Seroimmunological Characteristics of Korean Workers Exposed toToluene Diisocyanate1

Heon Kim, Yong-Dae Kim, and Jaewook Choi*

Department of Preventive Medicine, College of Medicine, Chungbuk National University, Cheongju, Korea; and *Department ofPreventive Medicine, College of Medicine, Korea University, Seoul, Korea

Received August 21, 1996

Since type I allergy caused by specific IgE anti-bodies may play principal roles and IgG antibody-mediated reactions have been thought to be in-volved in some parts of the pathogenesis, this studywas performed to investigate the role of IgE- or IgG-mediated hypersensitivity reactions in develop-ment of toluene diisocyanate (TDI) asthma inKorean workers. For 81 TDI spray painters, self-administrative questionnaires and direct inter-views on respiratory symptoms, chest auscultation,and measurements of forced vital capacity (FVC)and forced expiratory volume in 1 sec (FEV1.0) wereperformed. The TDI concentration in their workingenvironments was measured. Levels of serum IgEand IgG specific to TDI were estimated by radioal-lergosorbent test (RAST) and ELISA using p-tolylisocyanate–human serum albumin (TMI–HSA) asthe antigen. When sputum, cough, and dyspnea ag-gravated by work or wheezing existed, when FVC orFEV1.0 was less than 80% of the normal referencevalue, or when IgE RAST for TDI was positive, thepeak expiratory flow rate (PEFR) was recorded fourtimes per day for over 2 weeks. If decrease of PEFRwas over 20% of baseline PEFR and changing pat-tern of PEFR was closely related to workshift intime, then a diagnosis of TDI asthma was made.Changing patterns of PEFR of 8 (9.9%) workers cor-responded to the diagnostic criteria of TDI-relatedoccupational asthma. Levels of the specific IgEwere increased in 9 (11.1%) of the 81 subject work-ers and in 3 (37.5%) of the 8 PEFR-positive workers.Levels of the specific IgG were increased in 9(11.1%) workers, and in only 1 (12.5%) of the asth-matics sensitive to TDI. Neither elevated TDI-specific IgE levels nor PEFR test positivities were

associated with increased IgG levels. The mean titerof the PEFR-test-positive workers was slightlylower than that of the PEFR-negative workers andthat of the IgE RAST-positive workers lower thanthat of the test-negative workers, but there was nostatistical significance. These results suggest thatIgG is not deeply involved in the pathogenesis ofTDI-induced occupational asthma in Korean work-ers. © 1997 Academic Press

Key Words: toluene diisocyanate; occupational-asthma; IgE; IgG; PEFR.

INTRODUCTION

In Korea, over 30,000 tons of isocyanates are usedin a year, and more than 40,000 workers are underoccupational exposure to isocyanates (Chung et al.,1993). Toluene diisocyanate (TDI) is the most widelyused type of the isocyanates, and 2–13% of KoreanTDI workers suffer occupational asthma (Park et al.,1992; Kim et al., 1994).

Karol et al. (1978) reported that 75% of the indi-viduals with clinical hypersensitivity to TDI had apositive isocyanate-specific IgE radioallergosorbenttest (RAST). However, Baur et al. (1984) reportedthat 14% of symptomatic workers had specific IgE,and the proportions of workers with positive inhala-tion challenge who have specific IgE were less than20% in many other reports (Danks et al., 1981; Tse etal., 1985). The results of these studies suggest thatonly a minority of isocyanate-induced reactions canbe attributed to an IgE-mediated allergic reaction(Wass, 1989).

Some investigators reported that specific IgG an-tibody levels were increased in the sera of workerswho had occupational TDI asthma (Zeiss et al., 1980;Grammer et al., 1990) and workers who were ex-posed to isocyanates (Welinder et al., 1988; Selden et

1 The research protocol ‘‘Seroimmunological Characteristics ofKorean Workers Exposed to Toluene-Diisocyanate’’ by Heon Kim,Yong-Dae Kim, and Jaewook Choi has been approved by the In-stitutional Review Board of Chungbuk National University Hos-pital on October 25, 1994.

ENVIRONMENTAL RESEARCH 75, 1–6 (1997)ARTICLE NO. ER973763

1

0013-9351/97 $25.00Copyright © 1997 by Academic Press

All rights of reproduction in any form reserved.

al., 1989). In an animal study, TDI-specific IgG wasmore closely related to the incidence of asthmaticresponse than TDI-specific IgE (Sugawara et al.,1993). But there were some studies in which in-creased IgG levels could not be found in workersexposed to TDI (Paggiaro et al., 1983).

This study was performed to investigate the levelof TDI-specific IgE and IgG antibodies in sera ofKorean TDI workers, in relation to their clinical di-agnosis of occupational asthma in response to TDI.

MATERIALS AND METHODS

Subjects

Eighty-one workers (41 male and 40 female work-ers) who had been exposed to TDI in workshopsmanufacturing furniture or musical instruments orrepairing motor vehicles were studied. The mean oftheir age was 41.8 ± 8.1 years.

Design of the Studies

To estimate the exposure level, the TDI concen-tration in working environments was measured.Self-administrative questionnaires and direct inter-views regarding respiratory symptoms and chestauscultation were performed. Venous blood wassampled, and serum was separated for ELISA andRAST for TDI. Forced vital capacity (FVC) andforced expiratory volume in 1 sec (FEV1.0) were mea-sured.

Peak expiratory flow rate (PEFR) was recorded inthe following cases:

(1) If the subject complained of sputum, cough,and dyspnea aggravated by work.

(2) If wheezing was audible by auscultation.(3) If FVC or FEV1.0 was less than 80% of the

normal Korean reference value estimated from hisor her age, sex, and height.

(4) If the IgE RAST for TDI was positive.Baseline PEFR was measured when the subjects

were off duty. If the decrease in PEFR was over 20%of the baseline PEFR and the changing pattern ofPEFR was closely related to workshift in time, thena diagnosis of TDI asthma was made.

The results of the IgE RAST and the IgG ELISAwere compared between TDI asthmatic workers andthe others.

TDI Concentration Measurement

Area sampling was performed to measure thelevel of TDI in the air of working environments. The

standard method of National Institute of Occupa-tional Safety and Health was used for analysis ofTDI concentrations. In short, TDI was trapped bypassing air through a midzet impinger filled withsolution of 1-(2-methoxyphenyl)piperazine in tolu-ene at a rate of 1.0 liter/min. Acetic anhydride wasadded to acetylate the 1-(2-methoxyphenyl)piperaz-ine, and the sample was evaporated to dryness un-der a nitrogen stream on a hot plate. The residuewas redissolved in methanol, and 20-ml aliquotswere injected for high-pressure liquid chromatogra-phy. A Gilson Model 302 pump, a Gilson 802C mano-metric module, and a 5-mm-particle size, 10.0-cm ×8-mm octadecylsilylated silica column was used, andthe absorption peak was detected at 254 nm with anultraviolet detector (Gilson 112 UV/vis detector).The mobile phase was 7.3 mM sodium acetate in 50%acetonitrile adjusted to pH 6.0 with acetic acid, andthe flow rate was 1.0 ml/min.

Questionnaire Survey and Direct Interview

The workers were asked to fill in self-ad-ministered questionnaires on respiratory symptomsand their relations to workshifts which were modi-fied from that of British Rubber Manufacturers As-sociation. Symptoms marked in questionnaire wereconfirmed by direct interview.

Pulmonary Function Tests

FVC and FEV1.0 were measured three times witha Spiro Analyzer LAM-5 (Fukuda Company, Japan).

Peak Expiratory Flow Rate

PEFR was checked and recorded four times perday (before workshift, between 11 o’clock AM andnoon, between 4 and 5 o’clock PM, and before going tobed) for over 2 weeks. Diagnosis of TDI asthma wasmade if there was a decrease of PEFR over 20% ofbaseline and if the changing pattern was closely re-lated to workshift in time. If the changing pattern ofPEFR did not show exact correlation in time withworkshift, the subject was asked if there was anyinterruption of exposure by bad weather or a leave.

TMI–HSA Antigen Conjugation

A p-tolyl isocyanate–human serum albumin(HSA) was prepared after the method of Karol et al.(1978). p-Tolyl isocyanate (280 ml, TMI) was added

KIM, KIM, AND CHOI2

to 100 ml of 1% HSA in borate buffer (pH 8.0). Thesolution was mixed in ice for 30 min, centrifuged,and incubated for 1 h at 4°C. The supernatant wasdialyzed extensively with phosphate-buffered saline(PBS) and deionized water. Conjugation was identi-fied by the difference in the absorption pattern ofUV spectrophotometry (Fig. 1) and in the migrationin native polyacrylamide gel electrophoresis (Fig. 2).

Specific IgE and IgG Determination

Filter paper disks were activated with cyanogenbromide (Ceska et al., 1972) and were bound TMI–HSA conjugate and HSA (Wass, 1989). Cyanogenbromide-activated disks (200 mg) was incubated in20 mg/ml TMI–HSA or HSA in freshly prepared 0.1M NaHCO3 and rotated with a mixer at room tem-

FIG. 1. Ultraviolet spectrophotometric lambda scan of HSA (H) and TMI–HSA (T).

EXPOSURE TO TOLUENE DIISOCYANATE 3

perature for 7 hr. After incubation, filter paperswere washed three times in 0.1 M NaHCO3, and in0.1 M ethanolamine in 0.1 M NaHCO3, and incu-bated overnight at 10°C in the last washing solution.Disks were washed three times with phosphate-buffered saline (PBS)–0.05% sodium azide (pH 7.4)and incubated at 4°C in the same solution until test.

Specific IgE antibody was determined with Phad-ebas RAST radioimmunoassay (Pharmacia, Swe-den) following manufacturer’s direction. An aliquotof 50 ml of worker’s serum was added to each tubecontaining TMI–HSA- or HSA-coated disks, and ifthe count per minute of the reference disk wasgreater than two times the count of the control disk,the result of RAST was classified as positive.

In order to measure levels of anti-TMI–HSA andanti-HSA IgG antibodies, ELISA was performed.Wells of polystyrene microtiter plates (Nunc-Immuno Plate MaxiSorp, Nunc, Denmark) werecoated with 100 ml TMI–HSA or HSA solution in

PBS. After washing two times with PBS–0.05%Tween 20, the wells were blocked with 200 ml of 3%BSA–PBS–0.02% sodium azide solution for 2 hr at37°C. The wells were washed two times and dilutedsera were added and incubated for 2 hr. Sera of 90medical students who had not been exposed to TDIwere used as negative control sera. The wells werewashed extensively with PBS four times. After add-ing 100 ml of peroxidase-conjugated antihuman IgG,the plates were incubated for 1 hr at 37°C andwashed thoroughly. After adding substrate solutionmade of o-phenylenediamine and 30% H2O2 in phos-phate–citrate buffer (pH 5.0), the plates were leftuntouched at room temperature for 30 min. Onehundred microliters of 0.2 M H2SO4 was added tostop the reaction, and OD was measured at 492 nm.Ratios of OD for TMI–HSA to OD for HSA were cal-culated, and the specific IgG test was judged as posi-tive when the OD ratio was greater than the value ofthe mean ± 3 SD of the OD ratio of negative controlsera of medical workers.

RESULTS

TDI concentrations in the air of the workshopsranged from 0.0005 to 0.0100 ppm, and the meanwas 0.0035 ± 0.0023 ppm. Of 41 area samples, 4(9.8%) were higher than 0.005 ppm, the thresholdlimit value–time-weighted average (TLV–TWA) ofthe American Conference of Governmental Indus-trial Hygienists. The mean (±standard deviation) ofFVC was 2.22 (±0.70) liters and that of FEV1.0 was2.19 (±0.69) liters.

PEFR was checked for 15 workers among 81 sub-jects. There was only one worker whose FVC orFEV1.0 was less than 80% of normal reference value,and the level of specific IgE was elevated in 9 work-ers. PEFR of 5 workers were recorded because oftheir respiratory symptoms or wheezing sound. Bythe PEFR test, 8 workers (9.9%; 3 male and 5 femaleworkers) were diagnosed as having occupationalasthma in response to TDI.

The positive rate of IgE RAST in the PEFR-test-positive group was 37.5% (3⁄8), which was higherthan that of PEFR-negative group significantly(Table 1). The mean (±standard deviation) of the IgGODs ratio for sera of the workers was 1.267 (±0.319),and it was significantly higher than that of negativecontrol sera, 1.045 (±0.202) (P < 0.01, Table 2). Thespecific IgG test was classified as positive if ODsratio was greater than the value of the mean + 3 SDof the ODs ratio of negative control sera, 1.651. Ac-cording to this criteria, the specific IgG test waspositive for the sera of 9 workers (11.1%). The pro-

FIG. 2. Native polyacrylamide gel electrophoresis of HSA (H)and TMI-HSA (T). TMI–HSA migrated farther than HSA despitethe molecular weight of TMI–HSA possibly being greater thanthat of HSA.

KIM, KIM, AND CHOI4

portion of increased specific IgG antibody in PEFRpositive group was 12.5%, which was not statisti-cally different from that of the PEFR-negative group(Table 3).

The level of specific IgG and IgE did not show anystatistically significant correlation. The proportionof increased IgG levels in the IgE RAST-positivegroup was exactly same as that of the IgE RAST-negative group (Table 4).

DISCUSSION

By the results of the questionnaire and direct in-terview, the pulmonary function test, and the PEFRtest, nearly 10% of the workers exposed to TDI hadoccupational asthma caused by TDI. This was notsignificantly different from the results of the previ-ous Korean studies (Park et al., 1992; Kim et al., 1993).This suggests that about 4000 Korean workers whoare exposed to TDI suffer occupational asthma.

In diagnosing occupational asthma, bronchialprovocation tests have been used to confirm the cau-sal relationship between exposure to TDI and theasthmatic symptoms. Since the bronchial provoca-tion test is very time consuming, and can sometimesinduce a very severe asthma attack (Fabbri et al.,1988), prolonged recording of PEFR at work andaway from work was proposed as a valuable diag-nostic test in occupational asthma (Burge et al.,1979). Cote et al. (1990) assessed the sensitivity andspecificity of prolonged recording of PEFR with thespecific challenge test as the ‘‘gold standard.’’ The

sensitivity and specificity of PEFR recording were86 and 89%, respectively, and in combination withclinical history, sensitivity was 100%, but specificitywas relatively low. We cannot rule out the possibil-ity of high false-positive rate due to the use of PEFRand clinical history to confirm the TDI asthma inthis study.

Increased anti-TMI–HSA IgE levels were associ-ated with occupational asthma. However, the posi-tive rate of the IgE RAST is only 37.5% of the PEFR-positive workers, and 8.2% of PEFR negative work-ers showed increased specific IgE in response toTMI–HSA. These results suggest that IgE-mediatedhypersensitivity reaction plays only a minor role inthe occurrence of TDI asthma. However, a smallnumber of workers with occupational asthma in thisstudy made this suggestion less valid.

The level of TMI–HSA-specific IgG was not asso-ciated with the level of specific IgE determined withRAST and was not increased in PEFR-positive work-ers. It is suggested by these findings that IgG-mediated allergy is not closely related with develop-ment of TDI asthma in Korean workers.

CONCLUSIONS

In summary, the immunologic reaction in whichIgE or IgG is involved is not the only developingmechanism of TDI asthma of Korean workers, andother nonimmunologic processes (Davies et al., 1977;

TABLE 2Mean Absorbance for TMI–HSA and HSA in IgG ELISA of

Diluted Sera

Group N

Absorbance unitRatio of

absorbancesaTMI–HSA HSA

Control group 90 0.164 ± 0.038 0.164 ± 0.055 1.045 ± 0.202Worker group 81 0.204 ± 0.080 0.160 ± 0.029 1.267 ± 0.319

a P < 0.01.

TABLE 3Number of Increases in Specific IgG Levels as Detected

with the PEFR Test

SpecificIgG level

Occupational asthmaby PEFR test

Yes (%) No (%) Total

Increased 1 (12.5) 8 (11.0) 9 (11.1)Not increased 7 (87.5) 65 (89.0) 72 (88.9)Total 8 (100) 73 (100) 81 (100)

Note. P > 0.05 (by Fisher’s exact test).

TABLE 4Number of Increases in Specific IgG Levels as Detected

with the IgE RAST

SpecificIgG level

Specific IgE level

Increased(%)

Not increased(%) Total

Increased 1 (11.1) 8 (11.1) 9 (11.1)Not increased 8 (88.9) 64 (88.9) 72 (88.9)Total 9 (100) 72 (100) 81 (100)

Note. P > 0.05 (by Fisher’s exact test).

TABLE 1Number of Increases in Specific IgE Levels as Detected

with the PEFR Test

SpecificIgE level

Occupational asthmaby PEFR test

Yes (%) No (%) Total

Increased 3 (37.5) 6 (8.2) 9 (11.1)Not increased 5 (62.5) 67 (91.8) 72 (88.9)Total 8 (100) 73 (100) 81 (100)

Note. P value < 0.05 (by Fisher’s exact test).

EXPOSURE TO TOLUENE DIISOCYANATE 5

Butcher et al., 1979) should be considered simulta-neously.

REFERENCES

Bauer, X., Dewair, M., Fruhmann, G. (1984). Detection of immu-nologically sensitized isocyanate workers by RAST and intra-cutaneous skin test. J. Allergy Clin. Immunol. 73, 610–618.

Burge, S. P., O’Brien, I. M., and Harries, M. G. (1979). Peak flowrates in the diagnosis of asthma caused by isocyanate. Thorax34, 317–323.

Butcher, B. T., Karr, R. M., and O’Neil, C. E. (1979). Inhalationchallenges and pharmacologic studies of toluene diisocyanate(TDI)-sensitive workers. J. Allergy Clin. Immunol. 64, 146–152.

Ceska, M., and Lundkvist, U. (1972). A new and simple radioim-munoassay method for determination of IgE. Immunochemistry9, 1201–1230.

Chung, I.-S., Choi, J.-W., and Cha, C.-W. (1993). A study on theoccupational asthma in the workers exposed to toluene diiso-cyanate. Korea Univ. Med. J. 30, 95–107.

Cote, J., Kennedy, S., and Chan-Yeung, M. (1990). Sensitivity andspecificity of PC20 and peak expiratory flow rate in cedarasthma. J. Allergy Clin. Immunol. 85, 592–598.

Danks, J. M., Cromwell, O., Buckingham, J. A., Newman-Taylor,A. J., and Davies, R. J. (1981). Toluene-diisocyanate inducedasthma: Evaluation of antibodies in the serum of affected work-ers against a tolyl mono-isocyanate protein conjugate. Clin. Al-lergy 11, 161–168.

Davies, R. J., Butcher, B. T., O’Neil, C. E., and Salvaggio, J. E.(1977). The in vitro effect of toluene diisocyanate on lymphocytecyclic adenosine monophosphate production by isoproterenol,prostaglandin and histamine: A possible mode of action. J. Al-lergy Clin. Immunol. 60, 223–229.

Fabbri, L. M., Danieli, D., Crescioli, S., Bevilacqua, P., Meli, S.,Saetta, M., and Mapp, C. E. (1988). Fatal asthma in a subjectsentized to TDI. Am. Rev. Respir. Dis. 137, 1494–1498.

Grammer, L. C., Harris, K. E., Malo, J.-L., Cartier, A., and Pat-terson, R. (1990). The use of an immunoassay index for anti-bodies against isocyanate human protein conjugates and appli-

cation to human isocyanate disease. J. Allergy Clin. Immunol.86, 94–98.

Karol, M. H., Ioset, H. H., and Alarie, Y. C. (1978). Tolyl-specificIgE antibodies in workers with hypersensitivity to toluene di-isocyanate. Am. Ind. Hyg. Assoc. J. 39, 454–453.

Kim, Y.-Y., Cho, S.-H., Yoon, H.-J., Min, K.-Y., Paik, D.-M., andChung, K.-C. (1994). Isocyanate-induced occupational asthmain Korea. Korean J. Med. 47, 439–453.

Paggiaro, P. L., Filieri, M., Loi, A. M., Roselli, M. G., Cantalupi,R., Parlanti, A., Toma, G., and Baschieri, L. (1983). Absence ofIgG antibodies to TDI–HSA in a radioimmunological study.Clin. Allergy 13, 75–79.

Park, H. S., Park, J. N., Kim, J. W., and Kim, S. K. (1992). Clini-cal and immunological evaluation of isocyanate-exposed work-ers. J. Korean Med. Sci. 7, 122–127.

Selden, A. I., Belin, L., and Wass, U. (1989). Isocyanates exposureand hypersensitivity pneumonitis—Report of a probable caseand prevalence of specific immunoglobin G antibodies amongexposed individuals. Scan. J. Work Environ. Health 15, 234–237.

Sugawara, Y., Okamoto, Y., Sawahata, T., and Tanaka, K. (1993).An asthma model developed in the guinea pig by intranasalapplication of 2,4-toluene diisocyanate. Int. Arch. Allergy Im-munol. 101, 95–101.

Tse, K. S., Johnson, A., Chan, H., and Cheng-Yeung, M. (1985). Astudy of serum antibody activity in workers with occupationalexposure to diphenylmethane diisocyanate. Allergy. 40, 314–320.

Wass, U. L. (1989). Immunologic specificity of isocyanate-inducedIgE antibodies in serum from 10 sensitized workers. J. AllergyClin. Immunol. 83, 126–135.

Welinder, H., Nielsen, J., Densryd, I., and Skerfving, S. (1988).IgG antibodies against polyisocyanates in car painters. Clin.Allergy 18, 85–93.

Zeiss, C. R., Kanellakes, T. M., Bellone, J. D., Levitz, D., Pruzan-sky, J. J., and Patterson, R. (1980). Immunoglobin E-mediatedasthma and hypersensitivity pneumonitis with precipitatinganti-hapten antibodies due to diphenylmethane diisocyanate(MDI) mixture. J. Allergy Clin. Immun. 65, 346–352.

KIM, KIM, AND CHOI6