вход по аккаунту


Evidence for a novel picornavirus in human dermatomyositis.

код для вставкиСкачать
Evidence for a Novel Picornavirus
in Human Uermatomyosim
Neil L. Rosenberg, MD,"$ Harley A. Rotbart, MD,§II Mark J. Abzug, MD,l-I1
Steven P. Ringel, MD,S and Myron J. Levin, MD§#
We prepared RNA probes from cloned segments of human and murine enteroviruses (EVs) for in situ hybridization of
skeletal muscle biopsies from patients with dermatomyositis (DM), polymyositis, other inflammatory myopathies, and
nonidammatory muscle diseases, and from normal control subjects. A probe derived from Theiler's murine encephalomyelitis virus (TMEV) detected viral RNA within mononuclear cells of the interstitial connective tissue in 3 of 5
patients with adult-onset DM. None of these patients showed positive hybridization to probes derived from human
EVs (poliovirus type 1 and Coxsackie virus B3) applied to subjacent sections of the same biopsies. The remaining 2
adult DM patients, 4 patients with childhood-onset DM, and 24 non-DM patients did not react with either TMEV or
human enterovirus probes. Histochemical stains for esterase and immunoperoxidase stains for Mac-1 antigen in the 3
DM patients who reacted positively revealed positive cells in the same distribution as, but in far greater number than,
those positive by in situ hybridization. Immunoperoxidase staining for HLA-DR antigens revealed positive cells in the
same distribution and number as were seen with the TMEV probe. We conclude that an EV-like agent, more closely
related to TMEV than to human EVs, may be associated with DM and that this agent is probably localized within
muscle macrophages that express class I1 major histocompatibility complex antigens.
Rosenberg NL, Rotbart HA, Abzug MJ, Ringel SP, Levin MJ. Evidence for a novel picornavirus in human
dermatomyositis. Ann Neurol 1989;26:204-209
The cause of dermatomyositis (DM) and polymyositis
(PM), inflammatory myopathies that occur in both children and adults, is not known El}. Numerous studies
suggest that muscle injury in these disorders is mediated by cellular immune mechanisms [ 2 ] . Pathological
and immunopathological data suggest that DM and PM
may result from different immunopathogenic mechanisms [31.
Enteroviruses (EVs), members of the picornaviridae
family of single-stranded RNA viruses, have been implicated in both DM and PM by electron microscopic
[ 4 , 51 and serological {G} studies. Recently, dot-blot
hybridization of muscle extracts with a Coxsackie virus
probe also suggested an enteroviral cause for these
diseases [7}. Enteroviruses have not been isolated
from muscle tissues of DM/PM patients except for
occasional patients with hypoglobulinemia or agmmaglobulinemia [ 8 ] . Patients with these immune disorders are uniquely susceptible to chronic infections with
EVs and DM develops in greater than 50% of them
We investigated the role of EVs in inflammatory
muscle diseases with in situ hybridization using RNA
probes derived from poliovirus type 1 (PVl), CoxsacFrom the "Divisions of Research and Neurology, Veterans Administration Medical Center, and tDivision of Infectious Diseases, the
Of Denver' and the Departments Of SNeurology,
§Pediatrics (Section of Infectlous Diseases), "Microb~ologyand Immunology, and 'Medicine, University of Colorado School of Medicine, Denver, CO.
kie virus B3 (CB3), and Theiler's murine encephalomyelitis virus (TMEV), a naturally occurring EV pathogen of mice. The PV1- and CB3-derived probes are
broadly cross-reactive with other human EVs 19, lo},
while the TMEV-derived probe is more specific, hybridizing most avidly with strains of TMEV and encephalomyocarditis virus (EMC), another murine picornavirus [lo}.
Materials and Methods
Muscle biopsies were obtained from 33 patients with the
following diagnoses: adult DM (n = 5), childhood DM (n =
4), polymyositis (n = 5), inclusion body rnyositis (n = 5),
fasciitis (n = l), type 2 fiber atrophy (n = 4),chronic denervation (n = 2), Duchenne muscular dystrophy (n = 2),
rnitochondrial rnyopathy (n = l),and no muscle disease (n
= 4). Diagnosis was based on clinical and histopathological
features fl).
RNA Probes
Cloned sequences of PV1 (pDS14, provided by B. Sernler
and E. Wimmer), CB3 (pCBIII35, provided by S. Tracy),
and TMEV (~1302,provided by H. Lipton) were excised
from their original vectors and recloned into the SP65
Received Oct 6, 1988, and in revised form Nov 30, 1988, and Jan
11, 1989. Accepted for pubhcation Jan 13, 1989.
Address correspondence to Dr Rosenberg, Swedish Medical Center,
125 East Hampden Ave, Eng,ewood, co 80110-0101~
Copyright 0 1989 by the American Neurological Association
PV 1
Fig I. Locations and sizes of the 3 hybridization probes with
respect to the 7.5-kilabase(kb)picomaviral genome. PVl =
poliovims type 1; CB3 = Coxsackie virus B3; TMEV =
Theilw'smurine encephalomyelitis virus.
(pDS14) or PGEM2 (pCBIII35 and p1302) transcription
vectors (Promega, Madison, WI) at the Pstl site of the multiple cloning site linker. In the case of p1302, Pstl digestion
resulted in 2 fragments, the larger of which (p1302a) was
recloned into PGEM2. This new p1302a construct required
removal of the insert poly C tail to achieve subsequent transcription [lo}. Promoter/polymerase-specified RNA transcripts of negative sense were generated from the new
clones, labeled by incorporation of 3H-uridine triphosphatase {lo]. Specific activity of the probes was 2-4 x lo6 c p d
pg RNA. The locations and sizes of the probes with respect
to the generic picornavirus map are depicted in Figure 1.
In Situ Hybridization
Muscle biopsy specimens were quick frozen in isopentane
(cooled in liquid nitrogen) and sectioned (4 pm) with a cryostat microtome. Hybridization was performed on 6 sections
per patient in 2 separate experiments. Sections were individually applied to glass slides that had been treated and cleaned
with xylenes, ethanol, sodium dodecyl sulfate, water, and
hydrochloric acid and pretreated with poly-D lysine to reduce nonspecific binding of probe. Slides were then coated
with 1% 3-aminopropyltriethoxysilane (Sigma, St Louis,
MO) to promote retention of the specimen to the glass.
Tissue sections were fixed by immersion of slides for 20
minutes in a solution of paraformaldehyde-lysine-periodate,
prepared by mixing equal volumes of 8% paraformaldehyde
and 0.1 M lysine hydrochloride in 0.05 M Na2HP04 to
which 200 mg sodium periodate per 100 ml of the mixture
had been added. Tissues were then treated with 0.2 N hydrochloric acid for 20 minutes at room temperature, followed by 1 pdml proteinase K for 15 minutes at 37°C to
make them permeable to the probes; they were then dehydrated again. Coxsackie virus-infected and uninfected LLCMK2 cells and TMEV-infected and uninfected BHK-21 tissue culture cells were used as controls for the quality of the
hybridization procedure icself. Cytospin preparations of
these cells were applied to treated and cleaned (as previously
described) glass slides and hybridized using the tissue protocol already described.
Hybridization buffer was 50% deionized formamide, 1 x
Denhardt's solution, 0.6 M NaC1, 2 mM TFUS, 1 mM ethylenediaminetetraacetic acid (EDTA), and 300 pdml yeast
transfer R N A (tRNA) to which 4 x lo5 cpm per slide of
the TMEV probe alone or 2.5 x lo5 cpm per slide each of
the PV1 and CB3 probes was added. This hybridization mixture was applied to the tissues on the slides and covered with
siliconized coverslips, which were then sealed to the slides
with rubber cement. Incubation was at room temperature for
3 days. Coverslips were then removed and the slides dipped
in a 2O-pdml solution of ribonuciease A for 15 minutes at
37°C to digest nonhybridized probe. The slides were washed
for 72 hours in a solution of 50% formamide, 0.6 M NaC1, 1
mM EDTA, and 0.01 M TRIS at room temperature with
constant agitation. The tissues were dehydrated in graded
ethanols and coated with Kodak NTB 2 emulsion (Eastman
Kodak, Rochester, NY). Autoradiographic exposure was for
4 weeks at 4"C, and was followed by development and staining with hematoxylin and eosin.
Immzrnobistocbernical Staining
Muscle biopsy specimens were quick frozen in isopentane
cooled in liquid nitrogen. Nonspecific esterase staining was
performed, as previously described [l 13, on 10-p.m cryostat
sections. For monoclonal antibody studies acetone-fixed cryostat sections (6 pm) were used. These studies were performed with biotin-conjugated monoclonal antibody directed to HLA-DR, clone L243 (Becton-Dickinson, Mountain View, CA) 1123, and unconjugated monoclonal antibody
to the Mac-1 antigen, clone M1/70HL (Hybritech, Inc, San
Diego, CA) [ 131. Sections were rehydrated in phosphatebuffered saline (PBS) for 3 minutes and then incubated sequentially with the following reagents at appropriate dilutions for the times shown and rinsed with PBS after each
incubation: monoclonal antibody, 60 minutes; avidin-horseradish peroxidase (for biotin-labeled anti-HLA-DR) or peroxidase-labeled anti-rat IgG (for anti-Mac-1), 30 minutes;
chromogen incubation with 0.5% 3,3-diaminobenzidine in
0.01% H202 in PBS, 5 minutes. Sections were rinsed with
distilled water, blued in 3% copper sulfate (5 minutes), and
counterstained with hematoxylin and eosin.
Three of the 5 biopsy specimens from patients with
adult-onset DM (and none of the remaining 28 specimens) were positive by in :situ hybridization with the
TMEV-derived probe (Fig 2). Serial sections from
those 3 positive specimens, were negative when hybridized with the PV1- and CB3-derived probes, as
were the remaining 30 specimens. Cells found to be
positive by hybridization were mononuclear with
abundant cytoplasm located predominantly within the
perimysial, but also within 1he endomysial, connective
tissue. Grains, reflecting the presence of viral nucleic
acid, were concentrated in the cytoplasm with relative
nuclear sparing, consistent with the known replication
pattern of EVs [14}. The number of positive cells per
section ranged from 10 to greater than 50. Findings
were confirmed with a repeat experiment, which produced identical results. Infected tissue culture cell controls were positive with their respective probes; uninfected cells were negative (data not shown).
In subjacent sections, cells in the same distribution
as those found to be positive by hybridization were
both esterase positive (data not shown) and Mac-1
Rosenberg et
Picornavirus(5sin Dermatomyositis
Fig 2. In situ hybridization of muscle biopsy specimens with
Theilds marine encephalomyelitis virus ITMEV)-detivedprobe
fmm patients with dermatomyositis. Panels A-D, TMEV hybridization showing graiw in Lytoplasm of mononuclear cells. In
panel A severul positive ceIls are seen in the perimysaal connective
tissue (arrow on 2 of these cells)).( X 160 before 32% reduction.)
Panels B,C , and D are high-power views of mononuclear cells
with hybvidization grains concentrated in the cytoplmms. Panel
E is a high-power view ofa muscle fiber with hybridization
grains not associated with a mononuclear cell. I x 400 before
47% reduction.)
positive, but these latter proteins were found in many
more cells than those that contained hybridization
grains (Fig 3). Mononuclear cells staining for HLA-DR
antigens were also found within connective tissue, in
similar distribution and number to those identified by
hybridization with the TMEV-derived probe (Fig 3).
Immunohistochemical staining for CD4 and CD8 antigens, as well as for IgM, revealed mononuclear cells
bearing these surface antigens, but in different distribution and frequency from those identified by hybridization (data not shown).
Although muscle tissue itself was not consistently
found to be positive by in situ hybridization in any of
the patients we studied, rare accumulations of grains
were detected over muscle fibers (see Fig 2). These
206 Annals of Neurology Vol 26 No 2 August 1989
resembled the sporadic clumped distribution of EVlike particles seen by electron microscopy, as reported
previously C4, 57.
The finding of viral nucleic acid homologous to TMEV
in DM was not expected because TMEV has not been
associated with human disease. TMEV is a natural
pathogen of mice that produces either an acute poliomyelitis-like disease or a chronic demyelinating illness
{ 151. The hybridization results are unlikely to be
Fig 3 . Staining of muscle biopsiesfrom patients with dermatomyositis that werefound to be podive by in situ hybridization.
(A) Immunoperoxiduse stain for Mac-1 sbou?r that most perimysial cells (in brackets) as well as cells outside the perimysial regions are M a e l positive. ( X 160 before 32% reduction.) (B)
High-power view of Mac-1 -positive perimysial cells. ( x 400 before 32% redaction.) (C)Immunoperoxidase stazn far HLA-DR
shows f m e r positive perimysial cells (in brackets) and cells outside the perzmysial regions than Mac-1 -positive cells. ( x 160
before 32% wduction.) (0)High-power view of HLA-DRpositive perimysial cells. I x 400 before 32% reductzon.)
falsely positive for the following reasons: (1) PV1- and
CB3-derived RNA probes produced no positive hybridization signals; these reagents were prepared identically to the TMEV probe, with the CB3 probe overlapping the TMEV probe in location on the
picornaviral map (see Fig 1). (2) Esterase+/Mac-l+/
HLA-DR+ cells in other inflammatory muscle specimens did not hybridize with the probes, making
nonspecific binding to a particular inflammatory cell
type unlikely. (3) The hybridization reactions were carried out at a stringency that made them essentially free
of background grains, and positive cells were clearly
distinguishable from negative ones (see Fig 2).
Finding only 3 of 9 DM (adult and childhood cases)
biopsies positive can be explained in a number of ways.
First, because of the low number of hybridization-
positive cells and the small size of the muscle biopsy
specimen in relation to total muscle mass, the negative
results may be a result of sampling error. Alternatively, the detection of EVs may be related to the stage
of the disease in which the biopsy is performed. Finally, perhaps adult-onset DM differs from childhoodonset disease in cause as well as age of onset. The small
number of cases studied to date makes interpretation
of this observation difficult.
The specific identity of the nucleic acid detected in
these preliminary studies remains to be determined.
Nevertheless, the finding of stronger reactivity with
the TMEV-derived probe than with those from PV l
and CB3 may explain inconsistencies in previous studies. The association of EVs and inflammatory myopathies has been circumstantial, based on visualization of
EV-like particles by electron microscopy r4, 51 or
serology [GI. Picornaviruses are indistinguishable by
electron microscopy and serological cross-reactivity exists among many of them. Coxsackie virus B2-derived
cDNA has been used to detect presumed viral RNA
in extracts of muscle from patients with DM and PM
[7]. The probe was generated by reverse transcription
of the entire Coxsackie virus B2 genome and, therefore, probably hybridizes to both human and murine
picornaviruses, including TMEV and EMC {9, 101.
Rosenberg et al: Picornaviruses in Dermatomyositis 207
The failure of previous investigators to grow virus
from DM/PM patients may be a function of the choice
of tissue culture cell lines; for example, those routinely
used to isolate human EVs will not sustain TMEV
growth. Thus, a virus related to, but different from, the
recognized human EVs may explain all of these earlier
Encephalomyocarditis virus, another murine picornavirus, reportedly causes human neurological disease,
demonstrating some flexibility in the species specificity
of picornaviruses [161. Genetically, EMC is closer to
TMEV than to human EVs [ 171, and our TMEV probe
hybridizes more avidly with EMC than with human
EVs [lo]. EMC itself is unlikely to be the etiological
agent in our D M patients, as the CB3-derived probe
we used strongly hybridizes to EMC {lo}, and all studies with the CB3-derived probe produced negative results.
The results of our immunohistochemical studies implicate but do not prove that macrophages, or some
other class I1 major histocompatibility complex
(MHC) antigen-bearing cells, are involved in the interaction between virus and host in DM. Enterovirus
infection of macrophages may be related to the pathogenesis of DM by 2 possible mechanisms: viral persistence with ongoing immune response or viral-induced
autoimmunity. Human EVs are capable of persistent
infections of lymphoid cell lines in vitro [lS] and in
patients with antibody deficiency IS]. The cellular reservoir for infectious virus in human disease is not
known. The role of macrophages in the persistence of
EVs in human diseases and in TMEV murine diseases
has not been explored, but evidence is emerging that
macrophages are important sites of persistent infection
for other viral agents [19}. Macrophages may be
phagocytosing virus or virus-infected cells in the DM
patients, or they may be the primary focus of infection,
perhaps even transporting virus to muscle. In either
scenario, products within macrophages may be released, with resultant muscle injury [191.
The circumstantial evidence linking EVs to autoimmune muscle diseases is similar to that linking EVs to
other probable autoimmune disorders such as diabetes
mellitus [ZO} and myocarditis [Zl}. Autoimmunity and
viral (or other) infections are not necessarily mutually
exclusive as pathogenetic mechanisms {22). The presence of viral genome within cells expressing class I1
MHC antigens suggests the possibility that these cells
may be involved in an immune-mediated reaction that
results in bystander muscle injury in these diseases.
Antigenic processing by an antigen-presenting cell,
and its subsequent re-expression in a modified form on
the cell surface, is believed to be an absolute requirement for recognition by class I1 antigen-restricted
C D 4 + T cells [231. CD4+ T cells are found in large
numbers, primarily in the interstitial connective tissue
208 Annals of Neurology Vol 26 No 2 August 1989
in D M [24} and were also found in the interstitial
connective tissue of our DM patients. Antigenic mimicry, wherein viral (or other foreign) antigen is processed by and re-expressed on the macrophage surface
(in concert with class I1 MHC antigens) may be important in the pathogenesis of certain autoimmune diseases 1251. A similar mechanism may explain the relationship between EVs and DM.
Recent studies have found an autoantibody specific
for PM and directed against histidyl-tRNA synthetase
(Jo-1 antigen), possibly resulting from the binding of
histidyl-tRNA synthetase to a viral RNA [26J Sequence homology between a protein of EMC and the
Jo-1 antigen supports a hypothesis of antigenic mimicry and a viral cause in PM [27].
Supported in part by the Biomedical Research Grant Program, Division of Research Resources, NIH (NLR); Veterans Administration
(NLR); and Public Health Service grants POlAG07347-01 (HAR)
and R23A122945-03 (HAR).
Dr Rosenberg is the recipient of a Research Associate Career Development Award of the Veterans Administration.
We gratefully acknowledge Drs Ronald Murray and Lewis P. Rowland for helpful discussions and Neva Murphy for expert technical
1. Rosenberg NL, Ringel SP. Adult polymyositis and dermatomyositis. In: Mastaglia FL, ed. Inflammatory diseases of muscle.
London: Blackwell, 1988:87-106
2. Behan WMH, Behan PO. Immunological features of polymyositiddermatomyositis. Springer Semin Immunopathol 1985;8:
3. Kissel JT, Mendell JR, Rammohan KW. Microvascular deposition of complement membrane attack complex in dermatomyositis. N Engl J Med 1986;314:329-334
4. Chou SM, Gutmann L. Picornavirus-like crystals in subacute
polymyositis. Neurology 1970;20:205-2 13
5. Mastaglia FL, Walton JN. Coxsackie virus-like particles in skeletal muscle from a case of polymyositis. J Neurol Sci 1970;
6. Travers RL, Hughes GRV, Cambridge G, SewellJR. Coxsackie
B neutralisation titres in polymyositiddermatomyositis.Lancet
7. Bowles NE, Sewry CA, Dubowitz V,Archard LC. Dermatomyositis, polymyositis, and Coxsackie-B-virus infection. Lancet
1987;1:1004- 1007
8. McIClnney RE, Katz SL, Wilfert CM. Chronic enteroviral
meningoencephalitis in agammaglobulinemic patients. Rev Infect Dis 1987;9:334-356
9. Rorbart HA, Levin MJ, Vilarreal LP, et al. Factors affecting the
detection of enteroviruses in cerebrospinal fluid with coxsackievirus B3 and poliovirus 1 cDNA probes. J Clin Microbiol
10. Rotbart HA, Abzug MJ, Levin MJ. Development and application of RNA probes for the study of picornaviruses. Mol Cell
Probes 1988;2:65-73
11. Sheehan DC, Hrapchak BB. Enzyme histochemistry. In:
Sheehan DC, Hrapchak BB, eds. Theory and practice of histotechnology, 2nd ed. St. Louis: Mosby, 1980:292-309
12. Engleman EG, Warnke R, Fox RI, et al. Studies of a human T
lymphocyte antigen recognized by a monoclonal antibody. Proc
Natl Acad Sci USA 1981;78:1791-1795
13. Springer T, Galfre G, Secher DS, Milstein C . Mac-1: a macrophage differentiation antigen identified by monoclonal antibody. Eur J Immunol 1979;9:301-306
14. Rueckert RR. Picornaviruses and their replication. In: Fields
BN, Knipe DM, Chanock RM, et al, eds. Fundamental virology.
New York Raven Press, 1986:357-390
15. Lipton HL. Theiler’s virus infection in mice: an unusual biphasic
disease process leading to demyelination. Infect Immun
16. Gajdusek DC. Encephalomyocarditis virus infection in childhood. Pediatrics 1955;16:902-906
17. Pevear DC, Calenoff M, Rozhon E, Lipton HL. Analysis of the
complete nucleotide sequence of the picornavirus Theiler’s
murine encephalomyelitis virus indicates that it is closely related
to cardioviruses. J Virol 1987;61:1507-1516
18. Matteucci D, Paglianti M, Gkdngregorio AM, et al. Group B
coxsackieviruses readily establish persistent infections in human
lymphoid cell lines. Virology 1985;56:651-654
19. Ho DD, Pomerantz RJ, Kaplan JC. Pathogenesis of infection
with human immunodeficiency virus. N Engl J Med 1987;317:
2 78-2 86
20. B a r t - C o n n o r E. Is insulin-dependent diabetes mellitus caused
2 1.
by coxsackievirus I3 infection? A review of the epidemiologic
evidence. Rev Infect Dis 1985;7:207-215
Lerner AM, Wilson FM, Reyes MP. Enteroviruses and the heart
(with special emphasis on the probable role of coxsackieviruses,
group B, types 1-5). Mod Concepts Cardiovasc Dis 1975;44:710
Smith HR, Steinberg AD. Autoimmunity-a perspective. Ann
Rev Immunol 1983;1:175-2 10
Unanue ER. Antigen-presenting function of the macrophage.
Ann Rev Immunol 1984;2:395-428
Arahata K, Engel AG. Monoclonal antibody analysis of mononuclear cells in myopathies. I. Quantitation of subsets according
to diagnosis and sites of accumulation and demonstration and
counts of muscle fibers invaded by T cells. Ann Neurol 1984;
Theofilopoulos AN. Autoimmunity. In: Stites DP, Stobo JD,
Wells JV, eds. Basic and clinical immunology, 6th ed. Norwalk,
Cr: Appleton & h g e , 1987:128-158
Mathews MR, Bernstein RM. Myositis autoantibody inhibits
histidyl-tRNA synthetase: a model for autoimmunity. Nature
1983;304:177-1 79
Walker EJ, Jeffrey PD. Sequence homology between encephalomyocarditis virus protein VP1 and histidyl-tRNA synthetase
supports a hypothesis of molecular mimicry in polymyositis.
Med Hypotheses 1988;25:21-25
Rosenberg et al: Picornaviruses in Dermatomyositis 209
Без категории
Размер файла
742 Кб
evidence, dermatomyositis, picornavirus, novem, human
Пожаловаться на содержимое документа