toksikologi paraben

The Paraben DebateHosted by the Organic Pharmacy, London, 8th October 2008

Dr. Philippa DarbreSchool of Biological Sciences - The University of Reading

JOURNAL OF APPLIED TOXICOLOGYJ. Appl. Toxicol. 2008Published online in Wiley InterScience(www.interscience.wiley.com) DOI: 10.1002/jat.1358

Copyright 2008 John Wiley & Sons, Ltd.

John Wiley & Sons, Ltd.ReviewParaben esters: review of recent studies of endocrine toxicity, absorption, esterase and human exposure, and discussion of potential human health risksParaben esters: review of recent 3endocrine toxicityPhilippa D. Darbre1 and Philip W. Harvey2

1 School of Biological Sciences, The University of Reading, Reading RG6 6AJ, UK2 Covance Laboratories, Dept Toxicology, Otley Road, Harrogate, North Yorkshire HG3 1PY, UK

Received 24 January 2008; Revised 17 April 2008; Accepted 17 April 2008

ABSTRACT: This toxicology update reviews research over the past four years since publication in 2004 of the rst mea-surement of intact esters of p-hydroxybenzoic acid (parabens) in human breast cancer tissues, and the suggestion that theirpresence in the human body might originate from topical application of bodycare cosmetics. The presence of intact parabenesters in human body tissues has now been conrmed by independent measurements in human urine, and the ability ofparabens to penetrate human skin intact without breakdown by esterases and to be absorbed systemically has been demon-strated through studies not only in vitro but also in vivo using healthy human subjects. Using a wide variety of assay systemsin vitro and in vivo, the oestrogen agonist properties of parabens together with their common metabolite (p-hydroxybenzoicacid) have been extensively documented, and, in addition, the parabens have now also been shown to possess androgenantagonist activity, to act as inhibitors of sulfotransferase enzymes and to possess genotoxic activity. With the continued useof parabens in the majority of bodycare cosmetics, there is a need to carry out detailed evaluation of the potential for parabens,together with other oestrogenic and genotoxic co-formulants of bodycare cosmetics, to increase female breast cancer incidence,to interfere with male reproductive functions and to inuence development of malignant melanoma which has alsorecently been shown to be inuenced by oestrogenic stimulation. Copyright 2008 John Wiley & Sons, Ltd.

KEY WORDS: paraben; oestrogen; androgen; cosmetics; endocrine disruption; breast cancer; melanoma; male reproductive disorders;esterase; skin; absorption; carcinogenesis

Introduction

It has been suggested previously that chemicals with oes-trogenic and/or genotoxic properties applied in bodycarecosmetics around the breast area could be a contributoryfactor in the rising incidence of breast cancer (Darbre,2001; Darbre, 2003; Harvey and Darbre, 2004). Compel-ling evidence of a link comes from the disproportionatelyhigh numbers of female breast cancers which originatein the upper outer quadrant of the breast and which isthe area to which underarm and bodycare cosmetics aretargeted (Darbre, 2001, 2003). Analysis of the annualquadrant incidence of breast cancer in Britain publishedin 2005 showed not only that there were now 54% ofbreast cancers in the upper outer quadrant (subdivision ofthe breast into four quadrants and a central nipple areawould be expected to give no more than 20% in eachregion from random distribution), but that the relativeproportion in that region had risen linearly on an annual

basis since 1979 (Darbre, 2005a). This is inconsistentwith the dogma that the high incidence of breast cancerin the upper outer quadrant of the breast relates solely tothe slightly greater amount of breast epithelial tissue inthat region but would be consistent with the increasinguse of cosmetic products in the underarm area (McGrath,2003). Studies published in 2004 further challenged thisdogma by showing increased levels of genomic instabil-ity in outer regions of the human breast in histologicallynormal tissue (Ellsworth et al., 2004a). Instability of thegenome in human cells is an important contributor togenetic changes that drive tumorigenic processes (Lengaueret al., 1998) and in accordance with the cancer eld the-ory could provide a milieu where genetically altered cellswould then be more susceptible to the development ofcancer (Slaughter et al., 1953). The underlying mecha-nism of genetic instability found in outer regions of thebreast remains to be identied, but has been suggested toinvolve damage resulting from topically applied cosmeticchemicals (Ellsworth et al., 2004b).

A wide range of consumer products including underarmdeodorants, antiperspirants, skin moisturisers, body creams,body sprays and suncare products are applied topically

* Correspondence to: P. Darbre, School of Biological Sciences, TheUniversity of Reading, Whiteknights, Reading RG6 6AJ, UK. E-mail: [email protected]

Copyright 2008 John Wiley & Sons, Ltd. J. Appl. Toxicol. 2008DOI: 10.1002/jat

P. D. DARBRE AND P. W. HARVEY

to the breast and upper chest region on a frequent basis andleft on the skin, allowing for continuous dermal exposure,absorption and accumulation in underlying tissues (Harveyand Darbre, 2004; Darbre, 2006a). Over the past fewyears, chemical components used in these increasinglychemically complex products have been shown to possessoestrogenic properties and to be present in either humanbreast tissue or human milk (which has been secretedfrom breast epithelial cells; Darbre, 2006a; Donovanet al., 2007). The alkyl esters of p-hydroxybenzoic acid(parabens) are one such group of chemicals which areused extensively as preservatives in cosmetic consumerproducts, and the structures of the most commonly usedesters are shown in Fig. 1. In 1984, it was estimated thatparabens were used in 13 200 different cosmetic formula-tions (Elder, 1984) and a survey of 215 cosmetic prod-ucts in 1995 found parabens in 99% of leave-on productsand 77% of rinse-off products (Rastogi et al., 1995).More recent studies continue to show the presence ofparabens, primarily methylparaben and propylparaben(also in combination with phthalates) in the majority of

body care cosmetics analysed, including deodorants,creams and lotions (Shen et al., 2007). The EuropeanUnion permits the use of parabens in cosmetic productswith a maximum concentration of each one of 0.4% anda total maximum concentration of 0.8% (EU CosmeticsDirective 76/768/EEC). In 2004, the measurement in humanbreast cancer tissue of intact esters of the ve commonlyused parabens, methylparaben, ethylparaben, n-propylpa-raben, n-butylparaben and isobutylparaben (Darbre et al.,2004a), stimulated international discussion, and althoughthe presence of a chemical in a tissue does not imply anyfunctional role in disease processes, this nding did stim-ulate review of the safety of using parabens in so wide arange of consumer products (Harvey and Darbre, 2004;Bergfeld et al., 2005; Golden et al., 2005; Soni et al.,2005). Over the ensuing four years, there has been notonly publication of literature reviews but also of originalresearch, and this review attempts to summarize the newand signicant ndings which have added to the knowl-edge database on parabens and which have stimulatedregulatory action for some uses.

Figure 1. The chemical structures of seven alkyl esters of p-hydroxybenzoic acid (parabens) which are commonlyused in consumer products. Hydrolysis of the ester linkage (arrow) gives the common paraben metabolite p-hydroxybenzoic acid


PARABEN ESTERS: REVIEW OF RECENT ENDOCRINE TOXICITY

Measurement of Parabens in Human Tissues

Environmental Exposures

If a chemical component of body care cosmetics is toinuence human breast cancer incidence, then continuousdermal exposure must translate into absorption of thechemical through the dermal layers and into underlyingbreast tissues. Publication of results describing the meas-urement of paraben esters in human breast cancer tissue(Darbre et al., 2004a) caused substantial discussionbecause this was the rst time parabens had been shownto be present as intact esters within the human body andspecically within the human breast. Although critiquesunderlining the limitations of this pioneering studyhave been extensive from cosmetic industry associations(Golden and Gandy, 2004; Jeffrey and Williams, 2004;Flower, 2004; and see the discussion in Harvey, 2004and reply Darbre et al., 2004b), the systemic absorptionof parabens from environmental exposures has now beenconrmed by other groups through the measurement ofintact esters of parabens in raw sewage (Lee et al., 2005;Canosa et al., 2006a) and in human urine (Ye et al., 2006a;see Table 1). Although presence in raw sewage can befrom various sources, from human excretion to wash-offproducts (and non-cosmetics) entering the waste watersystem, the presence in urine conrms human systemicabsorption, and because intact paraben esters were found,these compounds have escaped metabolism by eitherskin esterases if exposure was dermal, or by intestinal andliver metabolic systems if exposure was oral (discussedlater). In the Canadian study, methylparaben, ethylparaben,n-propylparaben and n-butylparaben were all detectedin all sewage inuent samples with methylparaben (up to1.47 g ml1) and n-propylparaben (up to 2.43 g ml1)being detected at the greatest levels (Lee et al., 2005).Similar results were obtained in the Spanish study withmethylparaben detected in sewage up to 2.92 g ml1 andn-propylparaben up to 1.22 g ml1 (Canosa et al., 2006a).

The same group has also detected di-chlorinated formsof methylparaben and propylparaben in raw sewage watersamples (Canosa et al., 2006b) and it is currently unknownhow halogenation affects toxicity. The parabens detectedin human urine at the highest levels were also methylpa-raben and n-propylparaben at median concentrationsof 43.9 and 9.05 ng ml1 respectively (Ye et al., 2006a).In another report by the same group, methylparaben,ethylparaben and propylparaben were detected mostly asconjugated species in 22 urine samples from adults (Yeet al., 2006b), again conrming both systemic exposureand detection of intact esters that escaped metabolism.Evidence that parabens can enter the human body as intactesters has therefore now been conrmed in several studies,and over all six studies the parabens detected at highestlevels were consistently methylparaben and n-propylpa-raben (which is now subject to regulatory withdrawalfor food uses, discussed later). Whilst this may reect thegreater use of methylparaben in cosmetics (Rastogi et al.,1995), more recent work has shown that, of the com-monly used parabens, methylparaben penetrates the skinto the greatest extent (El Hussein et al., 2007), despitehaving the lowest lipophilicity. As a general note, recentwork predicting intestinal absorption in Caco-2 cellshas shown that various parabens were metabolized byesterases to p-hydroxybenzoic acid and the authors con-cluded that pre-systemic intestinal metabolism of orallyingested parabens may limit systemic exposure to parabenesters in vivo (Lakeram et al., 2007). The signicanceof this is that if intact paraben esters are detected inhuman tissue or urine, it is less likely that exposure wasvia the oral route (since oral ingestion results in bothintestinal and liver metabolism, reducing the probabilitythat paraben esters will survive intact) and implicatesdermal exposure as a potentially important exposureroute. The nature of systemic species of paraben esterswill inuence toxicity, not least the endocrine disruptingpotency of the different esters as one example of a rela-tively well-researched toxicological endpoint, and thismust be taken into account in toxicological evaluations.

Table 1. Levels of parabens measured in human tissues

Human breast tissue mean

ng g1 tumour

Human urine (USA) median

ng ml1

Human serum mean peak level (3 h) ng ml1

n-butyl applied as topical cream

Human urine mean level mg 24 h1

n-butyl applied as topical cream

n = 20 n = 100 n = 26 n = 26

Methylparaben 12.8 43.9Ethylparaben 2.0 1.0n-Propylparaben 2.6 9.1n-Butylparaben 2.3 0.5 0.135 2.6Isobutylparaben 0.9Benzylparaben 0.0 0.0Reference Darbre et al., 2004 Ye et al., 2006a Janjua et al., 2007 Janjua et al., 2008a



Dermal Absorption

Measurement of parabens in human tissues then posesthe question of the origin of the absorbed compoundsand whether parabens from topically/dermally appliedcosmetic products could be a source of the body burdensfound. Parabens are known to be readily absorbedthrough the skin and it has been suggested that hydrolysisof parabens by skin esterases could be incomplete in thecontext of increasing cosmetic useage and inter-individualvariations (Darbre et al., 2004a; Harvey and Darbre,2004). In vitro studies have shown that 30% of appliedpropylparaben penetrates the skin intact in rat skin(Bando et al., 1997), and after 8 h contact, penetrationof some esters can be even higher with up to 60% ofmethylparaben and 40% of ethylparaben crossing rabbitskin intact (Pedersen et al., 2007). In humans, variationsbetween individuals in hydrolysis of parabens have nowbeen shown in the case of human liver esterases (Jewellet al., 2007), although studies are still lacking for skinesterases. Furthermore, recent work has also revealedthat hydrolysis of parabens by esterases is slower inhuman skin than in rat skin (Prusakiewicz et al., 2006;Harville et al., 2007), suggesting that predictions basedon rat skin metabolism data may signicantly underesti-mate the level of paraben esters that can be absorbedfrom topical application into underlying tissues of humanskin. Another study (Ishiwatari et al., 2007) has demon-strated that methylparaben is not hydrolysed completelyby esterases of human skin. At 12 h after applicationof a test formulation of 0.15% methylparaben to humanvolunteers, the concentration in the stratum corneum was10 pmol cm2 (detection limit given as 0.02 pmol cm2)or 0.028% of the application, and repeated application(twice a day for one month) caused accumulation ofmethylparaben in the stratum corneum to 20 pmol cm2

after 1 week and 120 pmol cm2 after 4 weeks. Just 2 daysafter cessation of use, levels of methylparaben decreasedto



note that Janjua and colleagues have shown other oestro-genic chemicals commonly found in bodycare cosmeticsto also be absorbed systemically from whole body topicalapplication of cosmetic creams. This includes phthalates(Janjua et al., 2007) and the sunscreens benzophenone-3,octyl-methoxycinnamate and 3-(4-methyl-benzylidene)camphor (Janjua et al., 2004). Thus, real-life repeateddermal exposures to the range of paraben esters, and otheroestrogenic co-formulants in body care cosmetics, couldprovide signicant additional oestrogenic stimulation inthe human body.

Oestrogenic Activity of Parabens and its Main Metabolite p-hydroxybenzoic Acid

Oestrogen Agonist Properties

Epidemiological, clinical and experimental studies overmore than a century conrm that oestrogen plays a cen-tral role in the development, progression and treatment ofbreast cancer (Miller, 1996; Lonning, 2004), which bringsinto question potential interactions from environmentalchemicals which can enter the human breast and whichcan mimic oestrogen action. All the widely used parabenesters have now been shown to possess oestrogenic activ-ity in assay systems in vitro and in vivo and a list of themany studies from different laboratories was publishedin 2004 (Harvey and Darbre, 2004), updated in 2006(Harvey and Everett, 2006) and has been further updatedhere in Table 2.

The molecular basis of the genomic action of oestrogeninvolves binding of ligand to oestrogen receptor (ER), ERor ER, receptorligand dimerization, binding to oestro-gen response elements (ERE) in the DNA and transacti-vation of gene expression (Oettel and Schillinger, 1999).Strategies for determining whether a ligand possessesoestrogenic activity therefore begin with investigationof the ability to bind to ER and the ability to induceoestrogen-regulated gene expression, and then progressto the question of whether physiological responses can beinduced in cells in culture or in whole animal models(Soto et al., 2006; Clode, 2006). Growth of oestrogen-dependent cells, especially of the MCF7 human breastcancer cell line, has provided a sensitive and reliablemeasure of cell growth response (Soto et al., 2006) andincrease in uterine weight in the immature female rodenthas provided a reliable indicator of oestrogenic activityin a whole animal model (Clode, 2006). The majority ofstudies shown in Table 2 report oestrogen agonist activityof parabens, a minority of studies (one out of 25 in vitrostudies, and seven out of 30 in vivo studies in Table 2)reported negative ndings in oestrogenicity assays. Thesereports tended to be studies of the shorter alkyl groupparabens such as methylparaben and ethylparaben, butwere notably more frequent in the in vivo studies, which

probably reects variations in methodology, animal strainand dose between studies. In this respect, the database issmall for in vivo work on methylparaben and ethylpa-raben, and it would be useful to conduct more systematic,consistent and controlled studies, using a range of doselevels and dose routes to establish both no-effect-leveland lowest-effect-level doses for all the paraben esters.

Oestrogenic activity of parabens is known to increasewith increasing length of the linear alkyl chain frommethylparaben to n-butylparaben (Routledge et al., 1998;Byford et al., 2002) and with branching in the alkyl chainfrom n-propylparaben to isopropylparaben (Okubo et al.,2001) or from n-butylparaben to isobutylparaben (Darbreet al., 2002). Extension of the alkyl chain of methylpa-raben with a structure containing an aromatic ring in ben-zylparaben also increased oestrogenic activity (Darbreet al., 2003). Further studies over the past 3 years have beenpublished conrming the oestrogenic activity of theseesters in yet more diverse in vitro and in vivo assays whichare summarized in Table 2. In addition, studies haveshown that the common metabolite of paraben esters p-hydroxybenzoic acid (see Fig. 1) also possesses oestrogenicactivity in both in vitro and in vivo assays (Table 2). Thus,whilst shortening of the alkyl group in the paraben esterwas already known by 2004 to reduce oestrogenic activ-ity, more recent work has shown that complete removal ofthe alkyl grouping reduces activity still further but doesnot eradicate all oestrogenicity. This questions currentcontention that metabolic hydrolysis of parabens to theircommon metabolite, p-hydroxybenzoic acid acts to eradi-cate oestrogenic body burdens from paraben exposure inconsumer products.

Should Parabens be Termed Weak Oestrogens?

Whilst their ability to mimic oestrogen action is now welldocumented, the extent to which parabens can be labelledas weak oestrogens (e.g. Golden et al., 2005) needsfurther consideration. Receptor-mediated mechanisms ofligand activity are dependent on two fundamental events:ligand afnity and ligand efcacy (Strange, 2008). In thespecic case of oestrogen action, ligand afnity for theER can be estimated through in vitro techniques suchas ligand binding assays (Green and Leake, 1987), andthe weaker the ligand binding afnity for ER, the higherthe concentration of ligand needed to saturate the ER orto compete out radiolabelled oestradiol from binding toER. Ligand efcacy refers to the ability of the ligand toinuence receptor-mediated signalling pathways whichin the case of genomic mechanisms of oestrogen actioncould involve the ability of ER to bind to chaperone Hspproteins, to dimerize (to homodimers or heterodimers withER and ER ), to translocate to the nucleus, to bind to theERE in DNA and to transactivate gene expression (Oetteland Schillinger, 1999). Displacement of radiolabelled



Table 2. Summary of in vitro and in vivo studies published on the oestrogenic activity of parabens

Paraben Result in vitro Result in vivo Reference

Methylparaben + ve (yeast + receptor binding) Routledge et al., 1998; Nishihara et al., 2000; Miller et al., 2001; Schultis and Metzger, 2004; Morohoshi et al., 2005

+ ve (human MCF7) Byford et al., 2002; Schultis and Metzger, 2004; Pugazhendhi et al., 2005, 2007; Vanparys et al., 2007

ve (rat uterus receptor binding) Lemini et al., 2003+ ve (rat uterus receptor binding) Blair et al., 2000

ve (rat uterotrophic) Hossaini et al., 2000 ve (rat dietary repeat dose reproductive toxicity study)

Oishi, 2004

+ ve (rat uterotrophic) Lemini et al., 2003+ ve (mouse uterotrophic) Lemini et al., 2003

Ethylparaben + ve (yeast + receptor binding) Routledge et al., 1998; Nishihara et al., 2000; Miller et al., 2001; Schultis and Metzger, 2004; Morohoshi et al., 2005

+ ve (human MCF7) Okubo et al., 2001; Byford et al., 2002; Schultis and Metzger, 2004; Vanparys et al., 2007

+ ve (rat uterus receptor binding) Blair et al., 2000; Lemini et al., 2003+ ve (human HeLa overexpressing ER) Gomez et al., 2005

ve (rat uterotrophic) Hossaini et al., 2000 ve (rat dietary repeat dose reproductive toxicity study)

Oishi, 2004

+ ve (rat uterotrophic) Lemini et al., 2003+ ve (mouse uterotrophic) Lemini et al., 2003

n-Propylparaben + ve (yeast + receptor binding) Routledge et al., 1998; Nishihara et al., 2000; Miller et al., 2001; Schultis and Metzger, 2004; Morohoshi et al., 2005

+ ve (human MCF7) Okubo et al., 2001; Byford et al., 2002; Schultis and Metzger, 2004; Vanparys et al., 2007


ve (rat uterotrophic) Hossaini et al., 2000+ ve (rat dietary repeat dose reproductive toxicity study)

Oishi, 2002a

+ ve (rat uterotrophic) Lemini et al., 2003+ ve (mouse uterotrophic) Lemini et al., 2003+ ve (rainbow trout) Bjerregaard et al., 2003+ ve (male medaka sh) Inuie et al., 2003+ ve but antagonist (zebrash) Mikula et al., 2006

n-Butylparaben ve (rat teratogenicity study) Daston, 2004; Harvey 2005+ ve (yeast + receptor binding) Routledge et al., 1998; Nishihara et al., 2000;

Miller et al., 2001; Schultis and Metzger, 2004; Morohoshi et al., 2005

+ ve (human MCF7) Okubo et al., 2001; Byford et al., 2002; Schultis and Metzger, 2004; Pugazhendhi et al., 2007; Vanparys et al., 2007


+ ve (rat uterotrophic) Routledge et al., 1998+ ve (rat uterotrophic) Hossaini et al., 2000+ ve (rat dietary repeat dose reproductive toxicity study)

Oishi, 2001

+ ve (mouse dietary repeat dose reproductive toxicity study)

Oishi, 2002b

+ ve (rat development and reproductive toxicity study)

Kang et al., 2002

+ ve (rat uterotrophic) Lemini et al., 2003+ ve (mouse uterotrophic) Lemini et al., 2003+ ve (rainbow trout) Alslev et al., 2005

Isopropylparaben + ve (human MCF7) Okubo et al., 2001+ ve (yeast + receptor binding) Morohoshi et al., 2005

Isobutylparaben + ve (human MCF7) Okubo et al., 2001+ ve (human MCF7; ZR-75-1) Darbre et al., 2002+ ve (yeast + receptor binding) Morohoshi et al., 2005

+ ve (mouse uterotrophic) Darbre et al., 2002+ ve (rat uterotrophic) Koda et al., 2005

Benzylparaben + ve (human MCF7; ZR-75-1) Darbre et al., 2003; Schultis and Metzger, 2004+ ve (yeast + receptor binding) Miller et al., 2001; Schultis and Metzger, 2004;

Morohoshi et al., 2005+ ve (rat uterus receptor binding) Blair et al., 2000

+ ve (mouse uterotrophic) Darbre et al., 2003p-hydroxybenzoic acid + ve (human MCF7) Pugazhendhi et al., 2005(main metabolite) + ve (mouse uterotrophic) Lemini et al., 1997

ve (rat uterotrophic) Lemini et al., 2003+ ve (mouse uterotrophic) Lemini et al., 2003



oestradiol in competitive ER binding assays requireshigher concentrations of parabens than physiologicaloestrogens and some other xeno-/phyto-oestrogens (seereferences in Table 2; summarized for the paraben estersin Darbre, 2006a), demonstrating that parabens havelower binding afnity to ER than some other oestrogenicligands. However, this does not result in reduced efcacyif sufcient concentration of paraben is present. Indeed,in whole cell assays, with sufcient concentration, theparabens give responses in terms of increased gene expres-sion and cell proliferation in human breast cancer cellsof the same magnitude as 17-oestradiol (see referencesin Table 2, especially Byford et al., 2002; Darbre et al.,2002, 2003). Parabens are not partial agonists, as mightbe implied by the term weak, but give full agonistresponses in whole cells. Furthermore, the fact that, incell-based assays, the parabens might achieve full agonistresponse at lower concentrations given more time, hasnot been a considered parameter in any of the publishedstudies on parabens. Oestrogenic activity is usually denedrelative to the concentration required to achieve maximal(or half-maximal) effect in a set time in any given assaysystem, such that compounds with weaker activity requirehigher concentrations to achieve the response in the settime frame of the assay. The principle that partial agonisteffects can be enhanced over a longer time period hasbeen shown in the case of the oestrogen agonist propertiesof triclosan (Gee et al., 2008), and this should be repeatedwith the paraben esters since it is of high importance toenvironmental situations where the compound would bepresent over the long term and not only for a set time frame.The operational label of weak oestrogen therefore needsto be reconsidered in the context of whole cells and envi-ronmental studies.

How closely do Parabens Mimic Oestrogen Action?

Another uncertainty which has surrounded the oestrogenicproperties of parabens is the extent to which parabenaction is identical to that of 17-oestradiol. Studies pub-lished up to 2004 have shown that parabens can upregu-late a few single genes (reporter genes, pS2, progesteronereceptor) in a manner similar to that of oestradiol (seereferences in Table 2), but global gene expression prolinghas revealed that oestradiol regulates the expression ofmany hundred genes and that more genes are downregu-lated by oestradiol than upregulated (Frasor et al., 2003).Recent expression microarray studies have shown somesimilarities in global gene expression patterns betweenparabens and oestradiol when parabens were used atconcentrations sufcient to stimulate a growth response(Terasaka et al., 2006; Pugazhendhi et al., 2007), but notall genes were regulated in an identical manner (Pugazhendhiet al., 2007). Using a 20K expression array, some geneswere found to respond differently to parabens from oes-

tradiol whilst other genes could be regulated to differentextents even between individual parabens (Pugazhendhiet al., 2007). Furthermore, since methylparaben wasfound to inuence the expression of a greater numberof genes than n-butylparaben (Pugazhendhi et al., 2007),this further questions the classication of methylparabenas weaker than n-butylparaben.

Role of ERa and ERb

The molecular basis of oestrogen action involves bindingof the ligand to intracellular receptors which function asligand-activated transcription factors (Oettel and Schillin-ger, 1999), but two oestrogen receptors (ER and ER )have now been characterized which vary in their tissuedistributions and in their ligand binding afnities (Kuiperet al., 1997; Imamov et al., 2005). Whilst ER levels arerelatively higher in mammary and uterine tissue, andmost notably in ER positive breast cancers, ER is moreubiquitously distributed and may mediate oestrogenicactions across non-reproductive tissues. Furthermore, ERis now known to act as a growth inhibitor and the relativeproportion of ER in breast cancer cells is thought toinuence the outcome in breast cancer (Speirs et al., 2004),which suggests that altering ER actions could be asdetremental as enhancing ER activity. Recent work hassuggested that individual parabens may vary in their rela-tive binding afnity for ER and ER (Okubo et al.,2001; Gomez et al., 2005), which in turn suggests thatthere could be equally varied responses of the differentparabens in different tissues and during development ofdifferent breast cancers. Whether the parabens can alsobind to the oestrogen-related receptors will also be impor-tant in the light of the recent description of the binding ofbisphenol A to human oestrogen-related receptor- (Okadaet al., 2008).

Effects on Oestrogen Metabolism

Although many of the effects of environmental oestrogensare mediated via ER and ER, it is now known thatsome xenoestrogens may be able to exert endocrine dis-rupting properties through interfering with metabolicenzymes responsible for the synthesis of physiologicaloestrogens or for modication of their availability in free,unconjugated form (Whitehead and Rice, 2006). In thisrespect, a recent report that parabens may be able to inhibitsulfotransferases (Prusakiewicz et al., 2007) is anotherillustration of their varied oestrogen disrupting actions.Oestrogen action in vivo is regulated through a balancedinteraction between sulfotransferase enzymes (SULTs)which catalyse sulfate conjugation and sulfatases whichrelease free oestrogens. Many environmental oestrogendisrupters, such as hydroxylated polyhalogenated aromatic



hydrocarbons, are known to function as SULT inhibitors(Kester et al., 2002) and that parabens can also inhibitsulfation of oestrogens through inhibition of SULTs(Prusakiewicz et al., 2007) suggests that parabens mayalso indirectly enhance oestrogen effects through eleva-tion of free oestradiol levels.

Antiandrogenic Properties of Parabens and Male Reproductive Disorders

Many environmental chemicals which possess oestrogenicproperties have also been shown to display antiandrogenicactivity. Certain pesticides, the fungicide vinclozolin,bisphenol A, some phthalates and triclosan can antago-nize the action of androgens in assays in vitro and inanimal models (Kelce and Wilson, 1997; Sohoni andSumpter, 1998; Gee et al., 2008). It is therefore interest-ing that recent reports have documented the ability ofseveral parabens to bind to human androgen receptor(Satoh et al., 2005) and an antiandrogenic activity for allparabens tested in antagonizing the action of testosteroneon reporter gene expression (Satoh et al., 2005; Chenet al., 2007; see Table 3 for details). Although male repro-ductive abnormalities resulting in animal models fromexposure to endocrine-disrupting compounds have beenattributed to the oestrogenic activity of the chemicals, therelevance of antiandrogenic properties is now receivingmore serious consideration (Bay et al., 2006; Sharpe,2006; Filby et al., 2007). Repeat oral dosage of propylpa-raben and butylparaben in diet to juvenile rodents hasbeen reported to result in alterations to male reproductivefunctions including spermatogenesis, testosterone secre-tion and epididymal weights (Oishi, 2001, 2002a,b), and

this has been assumed to result from oestrogenic activityof the parabens. However, it is also possible that anti-androgenic actions of parabens through AR-mediatedpathways could have contributed to the effects observed,which brings into question whether this data should beincluded under Table 2 or Table 3 or both.

Over the past 50 years, male reproductive disorders havebeen documented as increasing in the human population,and four male reproductive health problems, cryptorchidism,hypospadias, reduced semen quality and testicular cancer,are considered indicators of the testicular dysgenesis syn-drome with a suspected origin in fetal or early postnatallife (Bay et al., 2006). However, whether causality involvesexposure to oestrogenic and/or antiandrogenic chemicalsat this sensitive time frame and the source of such expo-sure remain uncertain. Evidence has been documented forexposure to oestrogenic/antiandrogenic chemicals caus-ing endocrine disruption in aquatic species (Matthiessen,2003) and for exposure to phthalates producing malereproductive abnormalities in laboratory rodent modelssimilar to those observed in humans (Sharpe, 2006).Maternal exposure to butylparaben during gestation andlactation has been shown to result in reproductive disor-ders in male offspring by postnatal day 49 (Kang et al.,2002). Whether the effects resulted from butylparabencrossing the placenta during gestation or passing in milkduring suckling, or a combination of both, remain to bedetermined, but the sensitivity of early postnatal malerodents to development of reproductive disorders whenexposed to butylparaben via the dietary route has beenconrmed in subsequent studies (Oishi, 2001). Sinceparabens are now known to penetrate skin (see earlier), itremains in question as to whether similar consequencesmight result from topical application of parabens to early

Table 3. Summary of in vitro and in vivo studies published on the antiandrogenic activity of parabens

Paraben AR binding assayReporter gene assay (antagonist activity) In vivo assay Reference

Methylparaben ve (recombinant hAR) + ve (transfected CHO-K1 cells) Satoh et al., 2005+ ve (transfected HEK 293 cells) Chen et al., 2007

ve (rat dietary repeat dose male reproductive toxicity study)

Oishi et al., 2004

Ethylparaben ve (recombinant hAR) + ve (transfected CHO-K1 cells) Satoh et al., 2005 ve (rat dietary repeat dose male reproductive toxicity study)

Oishi, 2004

n-Propylparaben + ve (recombinant hAR) + ve (transfected CHO-K1 cells) Satoh et al., 2005+ ve (transfected HEK 293 cells) Chen et al., 2007

+ ve (rat dietary repeat dose male reproductive toxicity study)

Oishi, 2002a

n-Butylparaben + ve (recombinant hAR) + ve (transfected CHO-K1 cells) Satoh et al., 2005+ ve (transfected HEK 293 cells) Chen et al., 2007

+ ve (rat dietary repeat dose male reproductive toxicity study)

Oishi et al., 2001

Isopropylparaben + ve (recombinant hAR) + ve (transfected CHO-K1 cells) Satoh et al., 2005Isobutylparaben + ve (recombinant hAR) + ve (transfected CHO-K1 cells) Satoh et al., 2005p-Hydroxybenzoic acid (main metabolite)

ve (transfected HEK 293 cells) Chen et al., 2007



Tab

le 4

.O

estr

og

en a

nd

an

dro

gen

ic a

ctiv

ity

of

chem

ical

co

nst

itu

ents

of

cosm

etic

s ap

plie

d t

o t

he

un

der

arm

an

d b

reas

t ar

ea a

nd

evi

den

ce f

or

thei

r ab

sorp

tio

nin

to t

he

hu

man

bo

dy

Co

smet

ic c

hem

ical

Fun

ctio

n

in c

osm

etic

Mea

sure

men

t in

h

um

an b

od

yPe

net

rati

on

of

hu

man

ski

n in

vit

ro

Syst

emic

ab

sorp

tio

n

into

hu

man

s fr

om

to

pic

al a

pp

licat

ion

Oes

tro

gen

ic a

ctiv

ity

(ag

on

ist

and

/or

anta

go

nis

t)

An

dro

gen

ic a

ctiv

ity

(ag

on

ist

and

/or

anta

go

nis

t)

Par

aben

sP

rese

rvat

ive

Bre

ast,

urin

e (s

ee T

able

1)

Prus

akie

wic

z et

al.,

200

6;

Har

ville

et a

l., 2

007;

Is

hiw

atar

i et a

l., 2

007

Dor

sal s

kin

into

blo

od

(Jan

jua

et a

l., 2

007)

and

ur

ine

(Jan

jua

et a

l., 2

008a

)

see

Tab

le 2

see

Tab

le 3

UV

lt

ers

Abs

orb

UV

ligh

tU

rine

(H

ayde

n et

al.,

199

7;

Feli

x et

al.,

199

8), m

ilk

(Han

y et

al.,

199

5)

Tre

ffel

and

Gab

ard,

1996

; H

ayde

n et

al.,

199

7;

Cha

tela

in e

t al.,

200

3

Dor

sal s

kin

into

blo

od

(Jan

jua

et a

l., 2

004)

and

ur

ine

(Jan

jua

et a

l., 2

008b

)

Schl

umpf

et a

l., 2

001;

In

ui e

t al.,

200

3;

Kod

a et

al.,

200

5;

Hen

ewee

r et

al.,

200

5;

Kun

z an

d Fe

nt, 2

006

Poly

cycl

ic m

usks

Fra

gran

ce

Adi

pose

tiss

ue (

Kan

nan

et a

l.,

2005

), s

erum

/mil

k (K

ukle

nyik

et a

l., 2

007)

; m

ilk (

Rei

ner

et a

l., 2

007)

Gom

ez e

t al.,

200

5;

Schr

eurs

et a

l., 2

005;

M

ori e

t al.,

200

7

Schr

eurs

et a

l., 2

005;

M

ori e

t al.,

200

7

Alu

min

ium

ch

lorh

ydra

teA

ntip

ersp

iran

tB

reas

t tis

sue

(Exl

ey e

t al.,

200

7)U

nder

arm

ski

n in

to b

lood

(F

lare

nd e

t al.,

200

1;

Gui

llard

et a

l., 2

004)

Inte

rfer

e w

ith

bind

ing

of o

estr

ogen

to E

R a

nd

oest

roge

n-re

gula

ted

gene

ex

pres

sion

(D

arbr

e 20

05b)

Cyc

losi

loxa

nes

Con

diti

onin

g,

spre

adin

gZ

areb

a et

al.,

200

2H

ayde

n an

d B

arlo

w, 1

972;

M

cKim

et a

l., 2

001;

H

e et

al.,

200

3T

ricl

osan

Deo

dora

nt/

pres

erva

tive

Milk

, blo

od (

Ado

lfss

on-E

rici

et

al.,

200

2; H

ovan

der

et a

l.,

2002

; All

myr

et a

l., 2

006)

Mos

s et

al.,

200

0C

osm

etic

use

into

pla

sma/

milk

(A

llmyr

et a

l., 2

006)

Gee

et a

l., 2

008

Gee

et a

l., 2

008

Phth

alat

esPl

asti

cize

rM

ilk (

Cal

afat

et a

l., 2

004)

; am

niot

ic

uid

(Sil

va e

t al.,

200

4);

saliv

a (S

ilva

et a

l., 2

005)

; uri

ne

(Cal

afat

and

McK

ee, 2

006)

Cos

met

ic u

se in

to u

rine

(D

uty

et a

l., 2

005)

; dor

sal s

kin

into

blo

od (

Janj

ua e

t al.,

200

7)

and

urin

e (J

anju

a et

al.,

200

8a)

Jobl

ing

et a

l., 1

995;

H

arri

s et

al.,

199

7;

Oku

bo e

t al.,

200

3

Soh

oni a

nd S

umpt

er,

1998

; Lee

and

Koo

, 200

7



postnatal male rodents. This is an important questionbecause parabens are used in an extensive array of babywipes and baby creams where the products would be lefton the skin of the genital area of baby boys, allowing forabsorption and accumulation (especially following skinabrasions from nappy/diaper rash) into underlying tissuesof the reproductive organs throughout this sensitive timeframe.

Parabens and Skin Cancer

Although the oestrogenic properties of parabens havebeen extensively discussed in relation to the developmentof breast cancer and in particular from the topical appli-cation of paraben-containing cosmetics around the breastarea, many other body tissues apart from the mammarygland are also sensitive to oestrogen action, not leastreproductive organs, skin, bone and the cardiovascularsystem. With the continued use of parabens in so wide arange of skincare products (Elder, 1984; Rastogi et al.,1995; Shen et al., 2007) and with such ubiquitous dis-tribution of parabens in the domestic environment thatthey can now be detected not only in body tissues but inhouse dust (Canosa et al., 2007), it is important to con-sider whether there are wider implications, particularlyfor skin, which is the body tissue with which the topicalcosmetics will be in immediate contact. Oestrogenshave long been known to have a profound inuence onskin development and composition, and the reduction inoestrogen levels at menopause results in changes associ-ated with aging (Thornton, 2002; Hall and Phillips, 2005;Verdier-Sevrain et al., 2006). These aging effects on theskin can be reduced with topical treatment of skin withoestrogen-containing creams (Sator et al., 2004). Use ofparabens in so many cosmetic products which are appliedtopically to skin should be expected therefore to have aninuence on both the epidermis and the dermis of skin.Whether such an inuence is positive or negative at dif-ferent life stages and at different concentrations needsurgent investigation. Recent research has reported thatmethylparaben can indeed inuence the aging and differ-entiation of keratinocytes (Ishiwatari et al., 2007), sincelong-term application of methylparaben to keratinocytescould inuence proliferation rate, cell morphology andexpression of hyaluronan synthases and type IV collagen.Further work has shown that methylparaben and ethylpa-raben can induce oxidative stress in skin after reactionwith singlet oxygen (1O2) in visible light to produce glu-tathione conjugates of hydroquinone (Nishizawa et al.,2006). However, the report that methylparaben potenti-ates UV-induced damage of skin keratinocytes includingreactive oxygen species (ROS) and nitric oxide (NO)production and lipid peroxidation (Handa et al., 2006)poses more serious questions concerning potentially gen-otoxic effects of methylparaben applied in cosmetics to

human skin when exposed to sunlight. This is importantin the context of the use of methylparaben in sunscreenproducts and the continued uncertainty as to whetherthere is a positive or negative association between sun-screen use and development of human skin cancers (Diffey,2005; Francis et al., 2006).

There are two main forms of skin cancer: melanomaand non-melanoma skin cancer. Non-melanoma skin can-cers, including basal cell and squamous cell carcinomas,are the most common form of skin cancer, and in 2004,there were at least 72 000 new cases registered in the UK(Ofce of National Statistics, London). However, theypose a lower clinical problem than the melanomas due tosuccess rates of early treatment (Neville et al., 2007). Bycontrast, melanoma has become a major public healthproblem in many countries and since the 1960s has risenby 38% per year in most European countries (Thompsonet al., 2005) with an annual incidence rate in the UK in2004 of 13.0 [age standardized rate (European) per 100 000population], which is equivalent to 9000 new casesregistered in 2004 in the UK (Gavin and Walsh, 2005).Melanoma affects younger people more than most can-cers, with about 40% of cases in the under 50s (Gavinand Walsh, 2005), and there is currently a strikingincrease in incidence in youth (Strouse et al., 2005;Downard et al., 2007) and a strong inverse relationshipwith social deprivation (Gavin and Walsh, 2005). Thenatural history of human malignant melanoma has sug-gested that oestrogen might inuence the incidence anddevelopment of these tumours (Schmidt et al., 2006),partly because the incidence in females is low beforepuberty, rising steeply through the reproductive years(Strouse et al., 2005), and survival may vary betweenmen and women (Reintgen et al., 1984; Gavin and Walsh,2005). Numerous studies have attempted to identify arole for ER as a biomarker in melanoma (Tanemuraet al., 2007), in the way that ER serves as a prognosticmarker in breast cancer (Miller, 1996), but to date thishas been without clear success and the spurious distribu-tion of ER in melanoma cells has obscured the role ofoestrogen in melanoma. However, recent studies showthat ER, and not ER, is the predominant oestrogenreceptor type in melanocytic lesions, with ER beingdetected ubiquitously where ER was not (Schmidt et al.,2006). The discovery that ER immunoreactivity wasincreased in severely dysplastic nevi and lentigo malig-nas (in-situ melanoma) but decreased in melanomas pro-gressively deeper in the dermis suggests that ER may beplaying a role in the biology of these cancers and mightserve as a useful prognostic marker (Schmidt et al.,2006). In the light of an involvement of ER on melano-cytic pathophysiology (Schmidt et al., 2006), the abilityof methylparaben to potentiate UV-induced damage inkeratinocytes (Handa et al., 2006) and the ability of parabensto act via ER (Okubo et al., 2001; Gomez et al., 2005),a potential involvement of parabens (alone or together



with other oestrogenic chemicals in cosmetics includingUV lters) should now be considered in studies of thedevelopment of malignant melanoma. The higher rate ofmelanoma in younger people (Gavin and Walsh, 2005;Strouse et al., 2005; Downard et al., 2007), the increas-ing incidence in youth (Strouse et al., 2005; Downardet al., 2007) and the inverse relationship with socialdeprivation (Gavin and Walsh, 2005) could all correlatewith greater use of paraben-containing skincare/suncareproducts, be it through more lavish amounts at eachapplication, more frequent applications or a lifestylewhere products are required more often and at higherlevels. It is interesting to note the recently reportedleft-sided excess of invasive cutaneous melanoma in sixdifferent countries (Brewster et al., 2007), which echosremarkably the similar ndings of left-sided excess ofbreast cancers (Darbre et al., 2003). This contrasts to thedecit of left-sided tumours at many other sites (Brewsteret al., 2007). Handedness in application of skincare prod-ucts may yet explain this phenomenon in both cases(Darbre, 2003; Brewster et al., 2007).

Genotoxic Activity of Parabens

Whilst the ability of oestrogen to inuence the incidence,growth, progression and metastasis of breast cancer iswell established (Miller, 1996), the potential for oestrogento act through genotoxic mechanisms to initiate breastcancer has been recognized only more recently (Russoand Russo, 2006). Several mechanisms have been postu-lated, including increasing the error rate of DNA replicationthrough stimulation of cell proliferation via oestrogen-mediated activity and more direct genotoxicity throughcytochrome-P450-mediated metabolic activation produc-ing genotoxic metabolites. The identication of adverseeffects of some environmental oestrogens which cannotbe explained solely on the basis of ER-mediated endo-crine disruption alone has prompted studies to considerwhether xenoestrogens might also possess genotoxicactivity. Notwithstanding parabens being inactive in clas-sical assays for mutagenicity and carcinogenicity (Soniet al., 2005), recent research has reported the ability ofpropylparaben and butylparaben to cause DNA damagedetectable in Comet assays and induction of chromosomeaberrations together with sister-chromatid exchanges(Tayama et al., 2008). Although the effects observed werereported at high concentrations in the millimolar range inthe CHO cells used, it remains to be ascertained whetherthere could be effects at lower concentrations over thelonger term in mammary cells, whether there could beadditive/inhibitory effects from multiple genotoxic chem-icals present in the human breast (from other cosmeticchemicals and/or diet), or whether there might be particu-lar windows of sensitivity to such genotoxic activity suchas in the prepubertal breast (Darbre, 2006a).

Regulatory Status of Parabens

Cosmetics are less stringently tested and receive less regu-latory attention compared with other types of chemicalsto which the general population is exposed (see discussionin Harvey and Everett, 2006). Despite this, cosmetics andbodycare products represent potentially the commonestexposure scenario to chemicals in individuals and thepopulation as a whole. Recent reports of cosmetics safetyin use show a surprisingly high percentage of adversereactions to products, with 26.5% of women and 17.4%of men reporting an adverse event/reaction to cosmeticuse (Di Giovanni et al., 2006). Of the reactions reported,95.9% involved the skin and 4.1% were systemic reactions,and of these reports of systemic reactions, headache wasthe most common (40.3%) followed by nausea (24.2%).

The parabens are a group of chemicals that have been,and continue to be, used extensively in cosmetics andbodycare products. Elder (1984) estimated that parabenswere used in 13 200 different cosmetic formulations andindependent analyses of cosmetic products found parabensin 99% of leave-on products (Rastogi et al., 1995). Recentanalyses conrm the presence of methylparaben and pro-pylparaben in the majority of types of cosmetics testedincluding deodorants, creams and lotions (Shen et al.,2007). Despite this, the regulatory toxicology data pack-age does not have adequate carcinogenicity or reproductivetoxicology studies to meet modern regulatory standards.This situation, and the current regulatory status of theparabens as a group, is discussed below.

The European Union permits the use of parabens incosmetic products with a maximum concentration of eachone of 0.4% and a total maximum concentration of 0.8%(EU Cosmetics Directive 76/768/EEC) and they are alsoregistered for use in foods. The latter use is more highlyregulated and recent regulatory reviews have resultedin the withdrawal of ADIs (Acceptable Daily Intake) forseveral paraben esters on grounds of reproductive andendocrine toxicity. For example JECFA 2007 [The JointFood and Agriculture Organization (FAO) and WorldHealth Organization (WHO) Expert Committee on FoodAdditives] has recommended that in view of the adverseeffects in male rats [reproductive toxicity associated withendocrine effects] propylparaben should be excludedfrom the group ADI for parabens used in food, recom-mending its withdrawal. The same evaluation noted thewithdrawal of specication for butylparaben on similargrounds. Similarly, the EC Scientic Panel on Food Addi-tives, Flavourings, Processing Aids and Materials in Con-tact with Food of the European Food Safety Authority(EFSA) reviewed propylparaben (EFSA, 2004) and wasunable to establish a no-observed-adverse-effect-level forreproductive and endocrine toxicity, and consequently anADI, effectively recommending its exclusion/withdrawalfrom food use. These regulatory evaluations involveoral exposures but it has been previously suggested that



dermal application of paraben esters may represent aspecial case because of a higher probability of escapinglimited skin esterase action (Bando et al., 1997; Oh et al.,2002; Prusakiewicz et al., 2006; El Hussein et al., 2007;Harville et al., 2007; Janjua et al., 2007) and local sub-cutaneous tissue accumulation (see Harvey and Darbre,2004), indicating a need for regulatory risk assessmentharmonization. Harmonization would require withdrawalof butylparaben and propylparaben from cosmetics.

With regard to cosmetics uses, the European CommissionsScientic Committee on Consumer Products (SCCP) pro-vided an extended opinion on the safety evaluation ofparabens in 2005, noted the deciencies in the toxicologydatabase of the parabens and requested more information,specically on full descriptions of available in vitropercutaneous absorption studies and a complete dossierwith regard to the reproductive and developmental toxic-ity of propyl, isopropyl, butyl and isobutyl paraben, witha special focus on the male reproductive system. TheEuropean Cosmetic Toiletry and Perfumery Association(COLIPA) provided a submission in response to this request,but SCCP (2006) evaluated these data and concluded thatthe submission contained too many shortcomings to bescientically valid and therefore there still remain inade-quacies and deciencies in the paraben safety dataset,and concerns over paraben endocrine and reproductivetoxicity.

Concerning the specic hypothesis of the role ofparabens and breast cancer, the European CommissionsScientic Committee on Consumer Products (SCCP)concluded in the light of the current state of knowledge(in 2005) that there was no evidence of a demonstrablerisk for the development of breast cancer following theuse of underarm cosmetics containing parabens (SCCP,2005). However, to date there are only two studies thathave looked at breast cancer risk with the use of underarmcosmetics and the data set is therefore too sparse to formany conclusions. Mirick et al. (2002) conducted a retro-spective interview-based casecontrol study on a relation-ship between use of products for underarm perspirationand the risk for breast cancer, and found risk of breastcancer did not increase with use of antiperspirant (OR =0.9; P = 0.23) or with deodorant (OR = 1.2; P = 0.19).Although this was a sizeable population study of breastcancer case patients (n = 813) and control subjects with-out breast cancer (n = 793), the weakness remains in thenature of the accuracy of the self-reported results, and inparticular whether all subjects were clear about the dif-ference between the ingredients added as antiperspirantand as deodorant in the commercial products which theyhad used, not least in view of the numbers who claimedto use antiperspirant in the absence of deodorant, prod-ucts which are not freely available (usual combination isdeodorant alone or both together). Accurate reportingcould have been ascertained from a group of subjectswho had never used any such products, but unfortunately

such a group was not included in the study. By contrast,McGrath (2003) addressed the issue of frequency (intensity)of underarm product use within a cohort of breast cancerpatients and their age of diagnosis, and reported that fre-quency and earlier use of antiperspirant/deodorants togetherwith underarm shaving were associated with an earlierage of breast cancer diagnosis (up to 19 years earlier).Neither study identies particular chemicals/ingredientsin the vast range of products that can be used.

Harvey and Darbre (2004) and Harvey and Everett(2006) have noted that all types of bodycare cosmeticsapplied to the skin (not just underarm cosmetics) can be asource of local oestrogenic chemical input to the breastand should be considered in risk assessments. Further-more, there are also an increasing number of other ingre-dients in various cosmetics that have been shown to beendocrine active or oestrogenic [for example polycyclicmusks (Gomez et al., 2005; Schreurs et al., 2005), UVlters (Schlumpf et al., 2001; Inui et al., 2003; Kodaet al., 2005), aluminium chlorhydrate (Darbre, 2006b),triclosan (Gee et al., 2008), phthalates (Jobling et al., 1995;Harris et al., 1997; Okubo et al., 2003), cyclosiloxanes(McKim et al., 2001; He et al., 2003)] and risk assess-ments should take into account mixture and combinedrepeated exposure effects.

Conclusions and Further Research Needed

The principle rst documented in 2004 that parabens canenter the human body as intact esters measurable in humanbreast cancer tissue (Darbre et al., 2004a) has now beenconrmed through the measurement of paraben esters alsoin normal human urine (Ye et al., 2006a), and the principlethat the measured parabens could be derived from topicalapplication of cosmetic products (Harvey and Darbre,2004) has been vindicated through the demonstration thatparabens can penetrate into the human circulatory systemfrom a single topical cosmetic application to a humansubject (Janjua et al., 2007). With the continued use ofparabens in the majority of bodycare cosmetics (Shenet al., 2007) and this evidence for systemic absorption ofparabens following topical cosmetic application to humansubjects (Janjua et al., 2007; 2008a), there is a need to nowascertain total blood and urine levels for all the parabenesters and their common metabolite p-hydroxybenzoicacid in order to more clearly understand whole body bur-dens in the population. A risk assessment of the oestro-gen equivalents of paraben absorbed from a single dailyapplication of a paraben-containing lotion has been cal-culated to be signicant (Harvey and Everett, 2006), andYe et al. (2006a) noted that, in a demographically diversegroup of 100 US male and female adults with no knownunusual exposure to parabens, methylparaben and propyl-paraben were detected in 96% of the samples. With suchwidespread presence of parabens in urines across the



population, there is a need to equally understand distribu-tion in all body tissues, and beyond breast, to now investigatethe distribution of parabens across all endocrine-sensitivetissues which might be inuenced through topical exposureto the parabens, not least male reproductive organs in theearly years of life and the skin itself.

At the current time it remains unknown as to whetherthe paraben levels measured in human tissues resultfrom continuous exposure or accumulated chemicals,but with the use of multiple cosmetic products on adaily basis (Loretz et al., 2006), many of which may con-tain parabens as preservatives, there is a need to investi-gate the potential for paraben accumulation followingrepeated dermal applications. In vitro models suggest thepotential for accumulation in underlying regions of skin(Ishiwatari et al., 2007), and recent unpublished data werebroadcast on UK television (screened 11 October 2007.http://www.unrealitytv.co.uk/reality-tv/beauty-addicts-how-toxic-are-you-channel-4) indicating that body burdensof parabens could be reduced by eradicating use of cos-metics containing parabens. The ability to reduce bodyburden through alteration to cosmetic exposure needs tobe substantiated by a controlled scientic study in astatistically viable group of subjects. A systematic exam-ination in human and rat skin models of the rates ofabsorption and hydrolysis for all the paraben esterswould provide scientic grounding for understanding theextent of absorption and escape from esterase metabolismat current environmentally relevant exposures. Further-more, there is a need to investigate the degree to whichthere can be variation between individuals in the poten-

tial for paraben absorption and tissue accumulation fol-lowing repeated applications in more substantial studieswhich Janjua and coworkers have initiated (Janjua et al.,2007).

A variety of studies have now demonstrated the abilityof parabens to disrupt physiologically important functionsin both in vitro cell culture systems and in vivo animalmodels and these are summarized in diagrammatic formin Fig. 2. The most extensive disrupting activity to bedescribed has been that resulting from the property ofparabens to bind to human ER and then to act via ER-mediated mechanisms to regulate gene expression andcell growth in oestrogen-responsive cells (see Table 2).Further endocrine disrupting activity has been demonstratedin the ability of parabens to antagonize AR-mediated eventsin androgen-responsive cells and to act as SULT inhibitors.Other reports have suggested parabens can inuence thesecretion of lysosomal enzymes in lymphocytes (Biaratiet al., 1994), can impair mitochondrial function in rathepatocytes (Nakagawa and Moldeus, 1998), can causeDNA damage in CHO cells (Tayama et al., 2008), and canpotentiate UV-induced damage including ROS and NOproduction in keratinocytes (Handa et al., 2006). Giventhat intact paraben esters have been measured in humanbreast tissue (Darbre et al., 2004a), the possibility thattheir oestrogenic activity could inuence the growth ofoestrogen responsive breast cancers is evident. However,there is also the potential for the paraben esters to inuencemale reproductive functions through a combination oftheir oestrogenic and antiandrogenic properties. It is alsoa possibility that parabens could inuence the development

Figure 2. Parabens have now been shown to influence several molecular pathways within cells. As endocrine dis-rupting chemicals, they can act as oestrogen receptor (ER) agonists, as androgen receptor (AR) antagonists or asinhibitors of sulfotransferase enzymes (SULT). On wider cellular functions, they can disrupt lysosomal and mito-chondrial functions, can cause DNA damage and can potentiate UVB-induced damage through production ofreactive oxygen species (ROS) and nitric oxide (NO)



of malignant melanomas through oestrogenic and genotoxicactivity. It may be that the inuence of parabens (togetherwith other cosmetic chemicals) on human tissues is ratherwider than has been hitherto anticipated and research isneeded to ascertain how wide the implications may actu-ally be.

Finally, there remains the need for controlled anddetailed evaluation of breast cancer risk from bodycarecosmetics, taking into account product chemical ingredi-ents, effect of formulations and total quantities applied,especially in potentially highly sensitive subgroups suchas babies and children. There are still only two epidemio-logical studies in the database (Mirick et al., 2002; McGrath2003) and no reported studies in animal models. It isunfortunate that under current UK regulations, cosmeticproducts can no longer be tested in animal models withinthe UK but breast cancer is a global problem and the poten-tial for bodycare cosmetic formulations to cause breastcancer under dened conditions are in urgent need of inves-tigation. Furthermore, at the current time, there remains awide gap between knowledge about the oestrogenic activityof single paraben esters on their own (Table 2) and theenvironmental reality where body tissues are exposed toa mixture of oestrogenic chemicals including a mixtureof all the paraben esters combined [all esters except ben-zylparaben were measured in both human breast tissue(Darbre et al., 2004a) and human urine (Ye et al., 2006a)]and a mixture of paraben esters together with otheroestrogenic chemicals of both dietary or cosmetic origin(Darbre, 2006a). Recent research has demonstrated thatenvironmental oestrogens can act in an additive manneralone or in combination with physiological oestrogens togive responses at concentrations where each alone wouldhave little or no effect in either in vitro (Rajapakse et al.,2002) or in vivo (Brian et al., 2005) assays. Butylparabenhas been shown specically to give oestrogenic responseswhich are additive when combined with either the physi-ological oestrogen, oestradiol or the xenoestrogens, non-ylphenol or Bisphenol A (Kyung-Sun-Kang et al., 2002).Research is now needed to dene clearly whether thereare additive or even inhibitory effects in human breastcancer cells of combinations of all the paraben esters andcombinations of all the paraben esters together with othercosmetic oestrogens (see above) and also environmentaloestrogens known to enter the breast through diet (Darbre,2006a). Breast cancer epidemiology must face the real-ity of combined exposures from environmental sources(Kortenkamp, 2006) and this deserves to be incorporatedinto regulatory risk assessments. A full risk assessmenttaking into account the total oestrogenic burden of allchemicals in the human breast will require a prole ofmeasurements of the oestrogenic chemicals in an averagehuman breast today. To date, only limited and inadequatemeasurements have been made of individual chemicals(and only one set of measurements remain in the databasefor parabens in breast) and there are no data on proles

of chemical content in individual human breast samples.Furthermore, in view of the disproportionate number ofbreast cancers in the upper outer quadrant, it wouldseem appropriate to investigate variations in regionaldistribution of chemicals across the human breast suchas has been described recently for aluminium (Exleyet al., 2007).

References

Adolfsson-Erici M, Pettersson M, Parkkonen J, Sturve J. 2002. Triclosan,a commonly used bactericide found in human milk and in the aquaticenvironment in Sweden. Chemosphere 46: 14851489.

Allmyr M, Adolfsson-Erici M, McLachlan MS, Sandborgh-Englund G.2006. Triclosan in plasma and milk from Swedish nursing mothersand their exposure via personal care products. Sci. Total Environ.372: 8793.

Alslev B, Korsgaard B, Bjerregaard P. 2005. Estrogenicity of butylpa-raben in rainbow trout Onchorhynchus mykiss exposed via food andwater. Aquat. Toxicol. 72: 295304.

Bando H, Mohri S, Yamashita F, Takakura Y, Hashida M. 1997.Effects of skin metabolism on percutaneous penetration of lipophilicdrugs. J. Pharm. Sci. 86: 759761.

Bay K, Asklund C, Skakkebaek NE, Andersson AM. 2006. Testiculardysgenesis syndrome: possible role of endocrine disrupters. BestPract. Res. Clin. Endocrinol. Metab. 20: 7790.

Bergfeld WF, Belsito DV, Marks JG, Andersen FA. 2005. Safety ofingredients used in cosmetics. J. Am. Acad. Dermatol. 52: 125132.

Biarati C, Goi G, Lombardo A, Tettamanti G. 1994. The esters of p-hydroxy-benzoate (parabens) inhibit the release of lysosomal enzymesby mitogen-stimulated peripheral human lymphocytes in culture.Clinica Chimica Acta 224: 147157.

Bjerregaard P, Anderson DN, Pedersen KL, Pedersen SN, Korsgaard B.2003. Estrogenic effect of propylparaben (propylhydroxybenzoate)in rainbow trout Oncorhynchus mykiss after exposure via food andwater. Comp. Biochem. Physiol. C 136: 309317.

Blair RM, Fang H, Branham WS, Hass BS, Dial SL, Moland CL, TongW, Shi L, Perkins R, Sheehan DM. 2000. The estrogen receptor bind-ing afnities of 188 natural and xenochemicals: structural diversityof ligands. Toxicological Sciences 54: 138153.

Brewster DH, Horner MJD, Rowan S, Jelfs P, deVries E, Pukkala E.2007. Left-sided excess of invasive cutaneous melanoma in six coun-tries. Eur. J. Cancer 43: 26342637.

Brian JV, Harris CA, Scholze M, Backhaus T, Booy P, Lamoree M,Pojana G, Jonkers N, Runnalls T, Bonfa A, Marcomini A, Sumpter JP.2005. Accurate prediction of the response of freshwater sh to amixture of estrogenic chemicals. Environ. Health Perspect. 113:721728.

Byford JR, Shaw LE, Drew MG, Pope GS, Sauer MJ, Darbre PD. 2002.Oestrogenic activity of parabens in MCF7 human breast cancer cells.J. Steroid Biochem. Mol. Biol. 80: 4960.

Calafat AM, McKee RH. 2006. Integrating biomonitoring exposuredata into the risk assessment process: phthalates [diethyl phthalateand di(2-ehtylhexyl)phthalate] as a case study. Environ. Health Perspect.114: 17831789.

Calafat AM, Slakman AR, Silva MJ, Herbert AR, Needham LL. 2004.Automated solid phase extraction and quantitative analysis of humanmilk for 13 phthalate metabolites. J. Chromatogr. B Analyt. Technol.Biomed. Life Sci. 805: 4956.

Canosa P, Rodriguez I, Rubi E, Bollain MH, Cela R. 2006a. Optimisa-tion of a solid-phase microextraction method for the determination ofparabens in water samples at the low ng per litre level. J. Chromatog.A 1124: 310.

Canosa P, Rodriguez I, Rubi E, Negreira N, Cela R. 2006b. Formationof halogenated by-products of parabens in chlorinated water. Anal.Chim. Acta. 575: 106113.

Canosa P, Rodriguez I, Rubi E, Cela R. 2007. Determination of para-bens and triclosan in indoor dust using matrix solid-phase dispersionand gas chromatography with tandem mass spectrometry. Anal. Chem.79: 16751681.



Chatelain E, Gabard B, Surber C. 2003. Skin penetration and sun pro-tection factor of ve UV lters: effect of the vehicle. Skin Pharma-col. Appl. Skin Physiol. 16: 2835.

Chen J, Ahn KC, Gee NA, Gee SJ, Hammock BD, Lasley BL. 2007.Antiandrogenic properties of parabens and other phenolic containingsmall molecules in personal care products. Toxicol. Appl. Pharmac.221: 278284.

Clarke R, Leonessa F, Welch JN, Skaar TC. 2001. Cellular and molecularpharmacology of antiestrogen action and resistance. Pharmac. Rev.53: 2572.

Clode SA. 2006. Assessment of in vitro assays for endocrine disruption.Best Pract. Res. Clin. Endocrinol. Metab. 20: 3543.

Darbre PD. 2001. Underarm cosmetics are a cause of breast cancer. Eur.J. Cancer Prev. 10: 389393.

Darbre PD. 2003. Underarm cosmetics and breast cancer. J. Appl. Toxicol.23: 8995.

Darbre PD. 2005a. Recorded quadrant incidence of female breast cancerin Great Britain suggests a disproportionate increase in the upperouter quadrant of the breast. Anticancer Research 25: 25432550.

Darbre PD. 2005b. Aluminium, antiperspirants and breast cancer. J. Inorg.Biochem. 99: 19121919.

Darbre PD. 2006a. Environmental oestrogens, cosmetics and breastcancer. Best Pract. Res. Clin. Endocrinol. Metab. 20: 121143.

Darbre PD. 2006b. Metalloestrogens: an emerging class of inorganicxenoestrogens with potential to add to the oestrogenic burden of thehuman breast. J. Appl. Toxicol. 26: 191197.

Darbre PD, Byford JR, Shaw LE, Horton RA, Pope GS and Sauer MJ.2002. Oestrogenic activity of isobutylparaben in vitro and in vivo.J. Appl. Toxicol. 22: 219226.

Darbre PD, Byford JR, Shaw LE, Hall S, Coldham NG, Pope GS andSauer MJ. 2003. Oestrogenic activity of benzylparaben. J. Appl. Toxicol.23: 4351.

Darbre PD, Aljarrah A, Miller WR, Coldham NG, Sauer MJ, Pope GS.2004a. Concentrations of parabens in human breast tumours. J. Appl.Toxicol. 24: 513.

Darbre PD, Pope GS, Aljarrah A, Miller WR, Coldham NG, Sauer MJ.2004b. Reply. J. Appl. Toxicol. 24: 299306.

Daston GP. 2004. Developmental toxicity evaluation of butylparaben inSpragueDawley rats. Birth Defects Res., Part B, Dev. Reprod. Toxicol.71: 296302.

Diffey BL. 2005. Sunscreens and melanoma: the future looks bright.Brit. J. Dermatol. 153: 378381.

Di Giovanni C, Arcoraci V, Gambardella L, Sautebin L. 2006. Cosme-tovigilance survey: are cosmetics considered safe by consumers?Pharmac. Res. 53: 1621.

Donovan M, Tiwary CM, Axelrod D, Sasco AJ, Jones L, Hajek R,Sauber E, Kuo J, Davis DL. 2007. Personal care products that containestrogens or xenoestrogens may increase breast cancer risk. Med. Hypoth.68: 756766.

Downard CD, Rapkin LB, Gow KW. 2007. Melanoma in children andadolescents. Surg. Oncol. 16: 215220.

Duty SM, Ackerman RM, Calafat AM, Hauser R. 2005. Personal careproduct use predicts urinary concentrations of some phthalatemonoesters. Environ. Health Perspect. 113: 15301535.

EFSA. 2004. Opinion of the Scientic Panel on Food Additives,Flavourings, Processing Aids and Materials in Contact with Foodon a request from the Commission related to para hydroxybenzoates(E214E219). The EFSA Journal 83: 126.

Elder RL. 1984. Final report on the safety assessment of methylparaben,ethylparaben, propylparaben and butylparaben. Journal AmericanCollege Toxicology 3: 147209.

Ellsworth DL, Ellsworth RE, Love B, Deyarmin B, Lubert SM, MittalMD, Hooke JA, Shriver CD. 2004a. Outer breast quadrants demon-strate increased levels of genomic instability. Ann. Surg. Oncol. 11:861868.

Ellsworth DL, Ellsworth RE, Liebman MN, Hooke JA, Shriver CD.2004b. Genomic instability in histologically normal breast tissues:implications for carcinogenesis. Lancet Oncol. 5: 753758.

El Hussein S, Muret P, Berard M, Makki S, Humbert P. 2007. Assessmentof principal parabens used in cosmetics after their passage throughhuman epidermisdermis layers (ex-vivo study). Exp. Dermatol. 16:830836.

Exley C, Charles LM, Barr L, Martin C, Polwart A, Darbre PD. 2007.Aluminium in human breast tissue. J. Inorg. Biochem. 101: 13441346.

Felix T, Hall BJ, Brodbelt JS. 1998. Determination of benzophenone-3and metabolites in water and human urine by solid-phase micro-extraction and quadrupole ion trap GC-MS. Anal. Chim. Acta 371:195203.

Filby AL, Thorpe KL, Maack G, Tyler CR. 2007. Gene expressionproles revealing the mechanisms of anti-androgen- and estrogen-induced feminization in sh. Aquatic Toxicol. 81: 219231.

Flarend R, Bin T, Elmore D, Hem SL. 2001. A preliminary study ofthe dermal absorption of aluminium from antiperspirants usingaluminium-26. Food Chem. Toxicol. 39: 163168.

Flower C. 2004. Letter to the editor. J. Appl. Toxicol. 24: 304305.[Reply Darbre PD, Pope GS, Aljarrah A, Miller WR, Coldham NGand Sauer MJ. 2004. J. Appl. Toxicol. 24: 305306.]

Francis SO, Mahlberg MJ, Johnson KR, Ming ME, Dellavalle RP.2006. Melanoma chemoprevention. J. Am. Acad. Dermatol. 55: 849861.

Frasor J, Danes JM, Komm B, Chang KCN, Lyttle CR, Katzenellenbo-gen BS. 2003. Proling of estrogen up- and down-regulated geneexpression in human breast cancer cells: insights into gene networksand pathways underlying estrogenic control of proliferation and cellphenotype. Endocrinology 144: 45624574.

Gavin A, Walsh P. 2005. Melanoma of skin. In Cancer Atlas of theUnited Kingdom and Ireland 19912000, Quinn M, Wood H, CooperN, Rowan S (eds). Palgrave-Macmillan: London. (Ofce of NationalStatistics, Crown Copyright, available online at: www.statistics.gov.uk/downloads/theme_health/caUKl91_00/Ch14_Melanoma.pdf)

Gee RH, Charles A, Taylor N, Darbre PD. 2008. Oestrogenic andandrogenic activity of triclosan in breast cancer cells. J. Appl. Toxicol.28: 7891.

Golden R, Gandy J. 2004. Letter to the editor. J. Appl. Toxicol. 24:297299. [Reply Darbre PD, Pope GS, Aljarrah A, Miller WR,Coldham NG, Sauer MJ. 2004. J. Appl. Toxicol. 24: 299301.]

Golden R, Gandy J, Vollmer G. 2005. A review of the endocrine activityof parabens and implications for potential risks to human health. Crit.Rev. Toxicol. 35: 435458.

Gomez E, Pillon A, Fenet H, Rosain D, Duchesne MJ, Nicolas JC,Balaguer P, Casellas C. 2005. Estrogenic activity of cosmetic com-ponents in reporter cell lines: parabens, UV screens and musks.J. Toxicol. Environ. Health A 68: 239251.

Green B, Leake RE. 1987. Steroid Hormones: A Practical Approach.IRL Press: Oxford.

Guillard O, Fauconneau B, Olichon D, Dedieu G, Deloncle R. 2004.Hyperaluminemia in a woman using an aluminium-containing anti-perspirant for 4 years. Am. J. Med. 117: 956959.

Hall G, Phillips TJ. 2005. Estrogen and skin: the effects of estrogen,menopause, and hormone replacement therapy on skin. J. Am. Acad.Dermatol. 53: 555568.

Handa O, Kokura S, Adachi S, Takagi T, Naito Y, Tanigawa T,Yoshida N, Yoshikawa T. 2006. Methylparaben potentiates UV-induced damage of skin keratinocytes. Toxicology 227: 6272.

Hany J, Nagel R and Nachweis V. 1995. UV-Filtersubstanzen inmuttermilch. Deutsche Lebensmittel-Rundshau 91: 341345.

Harris CA, Henttu P, Parker MG, Sumpter JP. 1997. The estrogenicactivity of phthalate esters in vitro. Environ. Health Perspect. 105:802811.

Harvey PW. 2004. Editorial: Discussion of concentrations of parabensin human breast tumours. J. Appl. Toxicol. 24: 307310.

Harvey PW. 2005. Comment on developmental toxicity evaluation ofbutylparaben in SpragueDawley rats. Birth Defects Res. B Dev.Reprod. Toxicol. 74: 114115.

Harvey PW, Darbre P. 2004. Endocrine disrupters and human health:could oestrogenic chemicals in bodycare cosmetics adversely affectbreast cancer incidence in women? A review of evidence and call forfurther research. J. Appl. Toxicol. 24: 167176.

Harvey PW, Everett DJ. 2006. Regulation of endocrine-disruptingchemicals: critical overview and deciencies in toxicology and riskassessment for human health. Best Pract. Res. Clin. Endocrinol.Metab. 20: 145165.

Harville HM, Voorman R, Prusakiewicz JJ. 2007. Comparison of para-ben stability in human and rat skin. Drug Metab. Lett. 1: 1721.

Hayden JF and Barlow SA. 1972. Structureactivity relationships oforganosiloxanes and the female reproductive system. Toxicol. Appl.Pharmac. 21: 6879.

Hayden CGJ, Roberts MS, Bensen HAE. 1997. Systemic absorption ofsunscreen after topical application. Lancet 350: 863864.



He B, Rhodes-Brower S, Miller MR, Munson AE, Germolec DR,Walker VR, Korach KS, Meade BJ. 2003. Octamethylcyclotetrasi-loxane exhibits estrogenic activity in mice via ER. Toxicol. Appl.Pharmac. 192: 254261.

Heneweer M, Muusse M, van der Berg M, Sanderson JT. 2005. Additiveestrogenic effects of mixtures of frequently used UV lters on pS2-gene transcription in MCF-7 cells. Toxicol. Appl. Pharmac. 208:170177.

Hossaini A, Larsen JJ and Larsen JC. 2000. Lack of oestrogenic effectsof food preservatives (parabens) in uterotrophic assays. Food Chem.Toxicol. 38: 319323.

Hovander L, Malmberg T, Athanasiadou M, Athanassiadis I, Rahm S,Bergman A, Klasson Wehler E. 2002. Identication of hydroxylatedPCB metabolites and other phenolic halogenated pollutants in humanblood plasma. Arch. Environ. Contam. Toxicol. 42: 105117.

Inui M, Adachi T, Takenaka S, Inui H, Nakazawa M, Ueda M, Wata H,Mori C, Iguchi T, Miyatake K. 2003. Effect of UV screens and pre-servatives on vitellogenin and choriogenin production in male medaka(Oryzias latipes). Toxicology 194: 4350.

Imamov O, Shim GJ, Warner M, Gustafsson JA. 2005. Estrogen receptorbeta in health and disease. Biol. Reprod. 73: 866871.

Ishiwatari S, Suzuki T, Hitomi T, Yoshinl S, Matsukuma S, Tsuji T.2007. Effects of methyl paraben on skin keratinocytes. J. Appl. Toxicol.27: 19.

Janjua NR, Mogensen B, Andersson AM, Petersen JH, Henriksen M,Skakkebaek NE, Wulf HC. 2004. Systemic absorption of the sunscreensbenzophenone-3, octyl-methoxycinnamate, and 3-(4-methyl-benzyli-dene)camphor after whole-body topical application and reproductivehormone levels in humans. J. Invest. Dermatol. 123: 5761.

Janjua NR, Mortensen GK, Andersson AM, Kongshoj B, SkakkebaekNE, Wulf HC. 2007. Systemic uptake of diethyl phthalate, dibutylphthalate, and butyl paraben following whole-body topical applica-tion and reproductive and thyroid hormone levels in humans. Envi-ron. Sci. Technol. 41: 55645570.

Janjua NR, Frederiksen H, Skakkebaek NE, Wulff HC, Andersson AM.2008a. Urinary excretion of phthalates and paraben after repeatedwhole-body topical application in humans. Int. J. Androl. 31: 118130.

Janjua NR, Kongshoj B, Andersson AM, Wulff HC. 2008b. Sunscreensin human plasma and urine after repeated whole-body topical applica-tion. J. Eur. Acad. Dermatol. Venereol. 22: 456461.

JECFA. 2007. Evaluation of certain food additives and contaminants.Sixty-seventh report of the joint FAO/WHO Expert committee onFood Additives. WHO Technical Report Series 940. World HealthOrganiZation.

Jeffrey AM, Williams GM. 2004. Letter to the editor. J. Appl. Toxicol.24 301303 [Reply Darbre PD, Pope GS, Aljarrah A, Miller WR,Coldham NG, Sauer MJ. 2004. J. Appl. Toxicol. 24: 303304].

Jewell C, Bennett P, Mutch E, Ackermann C, Williams FM. 2007.Inter-individual variability in esterases in human liver. Biochem.Pharmac. 74: 932939.

Jobling S, Reynolds T, White R, Parker MG, Sumpter JP. 1995. A varietyof environmentally persistent chemicals, including some phthalateplasticizers, are weakly estrogenic. Environ. Health Perspect. 103:582587.

Kang KS, Che JH, Ryu DY, Kim TW, Li GX and Lee YS. 2002.Decreased sperm number and motile activity on the F1 offspringmaternally exposed to butyl p-hydroxybenzoic acid (butyl paraben).J. Vet. Med. Sci. 64: 227235.

Kannan K, Reiner JL, Yun SH, Perrotta EE, Tao L, Johnson-Restrepo B,Rodan BD. 2005. Polycyclic musk compounds in higher trophic levelaquatic organisms and humans from the United States. Chemosphere61: 693700.

Kelce WR, Wilson EM. 1997. Environmental antiandrogens: develop-mental effects, molecular mechanisms and clinical implications.J. Mol. Med. 75: 198207.

Kester MH, Bulduk S, van Toor H, Tibboel D, Meinl W, Glatt H,Falany CN, Coughtrie MW, Schuur AG, Brouwer A, Visser TJ.2002. Potent inhibition of estrogen sulfotransferase by hydroxylatedmetabolites of polyhalogenated aromatic hydrocarbons reveals alter-native mechanism for estrogenic activity of endocrine disrupters.J. Clin. Endocr. Metab. 87, 11421150.

Kitagawa S, Li H and Sato S. 1997. Skin permeation of parabens inexcised guinea pig dorsal skin, its modication by penetration

enhancers and their relationship with n-octanol/water partitioncoefcients. Chem. Pharm. Bull. (Tokyo). 45: 13541357.

Koda T, Umezu T, Kamata R, Morohoshi K, Ohta T, Morita M.2005. Uterotrophic effects of benzophenone derivatives and p-hydroxybenzoate used in ultraviolet screens. Environ. Res. 98: 4045.

Kortenkamp A. 2006. Breast cancer, oestrogens and environmentalpollutants: a re-evaluation from a mixture perspective. Int. J. Androl.29: 193198.

Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S,Gustafsson JA. 1997. Comparison of the ligand binding specicityand transcript tissue distribution of estrogen receptors alpha and beta.Endocrinology 138: 863870.

Kuklenyik Z, Bryant XA, Needham LL, Calafat AM. 2007. SPE/SPME-GC/MS approach for measuring musk compounds in serum andbreast milk. J. Chromatog. B Analyt. Technol. Biomed. Life Sci. 858:177183.

Kunz PY, Fent K. 2006. Estrogenic activity of UV lter mixtures. Toxicol.Appl. Pharmac. 217: 8699.

Kyung-Sun-Kang, Cho SD, Lee YS. 2002. Additive estrogenic activitesof the binary mixtures of four estrogenic chemicals in recombinantyeast expressing human estrogen receptor. J. Vet. Sci. 3: 15.

Lakeram M, Paine AJ, Lockley DJ, Sanders DJ, Pendlington R, ForbesB. 2006. Transesterication of p-hydroxybenzoate esters (parabens)by human intestinal (Caco-2) cells. Xenobiotica 36: 739749.

Lakeram M, Lockley DJ, Sanders DJ, Pendlington R, Forbes B. 2007.Paraben transport and metabolism in the biomimetic articial membranepermeability assay (BAMPA) and 3-day and 21-day Caco-2 cell systems.J. Biomol. Screen. 12: 8491.

Lee BM, Koo HJ. 2007. Hershberger assay for antiandrogenic effects ofphthalates. J. Toxicol. Environ. Health A 70: 13651370.

Lee HB, Peart TE, Svoboda ML. 2005. Determination of endocrine-disrupting phenols, acidic pharmaceuticals, and personal-care productsin sewage by solid-phase extraction and gas chromatographymassspectrometry. J. Chromatogr. A 1094: 122129.

Lemini C, Silva G, Timossi C, Luque D, Valverde A, Gonzalez-Marti M,Hernandez A, Rubio-Poo C, Chavez Lara B and Valezuela F. 1997.Estrogenic effects of p-hydroxybenzoic acid in CD1 mice. Environ.Res. 75: 130134.

Lemini C, Jaimez R, Avila ME, Franco Y, Larrea F, Lemus AE. 2003.In vivo and in vitro estrogen bioactivities of alkyl parabens. Toxicol.Ind. Health. 19: 6979.

Lengauer C, Kinzler KW, Vogelstein B. 1998. Genetic instabilities inhuman cancers. Nature 396: 643649.

Lonning PE (ed.). 2004. Endocrinology and treatment of breast cancer.Clin. Endocrinol. Metab. 18: 1130.

Loretz L, Api AM, Barraj L, Burdick J, Davis DA, Dressler W, Gilberti E,Jarrett G, Mann S, Pan YHL, Re T, Renskers K, Scrafford C, VaterS. 2006. Exposure data for personal care products: hairspray, sprayperfume, liquid foundation, shampoo, body wash, and solid antiper-spirant. Food Chem. Toxicol. 44: 20082018.

Matthiessen P. 2003. Historical perspective on endocrine disruption inwildlife. Pure Appl. Chem. 75: 21972206.

Mbah CJ. 2007. Studies on the lipophilicity of vehicles (or co-vehicles)and botanical oils used in cosmetic products. Pharmazie 62: 351353.

McGrath KG. 2003. An earlier age of breast cancer diagnosis related tomore frequent use of antiperspirants/deodorants and underarm shav-ing. European Journal of Cancer Prevention 12: 479485.

McKim JM, Wilga PC, Breslin WJ, Plotzke KP, Gallavan RH, Meeks RG.2001. Potential estrogenic and antiestrogenic activity of the cyclicsiloxane ocatamethylcyclotetrasiloxane (D4) and the linear siloxanehexamethyldisiloxane (HMDS) in immature rats using the uterotro-phic assay. Toxicol. Sci. 63: 3746.

Mikula P, Dobsikova R, Svobodova Z, Jarovsky J. 2006. Evaluation ofxenoestrogenic potential of propylparaben in zebrash (Danio rerio).Neuro. Endocrinol. Lett. 27: 104107.

Miller WR. 1996. Estrogen and Breast Cancer. Chapman and Hall.Miller D, Wheals BB, Beresford N, Sumpter JP. 2001. Estrogenic activ-

ity of phenolic additives determined by an in vitro yeast bioassay.Environ. Health Perspect. 109: 133138.

Mirick DK, Davis S, Thomas DB. (2002). Antiperspirant use and therisk of breast cancer. J. Natl Cancer. Inst. 94: 15781580.

Mori T, Iida M, Ishibashi H, Kohra S, Takao Y, Takemasa T, Arizono K.2007. Hormonal activity of polycyclic musks evaluated by reportergene assay. Environ. Sci. 14: 195202.



Morohoshi K, Yamamoto H, Kamata R, Shiraishi F, Koda T, Morita M.2005. Estrogenic activity of 35 components of commercial sunscreenlotions evaluated by in vitro assays. Toxicol. in Vitro 19: 457469.

Moss T, Howes D, Willimas FM. 2000. Percutaneous penetration anddermal metabolism of triclosan (2,4,4-trichloro-2-hydroxydiphenylether) Food Chem. Toxicol. 38: 361370.

Nakagawa Y, Moldeus P. 1998. Mechanism of p-hydroxybenzoateester-induced mitochondrial dysfunction and cytotoxicity in isolatedrat hepatocytes. Biochem. Pharmac. 55: 19071914.

Neville JA, Welch E, Leffell DJ. 2007. Management of nonmelanomaskin cancer in 2007. Nat. Clin. Pract. Oncol. 4: 462469.

Nishihara T, Nishikawa J, Kanayama T, Dakeyama F, Saito K,Imagawa M, Takatori S, Kitagawa Y, Hori S, Utsumi H. 2000. Estro-genic activites of 517 chemicals by yeast two-hybrid assay. J. HealthSci. 46: 282298.

Nishizawa C, Takeshita K, Ueda J, Nakanishi I, Suzuki KT, Ozawa T.2006. Reaction of para-hydroxybenzoic acid esters with singlet oxygenin the presence of glutathione produces glutathione conjugates ofhydroquinone, potent inducers of oxidative stress. Free Radic. Res.40: 233240.

Oettel M, Schillinger E (eds). 1999. Estrogens and Antiestrogens I.Springer: Berlin; 1397.

Oh SY, Fujii M, Takeda Y, Yoda K,

toksikologi paraben

Documents

Transcript of toksikologi paraben