a model for the formation and removal of hemoglobin adductssolutions to the analytical equations,...

Timothy R. Fennell,1 Susan C. J. Sumner, andVernon E. Walker2

Chemical Industry Institute of Toxicology, Research Triangle Park,North Carolina 27709

AbstractHemoglobin adduds formed by chemical carcinogenscan be used as biomarkers of exposure. The kinetics ofaddud formation and removal is complex and dependson the processes involved in erythrocyte removal,adduct stability, and the duration and extent ofexposure. In order to relate the formation of adduds tothe extent of exposure in complex exposure scenarios,a model has been developed to describe the kinetics ofaccumulation and removal of adduds formed in vivo.The exposure scenario, lifetime of erythrocytes, andextent of adduct formation following a single exposureare required input parameters. Predictions of addudaccumulation have been generated for a wide varietyof exposure scenarios and compared with both thesolutions to equations derived for adduct formationand removal and experimental observations. Loss ofaddud by removal of erythrocytes from circulation,both by senescence and random removal and as aresult of chemical instability, has been simulated.Equations have been derived to describe the removal ofhemoglobin adduds under conditions of exposure forless than the lifetime of the erythrocyte, when removalis initially a linear function of time. This model makespossible the comparison of data obtained fromdifferent exposure scenarios and in different species.

IntroductionAn important goal of research on biomarkers of exposureis the identification of individuals exposed to toxic andcarcinogenic chemicals. Relating the extent of exposureto an internal dosimeter and categorization of individualsaccording to their exposure or internal dose can increasethe power of epidemiological studies (1, 2). The meas-urement of hemoglobin adducts as dosimeters of expo-sure has been the subject of considerable investigationin the last two decades. Electrophilic chemicals or theirmetabolites can react with nucleophilic sites in hemogbo-bin to form stable adducts (3-5). Since erythrocytes are

readily obtained from humans and have a long lifetimein circulation, hemoglobin adducts have considerablepotential for use as long-term integrated dosimeters inpeople (6).

Many hemoglobin adducts have been found to bechemically stable and are not removed by repair. Forthese adducts, the kinetics of formation and removal invivo is determined primarily by the removal of erythro-cytes from circulation. Equations describing the accu-mulation of adducts following continuous exposure forup to the lifetime of the erythrocyte, or the removal ofadducts following exposures for the lifetime of the eryth-rocyte, have been derived (3, 7-9). With continuousexposure, the change in adduct level, dy/dt, can bewritten as

�=a-� (t�tt,r) (A)

where a is the daily increment in adduct formation, t is

the time since onset of exposure, and t,.r is the lifetimeof the erythrocyte (3). The accumulated extent of adductformation (3, 7), y, is obtained by integration:

Y=a(t_�_) (B)

As t approaches ter, the adduct levels reach a plateau,

and when t =

(C)

The removal of adducts on cessation of chronic exposurehas been described as a quadratic function, expressed asthe decline in adduct concentration from the plateau as

ate, ft�2\

Y=--�at’+a��--) (D)

where t’ is the time since the end of exposure (t � ter)

(8, 9). This behavior arises from the simultaneous fall inthe number of cells exposed, together with the fall inaverage cumulative exposure of the remaining cells.

Following a single exposure, linear decline in thehemoglobin adduct levels occurs, reaching zero at thelifetime of the erythrocyte, since the hemoglobin of theerythrocytes of all ages has been alkylated to a similarextent (3, 7).

With some compounds, the elimination kinetics forhemoglobin adducts are not zero-order as expected butare compatible with first-order behavior (10-12). Thehemoglobin adduct for 4-aminobiphenyl declined fasterthan expected from the erythrocyte lifetime in smokerswho had enrolled in a withdrawal program (9), althoughin rats exposed to 4-aminobiphenyl, the hemoglobin

Received 8/8/91.

1 To whom requests for reprints should be addressed, at Chemical Indus-

try Institute of Toxicology, P. 0. Box 12137, Research Triangle Park, NC

27709.2 Present address: University of North Carolina, Department of Pathology,Chapel Hill, NC 27599.

Vol. 1 , 2 13-2 1 9, March/April 1992 Cancer Epidemiology, Biomarkers & Prevention 213

A Model for the Formation and Removal ofHemoglobin Adducts

at,.,

on May 23, 2021. © 1992 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/

(E)

Adduct AdductFormation Formation

� �er � �,9/ter

�:1I�D-.- E�IIII::�- -�. �I., -� Removal

/k\ /\ /k\

Random Adduct Random AdductLoss Instability Loss Instability

Fig. 1. A model of hemoglobin adduct formation and removal. Each ofthe compartments, denoted by a circle, represents a pool 01 hemoglobin

in erythrocytes of increasing age. Daily adduct formation is divided

equally be;ween each compartment )a/t,.,). Adducts are lost by threeprocesses: senescence of erythrocytes; random loss of erythrocytes; and

adduct instability. Adduct instability and random loss were incorporated

as a single rate constant. The total level of adduct in circulation iscalculated as the sum of the adduct bevels in each compartment minusone-half the level of adduct in the oldest compartment. At each day

during the simulation, the contents of each compartment are advancedin age, a new compartment is added, the oldest compartment is removed,

and the adduct levels are recalculated.

which simplifies to give

at� aty=at-�----+�--- (C)

This differs from Equation (B) by at/2t,.,. The total adductconcentration was adjusted in the computer simulationsby subtracting one-half of the adducts accumulated inthe compartment corresponding to the oldest cellpopulation.

In addition to loss by senescence, both random lossof erythrocytes and adduct instability were incorporatedas a first-order loss with a rate constant k. The adductlevel in each compartment, y,,, where n is the age of eachcompartment, was computed for each day as

y,, = (�)e� + y,,e� (H)

Simulations have been carried out using valuesof erythrocyte lifetime of 120, 61, and 40 days forhumans (15), Sprague-Dawley rats (16), and mice (17),

respectively.The model was written in Simusolv (Mitchell and

Gauthier Assoc., Concord, MA), a Fortran-based contin-uous simulation language, and run interactively on a VAX4000 (Digital Equipment Corp., Maynard, MA).

Results

In order to test predictions of the model, simulations ofsingle exposure and continuous exposure for the lifetimeof the erythrocyte were conducted, and the model pre-dictions were compared with those of analytical equa-tions corresponding to the appropriate exposure sce-nario. In Fig. 2A, the simulation shown is for a singleexposure with an adduct increment (a) of 100 pmol/gglobin and an erythrocyte lifetime (t,.,) of 1 20 days, similar

(F) to the lifetime of human erythrocytes (1 5). The simulated

adduct level rises steeply on the first day and then

2 14 Modeling Hemoglobin Adducts

adduct appeared to be stable (1 3). This type of behaviormay arise from chemical instability of the adduct oraccelerated removal of the adducted erythrocytes. Theremoval of ester hemoglobin adducts produced from asingle dose of N-ethyl-N-nitrosourea in mice has beendescribed both by removal of erythrocytes and by add uctinstability:

I t\

y = YU�1

where Yo �5 the starting concentration of adduct, and therate constant for adduct instability k was estimated to be0.0023 h (14).

Many animal studies used to evaluate hemoglobinadducts as dosimeters are carried out using single or alimited number of exposures. It is important to be ableto relate observations in animal studies to those in hu-mans and to compare different exposure scenarios withinthe same species. From the work of others describedabove, the kinetics of accumulation and removal of stableadducts following single or continuous exposure can bereadily predicted from the lifetime of the erythrocyte (3,7, 9). The effects of intermittent exposures and exposuresfor less than the lifetime of the erythrocyte are less wellunderstood. In addition to influencing the extent of ac-cumulation of adducts, we have hypothesized that inter-mittent exposures will affect the shapes of both theadduct formation and removal curves. The objective ofthis study was to describe the kinetics of hemoglobinadduct formation and removal under conditions otherthan single or multiple exposures for the lifetime of theerythrocyte. To this end, a computer model was devel-oped that can simulate the accumulation and removal ofadducts under a variety of exposure scenarios, includingsingle, intermittent, or continuous exposures. This modelmakes possible the comparison of different exposurescenarios, the effect of exposure on erythrocyte lifetime,and adduct stability. Analytical equations were also de-veboped for exposures for less than the erythrocyte life-time. The model simulations were compared with thesolutions to the analytical equations, and data from ani-mal studies were evaluated using the model.

Materials and MethodsIn the model used for simulating the formation andremoval of hemoglobin adducts, the total erythrocytepopulation was described as a series of compartments ofsuccessively increasing age (Fig. 1 ). The number of com-partments was chosen to equal the lifetime of the eryth-rocyte (in days). Each compartment was encoded as anelement in an array, with the daily addition of a newcompartment, removal of the oldest compartment, andtracking of the remaining compartments by age. Inputrequired for the model included the lifetime of the eryth-rocyte, the adduct increment (the increase in adductconcentration) from a single exposure or daily exposure,and the exposure scenario. During the days of exposure,a fraction of the adduct increment is added to each arrayelement, and the total adduct level is then computeddaily. Summation of the array elements gives the accu-mulated adduct level:

y = � a(n - 1) � at(L, - t + 1)

n=1 t,.r t(.r



2.0 �__C

0

0)Di

� 1.2

I0.8� 0.4

Time (days)

25 50 75 100 125Time (days)

Fig. 5. Simulation of accumulation and removal of adducts on workplaceexposure for less than the lifetime of the erythrocytes (C,., = 120 days; a= 100 pmol/g hemoglobin/day). Exposures were simulated 5 days/week

for 20 days.

instability. The random loss of erythrocytes is speciesdependent and is normally low in humans (18). There-fore, the first-order loss process considered below essen-tially describes loss due to adduct instability. The effectsof a number of different rate constants for removal ofadducts are simulated in Fig. 6, A and B. The rate con-stants used were selected to provide a range from unsta-ble adducts (half-life of 2 days) to more stable adductswith half-lives (up to 50 days) shorter than t,.,. The re-moval is largely influenced by the rate constant (Fig. 6A),

with the first-order process predominating for the shorterhalf-lives. For single exposures, the simulated adductlevels were described by

y = [a(t - ,,)_ ��.�]e_&* (L)

where t is constant (t = 1). With continuous exposure(Fig. 68), the extent of adduct accumulation is decreasedwith increasing rate constant (decreasing half-life). Ad-ducts with long half-lives will accumulate until the life-time of the erythrocyte, whereas adducts with short half-lives reach a plateau at times much shorter than t,.1.

Few data sets are available in the literature that canbe used directly for the validation of this model. Theadduct loss measured in many of the single-dose studiesin experimental animals resembled that expected for lossdue to senescence (3, 7, 19). In many of the studieswhere multiple doses were administered over the life-time of the erythrocyte, the animals increased consider-ably in weight, and blood samples were withdrawn re-peatedly (10, 13). The data of Bergmark et a!. (14) on theformation of ethyl esters in mouse globin following asingle administration of ethylnitrosourea were analyzedusing the published rate constant for adduct hydrolysis(0.0023 h) and produced a simulation curve that wasin good agreement with the data (not shown). Anotherexample that was investigated was the formation andremoval of adducts produced by the administration ofMOCA3 to rats (11, 20). Information on adduct removalfollowing a single dose and adduct accumulation and

Time (days)

Fig. 4. Simulation of accumulation and removal of adducts on (A)intermittent exposure once every 7 days or (B) using a workplace expo-sure scenario with exposure for 5 days/week. Simulations were con-ducted with t�, 120 days and a = 100 pmol/g hemoglobin/day.

900 pmol/g on day 120, with the slope of the periodicdecline equal to 17.a/t,.,. The slope, calculated from thenumber of exposures during the lifetime of the erythro-cyte (1 7), was verified from the model output.

Exposure to chemicals in the workplace is generallyintermittent, and many of the carcinogenesis bioassaystudies involve exposure of experimental animals tochemicals for 5 days/week, 6 h/day, to mimic exposuresin the workplace. The effect of such an exposure scenariois depicted in Fig. 4B. The adduct level increases withincreasing time and decreases on the days of no expo-sure. The slope of the decrease on weekends increasesas the number of exposures increases and is at a maxi-mum when the adduct levels have reached a plateau. Inthe example shown, the maximum adduct level calcu-lated was achieved at day 124 and was equivalent to0.728 times the maximum calculated for continuous ex-posure (Fig. 2B).

Simulation of a workplace exposure for less than thelifetime of the erythrocyte is shown in Fig. 5, where theexposure was discontinued after 4 weeks. The accumu-lation of adducts is similar to that shown in Fig. 4B, andafter exposure the decline in adduct concentration islinear until the lifetime of the erythrocyte and quadraticthereafter (similar to that shown in Fig. 3).

The simulations shown in Figs. 2-5 do not incorpo-rate effects of random loss of erythrocytes or adduct

2 16 Modeling Hemoglobin Adduds

3 The abbreviation used is: MOCA, 4,4’-methylenebis(2-chloroaniline).

C

00)0)

0EC

0C0

0

U

VV



C

00)0)

E0

E

UC00U

VV

Time (days) Time (days)

Fig. 7. Comparison of simulation with formation and removal of adductsin hemoglobin following a single dose (281 �moI/kg p.o.) of MOCA toSprague-Dawley rats (12). Experimentally derived data (x), Simulationswere performed with a = 8.2 pmol/mg globin/day (estimated from theinitial data point) and k = 0 ( ) or 0.02772 day (- . -).: B

Time (days)

Fig. 6. Simulation of adduct instability following (A) a single exposure or(Bl continuous exposure (t,, = 120 days; a = 100 pmol/g hemoglobin/day). In addition to removal of erythrocytes from circulation, a first-orderloss term was included in the model to simulate chemical instability.Values used for the first-order rate constant k were 0 (1), 0.01386 (2),

0.03465 (3), 0.0693 (4), or 0.1386 (5) day_i, corresponding to half-livesof 50, 20, 1 0, and 5 days, respectively. Exposures were discontinued after120 days.

Cancer Epidemiology, Biomarkers & Prevention 217

removal following multiple doses has been published.The adduct levels following a single administration ofMOCA (281 �moI/kg p.o.), shown in Fig. 7 (11), couldnot be adequately described in model simulations withsenescence as the only mechanism of adduct removal(i.e., k = 0). From model simulations incorporating first-order loss in addition to senescence, a first-order rateconstant for adduct loss was estimated to be 0.02772day. This rate constant corresponds to a half-life of 25days and differs from the overall half-life of adduct re-moval of 14.3 days, which incorporates loss by senes-cence (1 1). Comparison of the rate constant of 0.0049day’, which has been reported for random loss of eryth-rocytes in rats (21), with that derived from simulation(0.02772 day) suggests that MOCA hemoglobin ad-ducts are unstable or that MOCA treatment causes anincrease in the random removal of erythrocytes. Theaccumulation of adducts following repeated exposureswas simulated using the rate constant of 0.02772 day�(Fig. 8) and compared with the experimental values ob-tamed by Cheever et a!. (20) following daily administra-tion of 28 zmol/kg MOCA for 28 days. An experimentallydetermined value for the binding of MOCA to globinfollowing a single dose was not available but was esti-

mated, by fitting the data, to be approximately 23 fmol/mg. The simulation did not match all of the observeddata points, suggesting that additional variables may needto be considered in the removal of MOCA adducts fromglobmn. Simulation without adduct removal by a first-order process (k = 0; a = 16) produced a curve in betteragreement with the experimental data.

Discussion

The potential role of exposure scenario and erythrocytebehavior in the accumulation and removal of hemoglobinadducts has been investigated using a computer modelin which the RBC population is described as a series ofindividual subpopulations that can be tracked by age.Simulation of adduct formation and removal suggeststhat the removal of adducts depends on the exposurescenario. For repetitive but discontinuous exposures, theaccumulated adduct level approximates to the fraction(exposure time/total time) of that found for continuousexposures. Adduct instability has a considerable effecton the levels of adduct that will accumulate. The extentof accumulation can be readily calculated and comparedwith a stable adduct. This information should make pos-sible an assessment of the dose from repeated exposureswhere the adduct measured is unstable.

The parameters required forthe simulation of adductbehavior in this model are the lifetime ofthe erythrocytesin the species of interest and the daily adduct incrementfor the adduct in question. Several simplifying assump-tions have been made in this discussion of adduct be-havior: the formation of adducts is constant during theperiods of exposure; the lifetime of the erythrocyte isconstant and does not vary with treatment; hemoglobinin old and young erythrocytes is alkylated to a similarextent; dilution due to growth has not been included.Several of these assumptions will be discussed in moredetail below.

increases in body weight together with an increasein blood volume following exposure to an adduct-form-ing chemical could lead to dilution of the adduct. This isa particular concern during animal experiments wherebody weights change by a significant factor compared



40 60

Time (days)

218 Modeling Hemoglobin Adduds

4 V. E. Walker, I. P. MacNeela, J. A. Swenberg, M. J. Turner, Jr., and T. R.Fennell, manuscript in preparation.

Fig. 8. Comparison of simulation with formation and removal of adductsin hemoglobin following multiple doses (28 daily consecutive doses of

28.1 �zmol/kg p.o.) of MOCA to Sprague-Dawley rats (20). Experimentallyderived data (x), Simulations were performed with k = 0.02772 dayand a = 23 pmol/g globin/day ( ) or k = 0 and a = 16 pmol/gglobin/day (- .

with the erythrocyte lifetime. Correction for the increasein the body weight of animals has been incorporated intothe model (not shown).

Measurement of the lifetime of the erythrocyte isgenerally accomplished by one of two methods: randomlabeling, in which cells of all ages become labeled; orcohort labeling, in which cells ofa specific age are labeledduring their production. The disappearance of label fromcirculation is then monitored, and the lifetime of theerythrocyte can be calculated from kinetic analysis of thedisappearance of the label (1 5). These methods producea measurement ofthe average lifetime of the erythrocyte,and a range of lifetimes is observed under normal circum-stances. The distribution of lifetimes of erythrocytes hasnot been incorporated into this version of this model.While these determinations suggest that the majority ofRBCs display a finite lifetime, a component of the losscan be due to random processes that can vary in differentspecies. This random loss can be modeled as a first-orderprocess (1 5). Complications may arise due to the pres-ence of two or more populations of cells, changes in therate of random destruction, and nonuniformity of thepotential life span (15). While treatment with manychemicals can cause anemia, there is a poor understand-ing of less severe effects that may cause an increase inthe turnover of erythrocytes without an apparent anemia.This model readily makes possible the simulation of theeffect of erythrocyte lifetime, random loss of erythro-cytes, and adduct instability on the accumulation ofadducts under intermittent exposure conditions. Com-parison of simulations with experimentally derived datamay make possible the distinction of factors associatedwith adduct removal. The analysis of hemoglobin add uctformation and removal for ethylene oxide in experimen-tal animals is currently underway.4

The constant formation of adduct during periods ofexposure was used in these examples for simplicity. Thedaily extent of formation of adducts in humans will vary.

The modeling of different daily rates of formation ofadducts, together with background rates of formation, isthe subject of further investigations. The description ofhemoglobin adduct formation and removal reported hereis being incorporated into a physiologically based phar-macokinetic model for ethylene oxide, to model theformation of adducts on repeated exposure, with theextent of adduct formation calculated from the modelparameters and related to exposure.� The incorporationof kinetic information, DNA adducts, and other biomark-ers in parallel is a goal of these studies (22, 23).

The modeling of hemoglobin adduct accumulationunder various exposure scenarios shows that the expo-sure regimen dramatically affects both the accumulationand removal of adducts and emphasizes the need forinformation on the exposure scenario for use of hemo-globin adduct data in dosimetry. Adduct levels measuredfollowing a single exposure decline in a linear fashionwith time after exposure. With information on the timeelapsed between exposure and the collection of a bloodsample for adduct measurement, the peak level of add uctformation could be calculated, and from this the dosecould be estimated. Similarly, for multiple exposures fortime greater than t,.,, the extent of adduct formation canbe related to the integrated dose, assuming a constantexposure (24, 25). The pattern of exposures from indus-trial and environmental sources or from cigarette smokingcan in many instances be established. In the absence ofexposure information, the estimation of dose from ad-duct levels involves considerable uncertainty. Amongother factors, the adduct levels are critically dependenton the time since the last exposure. The determinationof adduct levels in samples obtained at different timeswould aid in establishing whether the adduct measuredhas achieved a plateau, is rising, or is falling. This wouldreduce the uncertainty associated with the calculation ofdose.

A model such as this makes possible the comparisonof data obtained from experimental animals under avariety of exposure scenarios and should permit thedistinction of some of the factors involved in the loss ofadducts from circulation. Similarly, this model could beused to provide an improved assessment of dose fromadduct determinations in humans following different ex-posure scenarios and when information on adduct sta-bility is known.

Acknowledgments

The authors would like to thank Drs. M. Gargas, R. Conolly, and M.Andersen for education in the use of Simusolv and useful discussions. Inaddition we would like to thank Dr. I. Kimbell for helpful discussions on

the mathematical aspects of the model.

References

1 . Perera, F. P., and Weinstein, I. B. Molecular epidemiology and carcin-ogen-DNA adduct detection: new approaches to studies in human cancercausation. I. Chronic Dis., 35: 581-600, 1982.

2. Perera, F. P. Molecular cancer epidemiology: a new tool in cancerprevention. J. NatI. dancer Inst., 78: 887-898, 1987.

5 K. Krishnan, M. L. Gargas, T. R. Fennell, and M. E. Andersen. A phys-iobogically based description of ethylene oxide dosimetry in the rat,

submitted for publication.



Cancer Epidemiology, Biomarkers & Prevention 219

3, Osterman-Golkar, S., Ehrenberg, L., Segerback, D., and H#{228}llstrom,I.Evaluation of genetic risks of alkylating agents. II. Haemoglobin as a dosemonitor. Mutat. Res., 34: 1-10, 1976.

4. Farmer, P. B., Neumann, H-G., and Henschler, D. Estimation ofexposure of man to substances reacting covalently with macromolecules.Arch. Toxicol., 60: 251-260, 1987.

5. Skipper, P. L., and Tannenbaum, S. R. Protein adducts in the moleculardosimetry ofchemical carcinogens. Carcinogenesis (Lond.), 1 1: 507-518,1990.

6. Ehrenberg, L., and Osterman-Golkar, S. Alkylation of macromoleculesfor detecting mutagenic agents. Teratog. Carcinog. Mutagen., 1: 105-127, 1980.

7. Segerback, D., Calleman, C. J., Ehrenberg, L., Lbfroth, C., and Oster-man-Golkar, S. Evaluation of the genetic risks of alkylating agents. IV.Quantitative determination of alkylated amino acids in haemoglobin as ameasure of the dose after treatment of mice with methyl methanesulfo-nate. Mutat. Res., 49: 71-82, 1978.

8. Bryant, M. S. Estimation of human exposure to the carcinogenicaromatic amine 4-aminobiphenyl via hemoglobin dosimetry. Ph.D. dis-sertation, Massachusetts Institute of Technology, 1987.

9. Maclure, M., Bryant, M. S., Skipper, P. L., and Tannenbaum, S. R.Decline ofthe hemoglobin adduct of 4-aminobiphenyl during withdrawalfrom smoking. Cancer Res., SO: 181 -1 84, 1990.

10. Carmella, S. G., and Hecht, S. S. Formation of hemoglobin adductsupon treatment of F344 rats with the tobacco-specific nitrosamines 4-(methylnitrosamino)-1 -(3-pyridyl)-1 -butanone and N’-nitrosonornicotine.Cancer Res., 47: 2626-2630, 1987.

1 1. Cheever, K. L., Richards, D. E., Weigel, W. W., Begley, K. B., DeBord,D. C., Swearengin, T. F., and Savage, R. E., Jr. 4,4’-Methylenebis(2-chloroaniline) (MOCA): comparison of macromolecular adduct formationafter oral or dermal administration in the rat. Fundam. AppI. Toxicol., 14:273-283, 1990.

12. Neumann, H-C. Analysis of hemoglobin as a dose monitor for alkyl-ating and arylating agents. Arch. Toxicol., 56: 1-6, 1984.

13. Green, L. C., Skipper, P. L., Turesky, R. J., Bryant, M. S., andTannenbaum, S. R. In vivo dosimetry of 4-aminobiphenyl in rats via acysteine adduct in hemoglobin. Cancer Res., 44: 4254-4259, 1984.

14. Bergmark, E., Belew, M., and Osterman-Golkar, S. Separation andenrichment of alkylated globin chains as a means of improving thesensitivity of hemoglobin adduct measurements. Acta Chem. Scand., 44:

630-635, 1990.

15. Berlin, N. I., and Berk, P. D. The biological life of the red cell. In: D.M. Surgenor (ed), The Red Blood Cell, Ed. 2, Vol. 2, pp. 957-1019. NewYork: Academic Press, 1975.

16. Derelanko, M. I. Determination of erythrocyte life span in F-344,Wistar, and Sprague-Dawley rats using a modification of the [3H]diiso-propylfluorophosphate ([3H]DFP) method. Fundam. AppI. Toxicol., 9:

271-276, 1987.

1 7. Van Putten, L. M. The life span of red cells in the rat and the mouseas determined by labeling with DFP32 in vivo. Blood, 13: 789-794, 1958.

18. Eadie, C. S., and Brown, I. W. The potential life span and ultimatesurvival of fresh red blood cells in normal healthy recipients as studiedby simultaneous Cr5 tagging and differential hemolysis. J. Clin. Invest.,34:629-636, 1955.

19. Corelick, N. I., Hutchins, D. A., Tannenbaum, S. R., and Wogan, C.N. Formation of DNA and hemoglobin adducts of fluoranthene after

single and multiple exposures. Carcinogenesis (Lond.), 10: 1579-1587,1989.

20. Cheever, K. L., DeBord, D. C., and Swearengin, T. F. 4,4’-Methyl-enebis(2-chloroaniline) (MOCA): the effect of multiple oral administra-tion, route, and phenobarbital induction on macromolecular adductformation in the rat. Fundam. AppI. Toxicol., 16: 71-80, 1991.

21. Belcher, E. H., and Harriss, E. B. Studies of red cell life span in therat. I. Physiol., 146: 217-234, 1959.

22. Wogan, C. N. Summary: methods. In: H. Bartsch, K. Hemminki, andI. K. O’Neill (eds.), Methods for Detecting DNA Damaging Agents inHumans: Applications in Cancer Epidemiology and Prevention, Vol. 85,pp. 9-12. Lyon: International Agency for Research on Cancer, ScientificPublication, 1988.

23. Wogan, C. N. Detection of DNA damage in studies on canceretiology and prevention. In: H. Bartsch, K. Hemminki, and I. K. O’Neill(eds.), Methods for Detecting DNA Damaging Agents in Humans: Appli-cations in Cancer Epidemiology and Prevention, Vol. 85, pp. 32-51. Lyon:International Agency for Research on Cancer, 1988.

24. Osterman-Colkar, S., Farmer, P. B., Segerback, D., Bailey, E., Calle-man, C. J., Svensson, K., and Ehrenberg, L. Dosimetry of ethylene oxidein the rat by quantitation of alkylated histidine in hemoglobin. Teratog.

Carcinog. Mutag., 3: 395-405, 1983.

25. T#{246}rnqvist, M., Osterman-Colkar, S., Kautiainen, A., Jensen, S.,Farmer, P. B., and Ehrenberg, 1. Tissue doses ofethylene oxide in cigarettesmokers determined from adduct levels in hemoglobin. Carcinogenesis(Lond.), 7: 1519-1521, 1986.



1992;1:213-219. Cancer Epidemiol Biomarkers Prev T R Fennell, S C Sumner and V E Walker A model for the formation and removal of hemoglobin adducts.

Updated version

http://cebp.aacrjournals.org/content/1/3/213

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cebp.aacrjournals.org/content/1/3/213To request permission to re-use all or part of this article, use this link



http://cebp.aacrjournals.org/cgi/alerts

mailto:[email protected]



a model for the formation and removal of hemoglobin adductssolutions to the analytical equations,...

Documents