1441
frequency of beta-cell-specific T-cell reactivity in healthy controls may reflect absence of T lymphocytes reactive to beta-cell antigens because of clonal deletion. Alternatively, such T cells may be present but anergic or actively down-regulated by other T cells; this possibly warrants further investigation. It will be interesting to clone T cells from the lines generated in this study and to investigate further the subcellular localisation and molecular nature of the antigen or antigens involved. It might then be possible to design more specific and simpler diagnostic tests for changes in cellular immunity which lead to diabetes. Such studies could in turn lead to specific immunotherapy for the disease, based on peptide analogues of the antigenic epitopes or the T-cell receptor sequences involved in their recognition.16 The 38 kD islet-cell antigen appears useful in assessment of disease-related islet-cell-specific T-cell reactivity in type 1 diabetes patients at disease onset, and probably in prediabetic subjects before total loss of beta-cell mass and consequent insulin dependency, a stage at which The low
intervention is still feasible. We thank the physicians who helped to obtain patient material; and the Child Health and Well-being Fund and the British Diabetic Association for supporting the project. E. M. B. is a senior research fellow supported by the Wellcome Trust.
REFERENCES 1. Bottazzo GF, Dean BM, McNally JM, MacKay EH, Swift PG, Gamble DR. In situ characterization of autoimmune phenomena and expression of HLA molecules in the pancreas in diabetes mellitus. 2.
N Engl J Med 1985; 313: 353-60. Sibley RK, Sutherland DER, Goetz F, Michael AF. Recurrent diabetes mellitus in pancreas iso- and allograft: a light and electron microscopic and immunohistochemical analysis of four cases. Lab Invest 1985; 53:
132-44. 3. De Berardinis P, Londei M, James RFL, Lake
SP, Wise PH, Feldmann M. Do CD4-positive cytotoxic T cells damage islet beta-cells in type 1 diabetes?. Lancet 1988; ii: 823-24. 4. Bendelae A, Carnaud C, Boitard C, Bach JF. Syngeneic transfer of autoimmune diabetes from diabetic NOD mice to healthy neonates: requirement of both L2T4+ and Ly -2+ T cells. J Exp Med 1987; 166: 823-32. 5. Reich E-P, Sherwin RS, Kanagawa O, Janeway CA Jr. An explanation for the protective effect of the MHC class II I-E molecule in murine diabetes. Nature 1989; 341: 326-28. 6. Van Vliet E, Roep BO, Meulenbroek L, Bruining GJ, De Vries RRP. Human T cell clones with specificity for insulinoma cell antigens. Eur J Immunol 1989; 19: 213-16. 7. Roep BO, Arden SD, De Vries RRP, Hutton JC. T cell clones from type 1 diabetes patient targeted at insulin secretory granule proteins. Nature 1990; 345: 632-34. 8. Kampe O, Andersson A, Bjork E, Hallberg A, Karlsson FA. Highglucose stimulation of 64,000-Mr islet cell autoantigen expression. Diabetes 1989; 38: 1326-28. 9. Buschard K, Jørgensen M, Aaen K, Bock T, Josefsen K. Prevention of diabetes mellitus in BB rats by neonatal stimulation of beta cells. Lancet 1990; 335: 134-35. 10. Gotfredson CF, Buschard K, Frandsen EK. Reduction of diabetes incidence of BB Wistar rats by early prophylactic insulin treatment of diabetes-prone animals. Diabetologia 1985; 28: 933-35. 11. Bruining GJ, Molenaar JL, Grobbe DE, et al. Ten-year follow-up study of islet-cell antibodies and childhood diabetes mellitus. Lancet 1989; i: 1100-03. 12. Baekkeskov S, Landin M, Kristensen JK, et al. Antibodies to a human islet cell antigen precede the clnical onset of insulin dependent diabetes mellitus. J Clin Invest 1987; 79: 926-34. 13. Guest PC, Pipeleers D, Rossier J, Rhodes CJ, Hutton JC. Co-secretion of carboxypeptidase H and insulin from isolated rat islets of Langerhans. Biochem J 1989; 264: 503-08. 14. Hutton JC, Penn EJ, Peshavaria M. Isolation and characterization of insulin secretory granules from a rat islet cell tumour. Diabetologia 1982; 23: 365-73. 15. Hutton JC. The insulin secretory granule. Diabetologia 1989; 32: 271-81. 16. Janeway CA. Autoimmune disease: immunotherapy by peptides? Nature
1989; 341: 482-83.
SHORT REPORTS Genetic predisposition to iatrogenic Creutzfeldt-Jakob disease
The spongiform encephalopathy Creutzfeldt-Jakob disease (CJD) has been transmitted to man via administration of growth hormone and extracted from gonadotropin large pooled batches of human cadaveric pituitary glands. In the UK, 1908 individuals were exposed to potentially contaminated growth hormone, of whom 6 have so far manifested CJD. Examination of the prion protein genes of all these cases and of a single case of gonadotropin-related CJD showed that 4 had the valine 129
homozygous genotype indicating genetic susceptibility to prion infection. Such genetic susceptibility may be important in the uncommon
aetiology of sporadic CJD disease.
The association between Creutzfeldt-Jakob disease treatment with human cadaveric pituitaryderived hormones was recognised in 1985. Each growth hormone batch had been pooled from up to 3000 pituitary glands removed at routine necropsy, and each recipient was treated with growth hormone from more than one batch. Thus all hormone recipients may have been exposed to CJD infective material at least once. We wondered whether development of CJD in some recipients was because a sequence variation in their prion protein (PrP) gene had rendered them genetically more susceptible to prion infection. Genetic susceptibility or incubation time alleles of PrP exist in mice for experimentally transmitted scrapie.1
(CJD) and
We screened samples from all 7 UK pituitary-hormone-related iatrogenic CJD cases (6 growth hormone [from total of 1908 individuals exposed to potentially contaminated batches], 1 gonadotropin) for all known PrP gene variants. DNA was extracted or from fixed or fresh brain tissue with standard The PrP gene open reading frame was amplified by use techniques. of the polymerase chain reaction (PCR).2 PCR product was size fractionated by agarose gel electrophoresis to detect insertions or deletions. 32P-Iabelled allele-specific oligonucleotides were used to detect the other PrP gene variants by hybridisation to PCR product immobilised on nylon membranes.* Codon 102, 117, 178, and 200 or insertion mutations seen in inherited prion disease (see figure) were absent. All patients were homozygous for the common (frequency 90%) PvuII-positive polymorphism at codon 117.3There is another polymorphism at codon 129, where an ATG to GTG change results in a methionine to valine substitution.4 4 of 7 iatrogenic CJD patients were homozygous for the uncommon valine allele (VV) and 2 were heterozygous (MV); the remaining patient was homozogous for the common allele (MM). To obtain better population data on the codon 129 polymorphism we screened 106 healthy caucasians. There were 39 methionine 129 homozygotes and 54 heterozygotes and 13 valine 129 homozygotes. Thus there was a significant excess of valine 129 alleles in the patients (x2 4-90 [with Yates correction
from blood
=
*Full details of primers used are available from The Lancet.
1442
Drugs designed to bind selectively to the abnormal PrP isoform might be expected to inhibit the infective process.
Polymorphic
We thank the physicians and neuropathologists who provided clinicopathological details and tissue specimens for DNA analysis, especially Dr J. Anderson, Prof R. Weller, Prof L. Duchen, Dr D. Spencer, Prof M. Preece, Dr N. Hyman, and Prof A. Harding. This work was funded by the
vananons
Prion protein open variants.
Wellcome Trust.
reading frame with position of known
A=a!anine,D=asparncacid,E=g!utarmcacid.K=!ysine,L=teucine, M=methionine, N=asparagine, P = proline, V = valine.
Pathogenic mutations are shown above the box and polymorphic in the normal population below the box, Pvu I I restriction site polymorphism at position 117 does not alter the aminoacid encoded
variants
for continuity], p 0-027). Since patient numbers are small the data were also analysed with the G-test, giving a value of G=6 11, Distribution of genotypes among the patients and p
frequency of beta-cell-specific T-cell reactivity in healthy controls may reflect absence of T lymphocytes reactive to beta-cell antigens because of clonal deletion. Alternatively, such T cells may be present but anergic or actively down-regulated by other T cells; this possibly warrants further investigation. It will be interesting to clone T cells from the lines generated in this study and to investigate further the subcellular localisation and molecular nature of the antigen or antigens involved. It might then be possible to design more specific and simpler diagnostic tests for changes in cellular immunity which lead to diabetes. Such studies could in turn lead to specific immunotherapy for the disease, based on peptide analogues of the antigenic epitopes or the T-cell receptor sequences involved in their recognition.16 The 38 kD islet-cell antigen appears useful in assessment of disease-related islet-cell-specific T-cell reactivity in type 1 diabetes patients at disease onset, and probably in prediabetic subjects before total loss of beta-cell mass and consequent insulin dependency, a stage at which The low
intervention is still feasible. We thank the physicians who helped to obtain patient material; and the Child Health and Well-being Fund and the British Diabetic Association for supporting the project. E. M. B. is a senior research fellow supported by the Wellcome Trust.
REFERENCES 1. Bottazzo GF, Dean BM, McNally JM, MacKay EH, Swift PG, Gamble DR. In situ characterization of autoimmune phenomena and expression of HLA molecules in the pancreas in diabetes mellitus. 2.
N Engl J Med 1985; 313: 353-60. Sibley RK, Sutherland DER, Goetz F, Michael AF. Recurrent diabetes mellitus in pancreas iso- and allograft: a light and electron microscopic and immunohistochemical analysis of four cases. Lab Invest 1985; 53:
132-44. 3. De Berardinis P, Londei M, James RFL, Lake
SP, Wise PH, Feldmann M. Do CD4-positive cytotoxic T cells damage islet beta-cells in type 1 diabetes?. Lancet 1988; ii: 823-24. 4. Bendelae A, Carnaud C, Boitard C, Bach JF. Syngeneic transfer of autoimmune diabetes from diabetic NOD mice to healthy neonates: requirement of both L2T4+ and Ly -2+ T cells. J Exp Med 1987; 166: 823-32. 5. Reich E-P, Sherwin RS, Kanagawa O, Janeway CA Jr. An explanation for the protective effect of the MHC class II I-E molecule in murine diabetes. Nature 1989; 341: 326-28. 6. Van Vliet E, Roep BO, Meulenbroek L, Bruining GJ, De Vries RRP. Human T cell clones with specificity for insulinoma cell antigens. Eur J Immunol 1989; 19: 213-16. 7. Roep BO, Arden SD, De Vries RRP, Hutton JC. T cell clones from type 1 diabetes patient targeted at insulin secretory granule proteins. Nature 1990; 345: 632-34. 8. Kampe O, Andersson A, Bjork E, Hallberg A, Karlsson FA. Highglucose stimulation of 64,000-Mr islet cell autoantigen expression. Diabetes 1989; 38: 1326-28. 9. Buschard K, Jørgensen M, Aaen K, Bock T, Josefsen K. Prevention of diabetes mellitus in BB rats by neonatal stimulation of beta cells. Lancet 1990; 335: 134-35. 10. Gotfredson CF, Buschard K, Frandsen EK. Reduction of diabetes incidence of BB Wistar rats by early prophylactic insulin treatment of diabetes-prone animals. Diabetologia 1985; 28: 933-35. 11. Bruining GJ, Molenaar JL, Grobbe DE, et al. Ten-year follow-up study of islet-cell antibodies and childhood diabetes mellitus. Lancet 1989; i: 1100-03. 12. Baekkeskov S, Landin M, Kristensen JK, et al. Antibodies to a human islet cell antigen precede the clnical onset of insulin dependent diabetes mellitus. J Clin Invest 1987; 79: 926-34. 13. Guest PC, Pipeleers D, Rossier J, Rhodes CJ, Hutton JC. Co-secretion of carboxypeptidase H and insulin from isolated rat islets of Langerhans. Biochem J 1989; 264: 503-08. 14. Hutton JC, Penn EJ, Peshavaria M. Isolation and characterization of insulin secretory granules from a rat islet cell tumour. Diabetologia 1982; 23: 365-73. 15. Hutton JC. The insulin secretory granule. Diabetologia 1989; 32: 271-81. 16. Janeway CA. Autoimmune disease: immunotherapy by peptides? Nature
1989; 341: 482-83.
SHORT REPORTS Genetic predisposition to iatrogenic Creutzfeldt-Jakob disease
The spongiform encephalopathy Creutzfeldt-Jakob disease (CJD) has been transmitted to man via administration of growth hormone and extracted from gonadotropin large pooled batches of human cadaveric pituitary glands. In the UK, 1908 individuals were exposed to potentially contaminated growth hormone, of whom 6 have so far manifested CJD. Examination of the prion protein genes of all these cases and of a single case of gonadotropin-related CJD showed that 4 had the valine 129
homozygous genotype indicating genetic susceptibility to prion infection. Such genetic susceptibility may be important in the uncommon
aetiology of sporadic CJD disease.
The association between Creutzfeldt-Jakob disease treatment with human cadaveric pituitaryderived hormones was recognised in 1985. Each growth hormone batch had been pooled from up to 3000 pituitary glands removed at routine necropsy, and each recipient was treated with growth hormone from more than one batch. Thus all hormone recipients may have been exposed to CJD infective material at least once. We wondered whether development of CJD in some recipients was because a sequence variation in their prion protein (PrP) gene had rendered them genetically more susceptible to prion infection. Genetic susceptibility or incubation time alleles of PrP exist in mice for experimentally transmitted scrapie.1
(CJD) and
We screened samples from all 7 UK pituitary-hormone-related iatrogenic CJD cases (6 growth hormone [from total of 1908 individuals exposed to potentially contaminated batches], 1 gonadotropin) for all known PrP gene variants. DNA was extracted or from fixed or fresh brain tissue with standard The PrP gene open reading frame was amplified by use techniques. of the polymerase chain reaction (PCR).2 PCR product was size fractionated by agarose gel electrophoresis to detect insertions or deletions. 32P-Iabelled allele-specific oligonucleotides were used to detect the other PrP gene variants by hybridisation to PCR product immobilised on nylon membranes.* Codon 102, 117, 178, and 200 or insertion mutations seen in inherited prion disease (see figure) were absent. All patients were homozygous for the common (frequency 90%) PvuII-positive polymorphism at codon 117.3There is another polymorphism at codon 129, where an ATG to GTG change results in a methionine to valine substitution.4 4 of 7 iatrogenic CJD patients were homozygous for the uncommon valine allele (VV) and 2 were heterozygous (MV); the remaining patient was homozogous for the common allele (MM). To obtain better population data on the codon 129 polymorphism we screened 106 healthy caucasians. There were 39 methionine 129 homozygotes and 54 heterozygotes and 13 valine 129 homozygotes. Thus there was a significant excess of valine 129 alleles in the patients (x2 4-90 [with Yates correction
from blood
=
*Full details of primers used are available from The Lancet.
1442
Drugs designed to bind selectively to the abnormal PrP isoform might be expected to inhibit the infective process.
Polymorphic
We thank the physicians and neuropathologists who provided clinicopathological details and tissue specimens for DNA analysis, especially Dr J. Anderson, Prof R. Weller, Prof L. Duchen, Dr D. Spencer, Prof M. Preece, Dr N. Hyman, and Prof A. Harding. This work was funded by the
vananons
Prion protein open variants.
Wellcome Trust.
reading frame with position of known
A=a!anine,D=asparncacid,E=g!utarmcacid.K=!ysine,L=teucine, M=methionine, N=asparagine, P = proline, V = valine.
Pathogenic mutations are shown above the box and polymorphic in the normal population below the box, Pvu I I restriction site polymorphism at position 117 does not alter the aminoacid encoded
variants
for continuity], p 0-027). Since patient numbers are small the data were also analysed with the G-test, giving a value of G=6 11, Distribution of genotypes among the patients and p
