Toxic Interactions among EnvironmentalPollutants: Corroborating LaboratoryObservations with Human ExperienceKannan Krishnan and Jules BrodeurDepartement de medecine du travail et d'hygi6ne du milieu, Faculte de m6decine, Universit6 de Montreal,Montreal, Canada

Combined exposures to multiple chemicals may result in interactions leading to a significant increase or decrease in the overall toxicity of themixture compared to the summation of the toxicity of the components. A large number of chemical interactions have been described in animalstudies by administering high doses of chemicals by routes and scenarios often different from anticipated human exposures. Though limited, thereis some evidence for the occurrence of several supra-additive (the combined effects are greater than the simple summation of the individual effects)and infra-additive (the combined effects are smaller than the simple summation of the individual effects) chemical interactions in humans. For exam-ple, toxicokinetic interactions between several solvents have been found to occur in the workplace, whereas those involving pesticides have beenreported less frequently, especially during accidental exposures. Toxic interactions involving nutritionally important metals and metalloids appear tooccur more frequently, since several of them have an important role in a variety of physiological and biochemical processes. On thecontrary, there is not much evidence to confirm the occurrence of toxic interactions among the commonly encountered inorganic gaseouspollutants in humans. Overall, the majority of chemical interactions observed in animal studies have neither been investigated in humans nor beenextrapolated to humans based on appropriate mechanistic considerations. Future research efforts in the chemical interactions arena should addressthese issues by focusing on the development of mechanistically and biologically based models that allow predictions of the extent of interactionslikely to be observed in humans.-Environ Health Perspect 102(Suppl 9):11-17 (1994)

Key words: toxic interactions, chemical interactions, potentiation, synergism, antagonism

Introduction

Toxic interaction refers to the qualitativeand/or quantitative modification of the toxi-city of one chemical by another, the process

principally occurring within the organismafter the exposure phase (1). Interactions can

either result in greater-than-additive or less-than-additive toxic response. Over a thou-sand studies published to date report theoccurrence of supra- or infra-additive toxicityfrom combined exposure to two chemicals(2). The interactive toxicity resulting fromcombined exposures to chemicals is a conse-

quence of the alteration of the toxicokineticsand/or toxicodynamics of one chemical byanother. Interference at the kinetic levelwould imply a modulation of absorption,distribution, metabolism and/or excretion ofone chemical by another. Interference at thetoxicodynamic level might involve a compe-

tition between two chemicals for binding to

This article was presented at the IV European ISSXMeeting on Toxicological Evaluation of ChemicalInteractions: Relevance of Social, Environmental andOccupational Factors held 3-6 July 1992 in Bologna,Italy.

Address correspondence to Dr. K. Krishnan,Departement de Medecine du Travail et d'Hygienedu Milieu, Facult6 de M6decine, Universite deMontreal, Case Postale 6128, Succursale A,Montreal, P.O., Canada, H3C 3J7. Telephone (514)343-6581. Fax (514) 343-2200.

the target site or an alteration by one chemi-cal of the susceptibility of the target cells tothe effects of another agent.

Despite the report of the occurrence ofseveral significant interactions between envi-ronmental pollutants, their relevance forhumans and relative importance for regula-tion remain unclear and ill-defined. This sit-uation is a consequence of the lack ofchronic exposure studies with chemical mix-tures and the lack of understanding of themechanistic basis of interactions at a quanti-tative level. Until these issues are resolved,we probably would not be able to confi-dently use animal data on interactions tomake quantitative predictions for humans.However, there is some direct and/or epi-demiological evidence for the occurrence ofseveral supra-additive and infra-additivechemical interactions in humans. In thisarticle, we corroborate laboratory observa-tions with human experience as they relateto toxic interactions among environmentalpollutants and propose an approach to con-sider data on toxic interactions for humanhealth risk assessment.

Toxic Interactions among GaseousPoliutantsInteractions among gaseous air pollutants mostcommonly involve physicochemical mecha-nisms rather than toxicodynamic/kinetic inter-

ferences. A number ofanimal studies have sug-gested the occurrence of supra-additive andinfra-additive interactions among gaseous pol-lutants (Figures 1,2). Of those interactionsobserved in laboratory studies, there is evi-dence suggesting that a few of them may haveactually been experienced by humans. Forexample, the antagonistic interaction resultingfrom combined exposure to sulfuir dioxide andammoniacal compounds is thought to havebeen encountered during the London fog dis-aster of 1952 (3). The supra-additive toxicityresulting from combined exposures to sulfurdioxide and ozone may have been a causativefactor of the high mortality of Japanese chil-dren observed in the early seventies (4). Thereported higher incidence of cancer amongworkers in arsenic-smelting facilities has beenexplained by the supra-additive effects betweenarsenic and benzola]pyrene observed in labora-tory animals (5). However, the majority of theinteractions involving commonly occurringgaseous pollutants (i.e., NOx, SOx, ozone,aerosols) that have been well characterized inanimal studies do not seem to occur inhumans (6-11).

Toxic Interacdons amongPesticidal ChemicalsHumans ingesting food preparations con-taminated with one or more pesticides,workers in pesticide manufacturing and

Environmental Health Perspectives I11

KRISHNANAND BRODEUR

Sodium chloride Formaildehyde

/ Carbon

Ozone dioxkde

SuifricAmmonium acid

ufate /

Sulfur HydPotassium dioxide sulchloride /Z

23

/1 Zi>ncManganese oxidedioxide Ferric

oxide

Figure 1. Supra-additive interaction,and particulates demonstrated in anima

Methane Carbon monoxide

Ammonia Sulfur dioxide

Nitrogen dioxide

Figure 2. Infra-additive interactions arrparticulates demonstrated in animal stui

packaging units, and agricultuwho prepare, mix, and apply fthe fields are all potentially expcthan one pesticide on the same (

days. Interactions among pestinvolve either an induction or aof a particular metabolic pathwa

The supra-additive and inbinary interactions reported amercially important pesticidalanimal studies are summarizedand 4. The only instance wherrence of an interaction has beepertains to a poisoning episo(when hundreds of Pakistanibecame ill following the use ofpersible formulation of malathithis instance, isomalathion, anmalathion preparation, formquantities under improper stctions, accounted for the enharergic effects. Animal studies Ithat isomalathion is morS pmalathion as an inhibitorcholinesterase and also Xmalathion toxicity by interfericarboxylesterase pathway ofdetoxication (13,14).

Toxic Interactions amongMetals/MetaoidsHumans are exposed to low coof several metals and metallfrom fossil fuel burning, vehausts, metallic ore smeltingincineration. Occupational exjresult in substantial body burd

depending upon the type of operation and- Carbon facility. With metals and metalloids, toxi-

monoxide// cological interactions occur at bothNitrogen extremes of tissue exposure, i.e., deficiency-oxidea

and excess. This property is a consequencefrogen of several metals being required for aifide variety of physiological processes, e.g., as

cofactors of enzymes.The supra-additive and infra-additive

interactions among metals and metalloidss among gases described in animal studies are depicted instudies (2). Figures 5 and 6. Of these, the following

infra-additive interactions have been inves-tigated in humans: selenium and mercury,zinc and lead, and iron and mercury, cad-

,Chlorine mium or lead.Selenium-induced protection of mer-

~Acrolein cury toxicity in rats has been reportedwidely in the literature (15-17). There is

Ildshyde at least one case report of beneficial effectdue to selenium-mercury interaction in

nong gases and humans (18). These authors reported thatdies (2). allergic reactions, skin rash, incoordination

and memory loss in a 39-year-old womansensitive to mercury-based paints and

iral workers sprays, were markedly reduced after shepesticides in took periodic selenium supplements.)sed to more Suzuki et al. (19), measured the con-or successive centrations of trace elements in foodstuffs-icides often from a Japanese area with elevated intakeLn inhibition of methyl mercury and found comparableay. levels of selenium as well. These authorsLfra-additive considered that the alleviating effect ofmong com- selenium is the probable reason for thechemicals in toxic effects of mercury not being evidentin Figures 3 in that population. Some autopsy studies*e the occur- have reported that both mercury and sele-n suspected nium accumulated in tissues at an approxi-de in 1976, mate molar ratio of 1:1 suggesting ai spraymen natural protective effect by selenium ofa water-dis- mercury toxicity (20,21).ion (12). In Regarding the antagonistic interactionimpurity of between mercury and iron, there is oneled in great corroborative accidental report (22). A 5-)rage condi- year-old boy who ingested an open batteryiced cholin- containing iron and mercuric oxide experi-have shown enced no significant uptake of mercury.otent than This has been attributed to the reductionof acetyl- of HgO into insoluble elemental mercury

potentiates by iron present at the gastric site.ing with the Animal studies have indicated that sup-malathion plemental uptake of iron decreases the

absorption of lead, antagonizing lead-induced anemia; conversely, prevalence ofiron deficiency has been associated withincreasing severity of lead poisoning

tncentrations (23-28). It has therefore been suggestedoids arising that the correction of iron deficiencyhicular ex- among human populations can generally

;, and waste reduce lead toxicity (29). Interestingly, aposures may negative correlation between blood lead[en of metals levels and body burden of iron in human

Fen I trothion

Fenthion D PhosphomidanBPMC | Dimethoate \

/ MalathionCyanophos ieldrln

z x AidrinCarbaryl Endosulphan

Figure 3. Supra-additive interactions among pesti-cides demonstrated in animal studies (2). Abreviations:BPMC, 2-sec-butylphenylmethylcarbamate; DDT, 1,1,1-trichloro-2,2-bis-(4-chlorophenyl) ethane.

Malathion ~in/ da|ne~ pDT

Aldrin 7Parathion Toxaphene

DiazinonDieldrin Dlcrotophos Chlordane

IPhosphomidan Carbaryl

Carbofuran

Figure 4. Infra-additive interactions among pesticidesdemonstrated in animal studies (2).

populations has been reported by severalauthors (30-34).

Similar protective effects of zinc on leadtoxicity have been reported in certain ani-mal and human studies. In a survey ofindustrial workers, Dukiewicz et al. (35)found significantly reduced excretion ofaminolevulenic acid in workers exposed toboth lead and zinc compared to thoseexposed to lead alone. Though the protec-tive action of zinc in the case of lead poi-soning is suggested (36), its relevance forhumans still remains uncertain (37,38).

The absorption of zinc is reduced in thepresence of inorganic iron as a result of acompetition for intestinal carrier sites (39).According to this mechanistic description,25 mg of total ions (as the sum of zinc and

Chromium Lead

MercuryCadmium

Figure 5. Supra-additive interactions among metalsand metalloids demonstrated in animal studies (2).

Nicke

Molybdnum - Snenlmum. \cPP7I

Chfomium ~ u

Figure 6. Infra-additive interactions among metals andmetalloids demonstrated in animal studies (2).

Environmental Health Perspectives12

TOXIC INTERACTIONS IN HUMANS

Vinyl chloride

TrIchloroethyln - Styreno

1,1 - Dlchloroethyene Tetrachlornlhylene

Carbon tetrachlorlde Benzene

XyleneCarbon disulfide Toluene

Co| Mehylen ChlorideChloroform nlSIIuI

Figure 7. Infra-additive and inhibitory interactionsamong solvents demonstrated in animal studies (2).

Tribromo Toluenomethane

Carbon lerechlnrlde Acetontrilo

I AcrylonitrileTrichloJcF yln \

Slyrene

M BK[oChlnrnrm Acetone Mathylene chloride

n-Hene MnBKN,N-Dlmeihyl UK

AnilineME

Figure 8. Supra-additive interactions among solventsdemonstrated in animal studies (2).

iron administered orally) is the point of sat-uration, at which competitive effect willbegin to be expressed (39). Evidence insupport of interaction between iron andzinc/cadmium has been obtained in humanstudies (40).

Toxic Interactions among SolventsCombined exposure to organic solventsoften results in a mutual inhibition of theirmetabolism, since most of them appear tobe metabolized by cytochrome P450 2E1at low exposure concentrations. Figure 7presents the inhibitory interactions amongsolvents determined in laboratory animalstudies. Of the metabolic inhibition inter-actions observed in animal studies, the fol-lowing have been confirmed in humanvolunteer/occupational exposure studies:trichloroethylene and 1,1,1-trichloroethane(41), benzene and toluene (42), ethylben-zene and m-xylene (43), xylene andtoluene (44), trichloroethylene and tetra-chloroethylene (45), m-xylene and methylethyl ketone (46) and m-xylene and isobu-tanol (47). These inhibitory metabolicinteractions are characterized by reduced/delayed production and excretion of meta-bolites, and/or increased concentrations ofparent chemical(s) in the blood and expiredair. However, the health significance of sev-eral of these metabolic interactions has notbeen confirmed in workers occupationallyexposed to binary solvent mixtures(48,49). The inhibitory metabolic interac-tions can be expected to result in supra-additivity with respect to the toxicity of theparent chemicals.

Supra-additive toxicity during exposureto binary mixtures of solvents has oftenbeen observed in animal studies, wheneverone of the two solvents was a potentinducer of activating enzymes, or aninhibitor of detoxication enzyme system(50,51). Thus, most of the solvents thatinduce P4502E1 upon prior administration(e.g., ketones) have been shown to potenti-ate the toxicity of other solvents bioacti-vated by the same isoenzyme (Figure 8).

In human experience, isopropanol, aketogenic chemical, was the suspect poten-tiator of the hepato-renal toxicity of car-bon tetrachloride in two separate industrialaccidents (52,53). Similarly, methyl ethylketone-induced potentiation of n-hexane/MnBK neurotoxicity, through the en-hanced formation of 2,5-hexanedione,appears to have been responsible for theoutbreak of an occupational neuropathyamong the textile workers in Ohio (54,55)and among glue sniffers in West Berlin inthe seventies (56).

Relevance ofAnimal Data on ChemicalInteractions for Quantitative RiskAssessmentIn evaluating the relevance of chemicalinteractions for regulatory purposes, it isimportant to consider the underlyingmechanisms. Understanding the mechanis-tic basis enables us to determine whether ornot an interaction will occur at low expo-sure levels in animals, and if it will occur atall in other species, especially, humans.Table 1 presents some of the mechanismsof interaction unequivocally demonstratedin animal and/or human studies.

The data generated in most interactionstudies are qualitative, supporting or sug-gesting a specific mechanism as the possiblecause of the infra- and supra-additivity. Forquantitative risk assessment (QRA) pur-poses, we need quantitative mechanistic dataon chemical interactions. For example, whatis the dose-response for the induction orinhibition of metabolism of one chemical byanother? What is the quantitative relation-ship between two chemicals competing forprotein binding? Such quantitative analysesof chemical interactions can be conductedwhen the interrelationships among the criti-cal determinants of chemical disposition,action and interaction are identified andintegrated within a biologically based mod-eling framework. To date, this quantitativemodeling approach has been applied onlyfor a few binary chemical mixtures (72).

Since the quantitative mechanistic basisof chemical interactions has not been eluci-dated for majority of interacting chemical

combinations, their relevance for humanscannot be evaluated confidently. On theother hand, regardless of the level of mech-anistic understanding of interchemicalinteractions, there are instances that requirethe evaluation of the relevance of data onchemical interactions for specific purposes(e.g., for ensuring worker safety).

Under these circumstances, we areobliged to analyze the available data andcome up with a tentative conclusionregarding the importance of an interactionfor a given exposure situation. There arecertain types of data which suggest that aparticular interaction is unimportant forhumans. These relate primarily to instanceswhere the animal studies show the exis-tence of thresholds for interactions whichexceed the maximum allowable exposureconcentrations for humans, and where ani-mal studies confirm the occurrence ofinteractions which do not occur in humansexposed to low levels of both chemicals.

An example of the former case is theEPN (O-ethyl-0-4-nitrophenyl phenylphosphonothioate) malathion potentiation.This interaction, shown to be important athigh doses (73), was not apparent whenanimals were fed recommended daily toler-ance levels of both chemicals (74), suggest-ing that the potentiation may not occur inhumans exposed to much lower concentra-tions of these chemicals. Thus, Moeller andRider (75), giving small oral doses of EPNand malathion in combination to humanvolunteers, found no evidence of EPNpotentiation of malathion toxicity.

An example of the second type is theinteraction among several commonlyoccurring air pollutants. Pulmonary bio-chemical and morphological changesinduced by nitrogen dioxide have beenreported to be enhanced by coexposure toacidic aerosols (76). This interactionoccurs even at levels of 1 mg/m3 of sulfuricacid aerosol and 2 ppm of nitrogen diox-ide. But in human exposure studies, Staceyet al. (7) could not detect any synergisticeffect in people exposed to low levels ofboth chemicals. Similarly, synergistic inter-action between ozone and sulfuric acidaerosols has been demonstrated in animalsexposed to atmospheric concentrations of0.12 ppm and 5 mg/m3, respectively (77).Yet, Horvath et al. (10) found that a 2-hrexposure to a mixture of 0.2 ppm of ozoneand 1.2 to 1.6 mg/m3 of sulfuric acidaerosol did not induce effects beyond whatcould be attributed to human exposure toozone alone.

Detailed investigations of this kindhave not been conducted for the majority

Volume 102, Supplement 9, November 1994 13

KRISHNAN AND BRODEUR

of the several hundred interactionsreported to occur among environmentalpollutants at high doses (2). Neitherchronic animal exposure data nor humanvolunteer exposure data are available forthese binary mixtures, rendering the assess-ment of their relevance for humans diffi-cult. When confronted with a particularexposure situation involving chemicalinteractions, one should evaluate the avail-able information on that interaction (or fora similar chemical combination), on thebasis of factors such as dose administered

and species used, to determine the rele-vance for humans (Figure 9). Accordingly,if such an evaluation indicates a high prob-ability for occurrence of an interactionbetween two chemicals encountered in aparticular workplace setting, then necessaryprecautions should be taken to avoid such acombined exposure. In the following para-graphs, we discuss four categories of datathat should be given consideration indecreasing order of importance, while eval-uating the relevance of information onchemical interactions for humans.

Interactions Demonstrated inHumans Exposed At or BelowAllowable Exposure Concentrations([LV, RC, RID)Toxic interactions demonstrated inhumans, in most cases, pertain to occupa-tional exposures involving solvents morefrequently than other types of contami-nants. Toxicokinetic interactions amongvarious solvents often arise from themutual inhibition of their metabolism,thus resulting in increased blood levels.Such inhibitory metabolic interactions may

Table 1. Examples of mechanisms of toxicokinetic interactions among chemicals.

Basis of interaction Interacting chemicals Mechanism of interactive effects References

m-Xylene and isobutanol

Dimethyl sulfoxide (DMSO) andpesticides

Reduced absorption of both compounds due to dehydrationof skin elicited by isobutanol

Increased dermal absorption of pesticides and otherchemicals when they are mixed with DMSO, whichdisrupts the cellular permeability and acts as a "penetrantcarrier"

Rilhimaki (47)

Jacob et al. (57)Hayes and Pearce (58)

Pulmonary

Gastrointestinal

DistributionTissue distribution

Protein binding

MetabolismPhase

Induction

Inhibition

Phase II

Induction

Inhibition

ExcretionPulmonary

Biliary

Urinary

Hydrogen cyanide (HCN) andcarbon monoxide

Lead and iron

Lead and dithiocarbamates

Organochlorine andorganophosphate pesticides

Methyl n-butyl ketone(MnBK) and chloroform

Dithiocarbamates andchloroform

Sodium sulfate andcertain arylamines

Pentachlorophenol andarylamines

Ethanol andmercury

Arsenic and seleniumMercury and selenium

Sodium bicarbonate(NaHCO3) and fluoride

Increased pulmonary uptake of air contaminants when theyare present along with HCN, which at low concentrationsincreases the pulmonary ventilation rate

Decreased absorption of lead in the presence of iron due tocompetition for transport sites in the intestinal mucosa.

The lipophilic lead-dithiocarbamate complex has a greatercapacity than lead alone to penetrate the blood-brainbarrier and causes a greater accumulation in the lipid-richbrain components.

Organochlorine pesticides not only enhance the biotrans-formation of organophosphates, but also enhance theirbinding to plasma proteins and nonspecific esterases.

Increased bioactivation of chloroform due to induction ofhepatic microsomal P450 2E1 by MnBK.

Decreased bioactivation of chloroform due to inhibition ofhepatic microsomal enzymes by dithiocarbamates.

Increased sulfate supply might result in a greater amountavailable for conjugation of xenobiotics and their metabo-lites.

Pentachlorophenol, an inhibitor of cytosolic sulfotrans-ferases, renders such compounds as N-hydroxy-2-acetyl-aminofluorene (that are activated by the formation ofsulfate esters) less toxic.

Ethanol depresses the conversion of elemental mercury intoionic form. Their coexistance results in a diminution of thepulmonary retention as well as blood levels of mercury andenhances its pulmonary exhalation.

Arsenic increases the clearance of selenium from the liverinto the bile. Selenium, on the other hand, inhibits thebiliary excretion of mercury.

Alkalosis induced by NaHCO3 causes a more rapid renalclearance of fluoride.

Dreisbach (59)

Conrad and Barton (60)

Osakrsson (61)

Trioloand Coon (62)Cohen and Murphy (63)

Brady et al. (51)

Gopinath and Ford (64)

Meerman et al. (65,66)

Nielsen-Kudsk (67)Magos et al. (68)

Lavender and Baumann (69)

Reynolds et al. (70);Whitford and Pashley

Environmental Health Perspectives

AbsorptionPercutaneous

14

TOXIC INTERACTIONS IN HUMANS

hitrctlonoccurrence In humens

Y N

CONSIDER INREGULATION --- Lowdose High dose Occurrence Occurrence(Factor X) In enimel ?- In vitro-

Evaluateceee by Ceee

UCFiIILow do" High dose

Figure 9. A preliminary schematic for considering dataon chemical interactions for regulation. Abbreviations:Y, yes; N, no; UCF 1/11, uncertainty factors.

eventually cause supra-additive neurotoxiceffects even at exposure concentrations thatdo not exceed their respective TLVs.However, these toxicokinetic interactionsinvolving mainly the biotransformationsystem appear to have a threshold level foreach component.

Toxic interactions demonstrated innonoccupational settings, i.e., the generalenvironment, mainly involve nutritionally-important metals. In this regard, antago-nistic effects on the absorption of toxicmetals have often been reported to occurwith higher intake of iron and zinc.Importantly, these interactions occur evenwhen the dietary intake of one of thesemetals is insufficient, and therefore can beregarded as being more relevant forhumans than mrost of the other chemicalinteractions.

Another interaction that belongs to thiscategory is that of selenium and several car-cinogens. A number of epidemiologic stud-ies have reported an inverse relationshipbetween the selenium status and the inci-dence rates of various types of cancer andcardiovascular diseases in humans (78-80).Recently, Yu et al. (81) have reported thatthe primary liver cancer incidence in peo-ple (in the Chinese province of Jiangsu)consuming selenium supplemented tablesalt was lower than that in the populationnot receiving selenium supplementation.

Interacdons Demonstratedin Humans at High ExpoureConcentrations OnlyCertain chemical interactions occur only athigh doses or after unusual exposure sce-narios (e.g., accidental exposures). Anexample of this category is the ketone-haloalkane interaction. The suspectedsupra-additive interaction between iso-propanol and carbon tetrachlorideoccurred in two instances, both when theenvironmental concentrations exceeded theallowable exposure limits (52,53). It isunderstandable that when enzyme induc-

tion is the mechanism involved, thereprobably is a threshold for interaction.Particularly, this type of interaction can beexpected to be insignificant at low exposureconcentrations, where the rate-limiting fac-tor of metabolism is hepatic blood flowand not the capacity of the enzyme system.On the other hand, the enzyme inducers,on coexposure, may become inhibitors ofmetabolism of other chemicals, the impor-tance of such an interaction being deter-mined by the affinity and relative roles ofvarious isoforms of P450 involved in themetabolism of both chemicals.

Interacdons Demonstrated inAnimal Studies at Low Dosesbut Potential for Occurrence inHumans Not KnownSeveral chemicals acting as inhibitors ofhepatic and extrahepatic metabolismappear to have greater potential for interac-tion than inducers at low exposure concen-trations. Apparently, the quantitativeimportance of the influence of an inhibitoron another chemical depends on the mag-nitude of the inhibition constant, K1. Basedon experimental and simulation studies,trans-1,2-dichloroethylene (DCE) has beensuggested to be a potent suicide inhibitorof P450 (82). Therefore, metabolic inhibi-tion of a variety of chemicals by DCE canbe expected during combined exposures,even though direct evidence in humans isnot yet available. Based on mechanisticconsiderations, uncompetitive and non-competitive inhibitions are also quantita-tively more important than competitiveinhibition. Benzene-toluene interaction inanimals has been described to be the resultof noncompetitive interaction (83), indi-cating that this interaction can occur at lowexposure concentrations. Evidence formutual inhibitory interaction betweenthese chemicals has actually been obtainedin an occupational exposure monitoringstudy (42). Another example belonging tothis category is the toxicodynamic interac-tion between chlordecone and carbontetrachloride demonstrated at low doses inanimal studies (84).

Interacdons Demonstrated inAnimal Studies at High Doses butPotential fir Occurrence at Low DosesNot KnownMost of the chemical interaction studiesconducted to date belong to this category.Typically, these studies have involved theadministration of high doses of one or bothchemicals by routes and scenarios often dif-ferent from anticipated human exposures.

The potential of these interactions to occurduring low-dose, chronic exposures is notknown. Consideration of the number ofcombinations that need to be tested at lowdoses in chronic experiments, by at least oneexposure route, emphasizes the need to uti-lize alternative methodologies, particularlypredictive modeling strategy. This quantita-tive approach involves the integration ofmechanistic factors of chemical disposition,interaction, and effect into a biologicallybased framework, that provides the basis forpredicting the extent of interactions inuntested exposure situations (72).

There is also the possibility that aninteraction, not occurring in animals athigh or low doses, may in fact occur inhumans exposed to low concentrations.Based on mechanistic considerations, suchobservations would suggest differences inthe mechanistic determinants of interac-tion between the two species (e.g., tran-species variation in the relative roles ofdifferent P450 isozymes).

The preceding analysis of the relevanceand relative importance of experimentaldata on chemical interactions for humanscovered only certain of the importantaspects. For example, the influences ofexposure routes and sequence of chemicalexposures in the interaction studies are notexplicitly taken into account. These factorsare being considered in a weight-of-evi-dence approach formulated by theEnvironmental Criteria Assessment Officeof the U.S. Environmental ProtectionAgency (85).

In summary, there is some evidence forthe occurrence of chemical interactions inhumans, particularly in the occupationalenvironment and during accidental expo-sures. We need to investigate and utilizequantitative modeling approaches in thestudy of interactions to uncover the animalinteractions that are relevant to humanexposure situations. Such a quantitative,mechanistic approach to the study ofchemical interactions is fundamentallyimportant to achieve the ultimate goal ofassessing the health risks associated withhuman exposure to complex chemicalmixtures.

Volume 102, Supplement 9, November 1994 15

KRISHNANAND BRODEUR

REFERENCES

1. NAS. Principles of Toxicological Interactions Associated withMultiple Chemical Exposures. Washington:National Academyof Sciences, (1983).

2. Krishnan K, Brodeur J. Toxicological consequences of com-bined exposures to environmenta pollutants. Arch ComplexEnviron Studies 3(3): 1-106 (1991).

3. Waldbott GL. Pulmonary irritants. In: Health Effects ofEnvironmental Pollutants. St. Louis:CV Mosby and Co,1973;84.

4. Bates DV, Hazucha M. The effects of low levels of sulfur diox-ide and ozone in the same atmosphere on the pulmonary func-tion. In: Proceedings of an International Symposium on theRecent Advances in the Assessment of the Health Effects ofEnvironmental Pollution, Vol 4. Geneva:World HealthOrganization, 1974;1977-1988.

5. Pershageni G, Wall S, Taube A, Limman L. On the interactionbetween occupational arsenic exposure and smoking and itsrelationship to lung cancer. Scand J Work Environ Health7:302-309 (1981).

6. McJilton CE, Salawich RJ, Wofsy SC, Logan JA. Role of rela-tive humidity in the synergistic effects of a sulfur dioxideaerosol mixture in the lung. Science 182:503-504 (1986).

7. Stacey RM, Seal EG, Housie AF, Green J, Roger LJ, Reggio L.Effects of gaseous and aerosol pollutants on pulmonary func-tion measurement of normal humans. Arch Environ Health38:104-115 (1983).

8. Bell K, Linn W, Hazucha M, Hackney J, Bates D. Respiratoryeffects of exposure to ozone plus sulfur dioxide in southernCalifornia vs eastern Canadians. Am Ind Hyg Assoc J38:696-706 (1977).

9. Bedi JF, Folinsbee LJ, Horvath SM, Ebenstein RS. Humanexposure to sulfur dioxide and ozone: absence of a synergisticeffect. Arch Environ Health 34:233-239 (1979).

10. Horvath SM, Folinsbee LJ, Bedi JF. Combined effect of ozoneand sulfuric acid on pulmonary function in man. Am Ind HygAssoc J 48:94-98 (1987).

11. Frank R. Sulfur dioxide-particulate interactions: recent obser-vations. Am J Ind Med 1:427-434 (1980).

12. Baker EL, Zack M, Miles JW, Alderman L, Warren M,Dobbin RD, Miller S, Teekerr WR. Epidemic malathion poi-soning in Pakistan malaria workers. Lancet 1: 31-38 (1978).

13. Aldridge WN, Norseth T. The toxicological properties ofimpurities in malathion. Arch Toxicol 42:95-106 (1979).

14. Miles JW, Mount DL, Staiger MA, Teeters WR. The S-methylisomer content of stored malathion and fenitrothion water dis-persible powders and its relationship to toxicity. J Food AgriChem 27:421-425 (1979).

15. Parizek J, Ostadalova I, Kalouskova J, Babicky A, Benes J. Thedetoxifying effects of selenium interrelations between com-pounds of selenium and certain metals. In: Newer TraceElements in Nutrition (Metrz W, Cornatzer WE, eds). NewYork:Marcel Dekker, 1971;85-122.

16. Parizekf J, Kalouskova J, Babicky A, Benes J, Pavlik L.Interaction of selenium with mercury, selenium, cadmium, andother toxic metals. In: Trace Elements Metabolism in Animals,Vol 2 (Hoekstra WG, Sutties TF, Ganther HE, Metrz W, eds).Baltimore:University Park Press, 1974;1 19-131.

17. Ganther HE. Modifications of methyl mercury toxicity andmetabolism by selenium and vitamin E: possible mechanisms.Environ Health Perspect 25:71-76 (1978).

18. Stopford W, Donovan DH, Abou-Donia MB, Menzel DB.Glutathione peroxidase deficiency and mercury allergy: amelio-ration with selenium supplementation. In: Proceedings of theSymposium on Selenium-Tellurium in the Environment.Pittsburgh, PA:Industrial Health Foundation, 1976,105-112.

19. Suzuki T, Satoh H, Yamamoto R, Kashiwarzaki H. Seleniumand mercury in food stuff from a locality with elevated intakeof mercury. Bull Environ Contam Toxicol 24:805-811 (1980).

20. Kostaf L, Brynet AR, Zelenko V. Correlation between selenium

and mercury in man folowing exposure to inorganic mercury.Nature (London) 254:238 (1975).

21. Nylander M, Weiner J. Mercury, selenium concentrations andtheir interrelationships in the organs from dental staff and thegeneral population. Br J Id Med 48:729-734 (1991).

22. Barber TE. Menke RD. The relationship of ingested iron tothe absorption of mercuric oxide. Am J Emerg Med 2:500-505(1984).

23. Six KM, Goyer RA. The influence of iron deficiency on tissuecontent and toxicity of ingested lead in the rat. J Lab Clin Med79:128-136 (1972).

24. Ragan HA. Effects of iron deficiency on absorption and distrib-ution of lead and cadmium in rats. J Lab Clin Med 90:700-706 (1977).

25. Barton JC, Conrad ME, Nuby S, Harrison L. Effects of ironon the absorption and retention of lead. J Lab Clin Med92:937-947 (1978).

26. Flanagan PR, Hamilton DL, Haist J, Valberg L. Interrela-tionships between iron and lead absorption in iron-deficientmice. Gasteroenterology 77:1074-1081 (1979).

27. Dhir H, Sharma A, Talikar G. Alteration of cytotoxic effects oflead through interaction with other heavy metals. Nucleus28:69-89 (1985).

28. Mahaffey KR. Nutritional factors and susceptibility to lead tox-icity. Environ Health Perspect 8:107-112 (1974).

29. Sullivan JF, Reummler PS. Effect of excess iron on cadmium orlead absorption in rats. J Toxicol Environ Health 22:131-139(1987).

30. Ito Y, Niiya Y, Otani M, Sarai S, Shima S. Effect of foodintake on blood lead concentration in workers occupationallyexposed to lead. Toxicol Lett 37:105-114 (1987).

31. Chisolm JJ. Dose-effect relationships for lead in young chil-dren: evidence in children for interactions among zinc andiron. In: Environmental Lead: Proceedings of the SecondInternational Symposium on Lead Research (Lynam DR,Piantandia LG, Cole LG, eds). New York:Academic Press,1981; 1-7.

32. Watson WS, Hume R, Moore MR. Oral absorption of leadand iron. Lancet 2: 236-237 (1980).

33. Yip R, Norris TN, Anderson AS. Iron status of children withelevated blood lead concentrations. J Pediatr 98:922-925(1981).

34. Mahaffey KR, Annest JL. Assessment of erythrocyte protopor-phyrin with blood lead level and iron status in the SecondNational Health and Nutrition examination survey1976-1980. Environ Res 41:327-336 (1986).

35. Dukiewicz B, Dukiewicz T, Milkowska G. The effect of mixedexposure to lead and zinc on aminolevulenic acid level in urine.Int Arch Occup Environ Health 42:341-348 (1979).

36. Papaioannou R, Sohler A, Pfiffer CC. Reduction of blood leadlevels in battery workers by zinc and vitamin C. J OrthomolPsychiatry 7: 1-13 (1978).

37. Flanagan PR, Chamberlain MJ, Valberg LS. The relationshipbetween iron and lead absorption in humans. Am J Clin Nutr36:823-829 (1982).

38. Lauwerys R, Roelds H, Buchet JP, Bernard AA, Verhoeven L,Konings L. The influence of orally administered vitamin C orzinc on the absorption of and biological response to lead. JOccup Med 25:668-674 (1983).

39. Solomons NW. Competitive interaction of iron and zinc in thediet: consequences for human nutrition. J Nutr 116:927-935(1986).

40. Bunker VW, Lawson MS, Delves HT, Clayton BE. The intakeand excretion of lead and cadmium by the elderly. Am J ClinNutr 39:803 (1984).

41. Pfaffli P, Backman AL. Trichloroethylene concentrations inblood and expired air as indicators of occupational exposure: apreliminary report. Scand J Work Environ Health 9:140-144(1972).

16 Environmental Health Perspectives

TOXIC INTERACTIONS IN HUMANS

42. Inoue 0, Seiji K, Watanabe T, Kasahara M, Nakatsuka H, YinS, Jin C, Ikeda M. Mutual metabolic suppression between ben-zene and toluene in man. Int Arch Occup Environ Health60:15-20 (1988).

43. Angerer J, Lehnert G. Occupational chronic exposure toorganic solvents. 8. Phenolic compounds-metabolites of alkyl-benzenes in man: simultaneous exposure to ethylbenzene andxylenes. Int Arch Occup Environ Health 43:145-150 (1979).

44. Tardif R, Lapare S, Plaa GL, Brodeur J. Effect of simultaneousexposure to toluene and xylene on their respective biologicalexposure indices in humans. Int Arch Occup Environ Health63:279-284 (1991).

45. Inoue 0, Seiji K, Kawai T, Jin YT, Cai SX, Yin SN, Li GL,Nakatsuka H, Watanabe H, Ikeda M. Relationship betweenvapor pressure and urinary metabolite excretion among workersexposed to trichloroethylene. Am J Ind Med 15:103-110(1988).

46. Liira J, Riihimaki V, Engstrom K. Coexposure of man to m-xylene and methyl ethyl ketone. Scand J Work Environ Health14:322-327 (1988).

47. Riihimaki V. Percutaneous absorption of m-xylene from a mix-ture of m-xylene and isobutanol in man. Scand J Work EnvironHealth 5:143-150 (1979).

48. Olson BA, Gamberale F, Iregren A. Co-exposure to toluene andp-xylene in man: central nervous systems. Br J Ind Med,42:117-122 (1985).

49. Yin S, Li G, Hu Y, Zhang X, Jin C, Inoue 0, Seiji K, KasaharaM, Nakatsuka H, Ikeda M. Symptoms and signs of workersexposed to benzene, toluene, or their combination. Ind Health25:113-130 (1987).

50. Pass MA, Muir AD, Majak W, Yost G. Effect of alcohol andacetaldehyde dehydrogenase inhibitions on the toxicity of 3-nitropropanol in rats. Toxicol Appl Pharmacol 78:310-315(1985).

51. Brady JF, Li MJ, Ishizzaki H, Lee M, Ning S, Xiao F, Yang CS.Induction of cytochrome P-450IIE1 by secondary ketones andthe role of P-450IIE1 in chloroform metabolism. Toxicol ApplPharmacol 100:342-349.

52. Folland DS, Schaffner W, Grinn H, Crofford QB, McMurrayDK. Carbon tetrachloride potentiation by isopropanol: investi-gation of an industrial outbreak. J Am Med Assoc236:1853-1856 (1976).

53. Deng JF, Wang JD, Sih TS, Lan FL. Outbreak of carbon tetra-chloride poisoning in a color printing factory related to the useof isopropyl alcohol and air conditioning system in Taiwan. AmJ Ind Med 12:11-19 (1987).

54. Allen N, Mendell JR, Billmaier D, Fontaine RE, Oeill J. Toxicpolyneuropathy due to methyl n-butyl ketone. Arch Neurol32:209-218 (1975).

55. Billmaier D, Yee HT, Allen N, Craft B, Williams N, Epstein S,Fontaine R. Peripheral neuropathy in a coated fabrics plant. JOccup Med 16: 665-671 (1974).

56. Altenkirchf H. Schnuffelsucht Chronischer LosungsmittelMissbrauch bei Kindern und Jugendlichen mit neurotoxischenFolgen. Dtsch Med Wochenschr 104:935-938 (1979).

57. Jacob SW, Bischel M, Herschler RJ. Dimethyl sulfoxide: effectson the permeability of biologic membranes. Curr Ther Res6:193 (1964).

58. Hayes WJ, Pearce GW. Pesticide formulation: relation to safetyin use. J Agr Food Chem 1:466 (1953).

59. Dreisbach RH. Cyanides, sulfides and cobalt. In: Handbook ofPoisoning. 9 ed. Los Altos, CA:Lange Medical Publishing Co,1977;241.

60. Conrad ME, Barton JC. Factors affecting the absorption andexcretion of lead. Gasteroenterology 74:731-740.

61. Osakrsson A, Lind B. Increased Tead levels in brain after longterm treatment with lead and dithiocarbamate or thirum deriva-tives in rats. Acta Pharmacol Toxicol 56: 309-315 (1985).

62. Triolo AJ, Coon JM. Binding of paraoxon by plasma of aldrin-treated mice. Toxicol Appl Pharmacol 14:622 (1970).

63. Cohen SD, Murphy SD. A simplified bioassay for organophos-phate detoxication and interactions. Toxicol Appl Pharmacol

27:537-550.64. Gopinath C, Ford EJH. The role of microsomal hydroxylases in

the modification of chloroform hepatotoxicity in the rat. Br JExp Pathol 56:412-422 (1975).

65. Meerman JHN, van Doorn ABD, Mulder GJ. Inhibition of sul-fate conjugation of N-hydroxy 2-acetylfluorene in isolated per-fused rat iver and in the rat in vivo by pentachlorophenol.Cancer Res 40: 3772-3779 (1980).

66. Meerman JHN, Sterenborg D, Mulder GJ. Use of pen-tachlorophenol as longterm inhibitor of sulfation of phenolsand hydroxamic acids in the rat liver in vivo. BiochemPharmacol 32:1587-1593 (1983).

67. Nielsen-Kudsk F. The influence of alcohol in the absorption ofmercury vapor from the lungs in man. Acta Pharmacol Toxicol23:263-274.

68. Magos L, Clarkson TW, Grenwood MR. The depression of pul-monary retention of mercury vapor by ethanol: identification ofthe site of action. Toxicol Appl Pharmacol 26:180-183 (1973).

69. Lavender OA, Baumann CA. Selenium metabolism: effect ofarsenic on the excretion of selenium in the bile. Toxicol ApplPharmacol 9: 106-115 (1966).

70. Reynolds KE, Whitford GM, Pashley DH. Acute fluoride toxic-ity: the influence of acid-base status. Toxicol Appl Pharmacol45:415-427 (1978).

71. Whitford GM, Pashley DH. Fluoride reabsorption by nonionicdiffusion in distal nephron of the dog. Proc Soc Exp Biol Med196:178 (1991).

72. Krishnan K, Clewell HJ III, Andersen ME. Physiologicallybased pharmacokinetic modeling of simple mixtures. EnvironHealth Perspect 102(Suppl 9):151-155 (1994).

73. Frawley JP, Fuyat HN, Hagan EC, Blake JR, Fitzugh OG.Marked potentiation of mammalian toxicity from simultaneousadministration of two cholinesterase compounds. J PharmacolExp Ther 121:96-106 (1957).

74. DuBois KP. Combined effects of pesticides. Can Med Assoc J100:173-179 (1968).

75. Moeller HC, Clary JJ. Potentiation of red blood cellcholinesterase activity as indicators of the threshold of incipienttoxicity of EPN and malathion in human beings. Toxicol ApplPharmacol 4:123-130 (1962).

76. Last JA. Toxicological interactions between inorganic gases andparticulates. Fundam Appl Toxicol 12:9-11 (1989).

77. Last JA. Synergistic effects of air pollution: ozone plus a respira-tory aerosol. Research Report No. 38. Cambridge, MA:HealthEffects Institute, 43, 1990.

78. Schrauzer GN, White DA, Schneider CJ. Cancer mortalitystudies 3: statistical association with dietary selenium intakes.Bioinorg Chem 7:23-34 (1977).

79. Shamberger RJ, Tyoko SA, Willis CA. Antioxidant and cancer:4. Selenium and age adjusted mortality. Arch Environ Health31:231-235 (1976).

80. Kok FJ, Hofman A, Witteman JCM, De Bruin AM, KryossenDCM, De Bruin M, Valkenburg HA. Decreased selenium levelsin acute myocardial infarction. J Am Med Assoc261:1161-1164 (1989).

81. Yu SY, Zhu YJ, Li WG, Hou C. Chemoprevention trial of pri-mary liver cancer with selenium supplementation in Qidongcounty of China. In: Metal Ions in Biology and Medicine(Collery PH, Poirier LA, Manfait JC, Etienne JC, eds).Paris:John Libbey Eurotext, 1990;497-499.

82. Andersen ME, Gargas ML, Clewell HJ III. Suicide inactivationof microsomal oxidation of cis- and trans- dichloroethylene inmale Fischer-344 rats in vivo. Toxicologist 6:12 (1986).

83. Purcell KJ, Cason GH, Gargas ML, Andersen ME, Travis CC.In vivo metabolic interactions of benzene and toluene. ToxicolLett, 52:141-152 (1990).

84. Mehendale HM. Mechanism of lethal interaction of chlorde-cone and carbon tetrachloride at nontoxic doses. Toxicol Lett49: 215-241 (1989).

85. Mumtaz MM. The US EPA guidelines and current issues of thehealth risk assessment of chemical mixtures. Toxicologist 12:4(1992).

Volume 102, Supplement 9, November 1994 17