Document 9OnoraRaqxXJ49K8dYvB22Eq

V <* FUNDAMENTAL AND APPLIED TOXICOLOGY 16, 1-13(1991) 92 103 I 10 I 17 128 147 c si 161 il 172 ve . 198 210 21 1 SYMPOSIUM OVERVIEW Comparative Dosimetry of Inhaled Materials: Differences among Animal Species and Extrapolation to Man12 A lan R. Da h l .* Richard B. Schlesinger.t H enry d ' A. H eck.$ M ichele A. M ed in sky.J an d G eorge W. Lucier *Lovelace Inhalation Toxicology Research Institute. P O Bus 51100. Albuquerque. Sew Mexico S7IR5: Mnstitute <>/ Environmental Medicine. Sew York L'niversity Medical Center. 550 l-'irst Avenue. Sew York. S'ew York 10016: i Chemical Industry Institute o f Toxicology. Research Triangle Park. S'orth Carolina 27700: and National Institute ol Environmental Health Sciences. Research Triangle Park. S'orth Carolina 27700 Received duly 3 1. 1000; acceptai August 16. 1000 Dose is a fundamental concept in the science o f toxicology: yet. for all o f its importance, the determination o f dose for inhaled materials presents issues with which inhalation toxicol ogists are still grappling. Am id a myriad o f lesser factors, the two m ajor ones influencing dose for inhalants are the physicochemical properties o f the inhaled materials and the species o f the animal doing the inhaling. The speakers in this symposium presented the most recent advances in research related to descrip tions o f dose for inhaled particles, particle-as sociated organic compounds, reactive vapors, and metabolizable vapors. They also discussed methods for predicting inhaled doses in hu mans based on measured doses in test animals and advances in the search for biomarkers o f inhaled carcinogens. The factors that affect the fate o f inhaled particles were reviewed and the particle de position and clearance patterns o f experimen tal animals and people were compared. Ad vances in determining dose for reactive vapors that are largely absorbed in the nose (exem- ' Symposium held al the 29th Annual Meeting o f the Society of Toxicology (SOT) in Miami Beach. Florida. Sponsored by the Inhalation Specialty Section o f SOT. The U.S Government's right to retain a nonexclusive royalty-free license in and to the copyright covering this Paper, lor governmental purposes, is acknowledged. plified by formaldehyde) or in the lung (ex emplified by ozone) were discussed in terms o f relating experimental data to predictive models. The uptake o f metabolizable vapors, as affected by both the physicochemical prop erties o f the vapors and the metabolic capac ities o f test animals and people, was explored by using specific examples o f vapors com monly encountered in the environment. Problems in making interspecies comparisons were also addressed. Finally, methods for in dexing the dose o f inhaled carcinogens by us ing toxic metabolites. D N A or hemoglobin adducts, oncogene activation, gene mutations, and chromosomal changes as biomarkers were reviewed, with an emphasis on the use o f such markers in risk assessments. Comparative Deposition. Clearanee. and Re tention of Partiele- Borne To.xieanis (R B Settlesinger) Experimental animals are employed in tox icological studies o f inhaled particles, with the ultimate goal o f extrapolating the results to humans. To adequately apply the results o f such studies to human risk assessment, how ever. it is essential to consider interspecies d if ferences in particle disposition. Different spe cies exposed to the same particulate atmo- 0272-0590/91 $3.00 Copyright ic 1991 by the Society o f toxicology. All rights o f reproduction in any form reserved. ? DAHL ET AL. sphere may not receive identical doses in comparable respiratory tract regions. Extrap olation o f the results o f a study with any one species, therefore, may provide unrealistic es timates o f the human respiratory tract or sys temic dose and. thus, o f the relationship be tween exposure and potential human health effects. The fate o f inhaled particles is dependent both upon their pattern o f deposition-- the sites w ithin which such particles initially come into contact with airway surfaces and the amount removed from the inhaled air at such sites-- and their clearance-- the rates and routes by which deposited particles are phys ically translocated from the respiratory tract. For toxicants that exert their primary action on the surface contacted, the initial deposition pattern is the major predictor o f response. However, in many cases, it is the net result o f deposition and clearance processes, namely retention-- the number o f particles remaining in the respiratory tract at specific times after exposure-- that determines the degree of any hazard from exposure. The extent and loci o f particle deposition depend upon a number o f host factors. The m ajor ones are respiratory tract anatomy and ventilatory characteristics. The human respi ratory tract anatomy differs from that o f most other mammals used in inhalation toxicology studies, although the implications o f this d if ference for particle deposition have not been adequately appreciated. The respiratory tracts o f most mammals have comparable anatom ical components, but there are considerable differences in the structure o f some o f these components. In the upper respiratory tract, for example, there are species differences in the relative shapes and sizes o f nasal airways (Patra. 1986; Schreider. 1986). In general, labo ratory animals have much more convoluted nasal turbinate systems than do humans, and the length o f the nasopharynx in relation to the entire length o f the nasal passages also d if fers between species. Another major difference between humans and most other mammals commonly used in inhalation toxicology studies is the pattern o f tracheobronchial tree branching (Patra. 1986; Schlesinger and McFadden. 1981); the branching is much more symmetrical in hu mans than in most other laboratory animals. The branching pattern affects the depth of penetration o f inhaled panicles, because the number o f branch divisions from the trachea to each distal bronchiole may differ, and in fluences localized patterns o f deposition within individual airway generations. There are also large interspecies differences in the structure o f the pulmonary (alveolar) region (Gehr et ai. 1981; Tyler. 1983). including differences in the number o f generations o f bronchioles and the extent o f bronchiolar alveolarization. Alveolar size also differs between species; this may affect deposition efficiency, because of variations in the distance between airborne particles and airway walls. Breathing pattern during exposure to par ticle-containing gases or vapors influences the sites and relative amounts o f regional depo sition. Differences in tracheobronchial tree structures between species also affect airflow patterns. Furthermore, minute ventilation (even when adjusted for differences in body weight) differs between species and. in turn, affects both the number o f particles available for deposition and their depth o f penetration into the lungs (Schum and Yeh. 1980). Finally, most experimental animals are obligate nasal breathers. Coupled with the aforementioned greater complexity o f the nasal passages, the result is greater particle deposition in the upper respiratory tract o f experimental animals than occurs in humans breathing orally or even na sally. The extent o f upper respiratory tract re moval affects the number o f particles available for penetration into the lungs and. therefore, influences dose to the rest o f the respiratory system. Thus, the relationship between respi ratory tract deposition in obligate-nasal breathing experimental animals and orally breathing humans should be defined. The deposition o f inhaled particles occurs by similar physical mechanisms in both hu mans and experimental animals. However. there ; tionshi etficier as a pe inhalet region1 spirato But e\ pressec partick may di same a o f bod) will ret unit wi will lar ample, rat will fined o Pan: either c comple other mechai o f both deposit trachec cociliar lining c tion of the mm \eolar) anisms o f each chemic deposits insolub resident may tht gion aft mucoci, through tern. So lining fl recti v. Becat rates an sessmen l^aem of tra, 1986; '81); the cal in huanimals, depth o f cause the ie trachea and inon within e are also structure (Gehr et ifferences snchioles rrization. ;cies; this xause of airborne e to parcnces the >al depohial tree > ' rflow ition in body in turn, lvailable letration . Finally, ite' nasal rationed iges, the ie upper ials than even natract reivailable lerefore. piratory ra respie-nasalI orally , occurs oth huowever. COMPARATIVE DOSIMETRY OF INHALED MATERIALS 3 there are interspecies differences in the rela tionship between particle size and deposition efficiency (the amount o f deposition expressed as a percentage o f the total number o f particles inhaled for specific particle sizes) in different regions o f the respiratory tract, and for the re spiratory tract as a whole (Schlesinger. 1985). But even in cases for which deposition ex pressed as a percentage o f the total number o f particles inhaled is similar, the mass deposition may differ for different species exposed to the same atmosphere. When examined in terms of body or lung mass (weight), smaller animals will receive greater initial particle burdens per unit weight and per unit exposure time than will larger ones (Phalen et a/.. f 977). For ex ample, the deposition o f I-Mm particles in the rat will be 5 - I0 times that in humans, if de fined on a per unit-lung-weight basis. Particles that do deposit on airway surfaces either can be cleared from the respiratory tract completely or they may be translocated to other sites w ithin this system. Clearance mechanisms are regionally distinct, in terms ofboth routes and kinetics. Insoluble particles depositing in the upper respiratory tract and tracheobronchial tree may be cleared by mu cociliary transport, whereby a fluid (mucous) lining over the epithelium is moved by the ac tion o f cilia. Soluble material may dissolve in the mucus. Clearance from the respiratory (al veolar) region occurs by a number o f mech anisms and pathways; the relative importance of each appears to depend upon the physicochernical properties and amounts o f material deposited. A major pathway for clearance of insoluble particles involves phagocytosis by resident alveolar macrophages. These cells may then be cleared from the pulmonary re gion after reaching the distal terminus o f the mucociliary transport system or by migrating through the interstitium to the lymphatic sys tem. Soluble particles w ill dissolve in alveolar lining fluid and enter the blood or lymph di rectly. Because dosimetry depends upon clearance rates and routes, adequate toxicological as sessment necessitates relating clearance kinet ics in laboratory animals to those in humans. Although the basic mechanisms o f clearance from the respiratory tract are similar in most species, regional clearance rates may vary sub stantially. For example, clearance from the al veolar region o f mice and rats is much faster than that in dogs and humans, which have sim ilar clearance rates (Snipes, 1989). It is likely that dissolution rates, and rates by which dissolved substances are transferred into blood, are related solely to the properties o f the ma terial being cleared and are essentially inde pendent o f species (G riffith et aL. 1983; Bailey et aL. 1985). On the other hand, different rates o f mucociliary transport in the conducting airways (Felicetti et aL. 19 8 1) or o f macro phage-mediated clearance from the alveolar region (Bailey et at.. 1985) may result in spe cies-dependent rate constants for these path ways. In terms o f mucociliary transport, mu cous velocities, in the larger airways at least, seem to be proportional to body weight, while interspecies differences in functional proper ties o f macrophages include differences in phagocytic efficiency and differences in mo bility. both random and in response to chemotactic factors that may be released upon deposition o f certain particles. The end result o f deposition and clearance processes is retention, which may differ for the same particles inhaled by different species. Thus, especially during chronic exposures, identical lung burdens may not be obtained in all animals, and if the toxic response is re lated to the lung burden, the biological effects o f the inhaled material may also differ. The latter must be considered whenever experi mental animals are used in inhalation toxi cology studies, because such species differences may make it difficult to extrapolate from the results o f studies with laboratory animals to the results expected in humans. There is no one species that is an ideal surrogate for hu mans in terms o f being similar in all aspects o f those factors that affect particle disposition. Some species may deposit the particles sim i larly; others may deposit them differently but may have similar mucociliary or long-term clearance. There is. thus, a definite need to obtain reliable, baseline comparative data that w ill allow more precise interspecies extrapo lations o f deposition and clearance, so that better estimates o f risk to humans may be ob tained from particle inhalation studies with experimental animals. Comparative Dosimetry o f Inhaled Reactive Vapors (H. d 'A. Heck) Reactive vapors are defined as volatile compounds that can undergo nonenzymati cally catalyzed chemical reactions in biological systems. Such compounds include aldehydes, epoxides, isocyanates, halohvdrins. halome thyl ethers and ketones, d-lactones. and in organic oxidizing agents. Reactive vapors are frequently toxic to the respiratory tract, but systemic toxicity may be induced depending on the solubility and reactivity o f the vapor. Reactive vapors are inherently unstable: hence, to measure delivered dose, a fragment o f the reactive vapor covalently bound to a target molecule, rather than the vapor itself, is often identified in the tissue o f interest. The amount o f bound fragment is then related to the inhaled concentration o f the reactive vapor by using a pharmacokinetic model, thereby linking the delivered and administered doses. In principle, a variety o f target molecules, in cluding D N A. RNA. proteins, peptides, or lipids, could be used to measure delivered dose. The criteria for selecting an appropriate target include relevance to the toxic effect, biological half-life o f the target molecule, con centration o f target sites relative to the con centration o f the toxicant, and the sensitivity and facility o f the analytical method. In this article, we discuss two approaches to modeling the dosimetry o f reactive vapors. In the first approach, a physiologically based pharmacokinetic model is used to estimate the extent o f the reaction o f formaldehyde with D N A in the nasal mucosa o f F344 rats, rhesus monkeys, and adult humans. In the second approach, a convection-diffusion-chemical reaction model (Overton et ai. 1987)-is used to estimate the dose o f ozone delivered to the lower respiratory tract tissues o f rats and hu mans. These models differ significantly in their definitions o f dose and in their methods for calculating the quantity o f toxicant delivered to target macromolecules or tissues. Formaldehyde (HCHO) is an upper respi ratory tract (U R T) toxicant that is absorbed prim arily in the nasal cavity. Dosimetry mea surements o f HCHO have been performed by analyzing the concentration o f D N A-protein cross-links produced in the nasal mucosa of F344 rats and rhesus monkeys exposed to [ l4C]formaldehyde (Casanova et al,, 1989; Heck et ai. 1989). Cross-link formation in rats was described in terms o f a one-compartment pharmacokinetic model that combines the anterior nose and the nasal turbinates in a sin gle unit (Fig. I). Cross-linking in monkeys was interpreted by using a three-compartment model involving the anterior nose, turbinates, and nasopharynx as separate units (Fig. 1). HCHO was assumed to be absorbed from the nasal airstream into each compartment and, w ithin each compartment, to be eliminated by saturable reactions (metabolism) or by par allel. nonsaturable processes, one o f which is D N A -protein cross-link formation (Fig. 2). The kinetic constants relating delivered-toadministered concentrations were estimated by fitting the models to the nonlinear concen tration-response curves observed for cross-link formation. Lower concentrations o f cross-links were measured in the nasal tissues o f monkeys than in those o f rats, at all airborne concen trations. Because the pharmacokinetic models for HCHO are based on identifiable (and mea surable) physiologic parameters, the doses ob served should be predictable across species by applying the appropriate scaling factors. Cross link concentrations in monkeys were predicted from those measured in rats by using conven tional allometric scaling methods, but the pre dictions agreed only semiquantitatively with the experimental results. This was probably due. in part, to the marked differences in URT I abs or I anatorr monke similar to-hun rat-to-i eross-li from tl suggest co nee n in mor ofalloi pharm about . owing the inti for hig and pr Ozo toxicar tional ! ble boi in nasr simetn eontini etui.. I gionsa been c model s used to the nd hu ll their ids for livered respisorbed y meaned by protein cosa of >sed to 1989; 1 in rats irtment les the n a sineys was irtm e n t binates. Fig. 1). rom the md. mnated by parvhich is Fig. 2). ered-to.ti mated concenross-link oss-1inks nonkeys concen- >dels for nd mealoses oblecies by -s. Crossiredicted cortvent the pre/ely with probably s in URT COMPARATIVE DOSIMETRY OF INHALED MATERIALS 5 3-C.QMPABIMENT (MONKEY! r~ H INHALED V AIR ----1-a r L / TURBINATES \ / ERtTB. ANTERIOR NOSE ER. \ y LOWER AIRWAYS Fig. 1. Compart menial models o f the nasal conducting airways illustrating parameters pertinent to HCHO absorption: V. minute volume: <r. fraction o f airstream contacting the turbinates o f the monkey nose: ER or ER,. extraction ratio [fraction o f HCHO removed by a particular compartment from the airstream. where / = turbinates (TB). anterior nose (AN), or nasopharynx (NP)j. anatomy and physiology between rats and monkeys. The human nose is structurally similar to the monkey nose; hence, monkeylo-human scaling may be more accurate than rat-to-monkey scaling. Concentrations of cross-links in the human nose were predicted from those measured in monkeys; the results suggested that H CHO is likely to induce lower concentrations o f cross-links in humans than in monkeys. The calculations support the use ol allometric scaling in the absence o f suitable pharmacokinetic data, but there is still doubt about the quantitative validity o f the results, owing to uncertainties in the magnitudes o f the interspecies conversion factors, especially for highly divergent species such as rodents and primates. Ozone is a lower respiratory tract (LR T) toxicant that reacts rapidly with certain func tional groups, especially carbon-carbon dou ble bonds, in biological constituents present ln nasal mucus and cells. Measuring O i do simetry is experimentally challenging and continues to be an im portant goal (Santrock el ub, 1989). O i doses delivered to specific re gions and tissue components o f the LR T have been calculated by using a mathematical model that involves convection, diffusion, and chemical reactions in the lower airways (Fig. 3) (Overton el a i. 1987). Key elements o f the model are the dimensions o f the tracheobron chial airways and pulmonary regions and the thickness and composition o f the liquid lining (mucus or surfactant) covering the cells o f the LRT. The model predicts that a sharp maximum in the tissue dose o f O, w ill occur at the ter- METABOLITES METABOLITES + DNA-PROTEIN OTHER PRODUCTS CROSS LINKS Fig . 2. Reaction scheme for HCHO in compartment i: Cmu and airstream concentrations o f HCHO at anterior and posterior ends, respectively, o f compartment i'las.,. concentration o f HCHO in tissue of compartment i: 1mTMand Am. apparent maximal velocity and Michaelis constant, respectively, for elimination o f HCHO by a sat urable metabolism: kr. pseudo-first-order rate constant for removal o f HCHO by nonsaturable proces es other than DNA-protein cross-link formation: k h. pseudo-first-order rate constant for DNA-protein cross-link formation. 6 DAHL ET AL. ! LU M E N /A IR S P A C E C O N V E C T IO N . 03 O IF F U S IO N 1" LIQ U ID LIN IN G PROOUCT -- ---- HC T IS S U E H C -- - --1-- P R O D U C T BLOOO H C * ----- - P R O D U C T O IFFU SIO N C H E M IC A L R E A C TIO N S FlC. 3. Diagram illustrating some ol'the important physical and chemical processes taken into account by the dosimetry model ofO >: HC. a biological constituent that reacts with O j. Reprinted with permission from Overton et a t (1487). minai bronchiole-alveolar junction, which is the major site o f tissue damage in experimental animals exposed to O 3. The model also pre dicts that rats w ill receive a higher tissue dose than humans at a given airborne concen tration. This mathematical mode! pro vides a novel and potentially very useful ap proach to the problem o f interspecies com parisons and facilitates examinations o f different exposure scenarios. Unfortunately, the absolute accuracy o f dose predictions is presently unknown, and the doses calcu lated by using the model are not readily vali dated. Dosimetry measurements are very im por tant for understanding how species differ in their responses to reactive vapors and for pre dicting human risks. The database is currently quite lim ited, however, and much more in formation is needed with respect to other compounds and species (especially nonro dents). Further advances may be anticipated in several developing areas, including air flow characterization o f the upper respiratory tract, integration o f dosimetric pharma cokinetic models with models o f toxic re sponses (such as tum or growth models), and experimental validation o f model predic tions. Comparative Uptake and Pule ql Inhaled Or^anie l apors (M. 1. Medinsky) Volatile organic chemicals are o f interest to toxicologists, because they may have carci nogenic or other toxic effects. They are present in air. household and industrial products, cig arette smoke, and gasoline vapors, and ever ybody receives some degree o f exposure. The dosimetry o f these volatile organics is fre quently complicated by the fact that, for many organics, the metabolites are more toxic than the parent chemical. Because o f the great di versity in the biological activities and physical properties o f these volatile organics, it is useful to examine, in a general wav. the determinants most important for influencing their uptake, distribution, and elimination. These deter minants include those o f a physiological na ture. chemical factors, metabolic determi nants. and those governing the binding o f the parent chemical or metabolite to tissue mac romolecules. Physiological determinants. Physiological determinants, such as organ weights and physiological processes, have been shown to be related to body weight through the allometric expression, property oc (body weight)", where the property o f interest is proportional :ouni ission <aled Or- iterest to ve card- sent ucts. cigmd everure. The :s is frefor many )xic than great d i physical : is useful rminants r uptake, se deterigical nadeterming o f the sue mac- siological ghts and shown to the alloweight)" , portional COMPARATIVE DOSIMETRY OF INHALED MATERIALS 7 to a power function o f body weight (Adolph, 1949). For many tissue weights, such as those of liver, kidney, blood, and heart, the all meme component, a, is approximately equal to 1. Thus, the weight o f these organs increases in direct proportion to increasing body weight. For flow-related physiological processes, such as blood flows, clearance, or ventilation, the allmeme exponent lies between 0.65 and 0.8. These flow rates, when expressed per unit o f body weight, are slower in larger animals than in smaller ones. Differences in the allometric relationships for weights and flows have an impact on kinetics and tissue dosimetry that is important to take into account when ex trapolating across species. For example, the terminal half-time (l{) o f a chemical (Eq. (1)) tends to be shorter in smaller animals than in larger ones: ,, volume o f distribution t = 0.693 X ----------- --------------------- . (1) clearance The terminal f j decreases with body weight because, as size decreases, the volume o f dis tribution. a weight-related parameter, de creases faster than clearance, a measure o f flow. Thus, tissue exposure to volatile organics may be prolonged in larger animals such as man, in comparison to the exposure o f tissues in smaller animals such as rodents. For ex ample, after an inhalation exposure to a vol atile organic in which equivalent, steady-state blood concentrations are achieved, the clear ance o f the organic from blood w ill be slower in larger animals than in smaller ones. This interspecies difference in systemic clearance is illustrated by comparing the time course for blood concentrations o f styrene in rats and humans exposed to 80 ppm styrene for 6 hr (Ramsey and Andersen, 1984) (Fig. 4). Note the similar, achieved concentrations in the two species at the end o f exposure. However, the disappearance o f styrene from rat blood is much more rapid than the comparable decline of that in humans. The slower decline in hu mans is due to the slower rate o f processes such as blood flow and metabolism, in addi tion to the larger fat volume o f the human Hours After Start of Exposure FlG. 4. Blood concentrations o f styrene in rats and hu mans exposed to 80 ppm styrene for 6 hr. Circles are ex perimentally determined values. Lines are the results of physiological model simulations. Data taken from Ramsey and Andersen (1984). compared to that o f the rat. Differences in flow-related processes across species also in fluence the uptake o f inhaled organics into blood. It will take longer for steady-state blood concentrations to be achieved in larger animals than in smaller ones (National Academy o f Sciences. 1986). The prolonged tissue exposure to volatile organics in humans compared to that in smaller animals such as rodents may have im plications for volatile organics for which a metabolite is the toxic species. The total time interval over which metabolism can take place might be longer in larger animals compared to that in smaller ones. Partition coefficients. Partition coefficients are a measure o f the affinity o f a chemical for one medium compared to another at equilib rium . Partition coefficients o f organic vapors are often determined in vitro by using methods such as vial equilibration (Gargas et ai. 1989; Sato and Nakajima. 1979). Partition coeffi cients used in physiologically based models are often described in terms o f bloodiair (Eq. (2)). tissue:air (Eq. (3)) or tissue:blood (Eq. (4)) re lationships. Pb = [B lood]/[A ir], (2) Px = [Tissue]/[ Air], (3) Plh = PJP*.. (4) 8 DAHL ET AL. TABLE 1 Partition Coefficients for Some Volatile Organic Chemicals" Chemical Blood/air Fat/air Muscle/air Isoprene Benzene Styrene 3 72 18 500 40 2000 2 11 40 " Data taken from Gargas et al. ( 1989). where [A-,] = the concentration o f the volatile organic in that medium at equilibrium . Table 1 contains partition coefficients for several volatile organic chemicals. The blood: air partition coefficient is a critical determinant in the uptake and achieved blood concentra tions o f volatile organic chemicals (Fig. 5). As the blood:air partition coefficient increases from 3 for isoprene to 40 for styrene (Table 1). the concentration o f the volatile organic in the systemic circulation increases for expo sures at equivalent airborne concentrations. Similarly, the fat:air partition coefficients for isoprene. benzene, and styrene (Table 1) in dicate that the highest fat concentrations o f volatile organics w ill be achieved by very li pophilic chemicals such as styrene. The fat compartment plays an im portant role in ac cumulating and storing the volatile organic, both during and after exposure. This stored chemical becomes available for distribution by the systemic circulation to the metabolizing organs after the end o f exposure. The im por tance o f postexposure metabolism is demon strated in Fig. 6, which shows that for isoprene, the chemical with the smallest lat:air partition coefficient, most o f the metabolism occurs during the 6-hr exposure. For benzene, which has a larger partition coefficient than isoprene. the majority o f metabolism still occurs during exposure; however, approximately one-third o f the benzene metabolism occurs after the end o f the exposure. For the highly lipid-sol uble organic, styrene, over 50% o f metabolism takes place after the end o f exposure. Figure 6 illustrates the importance o f fat as a storage Fig . 5 Effect of blood/air partition coefficient on steadystate blood concentrations of three volatile organics, both during and alter a 6-hr inhalation exposure to 600 ppm. compartment for volatile organics and em phasizes that metabolism o f the volatile or ganic can continue for a significant period of time after the individual is removed from the exposure atmosphere. Metabolic determinants. Metabolic deter minants such as k max (maximum metabolic capacity) and Km(affinity o f the enzyme for a substrate) can be key determinants in the ex posure o f tissues to the toxic chemical species o f volatile organics. For many volatile organ ics. the metabolite o f the chemical, and not Fig . 6. Effect of lat/air partition coefficient on postex posure metabolism of three volatile organics, both during and after a 6-hr inhalation exposure to 600 ppm. 10 sure i sieadvics. both )0 ppm. id em ule orriod of om the deter'lie riv. .or a the ex species organmd not IRENE iPRENE NZENE COMPARATIVE DOSIMETRY OF INHALED MATERIALS 9 the chemical itself, is responsible for the tox icity. The importance o f metabolic determi nants is illustrated here for benzene. Sabourin ci al. (1988. 1989) and Medinsky el al. (1989) determined that the major, stable benzene metabolites produced by B6C3F, mice and F344 rats were sulfate and glucuronide con jugates o f phenol and hydroquinone. gluta thione conjugates o f phenol, and muconic acid, an opened-ring metabolite. Distinct spe cies differences in the formation o f these me tabolites were noted (Fig. 7). When rats were exposed to 600 ppm benzene for 6 hr. the pri mary metabolite produced was the sulfate conjugate o f phenol. Smaller amounts o f the glutathione conjugates and muconic acid were formed. A very small amount o f hydroquinone sulfate and virtually no hydroquinone gluc uronide were detected. In mice exposed to the same concentration o f benzene, the profile o f metabolites was very different. Significant quantities o f other metabolites in addition to phenyl sulfate were produced, including hy droquinone glucuronide and phenyl glucuro nide. Studies by Sabourin et al. (1989) and Medinsky el al. (1989) showed distinct, con centration-related differences in the profiles o f metabolites, in addition to the species-related differences. These studies indicated that, based on present knowledge, allometric scaling could not be used to predict the behavior o f benzene in one o f these rodent species based on its metabolic capacity in the other species. In summary, uptake o f volatile organic chemicals is a complex process that can be described in the context o f physiological, chemical, and biochemical determinants. Physiological factors that are im portant in the tissue dosimetry of volatile organics can gen erally be extrapolated across species. Because flow-related determinants tend to extrapolate across species as a fractional power o f body weight, the time course for systemic uptake and elim ination and for tissue exposure to ab sorbed chemicals w ill be prolonged for species with larger body weights (humans) compared to those o f species with smaller body weights (rodents) exposed to the identical concentra tion o f the chemical. Chemical determinants o f uptake can vary significantly among volatile organic chemicals, but generally are similar across species for particular chemicals. Chem ical determinants such as blood:air and tissue: blood partition coefficients govern the sys temic and tissue concentrations o f volatile organics and their metabolites. Metabolic capacity is probably the most important determinant o f tissue dosimetry, because me tabolites o f many volatile organics are more toxic than the parent chemical. Unfortunately, at the present time, metabolic rates are the most difficult to predict across species, because F 3 4 4 RATS - 6 0 0 ppm <uD n0m) o ~u B6C3F1 MICE - 6 0 0 ppm 0 25 )sure on postex>oth during >m Fig . 7. Differences in the urinary excretion of benzene metabolites by rats and mice exposed to 6(X) ppm benzene for 6 hr. IIQG. hydroquinone glucuronide; PPM A. pre-phenvl mereapiuric acid; MUC. muconic acid; PCil liC . phenyl glucuronide: C G LU C catechol glucuronide; HQS04, hydroquinone sulfate: PS04. phenyl sulfate: and PMA. phenylniereepaturic acid. 10 DAHL ET AL. o f our lack o f knowledge o f the factors that are critical for understanding species differ ences in the metabolism o f organic chem icals. Molecular Dosimetry of Inhaled Carcinogens: Implications for Epidemiology and Risk Assessment (G. W. Lacier) There is a great deal o f uncertainty in the estimation o f human risks from low-dose ex posure to chemical carcinogens when high dose animal data are used as the basis for ex trapolation. It is increasingly evident that mo lecular approaches can contribute a great deal to reducing the uncertainties that are inherent in the risk assessment process when gross bio logical endpoints such as tumors are used. This knowledge has led to the development o f ap proaches designed to incorporate biomark ers into toxicological and epidemiological studies. The framework for incorporating molecular data or "biom arkers." as they are frequently called, into the risk assessment process is il lustrated in Fig. 8. The essence o f this illus tration is that there are numerous biological and biochemical events that ultimately deter mine an adverse health effect following ex posure to a toxic chemical. The chemical must first be internalized, leading to its presence in blood or tissues. There are numerous cases where toxic chemicals have been detected at extraordinarily low concentrations in blood. For example, recent developments in analytic methodology have lowered the lim its o f de tection for 2,3.7.8-tetrachlorodibenzo-/wJioxin (TC D D ) to the low. part-per-trillion range, al lowing detection o f this compound in the gen eral population. Once internalized, many chemicals are metabolized by a wide variety o f drug-metabolizing enzymes that are present in virtually every tissue o f the body (Lucier el ai, 1979). These enzymes include the multiple forms o f cytochrome P450, epoxide hydrolase glutathione transferase, glucuronyltransferase, and sulfotransferase. among others. Depend ing on the chemical being metabolized, each o f these enzymes may play a role in either an activation or a detoxication pathway. For ex ample. glutathione transferase detoxifies elec trophilic arene oxides o f polycyclic aromatic hydrocarbons (Jerina and Bend. 1975). but catalyzes the formation o f a DNA-reactive metabolite o f ethylene dibromide (Guengerieh el at.. 1987). The balance between activating and detox ifying enzyme systems governs the rate o f de livery o f bioactive metabolite to the macro molecular target site. For carcinogens, which are initiating agents, the macromolecular in teraction o f interest could be a D N A adduct, and for chemicals that are tum or promoters, the critical interaction could be receptor oc cupancy. It should be noted that "in itia tio n " and "p ro m o tio n " are operational terms, not stages that have clearly defined mechanisms. In any event, the concentration o f DNA ad duct or occupied receptor has been termed the "biologically effective dose" and leads to BIOCHEMICAL MARKERS IN RISK ASSESSMENT Exposure------- Internal Dose Pharmacokinetics------- Biologically Effective Dose (P-450 isozymes (DNA-adducts. receptor conjugation Rxns) occupancy) Toxic Response (tumor incidence) Critical changes in gene expression (oncogenes; growth factor responses) Markers of exposure not associated with mechanism F ig . 8. S chem atic representation o f the sequence o f events p ro d u c in g a to x ic response a fte r exposure to a toxicant. change not be cinoge cinoge "mark arvlh) lowing drocar TCDL such a teratic have c carcin berg ei sible c chemn evalua publisi (NRC One risk as dose-r econo i imal b menta cidenc anima is a h such a^ can us much I increas reason: of bion exposu removi assessn able eh respon: of stuc simetn. The i on the simetei cell spe ers (i.e. ways o sample lits o f dev/j-dioxin range, aln the gen 'd, many je variety re present (Lucier et e multiple hydrolase ansferase. Dependzed. each either an y. For ex.ifies elecaromatic 975), but Vreactive uengerich nd detoxate o f deie macro- ffiich ecular in^ adduct, romoters. eptor oclitia tio n " ;rms, not .hanisms. DNA ad:rmed the leads to urc lo COMPARATIVE DOSIMETRY OF INHALED MATERIALS 11 changes in gene expression, which may or may not be associated with the mechanism o f car cinogenesis. Changes not involved in the car cinogenesis process may provide a reliable " marker o f exposure." such as occurs when ahylhydrocarbon hydroxylase is induced fol lowing exposure to polycyclic aromatic hy drocarbons such as benzo[]pvrene and T!CDD. Other changes in gene expression, such as activation o f proto-oncogenes and al terations in cell proliferation pathways, may hjave direct links to mechanisms o f chemical carcinogenesis (Reynolds et uL. 1987: Swenberg et a!.. 1987) and may also represent pos sible candidates for "markers o f effect" for cftemical carcinogens. A detailed review and evaluation o f biomarkers has recently been published by a National Research Council (NRC) Committee (1987). One o f the most compelling issues in the risk assessment process is the estimation o f dose-response relationships. For reasons o f economic necessity, high doses are used in an imal bioassavs for carcinogenicity o f environ mental chemicals. Extrapolation o f tum or in cidence arising from high-dose exposures in animals to low-dose exposure risks in humans is a highly uncertain exercise. Biomarkers, such as activated metabolites or D N A adducts, can usually be detected following exposure to much lower doses than those needed to detect increased tum or incidence. Therefore, it seems reasonable to assume that a careful evaluation of biomarkers in experiments covering a wide exposure dose range offers an opportunity to remove some o f the uncertainty in the risk assessment process by providing a more reli able estimation o f the shape o f the doseresponse curve at low doses. These kinds of studies have been termed molecular do simetry. There are several issues that have an impact on the use o f D N A adducts as molecular do simeters. These include adduct heterogeneity, cell specificity, and the use o f surrogate mark ers (i.e., lymphocytes). The two most sensitive w3ys o f detecting D N A adducts in human samples are immunochemistry (Perera, 1987; Harris et a/., 1985) and 32P-postlabeling (Reddy and Randerath. 1987). One o f the best examples o f a DNA adduct used as a molec ular dosimeter is N N K (a carcinogenic me tabolite o f nicotine) (Belinsky et a!.. 1987). O6Methylguanine. the promutagenic adduct o f N N K , is formed more efficiently in the lung at low exposure doses than at high exposure doses, and this finding is consistent with the dose-response relationships observed for the carcinogenicity o f this compound. Bond et at. (1989) characterized dose-response patterns for DNA adducts arising from exposure to polycyclic aromatic hydrocarbons present on particles such as diesel exhaust. These studies related relative adduct concentrations to lung tum or incidence. In another study. DNA ad ducts were detected in human lymphocytes by ^P-postlabeling. but the adduct profiles o f smokers and nonsmokers were not different (Jahnke et a!.. 1990). A great deal o f interin dividual variation exists in adduct concen trations. and this variation may reflect poly morphisms in metabolic activation/deactivation reactions in human populations. For example, one study showed that the polymorphism o f glutathione transferase u effects DNA adduct formation o f some carcinogens, but not o f others (Liu et a!.. 1990). The results o f a recent National Toxicology Program study (Tennant et at.. 1987) suggest that as many as 40% o f the chemicals that are positive for carcinogenicity in the lifetime bioassay are acting through nongenotoxic mechanisms. The im plication is that effects o f DNA adducts on signal transduction path ways. receptor-mediated proliferative re sponses, and cell cycle control are involved in the mechanism o f action o f many carcinogens. Therefore, one area o f research needed is the evaluation o f the quantitative relationships between biochemical events involved in tumor promotion and carcinogenic incidence. A central underlying need relative to molecular dosimetry studies is increased knowledge o f the diverse mechanisms o f chemical carcino gens. 12 DAHL ET AL. REFERENCES Adolph. E. F. (1449). Quantitative relations in the phys iological constitutions o f mammals. Science 109, 579 585. Ba il e y . M . R.. Ho d g s o n , a ., and Sm it h , H. (1985). Respiratory tract retention o f relatively insoluble par ticles in rodents. J. Aerosol Sci. 16, 279-295. Be lin s k y . S. a .. W h it e . C. M.. D e v e r e u x , T. R.. and A nderson. M. W. (1987). DN A adducts as a dosimeter for risk estimation. Environ Health hersped. 76, 3-8. Bo n d . J. a .. Harkema. J. R.. Henderson. R. F.. M auderly. J. L.. M cC lellan. R. O.. and Wolff. R. K. (1989). Molecular dosimetry o f inhaled diesel exhaust. In Assessment o f Inhalation Hazards, pp. 315-324. Springer-Verlag. New York/Berlin. Casan o va . M .. D e y o . D. F.. a n d H eck. H. d 'A. ( 1989). Covalent binding o f inhaled formaldehyde to DNA in the nasal mucosa o f Fischer 344 rats: Analysis of form aldehyde and DNA by high-performance liquid chro matography and provisional pharmacokinetic interpre tation. Enntlam. Appl. Toxicol. 12, 397-417. Felicetti. S. A.. Wolff. R. K... and M uggenburc,. B. A. (1981). Comparison o f tracheal mucous transport in rats, guinea pigs, rabbits, and dogs. J Appl. Physiol. 51.1612-1617 G argas. M. L.. Burgess. R. J.. V o is a r d . D. E . C ason. G. H.. a n d A ndersen. M. E. (1989). Partition coeffi cients o f low-molecular-weight volatile chemicals in various liquids and tissues. Toxicol Appl. Pharmacol 98, 87-99. G ehr. P.. M w a n g i. D. K,. A m m a n n . a .. M a lo iy . G. M. O.. T aylo r . C. R.. an d W eibfl. E. R. (1981). Design o f the mammalian system. V. Scaling morpho metric pulmonary diffusing capacity to body mass: Wild and domestic animals. Respir. Physiol. 44, 61-86 G r if f it h . W. C.. C u d d ih a . R. C ,, Bo ecker. B. B . G u il m e t t e . R. A.. M e d in s k y . M. A., a n d M e w h in ney. J. A. (1983). Comparison o f solubility o f aerosols in lungs o f laboratory animals. Health Phys. 45, 233. G u e n g e r ic h . F. P . Peterso n. L. a .. C m a r ik . J. L.. Ko g a . N.. a n d I nskeep. P B. (1987). Activation of dihaloalkanes by glutathione conjugation and formation of DNA adducts. Environ Health Perspect 76, 15-18 H arris. C. C.. V a h a k an g a s. K . N e w m a n . M. J.. T ri vers. G. E.. Sh a m s u d d in . a .. Sin o p o l i. N.. M a n n . D. L.. a n d W r ig h t . W. E. (1985). Detection of benzo(a)pyrene-diol epoxide-DNA adducts in peripheral blood lymphocytes and antibodies to the adducts in serum from coke oven workers. Proc Sal! Acati Set. USA 82, 6672-6676. H eck. H. d 'A.. Ca s a n o va . M .. St e in h a g e n . W. H.. Ev e r it t . J. I.. M o r g a n . K. T . a n d Popp. J. A. 119X9). Formaldehyde toxicity-- DNA-protein cross-linking studies in rats and nonhuman primates. In Nasal ( arcinoyenesis in Rodents Relevance to Haitian Health (V. J. Feron and M. C. Bosland. Eds.), pp. 159-164. Pudoc. Wageningen. The Netherlands. .. Jahnke. G. D . T hompson. C. L.. Walker. M, P,, Gal lagher. J. E.. Lucier. G. W , and D iAiigustine, R. P. (1990). Multiple DNA adducts in lymphocytes of smokers and nonsmokers determined by J'P-postlabeling analysis. Carcinoyenesis I I , 205-21 I. Jerina, D. M.. and Bend. J. R. (1975). Glutathione STranslera.se in Bioloyical Reactive Intermediates: For mation. Toxicity and Inactivation (D. J. Jollow. J. J. Kocsis. R. Snyder, and H. Vaimo. Eds ). Plenum. New York. Liu. Y. H.. M iller. C. R.. T aylor. J. A.. Nagorney. D.. Lucier. G. W.. and T hompson. C. L. (1990). G lu tathione S-transferse (GT) activity in human liver and lymphocytes and its role in regulating carcinogen-de rived DNA adduct formation. To.xicoloyisl 10. 145. Lucier. G. W.. Lu i. E. M. K . and Lamartiniere. C. A. (1979), Metabolic activation/deactivation reactions during perinatal development. Environ. Health Perspea 29. 7-16. M edinsky. M. a .. Sabourin. P. J.. Lucier. G ,, Birn baum. L. S.. and H enderson. R. F. (19X9). a physi ological model for simulation of benzene metabolism by rats and mice. Toxicol. Appl Pharmacol 99, 193 206. National Academy o f Sciences (19X6). Dose-route ex trapolations; Using inhalation toxicity data to set drink ing water limits. In Drinkiny II tiler and Health ( R. D. Thomas. Ed ). Vol. 6. Chap. 6, National Academy Press, Washington. DC. NRC Committee! 19X7). Biomarkers in Reproductive and Hevelopmental Toxicoloyy U S. Department of Health and Human Services. Public Health Service-- National Institutes of Health. National Institute of Environmental Health Sciences. Environ Health Perspect. 74. Overton. J. H.. G raham . R. C . and M iller. F J. (1987). A model o f the regional uptake of gaseous pol lutants in the lung. II. The sensitivity of ozone uptake in laboratory animal lungs to anatomical and ventilatory parameters. Toxicol. Appl Pharmacol 88. 418-432. Pa ERA. A. L. (1986). Comparative anatomy of mam malian respiratory tract: The nasopharyngeal region and the tracheobronchial tree. J Toxicol. Environ. Health 17, 163-174. Perera. F. P (14X7). The potential usefulness of biological markers in risk assessment. Environ. Health Perspect 76. 141-145. Ph ai en. R . Kenoyer. J.. and Davis. J. (1977). Depo sition and clearance of inhaled panicles: Comparison of mammalian species. In Prm eedinys o f the Annual Conlereiice on Environmental Toxicoloyy. Vol. 7. pp. 154-170. AMRL-TR-76-125. National Technical In formation Service. Springfield. VA. Ra m sey. J. C.. a n d A ndersen. M. E. (1984). A physi ologically based description of the inhalation pharma- cokii Phar Reddy belin .. medi lions REYNO M ar M. V licer 237. S aBOUI Luci in lh> rats a 12 8 Sabouf Birn ( I98c and r hy F: maco Santrc J. M, ygenToxii S\TO. a of SOI blood 9-164. GalNE. R. vies of tlabel- one Svv Forj. J. J. I, New RNEY. ). Gluer and en-de45. HERE. ictions rsped Bir n physivolism . 193- OnnkR. D. Press. eand lealth lional nental F. J. s polptake latory 32. mamn and lealth ogical sped. Tepoinson nnual I. pp. il In- ihysi- i- COMPARATIVE DOSIMETRY OF INHALED MATERIALS cokinetics o f styrene in rats and humans. Toxicol. Appi. Pharmacol. 73, 159-175. Reddy. M. V.. and Randerath. K. (198 7). 32P-postlabeling assay for carcinogen-DNA adducts: Nuclear P4mediated enhancement o f its sensitivity and applica tions. Environ. Health Perspect 76, 4 1-47. Reynolds, S. H.. Stowere. S. J.. Patterson. R. M.. Maronpot. R. R.. Aaronson. S. A., a nd Anderson. M. W. (1987). Activated oncogenes in B6C3F, mouse liver tumors: Implications for risk assessment. Science 237, jl 309-1316. Sabourin. P. J.. Bechtold. W. E,, Birnbaum. L. S.. LucipR. G.. and H enderson. R. F. (1988). Difference in th metabolism o f inhaled 'H-benzene by F344/N rats and B6C3F, mice. Toxicol. Appi Pharmacol. 94, 128-140. Sabourin. P. J.. Bechtold. W. E.. G riffith, w . C.. Birnbaum. L. S.. Lucier. G.. and H enderson. R. F. (1989). Effect ofexposure concentration, exposure rate, and route of administration on metabolism of benzene by F344 rats and B6C3F, mice. Toxicol Appi Phar macol. 99, 421 -444. Santrock. J.. Hatch. G. E.. Slade. R.. and Hayes. J. M. (1989). Incorporation and disappearance o f ox ygen-18 in lungs from mice exposed to I ppm "O ,. Toxicol Appi Pharmacol. 98, 75-80. Sato. Aj.. a n d Nakajima. T. 11979). Partition coefficients of some aromatic hydrocarbons and ketones in water, blood, and oil. Brit J. loci Meti 36, 231-234. Schlesinger. R. B. (1985). Comparative deposition of inhaled aerosols in experimental animals and humans: A review. J. Toxicol. Environ. Health 15, 197-214. Schlesinger, R. B.. and M c Fadden, L. (1981). Com parative morphometry o f the upper bronchial tree in six mammalian species. Anal. Rec. 199, 99-108. Schreider. J. P. (1986). Comparative anatomy and functions o f the nasal passages. In Toxicology o f the Nasal Passages (C. S. Barrow. Ed.), pp. 1-25. McGraw Hill. New York. Sc h u m . G. M.. a n d Y eh. H. C. (1980). Theoretical eval uation o f aerosol deposition in anatomical models of mammalian lung airways. Bull. Mailt. Biol. 42, 1-15. Snipes. M. B. (1989). Long-term retention and clearance o f particles inhaled by mammalian species. Crit Rev. Toxicol 20. 175-211. SWENBERG. J. A.. RICHARDSON. F. C.. BOUCHERON. J. A.. Deal, F. H.. Belinsky. S. a .. Charbonneau. M . and Short. B. G. (1987). High- to low-dose ex trapolation: Critical determinants involved in the dose response o f carcinogenic substances. Environ. Health Perspect 76. 57-63. T ennant. R. W.. M argolin. B H.. Shelby, M. D,, Z eiger. E.. Haseman. J. K.. Spalding. J.. Caspary. W.. Resnik. M.. Stasiew itz. S.. Anderson. B.. el ai (1987). Prediction of chemical carcinogenicity in rodents from in vitro genetic toxicity assays. Science 236(4804), 9 3 3 - 9 4 1. T yler. W. S. (1983). Comparative subgross anatomy o f lungs, .liner Rev Respir Dis !28(Suppl), S32-S36. r