular lesions

13-Week repeat-dose dietary study in F344 rats M: 176 M: 359; F: 356 ↑Liver and kidney weights5−6-Week-old at start of study, 10 rats/sex/group F: 177 ↑Liver and kidney weights (M) Hepatic lesionsDoses – M: 0, 176, 359, 720, 1540, 2964; F: 0, 177,

356, 712, 1413, 2943[14]Peroxisomal proliferation Changes in liver enzyme activityAnemia (M) Peroxisomal proliferation

Testicular lesionsHypospermia↓Testes weight↓Testicular testosterone levelsAnemia (M)

13-Week repeat-dose dietary study in B6C3F1 mice6-Week-old at start of study, 10 mice/sex/group

M: 353F: none

M: 812F: 238↑Kidney weight (F) (nodose response or histologicalchanges)↑Liver weight (M)↓Body weight gain (M)

↑Kidney weight (F) (no doseresponse or histological changes)

Doses – M: 0, 163, 353, 812, 1601, 3689; F: 0, 238,486, 971, 2137, 4278[14]

↑Liver weight↓Body weight gainMild histological liver effectsNo testicular lesions

data available are of very limited utility. In vitro compar-isons of DBP metabolism suggest that effects observed inrodents are relevant to humans.

5.1.3. Developmental toxicityThe are no data on the developmental toxicity of DBP

in humans. The most complete description of effects char-acterizing key aspects of the developmental toxicity ofDBP is contained in a series of publications by Ema et al.[34–36] and Mylchreest et al.[47,49,50]. Ema et al.[42]characterized the prenatal developmental toxicity of DBP inWistar rats and subsequently demonstrated that the metabo-lite MBuP caused developmental toxicity similar to DBP.These effects were produced at approximately equimolarconcentrations. For example, a maternal and developmentNOAEL and LOAEL of 500 and 630 mg/kg bw/day (1.80and 2.27 mmol/kg bw/day), respectively, were identified forDBP following gavage of Wistar rats on gd 7–15[34].Using a similar experimental design, a maternal and devel-opmental NOAEL and LOAEL for MBuP of 250 and 500mg/kg bw/day (1.13 and 2.25 mmol/kg bw/day), respec-tively, were identified[42]. Similar fetal effects in thesestudies included increased prenatal mortality, decreasedfetal weight, and cleft palate. Dose and time dependencystudies with DBP and MBuP resulted in similar findingsand are described inSection 3.2.1.

The most complete prenatal exposure study by Ema et al.from the perspective of group size and development of themale reproductive tract established a maternal and fetal

NOAEL and LOAEL of 331 and 555 mg/kg bw/day, respec-tively, in Wistar rats fed DBP-dosed diets on gd 11–21[36].Developmental effects at higher doses (≥555 mg/kg bw/day)included decreased fetal weight, cleft palates, fused sterne-brae, reduced anogenital distance in males, and cryp-torchidism.

Studies by Mylchreest et al. looked at postnatal effectsfollowing in utero exposure to DBP[47,49,50]. CD ratswere gavaged with DBP from gd 3 to pnd 20 or gd 12–21.Delayed preputial separation and retained nipples wereobserved at doses as low as 100 mg/kg bw/day. Effectsnoted at doses of 250 mg/kg bw/day or higher were consis-tent between studies and included hypospadia, agenesis ofepididymis or seminal vesicles, cryptorchidism, decreasedanogenital distance in males, and/or a low incidence ofinterstitial adenomas. A NOAEL of 50 mg/kg bw/day wasidentified. The three studies[47,49,50] exposed animalsduring the appropriate window of development, analyzedthe tissues appropriately, and combined them with otherindices of puberty and reproductive development. The con-cordance in dose-response in[38] is good.

The role of the monoester metabolite of DBP in develop-mental toxicity was elucidated by Saillenfait et al.[21], whogavaged Sprague–Dawley rats with 500 or 1500 mg/kg ofradiolabeled DBP/kg bw/day on gd 14. They demonstratedradioactivity in placentas and embryos at levels of 21–30%of those measured in maternal plasma. The majority of theradioactivity was associated with MBuP and its glucuronide.Postnatal effects following in utero exposure to the DBP


metabolite MBuP were studied in WKA rats that were gav-aged with 300 mg MBuP/day (∼1000 mg/kg bw/day) on gd15–18[52]. Testes descent was reduced on both gd 20 andpnd 30–40. Although only one dose was administered, thefindings are consistent with those observed in DBP devel-opmental toxicity studies conducted by Ema et al.[36] andMylchreest et al.[47,49,50], thus supporting the hypothesisthat MBuP is responsible for effects associated with DBPexposure.

The hallmark of developmental toxicity in the mousefollowing oral exposure to DBP appears to be primarilysystemic toxicity and death. In a study with ICR mice ex-posed to diet containing DBP on gd 0–18, Shiota et al.[32,33] reported a 98% incidence of fetal mortality at2100 mg/kg bw/day. Fetal body weight was reduced at660 mg/kg bw/day. The authors stated that the maximumnon-embryotoxic dose was 370 mg/kg bw/day. However, theexpert panel noted that delayed ossification occurred at alldose levels, and selected the lowest dose, 80 mg/kg bw/day,as a LOAEL. These data are from groups with small samplesize and have not been replicated. In a continuous-breedingprotocol with CD-1 mice, Lamb et al.[39] observed a de-crease in the number of pups, live pups per litter, and pupweight in dams that consumed a dose of 1750 mg/kg bw/dayin the diet. The developmental NOAEL was identifiedas 525 mg/kg bw/day. Effects of in utero and lactationalexposure to DBP were studied in B6C3F1 mice whereMarsman et al.[14] reported that length of gestation wasincreased at 2500 ppm (454 mg/kg bw/day) and higher. Atotal of 75 and 95% percent of litters were lost at 10,000(1816 mg/kg bw/day) and 20,000 (3632 mg/kg bw/day) ppm.Decreases were observed in litter size and pup body weightsat 2500, 7500, and 10,000 ppm. The expert panel is notconfident that these three studies fully assessed DBP de-velopmental toxicity, including reproductive function dueto limitations in study design that include small group size,failure to perform necropsies in critical dose groups, andfailure to assess appropriate landmarks of maturation.

NOAELs and LOAELs for the key developmental toxicitystudies for DBP are listed inTable 25. Ema et al.[36] ex-amined the most sensitive prenatal endpoints and allows fora comparison between maternal and developmental toxicity.The study of Ema et al.[34] DBP was also included to allowcomparison with the study of its metabolite, MBuP[42] thatwas evaluated according to the same protocol. Mylchreestet al. [50] study is considered key because it examined themost sensitive endpoints at the lowest doses (Table 25).

5.1.3.1. Utility of developmental toxicity data for a CERHRevaluation. The data in rats are adequate for an assessmentof developmental toxicity. Studies examined effects follow-ing dosing of dams through portions of or the entire periodof pregnancy. Fetuses were evaluated for prenatal mal-formations and postnatal effects. Evaluations included anexamination of reproductive organs and androgen-regulatedendpoints, which are thought to be the most sensitive

indicators of phthalate-induced toxicity. Prenatal effectsfollowing prenatal exposure to MBuP were also examined.A second rodent species, the mouse, was examined in aprenatal exposure and effect study. Based on the limitedparameters examined in the mouse it is not possible tocompare sensitivity in rats and mice.

5.1.4. Reproductive toxicity

5.1.4.1. Human data. The relationship of human spermdensity and total number of sperm to DBP concentrationwas studied in a group of unselected college students[53] Anegative correlation was found between sperm density andDBP concentration in the cellular fraction of ejaculate ofnon-occupationally exposed subjects, but the data are of littlevalue to the expert panel due to the insufficient evidence fora causal relationship of sperm characteristics to DBP levels.

5.1.4.2. Experimental animal studies.Reproductive stud-ies have been performed primarily in the rat and, to a lesserextent, the mouse. There are single reports of studies inguinea pigs and hamsters[55]. Collectively, the data aresufficient to show that oral exposure to DBP can causereproductive toxicity in male rats, mice, and guinea pigs.In contrast, the hamster failed to show testicular effects.Data that characterize effects on female reproduction arenot as complete and detailed interpretation is therefore lesscertain. The data do indicate a decrease in female fertilityin mice and rats.

5.1.4.2.1. Females.The Lamb et al.[39] data from acontinuous breeding study in mice clearly show adult fe-male functional effects at 1750 mg/kg bw/day. The limitedexamination of the lower dose groups (necropsies were notperformed) precludes the setting of a reliable NOAEL. Thecontinuous-breeding study by Wine et al.[38] in F344 ratsdid not show specific deficits in female parameters; how-ever, the data do not rule out that decreases in litter size atall doses may have a female component. In contrast, Grayet al. [41] reported that fertility in female Long Evans ratswas reduced following treatment with 500 mg/kg bw/dayfrom weaning through puberty, gestation, and lactation.The effect was apparently due to an increase in mid-termabortions. The F1 female pups in this study were alsomated and experienced a reduction in fecundity at doses of250 mg/kg bw/day and higher. Thus, clear effects on femalereproduction are seen in rats at doses of 250 mg/kg bw/day(LOAEL) and in mice at higher doses. NOAELs can not beestablished with any confidence (Table 26).

5.1.4.2.2. Males. Data from the continuous breeding study[38] clearly show functional and structural reproductive ef-fects in male Sprague–Dawley rats. In the F0 generationthere was clear indication that DBP, when administered inthe diet, affected total number of pups per litter in all treatedgroups. The F1 high-dose group had malformations of the


reproductive tract and a high rate of infertility. Dose relatedincreases in seminiferous tubular degeneration were seenat the 256 and 509 mg/kg bw/day doses. The LOAEL was52–80 mg/kg bw/day based on reductions in F0 litter size.Thus, a reproductive NOAEL was not established.

A delay in preputial separation was observed in LongEvans rats exposed at the lowest dose of 250 mg/kg bw/dayby gavage from the time they were weaned until the lit-ters they sired were weaned in[41]. At higher doses(500–1000 mg/kg bw/day), reductions in sperm counts andfertility and testicular lesions were also observed. The F1offspring exposed to DBP only during gestation and lacta-tion experienced a reduction in sperm counts.

Three studies by Mylchreest et al.[47,49,50], presentedin Sections 3.2.2 and 5.1.3of this document, indicated thatthe range of male structural abnormalities in Wine et al.[38] study could be reproduced with a much shorter dosingregime. Mylchreest et al.[49,50] also detected a significantincrease in testicular Leydig cell hyperplasia and a low in-cidence of Leydig cell adenomas in∼3-month-old animalsfollowing only a late gestational exposure (gd 12–21) of500 mg/kg bw/day. Wine et al.[38] dosed for 14 weeks withDBP in the diet, whereas Mylchreest et al.[50] exposedpregnant rats by gavage during gd 12–21. A NOAEL wasestablished at 50 mg/kg bw/day in[50].

The existing data show consistent effects (testicularpathology, reduced sperm numbers, effects on reproductivetract development), and are sufficient to conclude that DBPis a reproductive and developmental toxicant in male ratsat doses of 100 mg/kg bw/day and higher. Treatment ofrat weanlings with 250 mg/kg bw/day resulted in delayedpuberty and doses of 500 mg/kg bw/day induced testicularlesions. In general toxicity studies (Section 2.1.2), testicularlesions were observed in adult rats (6 weeks old) treatedwith 720 mg/kg bw/day, but not in adult mice treated withup to 3689 mg/kg bw/day for 3 months[14]. Histologicalchanges in testes of 4–6-week-old mice and guinea pigs ofa similar nature have also been observed following admin-istration of a single high dose (2000 mg/kg bw/day) for 7–9days, but hamsters were unaffected. The overall effects onthe testes indicate an age sensitivity with fetal sensitivity>pubertal sensitivity> adult sensitivity in male rats to theaction of DBP.

The responses that occur at the lowest doses appear toinvolve the development of the reproductive system. Theseresponses were seen with some consistency in the studiesby Mylchreest et al.[47,49]and Wine et al.[38]. The reportby Reel et al.[40] and the paper by Lamb et al.[39] did notreport on measures of reproductive system development.However, they are consistent with the Mylchreest et al. andWine et al. papers in that they show reproductive toxicityunder oral (dietary) exposure, and do so in a second species,the mouse (Table 26).

5.1.4.2.3. Mode of action.Gray et al.[55] investigated thereason for the lack of testicular lesions in hamsters admin-

istered DBP and MBuP orally at doses exceeding those thatproduced testicular lesions in rats. Using14C-labeled DBPand MBuP, it was determined that intestinal esterase activ-ities were similar in the two species and that the principalmetabolite in the rat and hamster was MBuP glucuronide[23]. However, the levels of unconjugated MBuP in urinewere 3–4-fold higher in the rat. Finding that the activity oftesticular beta-glucuronidase was significantly higher in therat than the hamster, the authors speculated that the testicu-lar damage might be associated with greater concentrationsof MBuP, the putative toxicant.

All phthalates that cause testicular toxicity produce acommon lesion characterized by alterations in Sertoli cellultrastructure and function[59–61]. It is known that someSertoli cell functions are mediated by FSH interactionwith membrane bound receptors. Lloyd and Foster[62]demonstrated that MEHP disturbs FSH interaction withthe FSH receptor. Further studies with MEHP using pri-mary rat Sertoli cell cultures revealed that the monoesterof DEHP inhibited FSH-stimulated cAMP accumula-tion. The MEHP-induced inhibition was specific for FSH[62].

Factors affecting increased sensitivity to phthalate-inducedtesticular toxicity in young animals were studied for DBP,DEHP, DnHP, and dipentyl phthalate. The monoesterderivatives of DBP and DEHP have been shown to causesimilar testicular effects. Sjoberg et al.[64] demonstratedthat gavage treatment with DEHP resulted in greater ab-sorption of MEHP, and hence, a greater systemic dose toyoung versus mature rats. Further, in vitro studies did notfind that FSH-stimulated cAMP accumulation and lactatesecretion were age related[65]. Lloyd and Foster[62] notedthat initiation of spermatogenesis was dependent on FSHinteraction with the Sertoli cell in young rats but was notnecessary for maintenance of spermatogenesis in adults.Their experiment in Sertoli cell cultures demonstratedthat MEHP interferes with FSH interaction at the receptorlevel and provided a hypothesis for increased sensitivityto testicular toxicity in young animals. The panel was notable to reach agreement that interfering with the FSH-signaling function was the accepted mode or mechanism ofaction.

The expert panel believes that data from studies withDEHP are relevant to a consideration of the mechanism bywhich DBP causes adverse effects. It is well understoodthat DEHP produces a range of hepatic effects (inductionof peroxisomes; increased Cyp4A1; PCoA) including hep-atic tumors in rat and mouse studies. The induction of theseeffects in rats is believed due to activation of PPAR-alpha.In PPAR-knockout mice, administration of DEHP does notresult in the induction of hepatic effects observed in thewild-type control animals. In humans, PPAR-alpha is acti-vated upstream of different enzymes from those noted in therat. Recently, an IARC[56] reanalysis of the cancer dataled them to conclude that DEHP is not classifiable as toits carcinogenicity in humans. IARC further concluded that


“ . . . the mechanism by which di (2-ethylhexyl) phthalateincreases the incidence of hepatocellular tumors in rats andmice is not relevant to humans.”

In studies with DEHP, a genetically modified strain ofmouse (the PPAR-alpha knockout mouse) cannot activatePPAR-alpha, but is susceptible to phthalate-induced devel-opmental toxicity and testicular toxicity. This mouse doesexpress PPAR-gamma in the testis which can be activatedby MEHP[57]. PPAR-gamma may conceivably play a rolein the reproductive toxicity of phthalates. PPAR-gamma hasbeen found in human hematopoetic cell populations andother organ systems including placenta and fetal tissue[58].Other members of the PPAR family (beta and gamma) havenot been extensively studied with regard to activation by ph-thalates.

Finally, the guinea pig, a non-responding species to theperoxisomal-proliferating effects of DBP, is susceptible tothe testicular effects of this phthalate.

Imajima et al. [52] suggests that the active metabo-lite for reproductive effects due to gestational exposure isMBuP. This pattern of effects induced in rodents by lategestational exposure (gd 12–21) is ‘anti-androgenic’ in thatflutamide mimics these effects[49]; however, DBP/MBuPdoes not bind to the androgen receptor[74]. In puber-tal and adult rodents, the Sertoli cell is the likely cellu-lar target for testicular injury mediated by the monoester[65,75].

DBP exhibited no or weak activity in in vitro assaysthat assess estrogenicity[66,67,70,72]. The assays did notinclude the addition of esterases or lipases to metabolizeDBP to its monoester. However, the DBP metabolite MBuPwas determined to be inactive in one assay[72]. There wasno synergism in estrogenic response with DBP and otherphthalates[72,73]. DBP was inactive in rodent in vivo as-says that measure endpoints such as increases in uterinewet weight, vaginal epithelial cell cornification, or uterinepermeability[66,71,73]. Malformations in reproductive or-gans and effects on androgen-mediated endpoints in malerats exposed to DBP or MBuP during prenatal develop-ment suggest antiandrogenic activity by DBP and MbuP[41,49,50,52].

5.1.4.3. Utility of reproductive data to the CERHR eval-uation. The data in rats are adequate for an assessmentof reproductive toxicity as several studies are availablethat evaluate both structure and reproductive function.Trans-generational effects were examined in many of thestudies. Animals were treated during gestational develop-ment, during lactation, and at weaning, thus ensuring thatthe most sensitive age for reproductive effects was assessed.The evaluation included androgen-regulated endpoints thatare believed to be the most sensitive indicators of DBPeffects. Reproductive organs were preserved in Bouin’s fix-ative, a method that reduces histological artifacts. Althoughstudies in other species are not as detailed, they do allowfor limited comparisons of interspecies sensitivity.

5.2. Integrated evaluation

DBP is used as a coalescing aid in latex adhesives, as aplasticizer in cellulose plastics, and as a solvent for dyes.General human exposure occurs primarily through food.All estimates place total DBP exposure in the general pop-ulation at less than 10�g/kg bw/day. Although infants andyoung children may have higher exposures than adults,primarily because of different dietary patterns, estimates oftheir exposure remain at approximately 10�g/kg bw/day,with the possible exception of non-dietary intake throughmouthing of phthalate-containing objects. Workplace ex-posure at phthalate production facilities is estimated to bebelow 143�g/kg bw/day. Exposure levels during incorpo-ration of DBP in plastics are not known.

Following oral exposure to rodents and humans, DBP isquickly metabolized in the small intestine to mono-n-butylphthalate, MbuP, andn-butyl alcohol. Several investigatorshave postulated that it is the monoester that is of toxico-logical interest. The panel finds logic and data to supportthis view. Absorption of the monoester into blood occursin both rats and humans. Although data for DBP is notavailable for humans or primates, it is reasonable to assumethat MBuP would be rapidly glucuronidated and excreted inthe urine in a manner analogous to DEHP in humans. Thetoxicokinetic data indicate that no tissue bioaccumulationwould be expected via the oral or dermal route.

There are no data on the developmental or reproductivetoxicity of DBP in humans. There are data in rats andmice that show oral exposure to DBP causes developmentaltoxicity. The developing male reproductive system is mostsensitive to the formation of structural and functional abnor-malities, with effects seen in rats whose mothers were ex-posed to 100 mg/kg bw/day during pregnancy. The NOAELfor male reproductive system developmental effects in ratsis 50 mg/kg bw/day. Breeding studies provide a good indi-cation of the potential for adverse functional reproductiveeffects from DBP exposure. Moreover, it is apparent thatDBP testicular exposure late in gestation can induce Ley-dig cell hyperplasia and a low incidence of Leydig celladenoma. Traditional teratogenicity protocols that evaluatefetuses just prior to birth were not effective in detectingthese effects on the developing male reproductive system.While a series of three recent studies have replicated andcharacterized the male reproductive system effects in rats,studies of similar design have not been performed in otherspecies. The panel is confident that these studies in ratscorrectly characterize the effects based on replication, gooddose response, and full reporting of study results. As a de-fault assumption, these data in rats are assumed relevant toa prediction of hazard to humans.

The expert panel notes that the male reproductive sys-tem is a sensitive target organ for effects in rodent studieswhere exposure is confined to the adult phase of life. Datain several species, including rat, mouse, and guinea pig,show such effects. The panel also notes that studies in


the hamster, although limited, do not show effects on thetestes.

There are indications that oral exposure of females duringthe adult phase of life impairs functional reproductive per-formance in rats at doses of 250 mg/kg bw/day and higher.There is also a report that exposure to similar doses duringgestation and nursing may impair fertility in female off-spring. However, the data are not of the scope and qualityfor the expert panel to confidently characterize these effects.

Data indicate that the monoester of DBP, MBuP, is theprincipal toxicant. Studies suggest that an antiandrogenicmechanism appears to be responsible for the most sensitiveendpoints observed in developing males rats (e.g. anogeni-tal distance, nipple retention, preputial separation). It is notcurrently known whether the target for DBP is similar ordifferent for gestational versus postnatal exposures.

The panel is aware of studies performed at CDC usingurine from human subjects. Results of these studies weregiven in an oral presentation in Copenhagen, Denmark,in May, 2000. MBuP values in the urine of women ofchild-bearing age were among the higher values. Such data,when published, should serve to improve our ability toassess phthalate exposure in the general populations.

5.3. Expert panel conclusions

DBP is used as a coalescing aid in latex adhesives, as aplasticizer in cellulose plastics, and as a solvent for dyes. Thebest estimate for exposures from all sources to the generalpopulation is 2–10�g/kg bw/day. There is significant uncer-tainty in the exposure database based on the age of many ofthe values/studies. The expert panel has high confidence inthe available studies to characterize reproductive and devel-opmental toxicity based upon a strong database containingstudies in multiple species using conventional and investiga-tive study designs. When administered via the oral route,DBP elicits malformations of the male rat reproductive tractvia a disturbance of the androgen status: a mode of actionrelevant for human reproductive development. This an-tiandrogenic mechanism occurs via effects on testosteronebiosynthesis and not androgen receptor antagonism. DBP isa testicular toxicant in three species of young adult labora-tory animals in high dose (>1000 mg/kg bw/day), sub-acuteoral exposure studies. In the rat, there is a life-stage sensi-tivity for testicular toxicity with the fetus most sensitive, pu-bertal less sensitive, and adult least sensitive. Adult femalefunctional reproductive toxicity (decreases in fertility) hasbeen noted in rats; however, the data do not permit confidentcharacterization of dose effects below 250 mg/kg bw/day.The expert panel has negligible concern for adult reproduc-tive toxicity.

DBP is developmentally toxic to both rats and mice bythe oral routes; it induces structural malformations. A con-fident NOAEL of 50 mg/kg bw/day by the oral route hasbeen established in the rat. Data from which to confidentlyestablish a LOAEL/NOAEL in the mouse are uncertain. The

expert panel has minimal concern about effects to humandevelopment and development of the reproductive systemfrom current estimated exposure to DBP. A modified dietarymultigeneration study is available, but did not establish aNOAEL. The LOAEL (M:52 F:80 mg/kg bw/day) is basedon decreases in litter size and pup weight.

5.4. Critical data needs

The multigeneration study for DBP in rodents, withsupport from other studies that incorporated more modernendpoints including developmental landmarks, indicatesno immediate data gaps. The potential effects of DBP onfemale rats warrant further investigation.

Although there are no critical data needs, studies in thefollowing areas would increase understanding about re-productive and developmental effects that occur followingDBP exposures. There is a need to extend the current PBPKmodel for DBP to include parameters for pregnant womenand their fetuses. There is a need to find out how broad ornarrow the window of prenatal exposure is that results inpostnatal male effects. The known current window in rats,12–20 days, is still quite wide from a rodent ontogenesisperspective. Greater precision as to size of the windowof sensitivity may be relevant to estimating the temporalbounds of human sensitivity.

Much of the recent focus on reproductive toxicity of ph-thalate esters has focused on the ability of certain esters toinduce effects on reproductive development. Significant pri-mate data exist to support the view that the high blood levelsof monoester necessary to achieve adult testicular toxicity inrodents will not occur in humans. Appropriate exposure tomonoester in blood from diester exposure could be achievedsuch that experiments could be conducted in primates toelucidate species sensitivity for equivalent exposures. Thiswould require exposure of pregnant animals during the crit-ical window of development of the reproductive system forthe species studied, followed by an examination of repro-ductive development in the resulting offspring. Such a studywould indicate if there is a species sensitivity. In the absenceof such a study, the rodent data must be considered relevantand critical for human risk examinations.

Acknowledgments

This work was supported by NIEHS Contract ES-85425at Sciences International Inc., Alexandria, VA, USA; JohnMoore (Principal Scientist), Annette Iannucci (Toxicolo-gist), and Ann Walker (Information Specialist and TechnicalEditor).

References

[1] CMA. Comments of the Chemical Manufacturers Associationphthalate esters panel in response to request for public input on


seven phthalate esters. FR Doc. 99-9484. Washington, DC: ChemicalManufacturers Association, 1999.

[2] Staples CA, Peterson DR, Parkerton TF, Adams WJ. The environ-mental fate of phthalate esters: A literature review. Chemosphere1997;35:667–749.

[3] IPCS. Environmental health criteria 189 Di-n-butyl phthalate, ISBN92 4 157189 6. Geneva: Health Organization, 1997.

[4] MAFF. Food surveillance information sheet – phthalates in food.Joint Food Safety And Standards Group, vol. 1999: MAFF, UK1996, 9p.

[5] MAFF. Food surveillance information sheet – phthalates in infantformulae – follow-up survey. Joint Food Safety and Standards Group,vol. 1999: MAFF, UK 1998, 13p.

[6] Chan PKL, Meek ME. Di-n-butyl phthalate evaluation of risks tohealth from environmental exposure in Canada. J Environ Sci Health1994;12:257–68.

[7] ATSDR AfTSaDR-di-n-butylphthalate. Toxicological Profile. Pre-pared by Life Systems Inc., under subcontract to: Clement AssociatesInc., US Department of Health and Human Services, Public HealthService, 1990.

[8] MAFF. Phthalates in infant formulae. Joint food safety and standardsgroup food surveillance information sheet, vol. 1999: MAFF, UK,1996, 7p.

[9] DHHS. Phthalates in infant formula – assignment summary. PublicHealth Service, 1996.

[10] Rastogi SC. Gas chromatographic analysis of phthalate esters inplastic toys. Chromatographia 1998;47:724–6.

[11] Elsisi AE, Carter DE, Sipes IG. Dermal absorption of phthalatediesters in rats. Fundam Appl Toxicol 1989;12:70–7.

[12] Scott RC, Dugard PH, Ramsey JD, Rhodes C. In vitro absorptionof someo-phthalate diesters through human and rat skin. EnvironHealth Perspect 1987;74:223–7.

[13] BASF. Study on the oral toxicity of dibutyl phthalate in Wistarrats. Administration via the diet over 3 months. Eastman–KodakCompany, 31S0449//89020, 1992.

[14] Marsman DS. NTP technical report on toxicity studies of dibutylphthalate (CAS no. 84-74-2) administered in feed to F344 rats andB6C3F1 mice NIH Publication 95-3353. Research Triangle Park:National Toxicology Program, 1995.

[15] Walseth F, Nilsen OG. Phthalate esters. Part II. Effects of inhaleddibutylphthalate oncytochrome P-450 mediated in rat liver and lung.Arch Toxicol 1984;55:132–6.

[16] Mes J, Campbell DS. Extraction efficiency of polychlorinatedbiphenyls, organochlorine pesticides and phthalate esters from humanadipose tissue. Bull Environ Contam Toxicol 1976;16:53–60.

[17] Tanaka A, Matsumoto A, Yamaha T. Biochemical studies on phthalicesters. Part III. Metabolism of dibutyl phthalate (DBP) in animals.Toxicology 1978;9:109–23.

[18] Lake BG, Phillips JC, Linnell JC, Gangolli SD. The in vitrohydrolysis of some phthalate diesters by hepatic and intestinal prepa-rations from various species. Toxicol Appl Pharmcaol 1977;39:239–48.

[19] Rowland IR, Cottrell RC, Phillips JC. Hydrolysis of phthalate estersby the gastro-intestinal contents of the rat. Food Cosmet Toxicol1977;15:17–21.

[20] Williams DT, Blanchfield BJ. The retention, distribution, excretion,and metabolism of dibutyl phthalate 7 sup 1sup 4C in the rat. JAgric Food Chem 1975;23:854–8.

[21] Saillenfait AM, Payan JP, Fabry JP, Beydon D, Langonne I, GallissotF, et al. Assessment of the developmental toxicity, metabolism, andplacental transfer of di-n-butyl phthalate administered to pregnantrats. Toxicol Sci 1998;45:212–24.

[22] Okada S, Tamemasa O. Distribution of metabolism of di-(n-butyl)-phthalate in mice and its interaction with nucleic acids and proteins.Yakugaku Zasshi 1978;98:1229–35.

[23] Foster PMD, Foster JR, Cook MW, Thomas LV, Gangolli SD.Changes in ultrastructure and cytochemical localization of zinc in rat

testis following the administration of di-n-pentyl phthalate. ToxicolAppl Pharmacol 1982;63:120–32.

[24] Astill BD. Metabolism of DEHP: effects of prefeeding and dosevariation, and comparative studies in rodents and the cynomolgusmonkey (CMA studies). Drug Metab Rev 1989;21:33–53.

[25] Keys DA, Wallace DG, Kepler TB, Conolly RB. Quantitativeevaluation of alternative mechanisms of blood disposition ofdi(n-butyl) phthlate and mono(n-butyl) phthalate in rats. Toxicol Sci2000;53:173–84.

[26] Keys DA, Wallace DG, Kepler TB, Conolly RB. Quantitativeevaluation of alternative mechanisms of blood and testes dispositionof di (2-ethylhexyl) phthalate and mono(2-ethylhexyl) phthalate inrats. Toxicol Sci 1999;49:172–85.

[27] Di Carlo F. Structure–activity relationships (SAR) and structure–metabolism relationships (SMR) affecting the teratogenicity ofcarboxylic acids. Drug Metab Rev 1990;22:441–9.

[28] Woodward KN, Smith AM, Mariscotti SP, Tomlinson NJ. Reviewof the toxicity of the esters ofo-phthalic acid (phthalate esters).London, UK: Health and Safety Executive, 1986.

[29] Douglas GR, Hugenholtz AP, Blakey DH. Genetic toxicology ofphthalate esters: mutagenic and other genotoxic effects. EnvironHealth Perspect 1986;65:255–62.

[30] Barber E, Cifone M, Rundell J, Przygoda R, Astill B, Moran E, etal. Results of the L5178Y mouse lymphoma assay and the Balb/3t3cell in vitro transformation assay for eight phthalate esters. J ApplToxicol 2000;20:69–80.

[31] Zeiger E, Haworth S, Mortelmans K, Speck W. Mutagenicity testingof di(2-ethylhexyl)phthalate and related chemicals inSalmonella.Environ Mutagen 1985;7:213–32.

[32] Shiota K, Chou MJ, Nishimura H. Embryotoxic effects ofdi-2-ethylhexyl phthalate (DEHP) and di-n-butyl phthalate (DBP) inmice. Environ Res 1980;22:245–53.

[33] Shiota K, Nishimura H. Teratogenicity of di (2-ethylhexyl) phthalate(DEHP) and di-n-butyl phthalate (DBP) in mice. Environ HealthPerspect 1982;45:65–70.

[34] Ema M, Amano H, Itami T, Kawasaki H. Teratogenic evaluation ofdi-n-butyl phthalate in rats. Toxicol Lett 1993;69:197–203.

[35] Ema M, Amano H, Ogawa Y. Characterization of the developmentaltoxicity of di-n-butyl phthalate in rats. Toxicology 1994;86:163–74.

[36] Ema M, Miyawaki E, Kawashima K. Further evaluation ofdevelopmental toxicity of di-n-butyl phthalate following adminis-tration during late pregnancy in rats. Toxicol Lett 1998;87–93.

[37] Ema M, Kurosaka R, Amano H, Ogawa Y. Comparative develop-mental toxicity ofn-butyl benzyl phthalate and di-n-butyl phthalatein rats. Arch Environ Contam Toxicol 1995;28:223–8.

[38] Wine R, H L, Barnes L-H, Gulati DK, Chapin RE. Reproductivetoxicity of di-n-butyl phthalate in a continuous breeding protocol inSprague–Dawley rats. Environ Health Perspect 1997;105:102–7.

[39] Lamb IV JC, Reproductive effects of four phthalic acid esters inthe mouse . Toxicol Appl Pharmacol 1987;88:255–69.

[40] Reel JR, Lawton AD, Feldman DB, Lamb JC. Di(N-butyl)phthalate: reproduction and fertility assessment in CD-1 mice whenadministered in the feed. NTP-84-411: National Toxicology Program,National Institute of Environmental Health Sciences, 1984.

[41] Gray Jr LE, Wolf C, Lambright C, Mann P, Price M, CooperRL, et al. Administration of potentially antiandrogenic pesticides(procymidone, linuron, iprodione, chlozolinate,p,p′-DDE, and keto-conazole) and toxic substances (dibutyl- and diethylhexylphthalate, PCB 169, and ethane dimethane sulphonate) duringsexual differentiation produces diverse profiles of reproductivemalformations in the male rat. Toxicol Ind Health 1999;15:94–118.

[42] Ema M, Kurosaka R, Amano H, Ogawa Y. Developmentaltoxicity evaluation of mono-n-butyl phthalate in rats. Toxicol Lett1995;78:101–6.

[43] Ema M, Kurosaka R, Harazono A, Amano H, Ogawa Y. Phasespecificity of developmental toxicity after oral administration ofmono-n-butyl phthalate in rats. Arch Environ Contam Toxicol1996;31:170–6.


[44] Tyl RW, Price CJ, Marr MC, Kimmel CA. Developmental toxicityevaluation of dietary di(2-ethylhexyl)phthalate in Fischer 344 ratsand CD-1 mice. Fundam Appl Toxicol 1988;10:395–412.

[45] Price CJ, Tyl RW, Marr MC, Myers CB, Sadler BM, Kimmel CA.Reproduction and fertility evaluation of diethylhexyl phthalate (CASno. 117-81-7) in CD-1 mice exposed during gestation. ResearchTriangle Park, NC, USA: National Toxicology Program, 1988.

[46] Price CJ, Tyl RW, Marr MC, Sadler BM, Kimmel CA. Reproductionand fertility evaluation of diethylhexyl phthalate (CAS no. 117-81-7)in Fischer 344 rats exposed during gestation NTP 86-309. ResearchTriangle Park, NC, USA: National Toxicology Program, 1986.

[47] Mylchreest E, Cattley RC, Foster PM. Male reproductive tractmalformations in rats following gestational and lactational exposureto di(n-butyl) phthalate: an antiandrogenic mechanism? Toxicol Sci1998;43:47–60.

[48] Chapin R, Sloane R, Haseman J. Reproductive endpoints in generaltoxicity studies: are they predictive? Reprod Toxicol. 1998;489–94.

[49] Mylchreest E, Sar M, Cattley RC, Foster PMD. Disruption ofandrogen-regulated male reproductive development by di(n-butyl)phthalate during late gestation in rats is different from flutamide.Toxicol Appl Pharmacol 1999;156:81–95.

[50] Mylchreest E, Wallace DG, Cattley RC, Foster P. Dose-dependentalterations in androgen-regulated male reproductive development inrats exposed to di(n-butyl)phthalate during late gestation. Toxicol Sci2000;55(1):143–51.

[51] Higuchi TT, Kane CM, Palmer JS, Veeramachaneni DNR. Develop-mental effects of digbutyl phthalate in frogs and rabbits. Abstract193 at the Society of the Study of Reproduction Meeting, 1999.

[52] Imajima T, Shono T, Zakaria O, Suita S. Prenatal phthalate causescryptorchidism postnatally by inducing transabdominal ascent of thetestis in fetal rats. J Pediatr Surg 1997;32:18–21.

[53] Murature DA, Tang SY, Steinhardt G, Dougherty RC. Phthalateesters and semen quality parameters. Biomed Environ Mass Spectrom1987;14:473–7.

[54] Cater BR, Cook MW, Gangolli SD, Grasso P. Studies on dibutylphthalate-induced testicular atrophy in the rat: effect on zincmetabolism. Toxicol Appl Pharmacol 1977;41:609–18.

[55] Gray T, Rowland I, Foster P, Gangolli S. Species differences in thetesticular toxicity of phthalate esters. Toxicol Lett 1982;11:141–7.

[56] IARC monographs on the evaluation of carcinogenic risks to humans:some industrial chemicals, vol. 77, 2000.

[57] Maloney EK, Waxman DJ. Trans-activation of PPAR? and PPAR?by structurally diverse environmental chemicals. Toxicol ApplPharmacol 1999;161:209–18.

[58] Greene ME, Blumberg B, McBride OW, Yi HF, Kronquist K,Kwan K, et al. Isolation of the human peroxisome proliferatoeactivated receptor gamma cDNA: expression in hematopoetic cellsand chromosomal mapping. Gene Expression 1995;4:281–95.

[59] Gray TJ, Butterworth KR. Testicular atrophy produced by phthalateesters. Arch Toxicol 1980;4:452–5.

[60] Creasy DM, Beech LM, Gray TJB, Butler WH. The ultrastructuraleffects of di-n-pentyl phthalate on the testis of the mature rat. ExpMol Pathol 1987;46:357–71.

[61] Creasy DM, Foster JR, Foster PMD. The morphological developmentof di-n-pentyl phthalate induced testicular atrophy in the rat. J Pathol1993;139:309–21.

[62] Lloyd SC, Foster PM. Effect of mono-(2-ethylhexyl) phthalate onfollicle-stimulating hormone responsiveness of cultured rat Sertolicells. Toxicol Appl Pharmacol 1988;95:484–9.

[63] Heindel JJ, Chapin RE. Inhibition of FSH-stimulated cAMPaccumulation by mono(2-ethylhexyl) phthalate in primary rat Sertolicell cultures. Toxicol Appl Pharmacol 1989;97:377–85.

[64] Sjoberg P, Bondesson U, Kjellen L, Linquist NG, Montin G,Ploen L. Kinetics of di(2-ethylhexyl) phthalate in immature andmature rats and effect on testis. Acta Pharmacol Toxicol 1985;56:30–7.

[65] Heindel JJ, Powell CJ. Phthalate ester effects on rat Sertoli cellfunction in vitro: effects of phthalate side chain and age of animal.Toxicol Appl Pharmacol 1992;115:116–23.

[66] Zacharewski TR, Meek MD, Clemons JH, Wu ZF, FieldenMR, Matthews JB. Examination of the in vitro and in vivoestrogenic activities of eight commercial phthalate esters. ToxicolSci 1998;46:282–93.

[67] Jobling S, Reynolds T, White R, Parker MG, Sumpter JP. Avariety of environmentally persistent chemicals, including somephthalate plasticizers, are weakly estrogenic. Environ Health Perspect1995;103:582–7.

[68] Bolger R, Wiese TE, Ervin K, Nestich S, Checovich W. Rapidscreening of environmental chemicals for estrogen receptor bindingcapacity. Environ Health Perspect 1998;106:551–7.

[69] Nakai M, Tabira Y, Asa D, Yakabe Y, Shimyozu T, Noguchi M,et al. Binding characteristics of dialkyl phthalates for the estrogenreceptor. Biochem Biophys Res Commun 1999;254:311–4.

[70] Petit F, Le Goff P, Cravedi JP, Valotaire Y, Pakdel F. Twocomplementary bioassays for screening the estrogenic potency ofxenobiotics: recombinant yeast for trout estrogen receptor and trouphepatocyte cultures. J Mol Endocrinol 1997;19:321–35.

[71] Coldham NG, Dave M, Sivapathasundaram S, McDonnell DP, ConnorC, Sauer MJ. Evaluation of a recombinant yeast cell estrogenscreening assay. Environ Health Perspect 1997;105:734–42.

[72] Harris CA, Henttu P, Parker MG, Sumpter JP. The estrogenic activityof phthalate esters in vitro. Environ Health Perspect 1997;105:802–11.

[73] Mulligan SR, Balasubramanian AV, Kalita JC. Relative potency ofxenobiotic estrogens in an acute in vivo mammalian assay. EnvironHealth Perspect 1998;106:23–6.

[74] Foster P. Personal communication with J. Moore 2000.[75] Gray TJ, Beamand JA. Effects of some pthalate esters and other

testicular toxins on primary cultures of testicular cells. Food ChemToxicol 1984;22:123–31.

NTP Center for the Evaluation of Risks to Human Reproduction: phthalates expert panel report on the...

Documents

Transcript of NTP Center for the Evaluation of Risks to Human Reproduction: phthalates expert panel report on the...