15619015 Hexavalent Chromium is Carcinogenic to F344N Rats and B6C3F1 Miceafter Chronic Oral...

7/29/2019 15619015 Hexavalent Chromium is Carcinogenic to F344N Rats and B6C3F1 Miceafter Chronic Oral Exposure

http://slidepdf.com/reader/full/15619015-hexavalent-chromium-is-carcinogenic-to-f344n-rats-and-b6c3f1-miceafter 1/7

716 volume 117 | number 5 | May 2009 • Environmental Health Perspectives

Research

Hexavalent chromium [Cr(VI)] compoundsare classied as human carcinogens [Cohenet al. 1993; International Agency or Researchon Cancer (IARC) 1990; National oxicology Program (NP) 1998], based on increasedincidences o lung cancers in animals and

humans ater inhalation exposures. Inhalationexposures occur primarily in occupational set-tings. Te highest exposure to Cr(VI) occursoccupationally among workers involved inchrome plating, chromate production, andstainless steel welding.

As a result o industrial contamination,concentrations o Cr(VI) in the drinking water and soil may be higher than concen-trations resulting rom natural sources alone,causing ingestion o higher concentrations by populations residing near these locations thanby the general population. A Cr(VI) concen-tration o 580 µg/L was ound in a ground-

water monitoring well in Hinkley, Caliornia(Pellerin and Booker 2000). Detectable lev-els o Cr(VI) have been reported in approxi-mately 30% o the drinking water sourcesmonitored in Caliornia, which has a 1 µg/Ldetection limit or purposes o reporting[Caliornia Department o Public Health(CDPH) 2007a]; 86% o those sources hadpeak concentrations o ≤ 10 µg/L. Te U.S.Environmental Protection Agency (EPA) hasset a maximum contaminant level o 100 µg/Lor total chromium in drinking water (U.S.EPA 2003). he limit in Caliornia and

numerous other states is 50 µg/L o total chro-mium in drinking water (CDPH 2007b).

Te toxicity o Cr(VI) in humans and ani-mals has been reviewed extensively (Agency oroxic Substances and Disease Registry 2000;Costa 1997; Costa and Klein 2006; U.S. EPA

1998). We identied only one lietime animalcarcinogenicity study in the literature in whichCr(VI) was administered in the drinking water(Borne et al. 1968). Further analysis o thedata (Sedman et al. 2006) revealed that inthree generations o emale NMRI mice, thecombined incidences o benign and malig-nant orestomach neoplasms were increasedover controls; this study has several limita-tions, including high early mortality in the F0 generation as a result o ectromelia (mouse-pox) virus. A review o the ew epidemiologicstudies that evaluated populations that wereexposed to Cr(VI) in drinking water or in soil

or slag ll concluded that these studies did notprovide denitive evidence o an increase incancer incidence or mortality rates (Proctoret al. 2002). However, a retrospective mortal-ity study in China ound higher incidences o lung and stomach neoplasms in people livingnear a chromium smelting plant compared with the general population (Zhang and Li1987); statistical analysis o these data sup-ported this conclusion (Beaumont et al. 2008;Sedman et al. 2006).

Because o concerns over its presence indrinking water source supplies, its potential

health eects, including carcinogenicity, andthe lack o adequate carcinogenicity studies by the oral route, the Caliornia CongressionalDelegation, Caliornia EnvironmentalProtection Agency, and Caliornia Departmento Health Services nominated Cr(VI) to theNP or toxicity and carcinogenicity testing. We selected sodium dichromate dihydrate(SDD) or testing because it is the primary basematerial or the production o chromium com-pounds, is widely used in industrial applica-tions, and is the most water-soluble chromate.

Te NP previously conducted 3-monthtoxicity studies o SDD administered indrinking water (NP 2007). F344/N ratsand B6C3F1 mice were exposed to 0, 62.5,125, 250, 500, or 1,000 mg SDD/L. In thesestudies, both rats and mice had reduced body weight and water consumption and microcytichypochromic anemia; the anemia was moresevere in rats. Exposure-related histopathologiclesions included ulcers, epithelial hyperplasia,and squamous metaplasia in the glandularstomach o rats; epithelial hyperplasia o theduodenum o mice; and histiocytic cellularinltration in the liver, duodenum, and pan-creatic lymph node o rats and in the duode-num and mesenteric lymph node o mice.

We selected exposure concentrations orthe 2-year studies o SDD ater review o the 3-month toxicity study data. Te highestexposure concentration or the 2-year stud-ies in rats and mice was limited by toxicity observed in the 3-month studies (NP 2007),and a wider spacing o exposure concentra-tions was selected to extend the dose–responsecurve. We added an additional low-exposuregroup [5 mg Cr(VI)/L] to the 2-year studiesto provide a concentration closer to humanexposure through contaminated drinking water. In this report we present the primary ndings o the NP chronic oral toxicity and

Address correspondence to M.J. Hooth, NIEHS,P.O. Box 12233, MD K2-13, Research rianglePark, NC 27709 USA. elephone: (919) 316-4643.Fax: (919) 541-4255. E-mail: [email protected] We thank P. Foster and S. Masten or critical review

o the manuscript.Tis research was supported in part by the Intramural

Research Program o the National Institutes o Health,National Institute o Environmental Health Sciences,under research project 1 Z01 ES045004-11 BB.

he authors declare they have no competingfnancial interests.

Received 19 September 2008; accepted 31 December2008.

Hexavalent Chromium Is Carcinogenic to F344/N Rats and B6C3F1 Miceafter Chronic Oral ExposureMatthew D. Stout, Ronald A. Herbert, Grace E. Kissling, Bradley J. Collins, Gregory S. Travlos, Kristine L. Witt,Ronald L. Melnick, Kamal M. Abdo, David E. Malarkey, and Michelle J. Hooth

National Toxicology Program, National Institute of Environmental Health Sciences, National Institutes of Health, Department of Healthand Human Services, Research Triangle Park, North Carolina, USA

B ackground: Hexavalent chromium [Cr(VI)] is a human carcinogen ater inhalation exposure.Humans also ingest Cr(VI) rom contaminated drinking water and soil; however, limited data existon the oral toxicity and carcinogenicity o Cr(VI).

oBjective: We characterized the chronic oral toxicity and carcinogenicity o Cr(VI) in rodents.

Methods: Te National oxicology Program (NP) conducted 2-year drinking water studies o Cr(VI) (as sodium dichromate dihydrate) in male and emale F344/N rats and B6C3F1 mice.

r esults: Cr(VI) exposure resulted in increased incidences o rare neoplasms o the squamous epi-thelium that lines the oral cavity (oral mucosa and tongue) in male and emale rats, and o the epi-thelium lining the small intestine in male and emale mice. Cr(VI) exposure did not aect survivalbut resulted in reduced mean body weights and water consumption, due at least in part to poor pal-atability o the dosed water. Cr(VI) exposure resulted in transient microcytic hypochromic anemia in rats and microcytosis in mice. Nonneoplastic lesions included diuse epithelial hyperplasia in theduodenum and jejunum o mice and histiocytic cell infltration in the duodenum, liver, and mesen-

teric and pancreatic lymph nodes o rats and mice.

conclusions: Cr(VI) was carcinogenic ater administration in drinking water to male and emalerats and mice.

k ey words: anemia, cancer, hexavalent chromium, histiocytic cellular infltration, National oxicology Program, oral cavity, small intestine. Environ Health Perspect 117:716–722(2009). doi:10.1289/ehp.0800208 available via http://dx.doi.org/ [Online 31 December 2008]



Oral carcinogenicity of hexavalent chromium

Environmental Health Perspectives • volume 117 | number 5 | May 2009 717

carcinogenesis studies o Cr(VI) in male andemale rats and mice; a more detailed reporton these studies is available as an NP techni-cal report (NP 2008a).

Materials and Methods

Chemical and dose ormulations. SDD (CAS7789-12-0) was obtained rom Aldrich

Chemical Company (Milwaukee, WI). hepurity was determined using dierential scan-ning calorimetry, titration o the dichromateion with sodium thiosulate and potassiumerrocyanide, speciation o the chromium ionsusing liquid chromatography/inductively cou-pled plasma-mass spectrometry, and poten-tiometric titrimetric analysis with sodiumthiosulate. Based on these analyses, the overallpurity was ≥ 99.7%. he dose ormulations were prepared approximately every 2 weeksby mixing SDD with tap water and stored atroom temperature in sealed containers pro-tected rom light. Te dose ormulations werestable or 42 days under these conditions.Periodic analysis by ultraviolet spectroscopy with detection at 350–390 nm conrmed thatall dose ormulations varied by < 10% o thetarget concentrations.

Animals and animal maintenance. hestudies were conducted at Southern ResearchInstitute (Birmingham, AL). Male and emaleF344/N rats and B6C3F1 mice were obtainedrom aconic Farms (Germantown, NY). Ratsand mice were quarantined or 14 days, and were 6–7 weeks o age at the beginning o thestudies. Animals were distributed randomly into groups o approximately equal initial meanbody weights and identied by tail tattoo. Rats

and mice were housed one (male mice), three(male rats), or ve (emale rats and mice) toa cage. Feed (irradiated NP-2000 waers;Zeigler Brothers, Inc., Gardners, PA) and tap water were available ad libitum. Animals werekilled by carbon dioxide asphyxiation.

Animal use was in accordance with theU.S. Public Health Service policy on humanecare and use o laboratory animals and theGuide for the Care and Use of Laboratory Animals (National Research Council 1996). Animal s were treated humane ly and withregard or alleviation o suering. hesestudies were conducted in compliance with

the Food and Drug Administration GoodLaboratory Practice Regulations (Food andDrug Administration 1987).

Study design. Groups o 50 male and50 emale rats and mice were exposed to SDDin drinking water at concentrations o 0, 14.3,57.3, 172, or 516 mg/L (male and emale ratsand emale mice) or 0, 14.3, 28.6, 85.7, or257.4 mg/L (male mice) or 105–106 weeks.able 1 shows corresponding Cr(VI) concen-trations. Water consumption was recorded weekly or the rst 13 weeks and every 4 weeksthereater, with each water consumption

measurement covering a 7-day period. Animals were weighed initially, weekly or the rst 13 weeks, at 4-week intervals thereater, and atthe end o the studies. Animals were observedtwice daily and clinical ndings were recordedat 4-week intervals beginning at week 5.

Complete necropsies and microscopicexaminations were perormed on all core study

rats and mice. At necropsy, all organs and tis-sues were examined or grossly visible lesions,and all protocol-required tissues were xed andpreserved in 10% neutral buered ormalin(eyes were initially ixed in Davidson’s solu-tion), trimmed and processed, embedded inparan, sectioned to a thickness o 4–6 µm,and stained with hematoxylin and eosin (H&E)or microscopic examination. For all pairedorgans (e.g., adrenal gland, kidney, ovary),samples rom each organ were examined. Teentire gastrointestinal tract was opened andpotential lesions were collected or microscopicevaluation. Oral mucosa and tongue are notprotocol-required tissues; however, becausegross lesions in these tissues were diagnosedas neoplasms, the oral mucosa and tongueo all animals were evaluated histologically. Additional details regarding the pathology datageneration, quality assurance review, and NPpathology working group are available in theNP technical report (NP 2008a). Details o these review procedures have been described, inpart, by Maronpot and Boorman (1982) andBoorman et al. (1985). For subsequent analyseso the pathology data, the decision o whetherto evaluate the diagnosed lesions or each tissuetype separately or combined was based on theguidelines o McConnell et al. (1986).

Hematology and clinical chemistry wereevaluated in additional groups o 10–16 male

rats and emale mice on days 4 (male ratsonly) and 22 and at 3, 6, and 12 months. Atthese time points, rats and mice were anes-thetized with CO2/O2, and blood was takenrom the retroorbital sinus.

Statistical methods. We est imated theprobability o survival by the product-limit pro-cedure o Kaplan and Meier (1958). Animals

ound dead o other than natural causes ormissing were censored rom the survival analy-ses; animals dying rom natural causes were notcensored. Statistical analyses or possible dose-related eects on survival used Cox’s (1972)method or testing two groups or equality andarone’s (1975) lie table test to identiy dose-related trends. All reported p-values or thesurvival analyses are two sided.

We used the poly-k test (Bailer and Portier1988; Piegorsch and Bailer 1997; Portier andBailer 1989) to assess neoplasm and non-neoplastic lesion incidence. Tis test is a sur-vival-adjusted quantal-response procedure thatmodiies the Cochran-Armitage linear trendtest to take survival dierences into account; we used a value o k = 3 in the analysis o site-specic lesions. ests o signicance includedpairwise comparisons o each exposed group with controls and a test or an overall exposure-related trend. We used continuity-correctedpoly-3 tests in the analysis o lesion incidence,and reported p-values are one sided. Sub- andsupralinear trends across doses or intestinaltumor rates were assessed with lack-o-t testsor the linear regression o intestinal tumor rateon dose (Neter et al. 1996).

We ana lyzed hematology and clinic alchemistry data, which typically have skewed

distributions, using the modiied (Dunn1964; Williams 1986) nonparametric multiple

Table 1. Concentrations in drinking water and average daily ingested doses o SDD and Cr(VI) ater expo-sure or 2 years.

Concentration in drinking water (mg/L) Average daily ingested dose (mg/kg)

Animals SDD Cr(VI)a SDDb Cr(VI)c

Male rats 0 0 0 014.3 5 0.6 0.257.3 20 2.2 0.8

172 60 6 2.1516 180 17 5.9

Female rats 0 0 0 014.3 5 0.7 0.257.3 20 2.7 0.9

172 60 7 2.4516 180 20 7.0

Male mice 0 0 0 014.3 5 1.1 0.428.6 10 2.6 0.985.7 30 7 2.4

257.4 90 17 5.9Female mice 0 0 0 0

14.3 5 1.1 0.457.3 20 3.9 1.4

172 60 9 3.1516 180 25 8.7

a Calculated using the drinking water concentration o SDD and the percent mass o Cr(VI) in SDD. b Calculated using bodyweight and water consumption data. c Calculated using the average daily dose o SDD and the percent mass o Cr(VI) in SDD.



Stout et al.


comparison methods o Shirley (1977). Weused Jonckheere’s test (Jonckheere 1954) toassess the signicance o the dose-related trendsand to determine whether a trend-sensitivetest (Shirley 1977; Williams 1986) was moreappropriate or pairwise comparisons than atest that does not assume a monotonic dose-related trend (Dunn 1964; Dunnett 1955).

Beore statistical analysis, we examined extremevalues identied by the outlier test o Dixonand Massey (1957) and eliminated implausiblevalues rom the analysis.

Results

Oral cavity neoplasms in rats. Male andemale rats showed signiicantly increasedincidences o highly aggressive neoplasms o the squamous epithelium that lines the oralcavity (oral mucosa and tongue) at 516 mg/L(able 2). Specically, we observed increasesor squamous cell carcinoma in the oralmucosa and or squamous cell papilloma orcarcinoma (combined) o the oral mucosa ortongue o male and emale rats at 516 mg/L.here were squamous cell carcinomas in theoral mucosa o two 172 mg/L emale rats; thisincidence exceeded the historical control rangesor drinking water studies and or all routes o administration (able 2). We did not observenon-neoplastic lesions in the oral mucosa.

Microscopically, the squamous cell carci-nomas were highly invasive, irregular masses.ypically, the carcinomas appeared to origi-nate in the oral mucosa o the palate adja-cent to the upper molar teeth (Figure 1A, B);in some animals they invaded the tongue,Harderian gland, and the sot tissues sur-

rounding the nose, and in one case it pene-trated the maxilla and invaded the brain. Tesquamous cell papillomas o the oral mucosaand tongue were exophytic masses that

projected rom the mucosa and consisted o irregular papillary prolierations o maturesquamous epithelium supported by a core o brovascular stroma (Figure 1C).

Small intestine neoplasms and hyperpla-sia in mice. Both males and emales showeda clear exposure concentration response orincreased incidences o adenoma or carcinoma

(combined) at all sites (combined) o the smallintestine (duodenum, jejunum, or ileum;able 3). Tese increases were signicant at85.7 and 257.4 mg/L in males and at 172 and516 mg/L emales, the two highest exposureconcentrations in each sex. In addition, theincidence in 57.3 mg/L emales exceeded thehistorical control ranges or drinking waterstudies and or all routes o administration(able 3). Tese increases were driven primar-ily by signicant increases in the incidenceso adenoma o the duodenum in 257.4 mg/Lmales and in 172 and 516 mg/L emales; thenumber o mice with multiple adenomas wasalso signicantly increased at the high dosein both sexes (data not shown). In emales,the incidence o carcinoma in the duodenum was signiicantly increased at 516 mg/L. Inthe jejunum, the incidence o adenoma wasincreased in 516 mg/L emales. For bothmales and emales, we tested the combinedincidence o adenoma and carcinoma ordepartures rom a linear dose response (datanot shown). For males, the trend was linear, with no signicant departure rom linearity.For emales, the response was supralinear witha signicant departure rom linearity.

Adenomas were discre te, broad based,ocally extensive, plaquelike areas o proli-

erating glandular epithelium that thickenedthe mucosa and protruded into the lumen(Figure 1D). Carcinomas were sessile, plaque-like neoplasms distinguished rom adenomas

by extensive invasion and eacement o themucosa, underlying submucosa, and musclelayers (Figure 1E).

Low incidences o ocal epithelial hyper-plasia occurred in the duodenum o exposedmale and emale mice (able 4). Althoughthe increased incidences were not exposureconcentration related or statistically signii-

cant, we considered this lesion a preneoplasticlesion related to exposure to SDD because o its morphologic similarities to adenoma. Focalepithelial hyperplasias were ocal areas o pro-lierating glandular epithelium distinguishedrom adenomas because they were smaller,less discrete, and conned to the supercialmucosal epithelium (Figure 1F).

Te incidences o diuse epithelial hyper-plasia were signicantly increased in the duo-denum o all exposed groups o male andemale mice (able 4). In the jejunum, theincidence o diuse epithelial hyperplasia wassignicantly increased in 516 mg/L emales. Incontrast to those o the controls (Figure 1G),the duodenums o exposed mice had short,broad, blunt villi and generalized mucosalhypercellularity that was particularly promi-nent in the villi (Figure 1H).

Histiocytic inltration in rats and mice. We ound signicant increases in histiocyticcell inltration in the duodenum and mesen-teric lymph node o rats and mice o bothsexes, in the jejunum o emale mice, in theliver o male and emale rats and emale mice,and in the pancreatic lymph node o emalerats and male and emale mice; these dataare presented in detail in the NP technicalreport (NP 2008a).

Te inltrating histiocytes were morpho-logically similar in tissues o rats and mice.Histiocytic iniltrates were characterized by the presence o individual, small clusters, and

Table 2. Squamous cell neoplasms o the oral cavity (oral mucosa and tongue) o F344/N rats exposed to SDD or 2 years in drinking water.

Historical control incidencea (range) Incidence/no. of animals necropsied (survival-adjusted percent incidence)b

Tissue/neoplasm Drinking water All routes 0 mg/L 14.3 mg/L 57.3 mg/L 172 mg/L 516 mg/L

MalesOral mucosa

Carcinoma 0/350 5/1,499 (0–2%) 0/50 # 0/50 0/49 0/50 6/49 (13.6)*Tongue

Papilloma NDc ND 0/50 0/50 0/49 0/50 1/49 (2.3)Carcinoma ND ND 0/50 1/50 (2.4) 0/49 0/50 0/49

Oral mucosa or tongue

Papilloma or carcinoma (combined) 1/350 (0–2%) 10/1,499 (0–2%) 0/50 #

1/50 (2.4) 0/49 0/50 7/49 (15.7)**FemalesOral mucosa

Carcinoma 0/300 5/1,400 (0–2%) 0/50 # 0/50 0/50 2/50d (4.6) 11/50 (23.9) #

TonguePapilloma ND ND 1/50 (2.2) 1/50 (2.3) 0/50 0/50 0/50Carcinoma ND ND 0/50 0/50 0/50 1/50 (2.3) 0/50

Oral mucosa or tonguePapilloma or carcinoma (combined) 4/300 (0–2%) 15/1,400 (0–6%) 1/50 (2.2) # 1/50 (2.3) 0/50 2/50e (4.6) 11/50 (23.9)**

ND, not determined.a The NTP historical database contains all studies that use the NTP-2000 diet with histopathology fndings completed within the most recent 5-year period, including the present study.b Calculated using the poly-3 test. c Historical control incidences are not determined or the tongue because it is not an NTP protocol-required tissue. d The incidence exceeded the his-

torical control range or both drinking water studies and all routes but was not signifcantly increased compared with the concurrent control. e The incidence exceeded the historicalcontrol range or drinking water studies but was not signifcantly increased compared with the concurrent control. * p ≤ 0.05, **p ≤ 0.01, and #p ≤ 0.001 compared with the control groupby poly-3 test or a signifcant trend i assigned to a control group.





sometimes syncytia o large histiocytes (mac-rophages) within the sinusoids o the liver andlymph nodes and the lamina propria at the tipso the duodenal and jejunal villi. In the lymphnodes, the histiocytic iniltrates occurred asexpansive sheets that in some cases replacedmuch o the lymph node parenchyma. hebiological signicance and cause o the histio-

cytic cellular inltrates are unknown.Hematology in rats and mice. In rats, anexposure-related decrease in mean cell vol-umes, mean cell hemoglobin concentrations,hematocrits, hemoglobin concentrations, anderythrocyte counts and increase in reticulocytecounts were indicative o erythrocyte micro-cytic, hypochromic, responsive anemia; thesedata are presented in detail in the NP tech-nical report (NP 2008a). he anemia wasmost prominent early in the study (22 days to3 months). Microscopic evaluations o bloodsmears demonstrated increased poikilocytes,erythrocyte ragments/schizocytes, keratocytes,erythrocyte hypochromia, and microcytes thatsuggested increased erythrocyte injury or turn-over. Tis eect was more prominent on day 22and at 3 months than at 6 or 12 months, whichmay have been either a result o the animalsadapting to exposure or a result o the decreasedCr(IV) ingestion per unit body weight withlonger exposure durations. Mice demonstrateda similar, but less severe, eect on erythron.

In lie efects in rats and mice. Survival,body weight, and water consumption dataare presented in detail in the NP techni-cal report (NP 2008a). Survival o exposedgroups o male and emale rats and mice was similar to that o the respective control

groups. Mean body weights compared withcontrols were decreased in male and emalerats and mice. In rats, mean body weights o 516 mg/L males and emales were less thanthose o controls throughout the study andby the end o the study were 12% (males) or11% (emales) less than the controls. Meanbody weights o 257.4 mg/L male mice wereless than controls or the rst 4 months o thestudy but were only slightly less (6%) by theend o the study. Mean body weights wereless than the controls rom months 3 to 12in 172 mg/L emale mice and rom month2 until the end o the study in 516 mg/L

emales. By the end o the study, mean body weights were < 8% o controls in 172 mg/Lemales and 15% less in 516 mg/L emales.

We attributed the lower body weights partly to poor palatability o the dosed water andconsequent reductions in water consumptionrather than direct toxic eects o SDD expo-sure. Water consumption by male and emalerats and mice exposed to the two highest con-centrations was that by the controls throughoutthe study. No clinical ndings were attributedto SDD exposure in rats or mice. When weadjusted water consumption or body weight

Figure 1. (A and B ) Male rat given 516 mg/L SDD or 2 years. (A) Low magnifcation o a section o the nasalcavity demonstrating the location o squamous cell carcinoma (arrows) arising rom the oral mucosa o

the sot palate; the neoplasm has invaded the adjacent submucosal tissue and surrounded a molar tooth.(B ) Higher magnifcation demonstrating the malignant eatures o the squamous cell carcinoma; islandsand cords o dysplastic squamous epithelium (arrows) are surrounded by dense prolierative connective tissue stroma. (C ) Female rat given 14.3 mg/L SDD or 2 years; squamous cell carcinoma projects rom thedorsal mucosal surace o the tongue (arrows). (D ) Female mouse given 516 mg/L SDD or 2 years; adenomao the duodenum (arrows) has distorted and replaced a segment o the mucosa and protruded into thelumen. Normal mucosa is visible opposite the adenoma. (E ) Male mouse given 257.4 mg/L SDD or 2 years;carcinoma o the duodenum (arrows) has eaced the mucosa, invading the submucosa, muscle layers, andpancreas. (F ) Male mouse given 172 mg/L SDD or 2 years; ocal hyperplasia o the duodenum (arrows) ispresent in the superfcial mucosa. (G ) Control emale mouse; duodenum demonstrates normal microscopicanatomy. Note tall, slender villi (arrows) lined by a single layer o tall columnar epithelial cells. (H ) Femalemouse given 516 mg/L SDD or 2 years; diuse hyperplasia is present in the duodenum. Duodenal villi areshort, wide, and blunt and lined by hyperplastic epithelial cells that are piling up along the villi (arrows). Notehistiocyte infltrates expanding the lamina propria at the tips o the villi (asterisks). All sections are stainedwith hematoxylin and eosin. Bars: A and C, 500 µm; B and F, 100 µm; D and E, 200 µm; G and H, 50 µm.



Stout et al.


(data not shown), dosed male and emale ratsand emale mice drank approximately the samequantities o water per gram o body weightas did the controls ater the rst 20 weeks onstudy. Male mice exposed to 257.4 mg/L drank less water per gram o body weight than did thecontrols throughout the study.

DiscussionHumans are exposed to Cr(VI) through inges-tion o contaminated water and soil; however,ew data exist on the oral toxicity and carcinoge-nicity o Cr(VI). Te NP conducted 3-month(NP 2007) and 2-year (NP 2008a) studieso SDD administered in the drinking waterto F344/N rats and B6C3F1 mice, to providedata on the potential or toxic and carcinogeniceects ater ingestion o Cr(VI).

Chronic administration o SDD in drink-ing water did not aect survival or produceclinical signs o toxicity in rats or mice. Weobserved exposure-related reductions in body weight gain and water consumption or ratsand mice in the highest exposure groupsand attributed these changes partly to poorpalatability o the dosed water. Several lineso evidence suggest that the animals were notdehydrated, including analysis o the water con-sumption data normalized to body weight andthe complete lack o clinical observations orhematologic or clinical chemistry eects (NP2008a) that typically indicate dehydration.

he NP concluded that the exposureconcentration-related signicant increases inepithelial neoplasms o the upper alimentary

tract (oral cavity) in male and emale rats ando the lower alimentary tract (small intestine)in male and emale mice provided clear evi-dence o carcinogenic activity o SDD in maleand emale rats and mice. We based this con-clusion on the increased neoplasm incidencesrelative to concurrent controls and the rarity o these neoplasms (ables 2 and 3) in his-

torical controls. In both rats and mice, thisconclusion was strengthened by similaritiesbetween the sexes. We observed no increasesin nonneoplastic histopathologic lesions ineither species suggestive o overt tissue dam-age due to the oxidant properties o Cr(VI).

We observed obvious species dierencesin the target tissues or the development o

neoplasms between rats and mice. O the21 chemicals that have caused neoplasms o the oral cavity in NP studies, none producedthese neoplasms in male mice and only one,1,2,3-trichloropropane (NP 1993), producedoral cavity neoplasms in emale mice, demon-strating a greater sensitivity to the develop-ment o oral cavity neoplasms in rats relative to

mice. Although slightly more common in rats,exposure-related increases o small intestineneoplasms in NP studies are relatively rarein both species. he 2-year study o captan(National Cancer Institute 1977) is the only other study perormed by the NP in B6C3F1mice in which both benign and malignantintestinal neoplasms o epithelial origin have

Table 3. Epithelial neoplasms o the small intestine in B6C3F1 mice exposed to SDD in drinking water or 2 years.

Historical control incidencea (range)

Tissue/neoplasm Drinking water All routes Incidence/no. of animals necropsied (survival-adjusted % incidence)b

MalesConcentration (mg/L) 0 14.3 28.6 85.7 257.4Duodenum

Adenoma (includes multiple) 6/299 (0–6%) 9/1,549 (0–6%) 1/50 (2.2) # 0/50 1/50 (2.3) 5/50c (10.8) 15/50 (32.9) #

Carcinoma 1/299 (0–2%) 3/1,549 (0–4%) 0/50* 0/50 0/50 2/50d (4.3) 3/50c (6.8)Jejunum

Adenoma 0/299 1/1,549 (0–2%) 0/50** 0/50 0/50 0/50 3/50c (6.8)Carcinoma (includes multiple) 5/299 (0–4%) 25/1,549 (0–8%) 0/50 2/50 (4.5) 0/50 1/50 (2.2) 2/50 (4.6)

Duodenum, jejunum, or ileum (combined)Adenoma 6/299 (0–6%) 10/1,549 (0–6%) 1/50 (2.2) # 1/50 (2.3) 1/50 (2.3) 5/50c (10.8) 17/50 (37.2) #

Carcinoma 6/299 (0–4%) 30/1,549 (0–8%) 0/50* 2/50 (4.5) 1/50 (2.3) 3/50d (6.5) 5/50 (11.4)*Adenoma or carcinoma (combined) 11/299 (0–10%) 39/1,549 (0–10%) 1/50 (2.2) # 3/50 (6.8) 2/50 (4.6) 7/50 (15.1)* 20/50 (43.8) #

Females

Concentration (mg/L) 0 14.3 57.3 172 516DuodenumAdenoma (includes multiple) 1/350 (0–2%) 3/1,648 (0–2%) 0/50 # 0/50 2/50c (4.2) 13/50 (27.8) # 12/50 (25.2) #

Carcinoma 0/350 1/1,648 (0–2%) 0/50 # 0/50 0/50 1/50d (2.1) 6/50 (12.6)*Jejunum

Adenoma (includes multiple) 0/350 0/1,648 0/50** 1/50c (2.2) 0/50 2/50c (4.3) 5/50 (10.6)*Carcinoma 2/350 (0–2%) 5/1,648 (0–2%) 1/50 (2.2) 0/50 2/50c (4.2) 2/50c (4.3) 1/50 (2.1)

Duodenum, jejunum, or ileum (combined)Adenoma 1/350 (0–2%) 3/1,648 (0–2%) 0/50 # 1/50 (2.2) 2/50c (4.2) 15/50 (32.0) # 16/50 (33.7) #

Carcinoma 3/350 (0–2%) 8/1,648 (0–2%) 1/50 (2.2) # 0/50 2/50c (4.2) 3/50c (6.4) 7/50 (14.7)*Adenoma or carcinoma (combined) 4/350 (0–4%) 11/1,648 (0–4%) 1/50 (2.2) # 1/50 (2.2) 4/50c (8.3) 17/50 (36.3) # 22/50 (45.9) #

a The NTP historical database contains all studies that use the NTP-2000 diet with histopathology fndings completed within the most recent 5-year period, including the present study.b Calculated using the poly-3 test. c The incidence exceeded the historical control range or both drinking water studies and all routes but was not signifcantly increased comparedwith the concurrent control. d The incidence exceeded the historical control range or drinking water studies but was not signifcantly increased compared with the concurrent control.*p ≤ 0.05, **p ≤ 0.01, and #p ≤ 0.001 compared with control group by poly-3 test or a signifcant trend i assigned to a control group.

Table 4. Focal and diuse hyperplasia o the small intestine in B6C3F1 mice exposed to SDD in drinkingwater or 2 years.

Tissue Incidence/no. of animals necropsied (mean severity)a

MalesConcentration (mg/L) 0 14.3 28.6 85.7 257.4

DuodenumEpithelium, hyperplasia

Focal 0/50 0/50 0/50 1/50 (3.0) 2/50 (3.5)Diffuse 0/50 11/50** (2.0) 18/50** (1.6) 42/50** (2.1) 32/50** (2.1)

FemalesConcentration (mg/L) 0 14.3 57.3 172 516Duodenum

Epithelium, hyperplasiaFocal 0/50 0/50 1/50 (2.0) 2/50 (3.0) 0/50Diffuse 0/50 16/50** (1.6) 35/50** (1.7) 31/50** (1.6) 42/50** (2.2)

JejunumEpithelium, hyperplasia

Diffuse 0/50 2/50 (2.0) 1/50 (2.0) 0/50 8/50** (1.9)

a Mean severity: 1, minimal; 2, mild; 3, moderate; 4, marked. **p ≤ 0.01 by poly-3 test.





been denitely attributed to chemical exposure(Shackelord and Elwell 1999).

Although the induction o neoplasms aterexposure to SDD was limited to the alimen-tary tract, other data, including the toxicity tothe erythron, provided evidence o systemicexposure and toxicity in male and emale ratsand mice exposed to Cr(VI) or 2 years. We

also observed these lesions in the 3-monthstudies (NP 2007). As par t o the NP 2-year studies on

SDD (NP 2008a) and chromium picolinatemonohydrate (CPM) (NP 2008b), whichcontains trivalent chromium [Cr(III)], totalchromium content was determined in selectedtissues and excreta o additional groups o male rats and emale mice; these data will bepresented in detail in an additional report. Tegoal o these studies was to examine the tissueuptake and distribution o Cr(VI) and Cr(III).Because Cr(VI) is reduced to Cr(III) bothintracellularly and extracellularly and becauseanalytical methods or the separate analysiso Cr(VI) or Cr(III) in biological samples arenot available, the speciation o the tissue chro-mium ater exposure to Cr(VI) was inerredby comparing total chromium concentrationsin tissues o rats and mice exposed to similardoses o Cr(VI) or Cr(III). Ater oral exposureto Cr(VI), chromium accumulation was cor-related with exposure concentration and dura-tion in several tissues (NP 2008a). Similardoses o Cr(VI) and Cr(III) resulted in signi-cantly higher tissue chromium concentrations with Cr(VI), indicating that chromium wasabsorbed and distributed to tissues o rats andmice as Cr(VI); these data are consistent with

previous studies (Costa 1997; Costa and Klein2006). he tissue concentration data wereconsistent with linear or supralinear (decreas-ing rate o response with increasing dose) doseresponses. In the present studies, neither theoral cavity nor the small intestine was collectedor total chromium analysis. However, otherreports suggest that Cr(VI) is also likely to beabsorbed in the small intestine to a greaterextent than Cr(III) (Donaldson and Barreras1966; Febel et al. 2001).

Reduction o Cr(VI) to the less perme-able and bioavailable Cr(III) is thought tooccur primarily in the stomach, as a mecha-

nism o detoxication. Gastric reduction hasbeen hypothesized to be ecient, such thatoral exposure to Cr(VI) would not result intoxicity or carcinogenicity, except perhaps inthe stomach (De Flora 2000; De Flora et al.1997; Proctor et al. 2002). Notably, in the2-year study, no neoplasms or nonneoplasticlesions were observed in the orestomach orglandular stomach o rats or mice. However,the observed increases in neoplasms o thesmall intestine o mice and the toxicity to theerythron, histiocytic inltration, and uptakeo Cr(IV) into tissues o rats and mice suggest

that, under the conditions o this study, atleast a portion o the administered Cr(VI) wasnot reduced in the stomach. Te signicantdisparity in the oral toxicity and carcinogenic-ity o Cr(VI) and Cr(III) in rodents, includ-ing the absence o increases in neoplasms ornonneoplastic lesions o the small intestine inrats or mice exposed to CPM (NP 2008b),

provides additional evidence that Cr(VI) isnot completely reduced in the stomach and isresponsible or the observed eects.

Recently, De Flora et al. (2008) havesuggested that increases in neoplasms o thesmall intestine observed in mice in the pres-ent study are the result o saturation o thegastric reduction capacity. I such a thresholdmechanism were to occur, the dose that satu-rated the reducing capacity would likely rep-resent an infection point on a sublinear doseresponse curve, with doses above the inlec-tion point demonstrating an increasing rateo response per unit dose, because unreducedchromium would be transported into tissues.However, when we tested tissue concentrationand mouse small intestine neoplasm data orlinearity, data that were statistically nonlinear were supralinear (decreasing rate o responseper unit dose).

A reduction capacity o about 84–88 mgCr(VI)/day has been estimated or human gas-tric juice (De Flora et al. 1997). Tis estimate was based on reported values o human secre-tion o gastric fuid per day during asting andater consuming three meals per day in com-bination with experimental data on reductiono Cr(VI)/mL o gastric juice produced duringthese periods. Similar data are not available

or Cr(VI) reduction by mouse gastric juice.However, assuming that Cr(VI) reduction isequally eective in mice and humans and thatgastric secretion scales across species by body weight3/4, then the Cr(VI) reduction capac-ity o gastric juice rom a 50-g mouse wouldbe approximately 0.4 mg/day (~ 8 mg/kg/day). Tis value is greater than all o the malemouse doses and is nearly equivalent to theaverage daily dose o Cr(VI) in the high-dosegroup o emale mice in the 2-year drinking water study o SDD (able 1). Collectively,the dose–response analysis and gastric reduc-tion capacity calculations indicate that SDD

induced neoplasms in the small intestine o mice at dose levels that did not exceed theestimated Cr(VI) reducing capacity or gastric juices in mice.

Cr(VI) is genotoxic in a number o in vitro and in vivo test systems (De Floraet al. 1990; IARC 1990); however, themechanisms o genotoxicity and carcinogenic-ity are not ully understood. Because Cr(VI)as chromate structurally resembles sulate andphosphate, it can be taken up by all cells andorgans throughout the body through non-speciic anion transporters (Costa 1997).

Once inside the cell, indirect DNA damagemay occur through the generation o oxy-gen radicals during intracellular reduction o Cr(VI) through the more reactive pentavalentand tetravalent chromium to Cr(III) (O’Brienet al. 2003); however, evidence o the role o reactive oxygen species in the genotoxicity o Cr(VI) is inconsistent (Chorvatovicova et al.

1991; O’Brien et al. 2003; Standeven and Wetterhahn 1991; Zhitkovich 2005). Cr(III),the nal product o intracellular reduction o Cr(VI), has been shown to interact directly with DNA and other macromolecul es toinduce chromosomal alterations and muta-tional changes (O’Brien et al. 2003; Quievrynet al. 2006; Reynolds et al. 2007; Zhitkovich2005). DNA adducts, DNA–protein cross-links, and DNA interstrand cross-links haveall been identiied as products o Cr(III)–DNA interactions. Te relative contributionso the multiple, complex pathways o chro-mium-induced genotoxicity continue to beinvestigated.

In conclusion, the NP 2-year study o SDD is the rst and only lietime study thatclearly demonstrates the carcinogenicity o Cr(VI) in rats and mice ater oral exposure.In addition, the hematology, histologic andtissue distribution data provide evidence o systemic exposure in rats and mice.

RefeRences

Agency or Toxic Substances and Disease Registry. 2000.

Toxicological Profle or Chromium. Atlanta, GA:Agency or

Toxic Substances and Disease Registry.

Bailer AJ, Portier CJ. 1988. Eects o treatment-induced mor-

tality and tumo r-induced mortality on tes ts o r carci no-

genicity in small samples. Biometrics 44:417–431.Beaumont JJ, Sedman RM, Reynolds SD, Sherman CD, Li LH,

Howd RA, et al. 2008. Cancer mortality in a Chinese popu-

lation exposed to hexavalent chromium in drinking water.

Epidemiology 19:12–23.

Boorman GA, Montgomery CA Jr, Eustis SL, Wole MJ,

McConnell EE, Hardisty JF. 1985. Quality assurance in

pathology or rodent carcinogenicity studies. In: Handbook

o Carcinogen Testing (Milman HA, Weisburger EK, eds).

Park Ridge, NJ:Noyes Publications, 345–357.

Borne J, Engelhardt K, Griem W, Kunte H, Reichert J. 1968.

Carcinogens in water and soil. XXII. Experiment with

3,4-benzopyrene and potassium chromate in mice drink [in

German]. Arch Hyg Bakteriol 152:45–53.

CDPH (Caliornia Department o Public Health). 2007a. Chromium-6

in Drinking Water Standard: Sampling Results. Available:

http://ww2.cdph.ca.gov/certlic/drinkingwater/Pages/

Chromium6sampling.aspx [accessed 10 November 2008].

CDPH (Caliornia Department o Public Health). 2007b.

Chromium-6 in Drinking Water: Regulation Update.Available: http://ww2.cdph.ca.gov/certlic/drinkingwater/

Pages/Chromium6.aspx [accessed 10 November 2008].

Chorvatovicova D, Ginter E, Kosinova A, Zloch Z. 1991. Eect o

vitamins C and E on toxicity and mutagenicity o hexa valent

chromium in rat and guinea pig. Mutat Res 262:41–46.

Cohen MD, Kargacin B, Klein CB, Costa M. 1993. Mechanisms

o chromium carcinogenicity and toxicity. Crit Rev Toxicol

23:255–281.

Costa M. 1997. Toxicity and carcinogenicity o Cr(VI) in animal

models and humans. Crit Rev Toxicol 27:431–442.

Costa M, Klein CB. 2006. Toxicity and carcinogenicity o chro-

mium compounds in humans. Crit Rev Toxicol 36:155–163.

Cox DR. 1972. Regression models and lie-tables. J R Stat Soc

B34:187–220.

De Flora S. 2000. Threshold mechanisms and site specifcity in

chromium(VI) carcinogenesis. Carcinogenesis 21:533–541.



Stout et al.


De Flora S, Bagnasco M, Serra D, Zanacchi P. 1990.

Genotoxicity o chromium compounds. A review. Mutat

Res 238:99–172.

De Flora S, Camoirano A, Bagnasco M, Bennicilli C, Corbett GE,

Kerger BD. 1997. Estimates o the chromium (VI) reducing

capacity in human body compartments as a mechanism

or attenuating its potential toxicity and carcinogenicity.

Carcinogenesis 18:531–537.

De Flora S, D’Agostini F, Balansky R, Micale R, Baluce B,

Izzotti A. 2008. Lack o genotoxic eects in hematopoietic

and gastrointestinal cells o mice receiving chromium(VI)

with the drinking water. Mutat Res 659:60–67.

Dixon WJ, Massey FJ Jr. 1957. Introduction to Statistical

Analysis. 2nd ed. New York:McGraw-Hill.

Donaldson RM Jr, Barreras RF. 1966. Intestinal absorption o

trace quantities o chromium. J Lab Clin Med 68:484–493.

Dunn OJ. 1964. Multiple comparisons using rank sums.

Technometrics 6:241–252.

Dunnett CW. 1955. A multiple comparison procedure or com-

paring several treatments with a control. J Am Stat Assoc

50:1096–1121.

Febel H, Szegedi B, Huszar S. 2001. Absorption o inorganic,

trival ent and hexaval ent chromium ollowi ng oral and

intrajejunal doses in rats. Acta Vet Hung 49:203–209.

Food and Drug Administration. 1987. Good laboratory practice regu-

lations. Final rule. 21 CFR 58. Fed Reg 52(172):33768–33782.

International Agency or Research on Cancer. 1990. Chromium.

IARC Monogr Eval Carcinog Risks Hum 49:49–256.

Jonckheere AR. 1954. A distribution-ree k -sample test against

ordered alternatives. Biometrika 41:133–145.

Kaplan EL, Meier P. 1958. Nonparametric estimation rom

incomplete observations. J Am Stat Assoc 53:457–481.

Maronpot RR, Boorman GA. 1982. Interpretation o rodent hepato-

cellular prolierative alterations and hepatocellular tumors in

chemical saety assessment. Toxicol Pathol 10:71–80.

McConnell EE, Solleveld HA, Swenberg JA, Boorman GA. 1986.

Guidelines or combining neoplasms or evaluation o rodent

carcinogenesis studies. J Natl Cancer Inst 76:283–289.

National Cancer Institute. 1977. Bioassay o Captan or

Possible Carcinogenicity (CAS No. 133-06-2). Technical

Report 15. Bethesda, MD:National Cancer Institute.

National Research Council. 1996. Guide or the Care and

Use o Laboratory Animals. Washington, DC:National

Academies Press.

Neter J, Kutner MH, Nachtsheim CJ, Wasserman W. 1996.

Applied Linear Statistical Models. 4th ed. Boston:WCB

McGraw-Hill, 115–124.

NTP. 1993. Toxicology and Carcinogenesis Studies o

1,2,3-Trichloropropane (CAS No. 96-18-4) in F344/N Rats

and B6C3F1 Mice (Gavage Studies). TR 384. Research

Triangle Park, NC:National Toxicology Program.

NTP. 1998. Report on Carcinogens. 8th ed. Research Triangle

Park, NC:National Toxicology Program.

NTP. 2007. Toxicity Studies o Sodium Dichromate Dihydrate

(CAS No. 7789-12-0) Administered in Drinking Water

to Mal e and Female F344/N Rats an d B6C3F 1 Mice and

Male BALB/c and am3 -C57BL/6 Mice. TOX 72. Research


NTP. 2008a. Toxicology and Carcinogenesis Studies o Sodium

Dichromate Dihydrate (CAS No. 7789-12-0) in F344/N

Rats and B6C3F1 Mice (Drinking Water Studies). TR 546.

Research Triangle Park, NC:National Toxicology Program.

NTP. 2008b. Toxicology and Carcinogenesis Studies o Chromium

Picolinate Monohydrate (CAS No. 27882-76-4) in F344/N Rats

and B6C3F1 Mice (Feed Studies) [drat]. TR 556. Research


O’Brien TJ, Ceryak S, Patierno SR. 2003. Complexities o chro-

mium carcinogenesis: Role o cellular response, repair,

and recovery mechanisms. Mutat Res 533:3–36.

Pellerin C, Booker SM. 2000. Relections on hexavalent

chromium: health hazards o an industrial heavyweight.

Environ Health Perspect 108:A402–A407.

Piegorsch WW, Bailer AJ. 1997. Statistics or Environmental

Biology and Toxicology. London:Chapman and Hall.

Portier CJ, Bailer AJ. 1989. Testing or increased carcino-

genicity using a survival-adjusted quantal response test.

Fundam Appl Toxicol 12:731–737.

Proctor DM, Otani JM, Finley BL, Paustenbach DJ, Bland JA,

Speizer N, et al. 2002. Is hexavalent chromium carcinogenic

via ingestion? A weight-o-evidence review. J Toxicol

Environ Health A 65:701–746.

Quievryn G, Messer J, Zhitkovich A. 2006. Lower mutagenicity

but higher stability o Cr-DNA adducts ormed during grad-

ual chromate activation with ascorbate. Carcinogenesis

27:2316–2321.

Reynolds M, Stoddard L, Bespalov I, Zhitkovich A. 2007. Ascorbate

acts as a highly potent inducer o chromate mutagenesis

and clastogenesis: linkage to DNA breaks in G2 phase by

mismatch repair. Nucleic Acids Res 35:465–476.

Sedman RM, Beaumont J, McDonald TA, Reynolds S,

Krowech G, Howd R. 2006. Review o the evidence

regarding the carcinogenicity o hexavalent chromium in

drinking water. J Environ Sci Health C 24:155–182.

Shackelord CC, Elwell MR. 1999. Small and large intestine, and

mesentery. In: Pathology o the Mouse (Maronpot RR, ed).

Vienna, IL:Cache River Press, 81–118.

Shirley E. 1977. A non-parametric equivalent o Williams’ test

or contrasting increasing dose levels o a treatment.

Biometrics 33:386–389.

Standeven AM, Wetterhahn KE. 1991. Is there a role or reac tive

oxygen species in the mechanism o chromium(VI) carcino-

genesis? Chem Res Toxicol 4:616–625.

Tarone RE. 1975. Tests or trend in lie table analysis. Biometrika

62:679–682.

U.S. EPA (Environmental Protection Agency). 2003. National

primary drinking water regulations. 40CFR141.62. Available:

http://edocket.access.gpo.gov/cr_2003/julqtr/40cr141.62.

htm [accessed 3 April 2009]

U.S. EPA 1998. Toxicological Review o Hexavalent Chromium

(CAS No. 18540-29-9). In Support o Summary Inormation

on the Integrated Risk Inormation System (IRIS).

Washington DC:U.S. Environmental Protection Agency.

Williams DA. 1986. A note on Shirley’s nonparametric test or

comparing several dose levels with a zero-dose control.

Biometrics 42:183–186.

Zhang JD, Li XL. 1987. Chromium pollution o soil and water

in Jinzhou [in Chinese]. Zhonghua Yu Fang Yi Xue Za Zhi

21:262–264.

Zhitkovich A. 2005. Importance o chromium—DNA adducts

in mutagenicity and toxicity o chromium (VI). Chem Res

Toxicol 18:3–9.

15619015 Hexavalent Chromium is Carcinogenic to F344N Rats and B6C3F1 Miceafter Chronic Oral...

Documents

Transcript of 15619015 Hexavalent Chromium is Carcinogenic to F344N Rats and B6C3F1 Miceafter Chronic Oral...