|
|
| |||||||||||||||||||||||||||||||||||||||||||||||
Background In patients with sporadic Creutzfeldt–Jakob disease, pathologic disease-associated prion protein (PrPSc) has been identified only in the central nervous system and olfactory-nerve tissue. Understanding the distribution of PrPSc in Creutzfeldt–Jakob disease is important for classification and diagnosis and perhaps even for prevention.
Methods We used a highly sensitive method of detection — involving the concentration of PrPSc by differential precipitation with sodium phosphotungstic acid, which increased the sensitivity of Western blot analysis by up to three orders of magnitude — to search for PrPSc in extraneural organs of 36 patients with sporadic Creutzfeldt–Jakob disease who died between 1996 and 2002.
Results PrPSc was present in the brain tissue of all patients. In addition, we found PrPSc in 10 of 28 spleen specimens and in 8 of 32 skeletal-muscle samples. Three patients had PrPSc in both spleen and muscle specimens. Patients with extraneural PrPSc had a significantly longer duration of disease and were more likely to have uncommon molecular variants of sporadic Creutzfeldt–Jakob disease than were patients without extraneural PrPSc.
Conclusions Using sensitive techniques, we identified extraneural deposition of PrPSc in spleen and muscle samples from approximately one third of patients who died with sporadic Creutzfeldt–Jakob disease. Extraneural PrPSc appears to correlate with a long duration of disease.
The most common prion disease in humans is Creutzfeldt–Jakob disease, which has been classified as sporadic, familial, iatrogenic, or variant. The causation of sporadic Creutzfeldt–Jakob disease is unclear, whereas biochemical, histopathological, and epidemiologic evidence suggests that the variant form results from the transmission of bovine spongiform encephalopathy prions to humans.5,6 PrPSc accumulates in tonsils7 and other lymphoreticular organs8 of patients with variant Creutzfeldt–Jakob disease,9 whereas the sporadic form of the disease is not thought to target the lymphoreticular system.10 The allegedly unique involvement of the lymphoreticular system in variant Creutzfeldt–Jakob disease serves as a diagnostic criterion to distinguish variant from sporadic Creutzfeldt–Jakob disease and is one of the principal pieces of evidence that these forms represent two distinct disease entities.6 Refinements in the sensitivity of the methods of detection11 prompted us to reinvestigate the distribution of PrPSc in humans with sporadic Creutzfeldt–Jakob disease.
Methods
Collection of Samples
We collected central nervous system, spleen, and muscle samples at autopsy from a cohort of Swiss patients who died between 1996 and 2002: 36 patients with prion disease (36 brain specimens, 32 muscle specimens, and 28 spleen specimens), 10 with Alzheimer's disease (10 muscle and 8 spleen samples), and 9 patients without neurologic disorders (9 muscle and 9 spleen samples). The utmost care was taken to avoid contamination of extraneural samples with central nervous system tissue. All tissues were processed according to established guidelines regarding safety and ethics.
Detection of PrPSc in Extraneural Tissues
All procedures were carried out in biosafety level 3 facilities with strict adherence to safety guidelines. The method of precipitation with the use of sodium phosphotungstic acid was adapted from published protocols.8 We prepared 10 percent tissue homogenates (weight per volume) in 2 percent sarkosyl in sterile phosphate-buffered saline using a RiboLyser (Hybaid). Cellular debris was removed by centrifugation at 80xg for one minute in a microcentrifuge. Occasional residual debris was removed by additional centrifugation at 2700xg for five minutes. Supernatants (500 µl) were mixed with equal volumes of 2 percent sarkosyl in phosphate-buffered saline and incubated for 10 minutes at 37°C. Benzonase (Benzon nuclease, Merck) and magnesium chloride were added (final concentration, 50 units per milliliter and 1 mmol per liter, respectively), and the mixture was incubated under constant agitation for 30 minutes at 37°C.
Samples were adjusted to a final concentration of 0.3 percent phosphotungstate, incubated at 37°C for 30 minutes under constant agitation, and centrifuged at 15,800xg for 30 minutes in a microcentrifuge. The pellets were resuspended in 20 µl of phosphate-buffered saline with 0.1 percent sarkosyl. Samples were adjusted to a final concentration of 20 µg of proteinase K per milliliter and incubated at 37°C for 30 minutes. Digestion was terminated by the addition of protease inhibitors (Complete protease inhibitor cocktail, Boehringer Mannheim). Samples were boiled in loading buffer (125 mM TRIS sodium chloride, 4 percent sodium dodecyl sulfate, 20 percent glycerol, and 0.02 percent bromophenol blue), subjected to electrophoresis in 12 percent TRIS–glycine gels, and blotted on nitrocellulose membranes. Membranes were blocked with 5 percent Topblock (Jura) in phosphate-buffered saline with 0.05 percent Tween and then incubated overnight with a monoclonal antibody against PrP (3F4; dilution 1:5000).12
The samples were then incubated with an alkaline-phosphatase–conjugated secondary antibody (1:20,000) and examined with use of 2-chlor-5-(4-methoxyspiro{1,2-dioxetan-3,2'-(5'-chlor) tricycl[3.3.1.13,7]decan}-4-yl)-1-phenylphosphate dinatrium salt chemiluminescent substrate (Amersham) and a VersaDoc digital imager (model 5000, Bio-Rad). Each blot included appropriate positive and negative controls (10 µg or 100 µg of brain homogenate from a patient with sporadic Creutzfeldt–Jakob disease or a patient without a prion disease added to 50 mg of muscle or spleen specimen from a patient without a prion disease). All analyses were carried out and all but six samples were examined at least twice by independent investigators. Samples were deemed positive if proteinase K–resistant diglycosylated, monoglycosylated, and unglycosylated bands with electrophoretic motility indicative of PrPSc were unambiguously identified.
Detection of PrPSc in Neural Tissues
Western blot analysis was carried out as described previously.13 PrP glycoforms were quantified with use of a Kodak image station (model 440) or a VersaDoc digital imager (model 5000), and patients were typed according to the size of the protease-resistant core PrP fragment14 and the prevalence of glycoforms.15,16
Genetic Analysis
Genomic DNA was extracted from frozen tissues. Polymerase chain reaction and analysis of the entire coding region of the prion protein gene (PRNP) were performed with use of standard techniques and software (SeqScape, Applied Biosystems). In addition, codon 129 polymorphisms were identified by restriction-fragment–length polymorphism analysis.17
Histologic Analysis
Tissue was fixed with 4 percent buffered formalin, inactivated by exposure to 98 percent formic acid for one hour, and embedded in paraffin. Sections (3 µm) were subjected to conventional staining and to immunostaining for glial fibrillary acid protein (Dako) and PrP (monoclonal antibody 3F4) after hydrolytic autoclaving.
Results
We analyzed a total of 45 spleen specimens and 51 skeletal-muscle samples from 36 patients with histopathologically, biochemically, and genetically confirmed sporadic Creutzfeldt–Jakob disease (mean [±SD] age, 66.8±8.7 years; 20 men and 16 women); 10 patients with Alzheimer's disease (mean age, 76.2±4.6 years; 6 men and 4 women); and 9 patients without neurologic disease (mean age, 59±19.2 years; 5 men and 4 women). The age distribution, sex ratio, and duration of disease (mean, 5.0±4.6 months) of the cohort of patients with sporadic Creutzfeldt–Jakob disease did not differ significantly from those in published series.18
Accumulation of PrPSc in Spleen and Skeletal Muscle
Western blot analysis of protease-resistant prion protein after phosphotungstate precipitation showed PrPSc in 10 of 28 spleen samples and in 8 of 32 skeletal-muscle samples from the patients with sporadic Creutzfeldt–Jakob disease (Figure 1). Three patients had PrPSc in both spleen and muscle (Table 1). Assays were repeated with independently homogenized tissue fragments: three spleen and three muscle samples always tested positive, whereas others yielded variable results (Table 1), possibly because of inhomogeneous peripheral distribution of PrPSc, as described previously.4,19
|
|
|
Genetic Analysis of PRNP
We sequenced the entire open reading frame of PRNP from all patients with extraneural PrPSc and found no disease-associated mutations. The common methionine–valine polymorphism at codon 129 was analyzed in all 10 patients with splenic PrPSc and all 8 patients with muscle PrPSc as well as in 15 patients without splenic PrPSc and 21 patients without muscle PrPSc. We observed a tendency toward overrepresentation of heterozygosity for methionine and valine and homozygosity for valine among patients with splenic PrPSc, whereas the majority of patients with muscle PrPSc were homozygous for methionine (Table 2).
Histologic and Biochemical Analyses
Typical histopathological features of sporadic Creutzfeldt–Jakob disease, consisting of spongiform changes, neuronal loss, and gliosis, were present in all patients (Figure 2). The electrophoretic mobility of unglycosylated, protease-digested PrPSc (also called the core fragment size, which can be 21 kD for type 1 or 19 kD for type 2)14 allows for the stratification of Creutzfeldt–Jakob disease into subgroups with specific clinical features, which may be caused by different prion strains.6 A finer degree of stratification is attained by including information on the codon 129 genotype and patterns of cerebral deposition of PrPSc; accordingly, we assigned each patient in our series to one of seven proposed groups.18 This analysis indicated an overrepresentation of uncommon variants of sporadic Creutzfeldt–Jakob disease such as MM2C (homozygous for methionine, type 2, with cortical deposition) and MV2 (heterozygous, type 2) in patients with extraneural PrPSc (Table 2), suggesting that peripheral deposition of PrPSc may identify a biochemically or genetically unique subgroup of patients.
|
|
Earlier studies in mice have shown that the PrPSc content may vary among muscle groups.4 In our series, the presence of PrPSc was not limited to specific muscle groups: PrPSc was found in four samples of pectoral muscle, two samples of psoas muscle, and two samples of intercostal muscle. It was absent in 10 pectoral-muscle specimens, 5 psoas-muscle specimens, 2 biceps-muscle specimens, and 4 intercostal-muscle specimens. In three patients without extraneural PrPSc, no information on the anatomical origin of muscle samples could be obtained.
Quantification of PrPSc in Extraneural Organs
In order to estimate the relative concentration of PrPSc in extraneural organs, we compared the intensity of proteinase K–digested Western blot signals with calibration curves obtained by diluting brain homogenate from a patient with Creutzfeldt–Jakob disease with muscle or spleen homogenate from a patient without a prion disease. Despite considerable variations in individual spleen and muscle specimens, the signal intensity of peripheral PrPSc was generally similar to that obtained when 10 µg of brain homogenate from a patient with sporadic Creutzfeldt–Jakob disease was diluted with 50 mg of spleen or muscle from a patient without a prion disease (Figure 1). Therefore, we estimate that the levels of PrPSc in extraneural organs are lower by a factor of approximately 1x10–4 than those typically found in the central nervous system of patients with sporadic Creutzfeldt–Jakob disease.
Discussion
Previous studies did not detect PrPSc in the tonsils, spleen, lymph nodes,10 or appendix of patients with sporadic Creutzfeldt–Jakob disease.21 However, after a systematic search for PrPSc in extraneural organs of patients referred to the Swiss National Reference Center for Prion Diseases, we found that extraneural deposits of PrPSc are much more frequent in patients with sporadic Creutzfeldt–Jakob disease than previously thought. There may be several reasons why this fact was not previously recognized.
First, our series of patients is much larger than previous ones. Second, repeated homogenization of distinct fragments of extraneural organs identified substantial variations in the PrPSc content of these tissues. Generic problems related to the technical reproducibility of our results are unlikely to account for this variation, since control samples consisting of brain homogenate from a patient with sporadic Creutzfeldt–Jakob disease diluted with muscle or spleen homogenate from a patient without a prion disease did not have such variations (data not shown). Instead, the presence of variability suggests that the distribution of PrPSc is not homogeneous in single muscle fibers and single splenic germinal centers. This finding parallels the finding that prion infectivity titers vary in spleen22 and muscle4 samples from experimentally infected mice.
Third, PrPSc levels in spleen and muscle were lower than those in brain by a factor of 1x10–4: extraneural PrPSc was never detectable by plain Western blot analysis (data not shown) and could be visualized only after sodium phosphotungstic acid precipitation, an approach that increases the sensitivity of Western blot analysis by three orders of magnitude by preferentially precipitating PrPSc from tissue homogenates.8,11 Although this method is not suitable for use with very small biopsy specimens (less than 100 mg), it is ideal for situations in which sample size is not a limiting factor.
Previous studies suggested the presence of abnormal PrP in the skeletal muscle of patients with inclusion-body myositis.23 However, this may represent denervation-dependent up-regulation of PrPC transcription.24 In one patient with coincident sporadic Creutzfeldt–Jakob disease and inclusion-body myositis, we found PrPSc levels that were 103 times as high as the levels in our study cohort, possibly because the overexpression of PrPC in muscle facilitates the deposition of PrPSc (Kovacs G: personal communication).
Does the presence of peripheral PrPSc correlate with any specific genetic, biochemical, or clinical factors? When plotting the incidence of codon 129 polymorphisms (MM, MV, or VV [homozygous for valine]), PrPSc core-fragment sizes (type 1, 21 kD, and type 2, 19 kD), and histologic characteristics (plaque and plaque-like lesions [P], cortical or synaptic deposits [C], or thalamic deposits [T]), we observed several uncommon phenotypes in patients with peripheral PrPSc, such as MM2C and MV2. In addition, one patient with PrPSc in lymphoid organs who was homozygous for valine at codon 129 could not be classified according to any of the existing schemes.18,25 These features may represent a new phenotype with atypical clinical signs,26 which we have termed VV1P because of the presence of plaque-like deposits of PrP in the cerebellum and a PrPSc core-fragment size of 21 kD.
The relative ratios of PrPSc glycoforms are different in sporadic and variant Creutzfeldt–Jakob disease. Glycotype analysis failed to reveal features specific to patients with peripheral PrPSc, and none of them had the glycoform distribution associated with variant Creutzfeldt–Jakob disease. Patients with splenic PrPSc had significantly longer durations of disease than did those without splenic PrPSc. Assuming that prions arise primarily in the central nervous system of these patients, one might speculate that protracted disease increases the likelihood of spillover of cerebral PrPSc to extraneural areas.
Although the amount of PrPSc found in spleen and muscle samples from patients with sporadic Creutzfeldt–Jakob disease was much lower than levels in lymphoid organs of patients with variant Creutzfeldt–Jakob disease,7 and the infectivity of prions from these tissues awaits verification, our findings arouse concern about the possibility of iatrogenic transmission of sporadic Creutzfeldt–Jakob disease. Brown et al. reported that extraneural tissues of patients with spongiform encephalopathy (including hereditary forms and kuru) occasionally transmitted disease when inoculated into nonhuman primates.27 Infectivity was detected in 2 of 4 lung-tissue samples, 4 of 35 liver specimens, 5 of 28 kidney specimens, 3 of 31 spleen samples, and 3 of 15 lymph-node samples.27 Five attempts to transmit the disease through the inoculation of skeletal muscle were not successful, possibly because the sample size was insufficient, because the infectivity of the samples was not uniform, or because transmission from human to nonhuman primates is not always highly efficient.
The causation of sporadic Creutzfeldt–Jakob disease is unknown. Some allegedly sporadic cases may in reality have an iatrogenic origin, and surgery of any kind was found to constitute a mild risk factor for sporadic Creutzfeldt–Jakob disease.28 Extensive epidemiologic surveys did not substantiate any risk of blood-borne transmission,29 suggesting that the presence of PrPSc in lymphoid organs whose cellular constituents migrate to and from blood may not lead to substantial contamination of blood donations by prions.
Many patients with sporadic Creutzfeldt–Jakob disease are subjected to extensive neurologic examinations in the prodromal stages, often including electromyography and muscle biopsies. The finding that PrPSc is prevalent in skeletal muscle reinforces calls for the use of single-use needle electrodes and of special protocols for the sterilization of surgical instruments used for muscle biopsies.
Our results suggest that muscle and lymph-node biopsies can be used as diagnostic procedures for sporadic Creutzfeldt–Jakob disease. Because of the unambiguous and extremely specific nature of the immunochemical detection of PrPSc, any positive results would firmly establish the diagnosis of Creutzfeldt–Jakob disease without the need for more invasive procedures. However, the diagnostic sensitivity of the detection of PrPSc in muscle-biopsy specimens is likely to be low, since less than one third of our patients were positive for PrPSc in muscle. We did not detect PrPSc in one muscle-biopsy specimen obtained because of suspected vasculitis 10 weeks before death in a patient with Creutzfeldt–Jakob disease (data not shown). The sensitivity of prion detection is likely to improve soon, in view of the considerable refinements in the technique that are being reported.30
In contrast to previous reports, our study focused on a homogeneous, narrowly defined group of patients. All the patients had died of sporadic Creutzfeldt–Jakob disease in Switzerland between 1996 and 2002. On the other hand, there has been an alarming increase in the incidence of Creutzfeldt–Jakob disease in Switzerland.13 Although the causes of this epidemiologic shift are not clear, the etiologic process of sporadic Creutzfeldt–Jakob disease in our cohort may differ from that in cohorts in other countries. It will therefore be important to extend the present study to additional countries, in order to determine whether our observations apply to all cases of sporadic Creutzfeldt–Jakob disease.
The Swiss Reference Center for Prion Diseases is funded by the Swiss Federal Office of Public Health. Dr. Glatzel is supported by a career development award (Forschungskredit) of the University of Zurich.
We are indebted to all the referring physicians; to Ivan Hegyi, Marco Prinz, and Judith Gottwein for assistance with tissue collection; to Mauri Peltola for technical help; to Lorenz Amsler for help with statistical analysis; to Johannes Streffer for sequencing; to Thorsten Lührs for proposing the use of triangular glycoplots; to Klaus Hess for critical reading of the manuscript; to Prionics for providing laboratory space; and to the University Hospital of Zurich for a generous infrastructural grant enabling the creation of the Swiss CJD tissue collection.
Source Information
From the Institute of Neuropathology and National Reference Center for Prion Diseases, University Hospital of Zurich, Zurich, Switzerland.
Dr. Glatzel and Mr. Abela contributed equally to the article.
Address reprint requests to Dr. Aguzzi at the Institute of Neuropathology, University Hospital of Zurich, Schmelzbergstr. 12, CH-8091 Zurich, Switzerland, or at adriano{at}pathol.unizh.ch.
References
| |||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Extraneural Pathologic Prion Protein
Favereaux A., Perret-Liaudet A., Vital A., Aguzzi A., Glatzel M.
Extract |
Full Text |
PDF
N Engl J Med 2004;
350:732-733, Feb 12, 2004.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
