1353

produce gliadin

antibodies. In small children, the gliadin

satisfactorily with mucosal atrophy.9 By contrast, adults with active coeliac disease have been found to be positive for gluten and gliadin antibodies in only 70%-80% of cases. 10,11 If gluten ingestion is extended over years in coeliac patients, the initially increased antibody litres have been reported to decline again and even become negative, despite gluten-containing food and mucosal abnonnality.2O Some healthy relatives might have had villous atrophy long before the present family study took antibody tests correlate

place. Healthy reticulin-antibody-positive first-degree relatives of coeliac disease patients, irrespective of the appearance of the small-bowel mucosa, are genetically similar to known coeliac disease patients. Reticulin or endomysium antibody positivity is an indicator of either silent or latent coeliac disease. The Coeliac Disease Study Project, Tampere, is supported by the Academy of Finland (M.M. and grant number 1061129), by the University of Tampere (K.H.), and by the Emil Aaltonen Foundation, Tampere, Finland.

REFERENCES 1. Logan RFA, Tucker G, Rifkind EA, Heading CG, Ferguson A. Changes in clinical features of coeliac disease in adults in Edinburgh and the Lothians 1960-79. Br Med J 1983; 286: 95-97. 2. Mäki M, Kallonen K, Lähdeaho M-L, Visakorpi JK. Changing pattern of childhood coeliac disease in Finland. Acta Paediatr Scand 1988; 77: 408-12. 3. Collin P, Hällström O, Mäki M, Viander M, Keyriläinen O. Atypical coeliac disease found with serological screening. Scand J Gastroenterol 1990; 25: 245-50. 4. Holmes GKT, Prior P, Lane MR, Pope D, Allan RN. Malignancy in coeliac disease: effect of a gluten-free diet. Gut 1989; 30: 333-38. 5. Freeman HJ, Chiu BK. Multifocal small bowel lymphoma and latent coeliac sprue. Gastroenterology 1986; 90: 1992-97.

6. Mäki M, Hällstrom O, Vesikari T, Visakorpi JK. Evaluation of a serum IgA-class reticulin antibody test for the detection of childhood celiac disease. J Pediatr 1984; 105: 901-05. 7. Hällström O. Comparison of IgA-class reticulin and endomysium antibodies in coeliac disease and dermatitis herpetiformis. Gut 1989; 30: 1225-32. 8. Chorzelski TP, Beumer EH, Sulej J, et al. IgA anti-endomysium antibody: a new immunological marker of dermatitis herpetiformis and coeliac disease. Br J Dermatol 1984; 111: 395-402. 9. Savilahti E, Viander M, Perkkio M, Vainio E, Kalimo K, Reuna T. IgA antigliadin antibodies: a marker of mucosal damage in childhood coeliac disease. Lancet 1983; i: 320-22. 10. Volta U, Molinaro N, Fratangelo D, Bianchi FB. IgA subclass antibodies to gliadin in serum and intestinal juice of patients with coeliac disease. Clin Exp Immunol 1990; 80: 192-95. 11. Scott H, Fausa O, Ek J, Valnes K, Blystad L, Brandtzaeg P. Measurements of serum IgA and IgG activities to dietary antigens: a prospective study of the diagnostic usefulness in adult coeliac disease. Scand J Gastroenterol 1990; 25: 287-92. 12. Grodzinsky E, Hed J, Liedén G, Sjögren F, Ström M. Presence of IgA and IgG antigliadin antibodies in healthy adults as measured by microELISA. Int Arch Allergy Appl Immunol 1990; 92: 119-23. 13. Beumer EH, Kumar V, Chorzelski TP. Screening for celiac disease. N Engl J Med 1989; 320: 1087. 14. MacDonald WC, Dobbins WO, Rubin CE. Studies of the familial nature of celiac sprue using biopsy of the small intestine. N Engl J Med 1965; 272: 448-56. 15. Stokes PL, Ferguson R, Holmes GKT, Cooke WT. Familial aspects of coeliac disease. Q J Med 1976; 45: 567-82. 16. Amos D, Badir H, Boyle W, McQueen M, Tiilikainen A. A simple microcytotoxicity test. Transplantation 1969; 7: 220-30. 17. Dupont B, ed. Immunobiology of HLA: immunogenetics and histocompatibility, vol II. New York: Springer Verlag, 1989. 18. Mäki M, Holm K, Collin P, Savilahti E. Increase of gamma/delta T cell receptor bearing lymphocytes in normal small bowel mucosa of a patient with latent coeliac disease. Gut (in press). 19. Mäki M, Hällström O, Marttinen A. Reaction of human non-collagenous polypeptides with coeliac disease autoantibodies. Lancet 1991; 338: 724-25. 20. Bürgin-Wolff A, Lentze MJ. Relation of antigliadin antibodies to glutensensitive enteropathy. In: Chorzelski TP, Beutner EH, Kumar V, Zalewski TK, eds. Serologic diagnosis of celiac disease. Boca Raton: CRC Press, 1990: 58-75.

Rapid diagnosis of hantavirus disease with an IgG-avidity assay KLAUS HEDMAN ANTTI VAHERI MARKUS BRUMMER-KORVENKONTIO

Nephropathia epidemica (NE), due to Puumala virus, is endemic in eastern Europe and Scandinavia. Serodiagnosis of NE relies on conventional indirect immunofluorescence to detect IgG against Puumala virus. However, in conventional serology, most patients with acute NE have raised but stable non-diagnostic antibody titres. For better serodiagnosis of NE, we have developed a test that measures the avidity (functional affinity) of IgG antibodies against Puumala virus. This new assay was highly specific and sensitive; a diagnosis of NE could be confirmed or excluded rapidly from an early single serum sample. With this test we have now verified the diagnosis of NE in more than 1300 Finnish patients during 22 months in 1989-91. Our findings point to an incidence of confirmed hantavirus disease much higher than previously shown.

Introduction

Nephropathia epidemica (NE) is one of the haemorrhagic fevers with renal syndrome.1 Clinically, there is fever of sudden onset accompanied by headache, conjunctival injection, sometimes acute myopia, myalgia, nausea, vomiting, back and abdominal pain, and signs of acute interstitial nephropathy.2,3 Haemodialysis or peritoneal dialysis may be required. Complete recovery is the usual outcome, and the mortality rate is low (below 1 %). NE is due to Puumala virus4,5 of the Hantavirus genus in the Bunyaviridae family. Other well-characterised related human pathogens are the Hantaan viruswhich is the agent of Korean haemorrhagic fever, and the Seoul viruswhich causes a milder illness in Asia. All hantavirus diseases are rodent-borne zoonoses;8 transmission between human ADDRESS: Department of Virology, University of Helsinki, Haartmaninkatu 3, SF-00290 Helsinki, Finland (K. Hedman, MD, Prof A. Vaheri, MD, M Brummer-Korvenkontio, MD). Correspondence to Dr K. Hedman.

1354

beings is not known. The predominant carrier of NE is the bank vole, Clethrionomys glareolus.4 NE is endemic in eastern Europe and in Scandinavia, including Finland, where it is an important cause of nephropathy and where IgG-prevalence in some areas is as high as 20%.9 However, almost as high seroprevalence rates are detected sporadically in western Europe; symptomatic illness is more common in forested areas.2The naturally acquired hantavirus infections in northern Europe are due to serotype III (Puumala virus), but the proportion of the other

serotypes

seems

to

increase

southwards.1O-13

According seroprevalence data, acute-stage hantavirus infection, including symptomless, mild, or atypical cases, to

should be more common than has been found so far.1,14 We describe a new IgG-avidity (functional affmity) assay for improved diagnosis of hantavirus disease.

Fig 1-Avidity of Puumala-virus IgG

in relation to time of

onset of NE.

Sequential sera of individual patients are shown by lines Solid circles= precise avidity values; open circles= upper limits of avidity (all low-titre IgG eluted). Note that antibody avidity is inversely related to IgG-titre ratio

Patients and methods For standardisation of the avidity assay we studied 198 serum samples from 149 subjects with current, recent, or previous NE, in whom symptoms with dates of onset were known. Diagnosis was based on a significant (greater than or equal to 4-fold) rise in immunofluorescence (IF) antibody titre of Puumala-virus IgG. To obtain viral antigen for the IF IgG-avidity test, cultures of Vero E6 cells were infected for 14 days with Puumala virus (Sotkamo strain) as described. Harvested cells were used to prepare antigen-coated glass slides by standard methods. 4-fold serial dilutions of samples in phosphate-buffered saline (PBS) were then placed (15 µ1/dot) onto two coated slides, one with the dilutions 1/40-1/40 960 and the other with dilutions 1/2-5-1/2560. After incubation in a moist chamber for 30 min at 37°C, the slides were washed in PBS for 5 min. The slide with the higher dilutions was washed twice more (5 min each) in PBS, and the slide with the lower dilutions was washed in PBS containing 8 mol/1 urea.1s Both slides were rinsed in distilled water, air dried, treated with anti-human

IgG-fluorescein isothiocyanate conjugate (Kallestad; 1/20 in PBS) for 30 min at 37°C, washed in PBS without urea and then in distilled water, dried, and observed under a fluorescence microscope. The samples were coded and scored according to intensity of fluorescence. Titre of Puumala-virus IgG was the last dilution of serum showing unequivocal fluorescence. Avidity of IgG was calculated as the ratio of IgG titres without urea/with urea. As a technical control, fixed antigen-coated slides, exposed to 8 mol/I urea, were examined by phase-contrast microscopy. The cells remained attached to glass and had an acceptable morphology. As antigen in indirect IF, such urea-pretreated cells produced an IgG signal similar in pattern and brightness to the cells not pretreated with urea. Similarly, in the avidity assay there was no difference between urea-pretreated cells and non-pretreated cells. These controls showed that the acetone-fixed cultures of infected cells tolerated the protein denaturant without substantial loss of cells or irreversible damage to viral antigenicity.

Results

patients who were followed up during acute phase and convalescence, with diagnostic (≥4-fbld) rises in IgG avidity. All samples studied for Puumala-virus antibodies since September, 1989 (usually on the basis of a clinical suspicion of NE), have been tested according to the protocol shown in fig 2. With this procedure, the detection of specific IgG leads to the measurement of its avidity. On the basis of the above data, IgG-titre ratios could be assigned to times of infection. Thus, high (32) titre ratios (low avidity) are diagnostic of acute NE. By contrast, low ( 4) ratios (high avidity) rule out NE during the previous month. Titre ratios 8 and 16 are borderline; whereas 8 is regarded as equivocal with respect to the time of infection, 16 suggests that the onset

of the illness could be recent.

During the first months of use of the avidity assay in routine serodiagnosis, we encouraged clinicians to continue sending paired serum samples, so that we could compare avidity results with conventional IgG-IF results. 140 patients from whom paired samples were taken and who were (or became) seropositive had low antibody avidity, 38 (27%) of those patients had diagnostic rises in IgG titre and 102 (73%) had stable (or only doubling) IgG titres. 41 other subjects with paired samples had high antibody avidity in the first (IgG-containing) serum of the pair. None of these 41 patients had a diagnostic rise in IgG titre, confirming the high predictive value of the negative results (high avidity) against acute NE. In January, 1991 (16 months after the NE-avidity test was adopted as a routine viral serological procedure at University of Helsinki), the number of Finnish patients with diagnostic low-avidity results reached 1000. As shown in fig 3, the samples with low avidity were usually collected

subjects who had had symptoms more than 2 previously (long-term immunity), serum antibodies bound to Puumala-virus antigen firmly (fig 1): on elution with 8 mol/I urea, 91 (99%) had a 4-fold or less reduction in IgG titre and 1 had an 8-fold reduction. The acute-phase antibodies had a weaker avidity: there was a 32-fold or more titre reduction in over 90% of samples taken within the first month of onset of NE, during which none had high IgG avidity. Fig 1 shows that low antibody avidity seldom persisted beyond 2 months, whereas intermediate avidity (16 or 8) was most prevalent during late convalescence, 2-4 In the 92

years

months after onset of illness. The time-course of maturation of antibody avidity was assessed in more detail by sequential samples of individual patients. The lines in fig 1 show

Fig 2-Protocol for rapid serodiagnosis of NE. AVI =avidity; IF= immunofluorescence, + = positive; - = negative

1355

serodiagnosis of hantavirus infections. The sensitivity of these tests during acute infection seemed to be very high. However, in those studies the numbers of patients followed up through late convalescence were not large enough to rule out the possibility of IgM persistence, which interferes with the diagnosis of other infections. 15,111 For serodiagnosis of such illnesses, IgG-avidity tests (ELISA) based on a similar principle as our assay are proving to be very usefuI.15,18-20 It is noteworthy that we and others1,2-4,21 have had less than ideal results with various indirect immunoassays for hantavirus IgM. The annual morbidity of confirmed hantavirus disease that we have found (15/100000 among the Finnish population of 5-0 million) is to our knowledge the largest known nationwide incidence, and exceeds both the rates of admission to hospital based on clinical diagnosis5.8 and the annual numbers of confirmed serodiagnoses22 even in the Far East. Based solely on results from "routine" serologyie, samples sent by clinicians for confirmatory testing-our calculated incidence of NE in northern Europe must still be an

Fig 3-Follow-up of avidity results. Included are all patients with reported dates of onset of symptoms, and with low ( 32) or borderline (16, 8) avidity results m the present or previous samples Results are for serodiagnosis during 18 months at our centre

during the first week of onset of symptoms. Generally, according to avidity results, acute NE could be diagnosed (,>32) or excluded ( 4) in more than 90% of seropositive subjects. During follow-up, the high-avidity ( 4) antibodies did not appear until about 2 months after onset of NE (fig 3). By the end of June, 1991 (at 22 months of avidity-based serodiagnosis), the number of patients with confirmed hantavirus disease was 1324.

Discussion The

diagnostic sensitivity of

our

new

single-serum

almost four-times higher than that with sample conventional IgG serology, which uses paired samples. In conventional serology, most patients with acute NE have raised but stable, non-diagnostic antibody titres.3,9,16 Despite its subjective element (IF), the avidity assay distinguished, with a very high specificity, patients with acute or recent hantavirus disease from those with previous infection. Moreover, the test is rapid; results are available within 1-2 days. The clinical importance of rapid serodiagnosis was emphasised by the fact that the diagnostic specimens usually were collected and sent for testing during the first week of symptoms. NE has a rapid IgG response. According to Settergren et aI,3 the seronegative period lasts at most until the 7th day of symptoms. In our experience, 85% of patients are IgG seropositive by the 3rd day and 100% by the 6th day. Therefore, the absence of specific IgG after the 6th day of illness rules out NE. Most NE patients in Finland and in Sweden3 seek medical attention on the 5th day of symptoms, according to our experience, with a 95 % probability of being IgG-seropositive on admission. Niklasson and Kjellsson16 and Zoller et al 17 have introduced p-capture IgM ELISAs for the rapid test was

underestimate.

Specific serodiagnosis of hantavirus infection is necessary because false diagnoses made on clinical grounds during the acute phase of the disease are common.1,22 The avidity assay is suitable for use because it is a straightforward test and uses simple reagents and equipment. For these reasons we recommend the measurement of IgG avidity as a sensitive and specific test for the rapid serodiagnosis of patients with nephropathy or other symptoms suggestive of hantavirus disease. We thank all the

patients and health personnel for the donated follow-up Hedman, Ms R. Leveelahti, Ms S. Luomanen, Ms I. Luoto, and Ms T. Manni for skilful assistance. This study was supported by the Sigrid Juselius Foundation. sera; and Ms L.

REFERENCES HW, van der Groen G. Hemorrhagic fever with renal syndrome. Prog Med Virol 1989; 36: 62-102. 2. Van Ypersele de Strihou C, Mery JP. Hantavirus-related acute interstitial nephritis in Western Europe: expansion of a world-wide zoonosis. QJ Med 1989; 270: 941-50. 3. Settergren B, Juto P, Trollfors B, et al. Clinical characteristics of nephropathia epidemica in Sweden: prospective study of 74 cases. Rev Infect Dis 1989; 11: 921-27. 4. Brummer-Korvenkontio M, Vaheri A, Hovi T, et al. Nephropathia epidemica: detection of antigen in bank voles and serologic diagnosis of human infection. J Infect Dis 1980; 141: 131-34. 5. Schmaljohn CS, Hasty SE, Dalrymple JM, et al. Antigenic and genetic properties of viruses linked to hemorrhagic fever with renal syndrome. 1. Lee

Science 1985; 227: 1041-44.

HW, Lee PW, Johnson KM. Isolation of the etiologic agent of Korean hemorrhagic fever. J Infect Dis 1978; 137: 298-308. 7. Lee P-W, Gibbs CJ, Gajdusek DC, Yanagihara R. Serotypic classification of Hantaviruses by indirect immunofluorescent antibody and plaque reduction neutralization tests. J Clin Microbiol 1985; 22: 6. Lee

940-44.

Yanagihara R. Hantavirus infection in the United States: epizootiology and epidemiology. Rev Infect Dis 1990; 12: 449-57. 9. Brummer-Korvenkontio M, Henttonen H, Vaheri A. Hemorrhagic fever with renal syndrome in Finland: ecology and virology of nephropathia epidemica. Scand J Infect Dis 1982; 36 (suppl): 88-91. 10. Osterhaus ADME, Groen J, UytdeHaag FGCM, et al. Hantavirus nephropathy in the Netherlands. Lancet 1989; ii: 338-39. 11. Pilaski J, Ellerich C, Kreutzer T, et al. Haemorrhagic fever with renal syndrome in Germany. Lancet 1991; 337: 111. 12. Avsic-Zupanc T, Cizman B, Gligic A, van der Groen G. Evidence for Hantavirus disease in Slovenia, Yugoslavia. Acta Virol 1989; 33: 8.

327-37. 13. Antoniadis A, LeDuc JW, Acritidis N, et al. Hemorrhagic fever with renal syndrome in Greece: clinical and laboratory characteristics. Rev Infect Dis 1989; 4: S891-96.

1356

14. Editorial. Hantavirus disease. Lancet 1990; 336: 407-08. 15. Hedman K, Lappalainen M, Seppälä I, Mäkelä O. Recent primary toxoplasma infection indicated by a low avidity of specific IgG. J Infect Dis 1989; 159: 736-40. 16. Niklasson B, Kjellsson T. Detection of nephropathia epidemica (Puumala virus)-specific immunoglobulin M by enzyme-linked immunosorbent assay. J Clin Microbiol 1988; 26: 1519-23. 17. Zöller L, Yang S, Zeyer M. Rapid diagnosis of haemorrhagic fever with renal syndrome due to hantavirus. Lancet 1991; 338: 183. 18. Thomas HIJ, Morgan-Capner P. The use of antibody avidity measurements for the diagnosis of rubella. Med Virol 1991; 1: 41-50.

p53 mutation

in

19. Pullen GR, Fitzgerald MG, determination by ELISA using

Hosking CS. Antibody avidity thiocyanate elution. J Immunol Meth

1986; 86: 83-87. Kangro HO, Manzoor S, Harper DR. Antibody avidity following varicella-zoster virus infections. J Med Virol 1991; 36: 100-05. 21 Settergren B, Juto P, Wadell G. Detection of specific serum immunoglobulin M in nephropathia epidemica (Scandinavian a biotin-avidin-amplified epidemic by nephropathy) immunofluorescence method. J Clin Microbiol 1987; 25: 1134-36. 22. Lee HW. Hemorrhagic fever with renal syndrome in Korea. Rev Infect 20.

Dis 1989; 11: S864-76.

hepatocellular carcinoma after

aflatoxin exposure MEHMET OZTURK AND COLLABORATORS*

Mutations of the p53 gene are found in hepatocellular carcinoma (HCC), the most common form of primary liver cancer. Specific mutations might reflect exposure to specific carcinogens and we have screened HCC samples from patients in 14 different countries to determine the frequency of a hotspot mutation at codon 249 of the tumour suppressor p53 gene. We detected mutations in 17% of tumours (12/72) from four countries in south Africa and the southeast coast of Asia. There was no codon 249 mutation in 95 specimens of HCC from other geographical locations including North America, Europe, Middle East, and Japan. Worldwide, the presence of the codon 249 mutation in HCCs correlated with high risk of exposure to aflatoxins and the hepatitis B virus (HBV). Further studies were completed in two groups of HBV-infected patients at different risks of exposure to aflatoxins. 53% of patients (8/15) from Mozambique at high risk of aflatoxin exposure had a tumour with a codon 249 mutation, in contrast with 8% of patients from Transkei (1/12) who were at low risk. HCC is an endemic disease in Mozambique and accounts for up to two thirds of all tumours in men. A codon 249 mutation of the p53 gene identifies an endemic form of HCC strongly associated with dietary aflatoxin intake.

Introduction Mutations of the p53 gene are frequently found in hepatocellular carcinoma (HCC).1,2 Tumour-specific mutations of the p53 gene often lead to synthesis of a faulty protein that has lost its normal growth regulatory functions.3 The mutation of the p53 gene in HCC suggests that loss of normal gene function may be a key step during malignant transformation of hepatocytes. In HCC, known mutations of the p53 gene include selective guanine to thymine (G to T) transversions, which tend to cluster at codon 249.2,3 This mutational specificity could reflect exposure to specific carcinogens. Our present study was designed to investigate the frequency of codon 249 mutations in patients with HCCs from different countries and to test whether there is a causal relation between this tumour-specific mutation and two suspected causal factors, hepatitis B virus (HBV) and aflatoxins.

Patients and methods 167 HCC specimens (122 fresh-frozen and 45 paraffmembedded tissues) were collected from patients who lived in 14 different countries (USA, Germany, Spain, Italy, Turkey, Israel, Saudi Arabia, India, China, South Korea, Vietnam, Japan, South Africa, and Mozambique) and analysed (table I). Genomic DNA was extracted directly from selected parts of the tissue blocks as previously described for fresh-frozen1 and paraffinembedded tissues.4 Eighteen tumours were analysed with RNA as starting material. RNA was extracted, by means of a commercial kit (’RNASOL’, Cinna-Biotecx Lab, Friendswood, Texas, USA), and reverse transcribed. Exon 7 of the p53 gene with the codon 249 sequence was amplified by the polymerase chain reaction (PCR) from tumour DNA or cDNAs, as described. Separate PCR products were studied by restriction-enzyme analysis with two different enzymes. Digestion of exon 7 of the wild-type p53 gene with HaelII results in two fragments of 75 bp and 35 bp lengths. In the presence of a mutation, the Hae III recognition site at codon 249 is lost and only one 110 bp fragment is detected after Hae III digestion.A G to T mutation at the third base of codon 249 creates a new PleI recognition site in exon 7, and digestion with this enzyme yields three fragments of 80 bp, 18 bp, and 12 bp. In the absence of a codon 249 mutation, Plel digestion yields only two fragments of 98 bp and 12 bp. Restriction-enzyme analysis with Hae III was used for the initial screening of tumours. All of the identified mutations were confirmed by additional restriction-enzyme analysis with Hae III and Plel (fig 1). The validity of restriction-enzyme analysis was confirmed by direct sequencing of another PCR product from six different tumours, as previously described.l

ADDRESSES:

*Brigitte Bressac, MS, Alain Puisieux, MS (Massachusetts General Hospital Cancer Center, USA); Michael Kew, MD, (University of the Witwatersrand, South Martin MD Volkmann, Africa); (Deutsches Krebsforschungszentrum, Germany); Sema Bozcall, MS, Jessika Bella Mura, MS, Suzanne de la Monte, MD, Rolf Carlson, MS, Hubert Blum, MD, Jack Wands, MD, Hiroshi Takahashi, MD, Fritz von Weizsacker, MD, Elthan Galun, MD (Massachusetts General Hospital Cancer Centre, USA); Siddhartha Kar, PhD, Brian I. Carr, MD (University of Pittsburgh School of Medicine, USA); Claus H. Schroder, MD (Deutsches Krebsforschungszentrum, Germany); Eren Erken, MD, Seyhan Varinli, MD (Cukurova University Medical School, Turkey); Vinod K Rustgi, MD (Fairfax Hospital, USA); Jaime Prat, MD (Hospital de la Santa Creu I San Pau, Spain); Gotaro Toda, MD (Jikei University School of Medicine, Japan); Herbert K Koch, MD (Institut für Pathologie, Germany); Xiao Huan Liang, MD, Zhao-you Tang, MD (Shanghai Medical University, China); Daniel Shouval, MD (Hadassah University Hospital, Israel); Hyo-Suk Lee, MD (Seoul National University School of Medicine, South Korea); Girish N Vyas, MD, lldiko Sarosi, MD (University of California, USA). Correspondence to Dr M Ozturk, PhD, Molecular Hepatology Laboratory, Massachusetts General Hospital Cancer Center, Charlestown, Massachusetts 02129, USA