Oxygen free radical scavenger enzyme polymorphisms in systemic sclerosis

Free Radical Biology & Medicine, Vol. 36, No. 11, pp. 1403 – 1407, 2004 Copyright D 2004 Elsevier Inc. Printed in the USA. All rights reserved 0891-5849/$-see front matter

doi:10.1016/j.freeradbiomed.2004.02.079

Original Contribution OXYGEN FREE RADICAL SCAVENGER ENZYME POLYMORPHISMS IN SYSTEMIC SCLEROSIS MOHAMMED TIKLY,* SARA E. MARSHALL, y NEIL A. HALDAR,z MARY GULUMIAN, § PAUL WORDSWORTH, b and KEN I. WELSH // *Department of Medicine, Chris Hani Baragwanath Hospital, and the University of the Witwatersrand, Johannesburg, South Africa; y Department of Immunology, Wright Fleming Institute, Imperial College School of Medicine, London, UK; z Nuffield Department of Surgery, University of Oxford, Oxford, UK; § Department of Biochemistry, National Centre for Occupational Health, Johannesburg; b Nuffield Orthopaedic Centre, Oxford, UK; and // Clinical Genomics, National Heart and Lung Institute, College School of Medicine, London, UK (Received 7 October 2003; Revised 29 January 2004; Accepted 27 February 2004) Available online 9 April 2004

Abstract—We performed a case – control study of polymorphisms of glutathione S-transferase (GST) isoenzymes and manganese superoxide dismutase (MnSOD) in black South Africans with systemic sclerosis (SSc). The frequency of the GSTM1*B phenotype was significantly decreased in the overall SSc group compared with controls (OR = 0.19, pcorr < .05), implying a possible protective effect against development of the disease. There was also a trend toward increased MnSODAla allele and phenotype frequencies in the diffuse cutaneous SSc subset compared with controls (OR = 2.11 and 3.15, respectively, pcorr < .1). Our findings provide new data on the distribution of GST and MnSOD polymorphisms in healthy Africans and further evidence that genetic factors may have a contributory role to play in predisposing to oxidative stress in SSc. D 2004 Elsevier Inc. All rights reserved. Keywords—Scleroderma, Free radicals, Glutathione S-transferase, Manganese superoxide dismutase, Africa

tene, a-tocopherol, ascorbic acid, and selenium levels have all been found to be lower in patients than in healthy controls [1,4]. Antioxidant enzymes are an important mechanism by which cells limit the damage caused by oxygen free radicals. The glutathione S-transferases (GSTs) are a group of dimeric enzymes that catalyze the conjugation of glutathione to a wide range of electrophilic compounds, thereby allowing them to be excreted in the bile and urine [5]. The GST supergene family now comprises at least 16 genes subdivided into 8 separate classes based on their amino acid sequences [6]. Many of these enzymes are polymorphic, and the existence of isoenzymes within each gene family may underlie differential responses to OS. Three polymorphic GSTs have been widely studied: GSTP1, GSTM1, and GSTT1. GSTM1 activity varies widely between individuals, and 45% of normal Caucasians fail to express a transferase at the GSTM1 locus due to a gene deletion [5]. In addition, two forms of GSTM1 may be expressed, GSTM1*A and GSTM1*B, which differ by a single amino acid at position 172 [7]. GST

INTRODUCTION

Systemic sclerosis (scleroderma, SSc) is a disease characterized by fibrosis of the skin and visceral organs, microvascular disturbance, and disease-specific antinuclear antibodies (ANAs). Oxidative stress (OS) is a major factor implicated in the causation and perpetuation of tissue damage in SSc [1] and is probably the result of many different processes. Most importantly, episodic vasoconstriction of the microvasculature in Raynaud’s phenomenon results in ischemia –reperfusion injury and increased production of reactive oxygen species (ROS) [2]. Environmental factors may also trigger OS through the stimulation of NADPH production, which increases free radical production [3]. In addition, there is evidence of reduced antioxidant capacity in SSc, and serum caro-

Address correspondence to: Dr. Mohammed Tikly, Division of Rheumatology, Department of Medicine, Chris Hani Baragwanath Hospital, P.O.Bertsham 2013, South Africa; Fax: +27 11 938-8738; E-mail: [email protected]. 1403

1404

M. TIKLY et al.

theta enzymes also act as a detoxification sink. Like the mu class, the GSTT1 null genotype is due to a gene deletion and is associated with decreased ability to detoxify metabolites of several carcinogens [8]. Two singlenucleotide polymorphisms in the GSTP1 gene have been identified, at positions 105 and 114, and the resulting isoenzymes have been associated with variation in substrate selectivity and activity both in vivo and in vitro [9]. GST polymorphisms have been widely studied in a variety of diseases, especially in determining cancer susceptibility [10] and response to chemotherapy [11]. In the autoimmune rheumatic diseases, the homozygous GSTM1 null genotype has been associated with higher Larsen bone erosion scores in smokers with rheumatoid arthritis (RA) [12], with Sjo¨gren’s syndrome [13], and with anti-Ro antibodies in systemic lupus erythematosus, where anti-Ro antibodies are thought to be produced in response to the ROS-associated toxic effects of UV light [14]. Two studies have addressed the role of GST polymorphisms in SSc [15, 16]. The GSTP1Val105 allele was found to be weakly associated with disease susceptibility in Caucasians [16], and patients who carried a combination of nonfunctional or poorly functional enzymes (i.e., GSTM1null, GSTT1 null, and GSTP1Val105/Val105) were at increased risk of carotid atherosclerosis as defined by duplex scanning. Manganese superoxide dismutase (MnSOD) is a homotetrameric enzyme that protects mitochondria against OS by scavenging superoxide anions produced from the electron transport system. Two polymorphic variants of this scavenger enzyme have been identified, and are the consequence of a point mutation in the genetic sequence resulting in a valine to alanine change at amino acid 16 [17,18]. The MnSODVal/Val genotype has previously been associated with increased severity of RA [19]. There are significant and interesting interethnic differences in the clinical phenotype of SSc. In black South Africans diffuse cutaneous SSc (dcSSc) is the predominant subset and anticentromere antibodies (ACAs) are very rare [20], whereas in northern Europeans in limited cutaneous SSc (lcSSc) subset and ACAs are common [21]. Little is known about the distribution of genetic polymorphisms in GST and MnSOD enzymes in black South Africans in health and disease. A case – control study was therefore undertaken to investigate whether variation in these genes is associated with the development and clinical phenotype of SSc in black South Africans. PATIENTS AND METHODS

The research protocol was approved by the Ethics Committee of the Faculty of Health Sciences, University

of the Witwatersrand. Black South African patients from the connective tissue disease clinics at Chris Hani Baragwanath and Johannesburg hospitals who met the American College of Rheumatology criteria for SSc [22] were recruited for the study. Subset classification of lcSSc and dcSSc was based on the extent of skin involvement [23], and evidence of specific clinical features, as defined previously [20], was based on the findings at the time of examination and review of case records. The 61 geographically and ethnically matched controls (47 women, 14 men) were either healthy hospital staff or patients attending the casualty department for minor trauma. Sera from all patients were screened for ANAs by indirect immunofluorescence. Specific autoantibodies, with the exception of ACAs, were identified by a combination of the immunoprecipitation assay using [ 35S]methionine-radiolabeled cell extracts [24] and Ouchterlony double immunodiffusion technique. Identification of ACAs was based on the indirect immunofluorescence staining pattern. Genotyping was performed for the following polymorphisms using polymerase chain reaction with sequence specific primers (PCR-SSP): GSTM1*A , GSTM1*B, GSTT1*1, G STP1*A, GSTP1*B, GSTP1*C, GSTP1*D, MnSODAla, and MnSODVal. Primers and amplification conditions have previously been described [25,26]. The genetic parameters of both the GSTM1 deletion and the GSTT1 deletion are unknown. PCR primers for GSTM1 were therefore designed specifically to amplify the GSTM1*A or GSTM1*B allele. The homozygous null genotype was assumed from the absence of an amplification product for both GSTM1*A and GSTM1*B. Primers for GSTT1 were based on those described by Pemble et al. [8], which determine the presence or absence of the GSTT1 gene. Using these primers, heterozygosity and homozygosity for the presence of the gene cannot be distinguished. To genotype GSTP1 variants, an assay was developed that uses both forward and reverse allelespecific primers, enabling identification of cis/trans orientation [27]. Four alleles are identifiable in this way, namely, GSTP1 105 Ile – 114Ala (GSTP1*A), GSTP1 105Val – 114Ala (GSTP1*B), GSTP1 105Val – 114Val (GSTP1*C), and GSTP1 105Ile – 114 Val (GSTP1*D). Allele, genotype, and phenotype frequencies were measured for all polymorphisms where appropriate. The m2 test and Fisher’s exact test were used to test for associations using Statisca 6 (Statcom Inc) and EpiInfo 6 software. To correct for multiple comparisons, a Bonferroni correction (BC) of 5 was applied, which corresponds to the number of polymorphic sites investigated. A corrected p value ( pcorr) of < .05 was defined as significant.

O2 free radical scavenger enzyme polymorphisms

Table 2. Comparison of MnSODAla Polymorphism in dcSSc Subset and Control Group

RESULTS

The mean (FSD) age and disease duration of the 51 patients (45 women, 6 men) with SSc were 40.7 (11.4) and 6.2 (7.7) years, respectively. Thirty-three patients had dcSSc and 18 had lcSSc. The cumulative frequencies of the salient clinical features were Raynaud’s phenomenon in 48 of 50 (96.0%), digital gangrene in 13 of 48 (27.1%), pulmonary fibrosis in 19 of 48 (39.6%) and pulmonary hypertension in 10 of 50 (20.0%) patients. ANA was positive in 49 of 51 (96.1%) patients. Identifiable specific autoantibodies were detected in 28 of the 39 patients tested, of whom 6 had more than one specific autoantibody. These included anti-U1RNP and anti-Ro each in 7 of 39 (17.9%), anti-topoisomerase I in 6 of 39 (15.4%), anti-fibrillarin (U3RNP) in 5 of 39 (12.8%), ACA and anti-La each in 4 of 39 (10.3%), anti-RNA polymerase I/III in 2 of 39 (5.1%), and anti-RNA polymerase II and anti-Pm-Scl antibodies each in 1 of 39 (2.6%) of the patients. Table 1. Allele, Genotype, and Phenotype Frequencies of Glutathione S-Transferase and Manganese Superoxide Polymorphisms All SSc (n = 51)

GSTM1 genotype frequency GSTM1*A/*A or *A/null GSTM1*B/*B or *B/null GSTM1*A/B GSTM1*null GSTM1 phenotype frequency GSTM1*A present GSTM1*B present GSTT1 genotype frequency GSTT*1/*1 or *1/null GSTT*null GSTP1 allele frequency GSTP*1A GSTP*1B GSTP1 genotype frequency GSTP*1A/A GSTP*1A/B GSTP*1B/B GSTP1 phenotype frequency GSTP*1A present GSTP*1A absent GSTP*1B present GSTP*1B absent MnSOD allele frequency MnSODAla MnSODVal MnSOD genotype frequency MnSODAla/Ala MnSODAla/Val MnSODVal/Val MnSOD phenotype frequency MnSODAla present MnSODVal present

1405

Controls (n = 61)

No.

Freq

No.

Freq

30 3 0 18

0.58 0.06 0 0.34

32 6 9 14

0.52 0.1 0.15 0.23

30 3a

0.58 0.06

41 15

0.67 0.25

40 11

0.78 0.22

45 16

0.74 0.26

51 51

0.5 0.5

60 60

0.5 0.5

12 27 12

0.24 0.52 0.24

17 26 17

0.28 0.44 0.28

39 12 39 12

0.76 0.24 0.76 0.24

43 17 43 17

0.72 0.28 0.72 0.28

50 54

0.48 0.52

45 79

0.31 0.69

14 22 16

0.27 0.42 0.31

11 23 28

0.18 0.37 0.45

36 38

0.69 0.73

33 51

0.54 0.83

a Overall SSc group versus controls: p < .01, pcorr < .05, OR = 0.19 (95% CI: 0.04 – 0.77).

p

pcorr

OR (95% CI)

45/122

< 0.03

< 0.1

2.11 (1.1 – 4.05)

33/61

< 0.03

< 0.1

3.15 (1.09 – 9.42)

MnSODAla

dcSSc

Controls

Allele frequency Phenotype frequency

36/66 26/33

Allele, genotype, and phenotype frequencies of GST and MnSOD polymorphisms are summarized in Table 1. GSTP1*C and GSTP1*D were not detected in either the patient or control groups. The major finding was that the GSTM1*B phenotype was significantly increased in the overall SSc group compared with controls after Bonferroni correction ( pcorr < .05). Subgroup analysis revealed no significant associations with specific clinical features after correction for multiple comparisons, although the following trends were observed: increased MnSODAla allele and phenotype frequencies in the dcSSc subset compared with controls (Table 2): decreased MnSODAla phenotype in anti-Ro antibody-positive patients [MnSOD Ala polymorphism present in 2 of 7 (28.6%) anti-Ro positive patients vs. 26 of 32 (81.3%) anti-Ronegative patients, OR = 0.09 (95%CI: 0.01 – 0.75), p < .02, pcorr < .07]; decreased GSTP1*A phenotype frequency in anti-La-positive patients [GSTP1*A polymorphism present in 1 of 4 (25%) anti-La-positive patients vs. 29 of 35 (82.9%) anti-La-negative patients, OR = 0.07 (95% CI = 0.0– 0.99), p < 0.04, pcorr < .2]. DISCUSSION

The majority of previous studies on the frequency of polymorphisms of free radical scavenging enzymes have been performed in Caucasian populations. The frequency of the homozygous GSTM1 null genotype of 24% in the black South African control cohort presented here is virtually the same as that reported in healthy Tanzanians, Zimbabweans, the Venda, and African-Americans [28,29]. Thus the frequency of the homozygous GSTM1 null genotype is approximately half that found in Caucasians and the Japanese [8]. By contrast, the frequency of the homozygous GSTT1 null genotype of 26.6% in the control group of the present study, similar to the 20– 26% reported in southern and East Africans [29], is about double the 10 – 15% reported in Caucasians [8,30]. No previous studies of GSTP1 polymorphisms in African individuals have been described, and only the GSTP1*A and GSTP1*B alleles were observed in either controls or patients. The significant negative (protective) association of the GSTM1*B phenotype with SSc found in this study is noteworthy. A recent study of delayed graft function after

1406

M. TIKLY et al.

renal transplantation demonstrated that patients who received a kidney from a donor who expressed GSTM1*B either alone or in combination with GSTM1*A experienced significantly lower rates of delayed graft function [26]. Interestingly, the GSTM1*A allele did not show any such association. The reason for this distinction is not clear: GSTM1*A and GSTM1*B differ at a single amino acid, but previous in vitro studies have not shown any functional differences for these alleles which encode monomers that form active homoand heterodimeric enzymes [31]. However, in vitro enzyme kinetics may not accurately reflect in vivo activity, especially when a range of substrates may be available. Alternatively, the association described may be the result of linkage disequilibrium with another, unidentified polymorphism either in the GSTM1 gene itself, or in a related gene. A number of additional trends were noted within clinical subgroups, although these did not reach statistical significance after application of a correction factor for multiple comparisons. In particular, we failed to confirm the findings of a larger study of SSc in Americans of various ethnic backgrounds, where an association between the null GSTT1 genotype and pulmonary involvement (defined as pulmonary fibrosis and pulmonary hypertension), a reduced frequency of anti-RNP antibodies in GSTM1 null allele homozygotes, and a trend toward a higher frequency of ACA in patients with homozygous null alleles for both GSTM1 and GSTT1 were observed [15]. Several factors other than true ethnic differences might explain these differences, including differences in environmental triggers, assays used to test for specific autoantibodies, and sample sizes. In summary, our findings confirm that the frequencies of polymorphisms in oxygen free radical scavenger enzymes in Africans may differ from those in other populations. The study also provides further evidence that genetic factors may play an important contributory role in predisposing to OS and, ultimately, to tissue injury, in SSc. Future work including large subgroups of patients of different ethnicity is needed to further elucidate the role of these polymorphisms in SSc. Acknowledgments — We thank Dr. Alison Rand and Dr. Neil McHugh, University of Bath, United Kingdom, for performing the specific autoantibody tests. REFERENCES [1] Herrick, A. L.; Rieley, F.; Schofield, D.; Hollis, S.; Braganza, J. M.; Jayson, M. I. V. Micronutrient antioxidant status in patients with primary Raynaud’s phenomenon and systemic sclerosis. J. Rheumatol. 21:1477 – 1483; 1994. [2] Belch, J. J. F. Raynaud’s phenomenon: its relevance to scleroderma. Ann. Rheum. Dis. 50:839 – 845; 1991. [3] Murrell, D. F. A radical proposal for the pathogenesis of scleroderma. J. Am. Acad. Dermatol. 28:78 – 85; 1993.

˚ kesson, A.; A ˚ kesson, B. Dietary intake and [4] Lundberg, A.-C.; A nutritional status in patients with systemic sclerosis. Ann. Rheum. Dis. 51:1143 – 1148; 1992. [5] Seidergard, J.; Vorachek, W. R.; Pero, R. W.; Pearson, W. R. Hereditary differences in the expression of the human glutathione S-transferase activity on trans-stilbene oxide are due to a gene deletion. Proc. Natl. Acad. Sci. USA 85:7293 – 7297; 1988. [6] Hayes, J. D.; Strange, R. C. Glutathione S-transferase polymorphisms and their biological consequences. Pharmacology 61: 154 – 166; 2000. [7] DeJong, J. L.; Chang, C. M.; Whang-Peng, J.; Knutsen, T.; Tu, C. P. The human liver glutathione S-transferase gene superfamily: expression and chromosome mapping of an Hb subunit cDNA. Nucleic Acids Res. 16:8541 – 8554; 1988. [8] Pemble, S.; Schroeder, D. R.; Spencer, S. R.; Meyer, D. J.; Hallier, E.; Bolt, H. M.; Ketterer, B.; Taylor, J. B. Human glutathione S-transferase theta (GSTT1): cDNA clonong and the characterisation of a genetic polymorphism. Biochem J. 300: 271 – 276; 1994. [9] Watson, M. A.; Stewart, R. K.; Smith, G. B. J.; Massey, T. E.; Bell, D. A. Human glutathione S-transferase P1 polymorphisms: relationship to lung tissue enzyme activity and population frequency distribution. Carcinogenesis 19:275 – 280; 1998. [10] Ryberg, D.; Skaug, V.; Hewer, A.; Phillips, D. H.; Harries, L. W.; Wolf, C. R.; Øgreid, D.; Ulvik, A.; Vu, P.; Haugen, A. Genotype of glutathione transferase M1 and P1 and their significance for lung DNA adduct levels and cancer risk. Carcinogenesis 18: 1285 – 1289; 1997. [11] Watters, J. W.; McLeod, H. L. Recent advances in the pharmacogenetics of cancer chemotherapy. Curr. Opin. Mol. Ther. 4: 565 – 571; 2002. [12] Mattey, D. L.; Hutchinson, D.; Dawes, P. T.; Nixon, N. B.; Clarke, S.; Fisher, J.; Brownfield, A.; Alldersea, J.; Fryer, A. A.; Strange, R. C. Smoking and disease severity in rheumatoid arthritis: association with polymorphism at the glutathione Stransferase M1 locus. Arthritis Rheum. 46:640 – 646; 2002. [13] Morinobu, A.; Kanagawa, S.; Koshiba, M.; Sugai, S.; Kumagia, S. Association of the glutathione S-transferase M1 homozygous null genotype with susceptibility to Sjogren’s syndrome in Japanese individuals. Arthritis Rheum. 42:2612 – 2615; 1999. [14] Ollier, W.; Davies, E.; Snowden, N.; Alldersea, J.; Fryer, A.; Jones, P.; Strange, R. Association of homozygosity for glutathione-S-transferase GSTM1 null alleles with the Ro+/La- autoantibody profile in patients with systemic lupus erythematosus. Arthritis Rheum. 39:1763 – 1764; 1996. [15] Tew, M. B.; Reveille, J. D.; Arnett, F. C.; Friedman, A. W.; McNearney, T.; Fischbach, M.; Ahn, C.; Tan, F. K. Glutathione S-transferase genotypes in systemic sclerosis and their association with clinical manifestations in early disease. Genes Immun. 2:236 – 238; 2001. [16] Palmer, C. N. A.; Young, V.; Ho, M.; Doney, A.; Belch, J. J. F. Association of common variation in glutathione S-transferase genes with premature development of cardiovascular disease in patients with systemic sclerosis. Arthritis Rheum. 48:854 – 855; 2003. [17] Rosenblum, J. S.; Gilula, N. B.; Lerner, R. A. On signal sequence polymorphisms and diseases of distribution. Proc. Natl. Acad. Sci. USA 93:4471 – 4473; 1996. [18] Shimoda-Matsubayashi, S.; Matsumine, H.; Kobayashi, T.; Nakagawa-Hattori, Y.; Shimuzu, Y.; Mizuno, Y. Structural dimorphism in the mitochondrial targeting sequence in the human manganese superoxide dismutase gene. Biochem. Biophys. Res. Commun. 226:561 – 565; 1996. [19] Mattey, D. L.; Hassell, A. B.; Dawes, P. T.; Jones, P. W.; Yengi, L.; Alldersea, J. E.; Strange, R. C.; Fryer, A. A. Influence of polymorphism in the manganese superoxide dismutase locus on disease outcome in rheumatoid arthritis: evidence for the interaction with glutathione S-tranferase genes. Arthritis Rheum. 43:859 – 864; 2000. [20] Tager, R. E.; Tikly, M. Clinical and laboratory manifestations of systemic sclerosis (scleroderma) in black South Africans. Rheumatology 38:397 – 400; 1999.

O2 free radical scavenger enzyme polymorphisms [21] Jacobsen, S.; Halberg, P.; Ullman, S.; van Venrooij, W. J.; HoierMadsen, M.; Wiik, A. Clinical and serum antinuclear antibodies in 230 Danish patients with systemic sclerosis. Br. J. Rheumatol. 37:39 – 45; 1998. [22] Subcommittee for Scleroderma Criteria of the American Rheumatism Association Diagnostic and Therapeutic Criteria Committee. Preliminary criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum. 23:581 – 590; 1980. [23] LeRoy, E. C.; Black, C.; Fleischmajer, R.; Jablonska, S.; Krieg, T.; Medsger, T. A. Jr.; Rowell, N.; Wollheim, F. Scleroderma (systemic sclerosis): classification, subsets, and pathogenesis. J. Rheumatol. 15:202 – 205; 1988. [24] Craft, J.; Mimori, T.; Olsen, T. L.; Hardin, J. A. The U2 small ribonucleoprotein particle as an autoantigen. J. Clin. Invest. 81:1716 – 1724; 1988. [25] Marshall, S. E.; Bordea, C.; Haldar, H. A.; Mulligan, C. G.; Wojnarowska, F.; Morris, P. J.; Welsh, K. I. Glutathione S-transferase polymorphisms and skin cancer after renal transplantation. Kidney Int. 58:2186 – 2193; 2000. [26] St Peter, S. D.; Imber, C. J.; Jones, D. C.; Fuggle, S. V.; Watson, C. J.; Friend, P. J.; Marshall, S. E. Genetic determinants of delayed

[27]

[28]

[29]

[30] [31]

1407

graft function after kidney transplantation. Transplantation 74: 809 – 813; 2002. Lo, Y.-M.; Patel, P.; Newton, C. R.; Markham, A. F.; Fleming, K. A.; Wainscoat, J. S. Direct haplotype determination by double ARMS: specificity, sensitivity, and genetic applications. Nucleic Acid Res. 19:3561 – 3567; 1991. Mukanganyama, S.; Masimirembwa, C. M.; Naik, Y. S.; Hasler, J. A. Phenotyping of the glutathione S-transferase M1 polymorphism in Zimbabweans and the effect of chloroquine on blood glutathione S-transferase M1 and A. Clin. Chim. Acta 265: 145 – 155; 1997. Dandara, C.; Sayi, J.; Masimirembwa, C. M.; Magimba, A.; Kaaya, S.; De Sommers, K.; Snyman, J. R.; Hasler, J. A. Genetic polymorphism of cytochrome P450 1A1 (Cyp1A1) and glutathione transferases (M1, T1 and P1) among Africans. Clin. Chem. Lab. Med. 40:952 – 957; 2002. Rebbeck, T. R. Molecular epidemiology of the human glutathione S-transferase genotypes GSTM1 and GSTT1 in cancer susceptibility. Cancer Epidemiol. Biomarkers Prev. 6:733 – 743; 1997. Ketterer, B.; Fraser, G.; Meyer, D. J. Nuclear glutathione transferases which detoxify irradiated DNA. Adv. Exp. Med. Biol. 264:301 – 310; 1990.