Eur Respir J 1997; 10: 1433–1437
Printed in UK - all rights reserved
Copyright ERS Journals Ltd 1997
European Respiratory Journal
ISSN 0903 - 1936
Viruses and idiopathic pulmonary fibrosis
J.J. Egan*, A.A. Woodcock*, J.P. Stewart#
Idiopathic pulmonary fibrosis (IPF) is an increasing-
ly recognized problem with a generally poor prognosis
. Fifty per cent of patients diagnosed as having IPF
may be expected to die within 4 yrs of the diagnosis .
Although termed idiopathic or cryptogenic it is increas-
ingly acknowledged that an occult inhaled environmen-
tal injury precipitates a common histological response,
which manifests as the heterogeneous clinical syndrome,
IPF [3, 4].
Since many patients date the onset of their symptoms
to a viral infection or a "cold", there has been a sugges-
tion that IPF may be an expression of injury precipitated
or potentiated by a diagnostically occult viral infection.
Previously, hepatitis C virus (HCV)  and Epstein-Barr
virus (EBV)  have been implicated in the aetiology of
IPF, and in this issue KUWANO et al.  present data
evaluating adenovirus as a factor in the aetiology of IPF.
Such viruses have the ability to actively replicate and
to assume an endogenous latent state (adenovirus, EBV)
or a state of chronic infection (HCV) . The study by
KUWANO et al.  highlights the difficulties in investi-
gating and interpreting findings related to viruses and
IPF. With the advent of highly sensitive molecular based
diagnostic techniques, there is the potential for errone-
ous associations between virus and conditions of unclear
aetiology, such as IPF. This review will attempt to place
in context the identification and potential role of viruses
in patients with IPF.
Hepatitis C virus
HCV, a ribonucleic acid (RNA) virus, is now recogni-
zed as the major cause of non-A, non-B transfusion-asso-
ciated hepatitis . With improved diagnostic techniques,
HCV infection is increasingly identifiable and is rec-
ognized as being associated with conditions other than
hepatitis. The background rate of HCV infection is im-
portant in interpreting the relationship between HCV
and extra-hepatic disease. Of people exposed to HCV,
50–70% may be expected to develop chronic infection
. This contrasts with viruses such as adenovirus and
EBV where almost 100% of the population would be
expected to have experienced an acute infection.
There is controversy as to whether IPF is associated
with HCV. UEDA et al.  first suggested a relationship
between IPF and HCV infection by studying 66 Japanese
patients with a clinical diagnosis of IPF, using the first
generation enzyme-linked immunosorbent assay (ELISA)
tests for evidence of HCV infection. Twenty eight per
cent of the IPF patients studied were found to be HCV
antibody positive. Confirmatory testing using recombi-
nant immunoblotting assay (RIBA) was used, and 12 of
the 19 index cases were found to be RIBA positive.
Unfortunately, in this study it was not specified which
RIBA assay was used. The study was then refuted by
IRVING et al.  who studied stored sera from 62 pati-
ents in the UK with a clinical diagnosis of IPF. This
study utilized the next, or "second generation" ELISA
tests and RIBA for four HCV antigens, C100, 5-1-1,
C33 and C22. IRVING et al  found equivocal evid-
ence of HCV antibody in only one patient, and confir-
matory reverse transcriptase polymerase chain reaction
(RT/PCR) was negative. The apparent disparity bet-
ween these two studies can be explained by the high
false-positive rates for the first generation ELISA tests
used by UEDA et al. . False-positive results can occur
with raised immunoglobulin G (IgG) levels, which are
recognized in approximately 50% of IPF patients [5,
The difficulties in interpretation of HCV and IPF are
further emphasized by MELICONI et al. , who stud-
ied 60 Italian patients. Control groups included 130 pati-
ents with obstructive and suppurative lung disease and
4,614 blood donors. The blood donors, as expected, had
a lower rate (0.3%) of HCV positivity than the general
population. Of the IPF patients, eight of the 60 (13%)
had evidence of HCV infection, using second genera-
tion ELISA tests. All eight IPF patients were RIBA pos-
itive and had evidence of HCV RNA by RT/PCR which
did not differ significantly from the 6% of patients in
the control group of mixed lung diseases who were also
HCV positive. Multiple hospital admissions, a risk fac-
tor for HCV exposure, was common to both the IPF
group and the mixed lung disease group .
The study of MELICONI et al.  clarified the issue of
diagnostic techniques, but emphasized that the baseline
prevalence of HCV infection in the population studied
determines the results. In the USA and western Europe,
the prevalence of HCV positivity is approximately 1%,
in northern Italy it is 3.4% and in some parts of Japan
it is as high as 30% [11, 12]. Therefore, it is most like-
ly that the relationship between IPF and HCV is spuri-
ous and reflects the background rate of HCV infection.
Adenovirus is a ubiquitous virus recognized as caus-
ing a spectrum of clinical disease, including both upper
and lower respiratory tract infections from which most
*North West Lung Centre, Wythenshawe Hospital, Manchester, UK.
#Dept of Veterinary Pathology, Royal (Dick) School of Veterinary
Studies, University of Edinburgh, Edinburgh, UK. Correspondence:
J.J. Egan, North West Lung Centre, Wythenshawe Hospital, Southmoor
Road, Manchester M23 9LT, UK.
people recover. MATSUSE et al.  have demonstrated
using the polymerase chain reaction (PCR) that aden-
oviral deoxyribonucleic acid (DNA) exists in the lower
respiratory tract of human lungs. KUWANO et al. , with
nested PCR, have studied the prevalence of adenovirus
DNA in the lung tissue of patients with IPF, using primers
directed against the early gene E1A, a gene that facili-
tates cell transformation . In this study, 16% of IPF
cases had evidence of adenovirus DNA. Using nested
PCR with an in vitro sensitivity of one single copy of
adenovirus DNA, the prevalence is unexpectedly low,
because the lower respiratory tract is a potential reser-
voir for adenovirus .
The low rate of adenovirus DNA positivity may be
explained firstly by enhanced clearance of infected cells
by the background inflammatory and repair process. This
process has been proposed to explain the low preva-
lence of adenovirus DNA in follicular bronchiectasis
. Secondly, the low prevalence of adenovirus DNA
in IPF may be due to sampling error by the use of trans-
bronchial biopsy for the acquisition of lung tissue. Three
transbronchial biopsies in the setting of lung fibrosis are
probably not comparable to studying surgically resec-
ted tissue specimens. In the study by MATSUSE et al. 
using primers for E1A adenoviral DNA, serial tissue
sections alternated between PCR positivity and PCR neg-
ativity. This suggested that E1A DNA was randomly
present in the bronchial epithelial cells. A similar dis-
tribution in the lung of IPF patients would, therefore,
explain the low prevalence in tissue obtained by trans-
bronchial biopsy specimens despite the sensitivity of
The finding by KUWANO et al.  that adenovirus DNA
is more common in those patients receiving corticos-
teroid therapy (67%) compared to those not receiving
therapy (10%) is interesting. Considering that transbron-
chial biopsies may have underestimated the incidence
of adenovirus DNA positivity, this may indicate that the
viral load in such immunocompromised patients is high-
er than normal. Potentially, this is an important finding,
which suggests that adenovirus replication is promoted
by immunosuppression. Whether viral replication in the
presence of immunosuppression then contributes to dis-
ease progression, could only be proven by demonstrat-
ing a response to antiviral therapy.
Conflicting data have been presented on the applica-
tion of antiviral therapy against adenovirus in patients
with IPF. PRIETO et al.  described a case in which
the patient responded to the administration of nebulized
ribivarin. The case appeared to be consistent with a pat-
tern of cellular, usual interstitial pneumonia (UIP). If it
is assumed that cellular UIP represents an "early" phase
of IPF, then such a response to ribivarin would support
the hypothesis that adenovirus infection was involved
in the initiation of the process that resulted in IPF. In
contrast, AGUSTI et al. , studied 10 immunocom-
promised patients with advanced IPF and observed no
benefit from aerosolized ribivarin. Although one patient
exhibited an objective improvement, it is often difficult
to demonstrate any therapeutic benefit in patients with
advanced IPF because of the degree of pulmonary re-
modelling that has occurred . Therefore, to date
there is no clear evidence of a response to therapy for
adenovirus in IPF patients.
EBV is a gamma herpes virus that is present in >90%
of the general population . Acute infection may man-
ifest as the syndrome of infectious mononucleosis. Follow-
ing acute infection, EBV assumes a latent state. Whether
epithelial cells or B-lymphocytes are the site of EBV
persistence is an area of controversy . Host T-cell
mediated immune surveillance maintains EBV latency.
Despite apparent host immune competence, EBV has
oncogenic potential and is associated with Burkitt's lym-
phoma, Hodgkin's lymphoma and nasopharyngeal car-
cinoma . Overt modification of the host immune
state (human immunodeficiency virus (HIV), follow-
ing transplantation) can allow latent EBV-driven B-cell
proliferation, which manifests as lymphoproliferative
disease. Immunosuppression can also give rise to a pro-
ductive EBV replication in epithelial cells, as seen in
oral hairy leukoplakia (OHL) .
EBV was first implicated in the aetiology of IPF in
1984 by a French study based on EBV blood serology
. Elevated levels of IgG and immunoglobulin A (IgA)
against viral capsid antigen (VCA) were observed in 11
patients with IPF. Although, in general, serological test-
ing for EBV infection has a low sensitivity and speci-
ficity, the specificity of IgA against VCA is high and
suggests the presence of active EBV replication at an
epithelial site . Subsequently, a tissue-based study
was undertaken using immunohistochemistry . Of
20 open lung biopsies taken from patients with IPF,
14 had positive staining for both VCA and the mem-
brane antigen gp340/220, EBV antigens expressed dur-
ing viral replication. These findings were compared to
resected tissue taken from a control group that was pre-
dominantly composed of patients undergoing surgical
treatment of lung cancer. In the control group, lung
tissue remote from the neoplasia was studied for EBV.
Of 21 controls studied, two were EBV VCA and gp340/
220 positive .
EBV antigen-positive staining needs to be interpret-
ed in the context of a number of factors, including whether
the viral antigens studied are expressed during viral lat-
ency or viral replication . The most commonly used
reagents for the diagnosis of EBV are those antibodies
specific for proteins associated with EBV transforma-
tion, which are expressed during virus latency. These
include monoclonal antibodies specific for EBV nuclear
antigen (EBNA) 2 and latent membrane protein (LMP),
which are able to detect EBV-transformed proliferating
cells . Examples of this include the detection of EBV
in association with the post-transplant lymphoma and
lymphocytic interstitial pneumonia .
In contrast, antibodies specific for cells containing
EBV that is undergoing productive replication, i.e. pro-
ducing viral particles, are also available in a diagnostic
context. Examples of these include antibodies specific
for immediate early protein BZLF1, components of the
VCA or the membrane antigen gp340/220 . These
reagents are used in the detection of productively repli-
cating EBV in epithelial lesions such as oral hairy leu-
koplakia, and have been studied in IPF [21, 25].
The interpretation of EBV positivity is further influ-
enced by the target cell type of interest . The infected
host cell characteristics influence the phase of the EBV
J.J. EGAN ET AL.
cell cycle. EBV infection of undifferentiated basal epi-
thelial cells or B-lymphocytes results typically in EBV
establishing a latent phase characterized by expression
of EBERs (EBV-related small messenger RNA detected
by in situ hybridization) and the EBNA/LMP antigens
[8, 26]. Conversely, viral replication characterized by
VCA and membrane antigen expression typically oc-
curs in terminally differentiated epithelial cells, includ-
ing the parotid gland, OHL and, it would appear, in IPF
A further difficulty in interpreting immunohistoche-
mistry is that there is the potential for antibodies direct-
ed against viral antigens to cross react with host antigens
("molecular mimicry") . Up to 5% of all monclonal
antibodies developed for the identification of viruses re-
act against autoantigens. In order to minimize this pos-
sibility, two antibodies directed against two different
EBV antigens (VCA, gp340/220) were used in the tissue-
based study of EBV in lung tissue .
Another technique for EBV detection and cellular loc-
alization is the use of in situ hybridization (ISH) for
EBERs . Whilst ISH for EBER and immunohisto-
logy have the advantage of allowing localization of EBV,
they do not necessarily complement or mirror each other.
EBERs are not expressed in lesions of active viral repli-
cation such as OHL, where the use of antibodies are
more informative . Such a biological model as OHL,
in which EBV viral replication occurs in differentiated
epithelial cells in the absence of EBER expression may
be comparable to EBV replication in IPF, in which EBER
expression is absent in cells expressing VCA and mem-
brane antigen (unpublished observations). The identifi-
cation of EBV-related antigens, therefore, demands the
application of corroborative diagnostic techniques.
The PCR is an extremely sensitive technique that has
been applied in the diagnostic context for a number of
viruses. Preliminary data has been presented on PCR
for EBV DNA in IPF. In a small study of lung tissue
(n=4), PCR for the EBV gene BAM W was negative .
In contrast, in a study of 20 cases of IPF using nested
PCR with a sensitivity of 1–10 copies of EBV DNA
(RAJI I), 11 cases had evidence of EBV DNA (versus
1 of 8 normals and 1 of 8 patients with sarcoidosis) .
Another group has also observed that six of 12 cases
with IPF were EBV DNA PCR positive . These study
differences may reflect: EBV activity in discrete areas
of alveolar tissue; patient selection; or the sensitivity of
the PCR assay. A major problem with PCR for the diag-
nosis of EBV is the ubiquitous nature of the virus. EBV
DNA can be detected in the blood of most individuals
and the lung is recognized as a reservoir for EBV .
Quantitation of the EBV DNA load by PCR may
allow a study of the relationship between the viral load
and IPF, and has proven useful in assessing individuals
in the context of post-transplant lymphoproliferative
disease . A study of quantitative PCR in IPF would
be a laborious and difficult study in a heterogeneous
group of individuals. Therefore, the study of EBV in
animal models may give insight into the behaviour of
EBV in lung tissue. EBV, like all gamma herpes viruses,
has an extremely narrow host range. The best-known
model for EBV infection, the cotton top tamarin, does
not breed well in captivity and is an endangered species.
The mouse model of murine gamma herpes virus 68
(MHV68) infection, which, like EBV, is transmitted via
the nasal route, results in an interstitial pneumonia and
transient splenomegaly reminiscent of infectious mono-
nucleosis. It then becomes latent in B-cells, persisting
for the lifetime of the mouse, and is associated with the
development of lymphoma [33, 34].
The presence of EBV, whether detected by immuno-
histochemistry or PCR, has to be judged in the context
of the presence or absence of immunosuppression. EBV-
associated pulmonary fibrosis has also been described
in a heart-lung transplant recipient . Such an asso-
ciation in an immunocompromised transplant recipient
emphasizes the issue of whether a ubiquitous herpes
virus like EBV represents "a passenger or a pathogen"
when localized in lung tissue. Further examples of this
controversy are with cytomegalovirus (CMV) in acqui-
red immune deficiency syndrome (AIDS) patients and
human herpes virus 6 in bone marrow transplant recip-
ients [36, 37]. In the study by KUWANO et al. , an
increased incidence of adenovirus seen in patients re-
ceiving corticosteroid therapy is a further example of
If latent viruses such as adenovirus and EBV exist in
groups of IPF patients, how might such viruses con-
tribute to disease progression? Firstly, viral proteins
that are already present in cells can promote chronic in-
flammation and repair. For instance, EBV LMP increas-
es class II antigen expression in previously EBV negative
B-lymphocytes . Furthermore, viral proteins may pro-
mote the persistence of inflammation that was origin-
ally initiated by a different (environmental) injury. The
repeat GPPAA sequence is observed both in the HLA-
DQ8 β-chain and the EBV EBNA 3 antigen . This
could promote antibody cross-reactivity between viral
and host antigen as is seen with CMV infection follow-
ing solid organ transplantation . Secondly, viral in-
fection may activate type 1 collagen genes in alveolar
epithelium. It has been demonstrated that rat neonatal
alveolar cells immortalized by the adenoviral 12SE1A
gene produce large amounts of type 1 collagen .
Finally, viral genes may act as transactivating factors.
Transactivating factors are nuclear proteins which, be-
cause of specific census sequences, bind to or interact
with DNA, thereby regulating RNA protein transcription
and modifying cell behaviour . Proto-oncogenes are
an example of such target genes for viral proteins. Proto-
oncogenes are normal, "wild" growth regulating genes
while "mutant" proto-oncogenes are expressed in a dere-
gulated manner and play a role in tumorigenesis. Such
oncogenes are expressed in lung tissue. A recent report
(KUWANO et al. ) suggests that overexpression of
wild p53 gene occurs in the lung tissue of IPF patients.
This may reflect regulation of cells with damaged DNA.
Fibroblasts obtained from patients with Li-Fraumeni
syndrome which inherently have p53 mutations, dem-
onstrate in vitro prolongation of their fibroblast life span
and, sometimes immortalization of the fibroblast .
EBV has the potential to interact with oncogenes. The
EBV nuclear antigen (EBNA 5) can form a molecular
complex with both the retinoblastoma (RB) gene and
p53 gene . The EBV immediate early protein BZLF1
interacts with p53 both in vivo and in vitro . BZLF1
inhibits p53-dependent transactivation. Adenovirus also
has the ability to interact with p53 tumour suppressor
VIRUSES AND IDIOPATHIC PULMONARY FIBROSIS
gene . The adenoviral protein E1A has the potential
to bind to the RB oncogene . The RB gene controls
the genesis of the cell replication phase. E1A promotes
cell replication and growth by binding to RB.
Therefore, the interaction of viral proteins (transacti-
vating factors) with genes involved in the regulation of
cell growth may provide a common mechanism by which
different viruses play a role in a single fibroprolifera-
tive disease process such as IPF. However, further study
is required in this area.
How can one reconcile the concept of virus-mediat-
ed IPF and IPF precipitated by an inhaled environmen-
tal injury [3, 4, 6, 21]? Within the heterogeneous group
of patients labelled as IPF, it is possible that different
aetiological groups exist, some with an environmental
trigger, and others with a viral trigger. In some IPF pati-
ents, a primary environmental injury may be potentiated
at a later phase in the disease process by viral replica-
tion within the injured tissue. Ultimately each indivi-
dual or combination of aetiological factors may contribute
to a common clinical and histological response to injury.
The application of sensitive diagnostic techniques has
allowed the detection of ubiquitous viruses in the lung
tissue of patients with IPF. This does not prove cause
and effect. The identification of viruses is influenced by
many factors, particularly the sensitivity and specificity
of the diagnostic techniques, the distribution of the virus
within the lung and the presence or absence of immuno-
Mechanisms by which viruses may promote an in-
flammatory response and influence cell behaviour have
been identified. IPF may not necessarily be related to
one single virus; rather, different ubiquitous viruses may
potentiate the disease process via a common pathway.
As many patients deteriorate while receiving immuno-
suppressant therapy, at issue is whether an increasing
viral load may be a cofactor contributing to disease pro-
gression. In idiopathic pulmonary fibrosis, a pathogenic
role can only be ascribed to viruses if a clinical response
is demonstrated by the administration of antiviral ther-
1. Johnston I, Britton J, Kinnear W, Logan R. Rising mor-
tality from cryptogenic fibrosing alveolitis. Br Med J
1990; 301: 1017–1021.
Stack BHR, Choo-Kang YFJ, Heard BE. The prognosis
of cryptogenic fibrosing alveolitis. Thorax 1972; 27:
Scott J, Johnston I, Britton J. What causes cryptogenic
fibrosing alveolitis? A case control study of environ-
mental exposure to dust. Br Med J 1990; 301: 1015–1017.
Hubbard R, Lewis S, Richards K, Johnston I, Britton J.
Occupational exposure to metal or wood dust and aeti-
ology of cryptogenic fibrosing alveolitis. Lancet 1996;
Ueda T, Ohta K, Suzuki N, et al. Idiopathic pulmonary
fibrosis and high prevalence of serum antibodies to hep-
atitis C virus. Am Rev Respir Dis 1992; 146: 266–268.
Vergnon JM, Vincent M, De The G, Mornex JF, Weynets
P, Brune J. Cryptogenic fibrosing alveolitis and Epstein-
Barr virus: an association. Lancet 1984; ii: 768–770.
7. Kuwano K, Nomoto Y, Kunitake R, et al. Detection of
adenovirus E1A DNA in pulmonary fibrosis using nest-
ed polymerase chain reaction. Eur Respir J 1997; 10:
Hogg JC, Hengle RG. Adenovirus and Epstein Barr Virus
in lung disease. Semin Respir Infect 1995; 10: 224–253.
Gumber S, Chopra S. Hepatitis C: a multifaceted dis-
ease. Ann Intern Med 1995; 123: 615–620.
Irving WL, Day S, Johnston IDA. Idiopathic pulmonary
fibrosis and hepatitis C virus infection. Am Rev Respir
Dis 1993; 148: 1683–1684.
Meliconi R, Andreone P, Fasano L, et al. Incidence of
hepatitis C virus infection in Italian patients with idio-
pathic pulmonary fibrosis. Thorax 1996; 51: 315–317.
Kiyosawa K, Tanka E, Sodeyama T, et al. Transmission
of hepatitis C in an isolated area in Japan: community
acquired infection. Gastroenterology 1994; 106: 1596–
Matsuse T, Hayashi S, Kuwano K, Keunecke H, Jefferies
WA, Hogg JC. Latent adenoviral infection in the patho-
genesis of chronic airways obstruction. Am Rev Respir
Dis 1992; 146: 177–184.
Bateman ED, Hayashi S, Kuwano K, Wilke TA, Hogg
JC. Adenoviral infection in follicular bronchiectasis. Am
J Respir Crit Care Med 1995; 151: 170–176.
Prieto J, Sangro B, Beloqui O. Ribavirin in desquama-
tive interstitial pneumonia. Chest 1988; 93: 446–447.
Agusti C, Xaubet A, Ballester E, Alarcon A, Picado C.
Aerosolised ribavirin in patients with advanced crypto-
genic fibrosing alveolitis: a pilot study. Thorax 1993;
Cherniack RM, Colby TV, Flint A, et al. Correlation
of structure and function in idiopathic pulmonary fibro-
sis. Am J Respir Crit Care Med 1995; 151: 1180–1188.
Niedobitek G, Young LS. Epstein-Barr virus persis-
tence and virus-associated tumours. Lancet 1994; 343:
Straus SE, Cohen JI, Tosato G, Meier J. Epstein-Barr
virus infections: biology, pathogenesis, and manage-
ment. Ann Intern Med 1993; 118: 45–48.
20. Henle W, Henle G. Epstein Barr virus specific serolo-
gy in immunologically compromised individuals. Cancer
Res 1981; 41: 4222–4225.
21. Egan JJ, Stewart JP, Hasleton PS, Arrand JR, Carroll
KB, Woodcock AA. Epstein - Barr virus replication with-
in pulmonary epithelial cells, in cryptogenic fibrosing
alveolitis. Thorax 1995; 50: 1234–1239.
22.Stewart JP, Arrand JR. Expression of the Epstein-Barr
virus latent membrane protein in nasopharyngeal car-
cinoma biopsy specimens. Hum Pathol1993; 24: 239–242.
23.Egan J, Stewart JP, Yonan N, Arrand JP, Woodcock AA.
Non Hodgkin lymphoma in heart/lung transplant recip-
ients. Lancet 1994; 1: 481.
24.Hoffman GJ, Lazarowitz SG, Hayward SD. Monoclonal
antibody against a 250,000 dalton glycoprotein of
Epstein-Barr virus identifies a membrane antigen and a
neutralising antigen. Proc Natl Acad Sci USA 1980; 77:
25. Niedobitek G, Young LS, Lau R, et al. Epstein-Barr
infection in oral hairy leukoplakia: virus replication in
the absence of a detectable latent phase. J Gen Virol
1991; 72: 3035–3046.
26.Howe JG, Steitz JA. Localisation of Epstein-Barr virus-
encoded small RNAs by in situ hybridization. Proc Natl
Acad Sci 1986; 83: 9006–9010.
27.Parkkonen P, Hyoty H, Ilonen J, Reijonen H, Yla-Hert-
tuala S, LeinikkI P. Antibody reactivity to an Epstein
J.J. EGAN ET AL.
barr virus BERF4-encoded epitope occurring also in Asp-
57 region of HLA-DQ8 B chain. Clin Exp Immunol
1994; 95: 287–293.
Wangoo A, Nicholson AG, Diss TC, Farrell PJ, du Bois
RM, Shaw RJ. Cryptogenic fibrosing alveolitis does not
appear associated with Epstein-Barr virus infection.
Thorax 1996; 51: s3: A11.
Egan JJ, Stewart JP, Hasleton PS, Nash AA, Woodcock
AA. EBV PCR (Raji I) in idiopathic pulmonary fibro-
sis and sarcoidosis. Am J Respir Crit Care Med 1997;
Quddus MR, Nayer A, King TC, Jackson CL, Kuhn C.
EBV in idiopathic pulmonary fibrosis. United States and
Canadian Academy of Pathology 1997; 171A.
Lung ML, Lam WK, So SY, Lam WP, Chan KH, Ng
MH. Evidence that respiratory tract is major reservoir
for Epstein-Barr virus. Lancet 1985; 1: 889–892.
Rooney CM, Smith CA, Ng CYC. Use of gene-modi-
fied virus specific T-lymphocytes to control Epstein-
Barr virus related lymphoproliferation. Lancet 1995;
Sunil-Chandra NP, Efstathiou S, Arno J, Nash AA.
Virological and pathological features of mice infected
with murine gamma herpes 68. J Gen Virol 1993; 73:
Nash AA, Usherwood EJ, Stewart JP. Immunological
features of murine gamma herpes virus infection. Semin
Virol 1996; 7: 125–131.
Egan JJ, Stewart JP, Hasleton PS. Epstein Barr Virus
associated graft failure following heart lung transplan-
tation. Thorax 1996; 51: 1160–1162.
Millar AB, Patou G, Miller RF, et al. Cytomegalovirus
in the lungs of patients with AIDS. Respiratory patho-
gen or passenger? Am Rev Respir Dis 1990; 141: 1474–
Carrigan DR, Drobyski WR, Russler SK, Tapper MA,
Knox KF, Ash RC. Interstitial pneumonitis associated
with human herpes virus-6 infection after marrow trans-
plantation. Lancet 1991; 338: 147–149.
38. Zhang Q, Brooks L, Busson P, et al. Epstein - Barr virus
latent membrane protein 1 increases HLA class II ex-
pression in an EBV-negative B-cell line. Eur J Immunol
1994; 24: 1467–1470.
Fujinami RS, Nelson JA, Walker L, Oldstone MA. Seq-
uence homology and immunologic cross-reactivity of
human cytomegalovirus and HLA-DR B chain: a means
for graft rejection and immunosuppression. J Virol
1988; 62: 100–105.
Matsui R, Goldstein RH, Mihal K, Brody JS, Steele
MP, Fine A. Type I collagen formation in rat type II
alveolar cells immortalised by viral gene products. Thorax
1994; 49: 201–206.
Pollack PS. Proto-oncogenes and the cardiovascular
system. Chest 1995; 107: 826–835.
Kuwano K, Kunitake R, Kawasaki M, et al. p21 WAF1/
CIP1/Sdi11 and p53 expression in association with
DNA strand breaks in idiopathic pulmonary fibrosis. Am
J Respir Crit Care Med 1996; 154: 477–483.
Bischolff FZ, Yim SO, Pathak S, et al. Spontaneous
abnormalities in normal fibroblasts from patients with
Li-Fraumeni cancer syndrome: aneuploidy and immor-
talization. Cancer Res 1990; 50: 7979–7984.
Szekely L, Selivanova G, Magnusson KP, Klein G, Wiman
KG. EBNA-5, an Epstein barr virus encoded nuclear
antigen, binds to the retinoblastoma and p53 proteins.
Proc Natl Acad Sci USA 1993; 90: 5455–5459.
Zhang Q, Gutsch D, Kenny S. Functional and physical
interaction between p53 and BZLF1: implications for
Epstein Barr virus latency. Mol Cell Biol 1994; 14:
Sabbatini P, Chiou SK, Roa L, White E. Modulation of
p53-mediated transcriptional repression and apoptosis
by the adenovirus E1B 19K protein. Mol Cell Biol 1995;
Whyte P, Buchovich KG, Horowitz JM, et al. Associa-
tion between an oncogene and an anti-oncogene: the
adenovirus E1A protein binds to the retinoblastoma
gene product. Nature 1988; 334: 124–129.
VIRUSES AND IDIOPATHIC PULMONARY FIBROSIS