вход по аккаунту


Detection of human T-lymphotropic virus type I (HTLV-I) tax RNA in the central nervous system of HTLV-IЦassociated myelopathytropical spastic paraparesis patients by in situ hybridization.

код для вставкиСкачать
Detection of Human T-Lymphotropic Virus
Type I (HTLV-I) Tax RNA in the Central
Nervous System of HTLV-I-associated
Myelopathy/Tropical Spastic Paraparesis
Patients by In Situ Hybridization
Tanya J. Lehky, MD,* Cecil H. Fox, PhD,? Scott Koenig, MD, PhD,S Michael C. Levin, MD,'
Nick Flerlage, BS,' Shuji Izumo, MD,$ Elichi Sato, MD,$ Cedric S. Raine, PhD,"
Mitsuhiro Osame, MD,$ and Steven Jacobson, PhD"
Autopsy specimens from 3 patients with human T-lymphotropic virus (HTLV-I)-associated myelopathyltropical spastic paraparesis (HAMITSP) were examined for the presence of HTLV-I in the central nervous system (CNS). In situ
hybridization using an HTLV-I tax RNA probe detected cells containing HTLV-I RNA in spinal cord and cerebellar
sections. HTLV-I infected cells were located within the white matter and, in particular, within the anterior and
lateral funiculi of the spinal cord. Consistent with previously described HAMlTSP pathology, there were perivascular
infiltrates in these CNS specimens. Significantly, HTLV-I RNA was not localized to these infiltrates but was detected
deeper within the neural tissue. Furthermore, phenotypic analysis demonstrated that at least some of the infected cells
were astrocytes. While previous polymerase chain reaction studies have demonstrated the presence of proviral HTLV-I
in CNS specimens, here we provide evidence for the in situ expression of HTLV-I RNA in the CNS of HAM/TSP
Lehky TJ, Fox CH, Koenig S, Levin MC, Flerlage N, Izumo S, Sat0 E, Raine CS, Osame M, Jacobson S.
Detection of human T-lymphotropic virus type I (HTLV-I) tax RNA in the central nervous
system of HTLV-I-associated myelopathyltropical spastic paraparesis patients
by in situ hybridization. Ann Neurol 1995;37:167-175
, Human T-lymphotropic virus (HTLV-I), a human ret;ovirus endemic throughout the world, is etiologically
associated with a hematological disease, adult T-cell
leukemia {I), and a neurological disorder, HTLV-Iassociated myelopathy/tropical spastic paraparesis
(HAM/TSP) [ 2 , 31. HAM/TSP affects about 1 to 5%
of all HTLV-I-infected individuals 141 and is clinically
similar to other chronic progressive myelopathies. The
predominant pathological finding is atrophy of the thoracic spinal cord with axonal degeneration and demyelination of the lateral and anterior spinal tracts {5-8).
In addition, perivascular T-lymphocyte infiltrates and
inflammation of the leptomeninges are noted throughout the central nervous system (CNS). The phenotypic
distribution of these T-lymphocytic infiltrates changes
with chronicity of the disease [7-9). A combination of
From the "Neuroimmunology Branch, National Instirute of Neurological Diseases and Stroke. National Insrirutes of Health. Bethesda.
MD; ?Molecular Histology Laboratories, and $Medimmune, Inc.,
Gaithersburg, MD; $The Third Department of Internal Medicine,
Kagoshima Universiry, Kagoshima, Japan; and "Departments of Pathology. Neurology, and Neuroscience, Albert Einstein College of
Medicine, Bronx, NY.
both CD4+ and CD8' T-cell subsets are present in
early disease with a shift to predominantly C D 8 + T
cells in disease of long duration [9). One report has
demonstrated HTLV-I env protein in spinal cord autopsy samples although the reactivity was weak {b].In
general, however, immunohistochemical, in situ hybridization, and electron microscopy studies have
failed to localize HTLV-I in CNS cells of autopsy specimens from HAM/TSP patients despite the presence
of proviral D N A by polymerase chain reaction (PCR)
{ 10-12).
The presence of inflammatory cells including macrophages and T cells in HAM/TSP spinal cord material
in addition to the upregulation of HLA and cytokine
molecules [13, 141 suggests that such an increase in
immunologic activity in the CNS may be related to
Received Jul 22, 1994, and in revised form Oct 12. Accepted for
uubkdtion Ocr 13. 1994.
Address correspondence to Dr Lehky, Neuroimmunology Branch,
BidR 1 0 1 5 ~ 4 6NINDS,
NIH, 9000 Rockville pike, Bethesda, MD
Copyright 0 1995 by the American Neurological Association
press HTLV-I in the CNS of HAMiTSP patients. In
this report, we use in situ hybridization techniques to
detect HTLV-I RNA in CNS tissue of HAM/TSP patients. Using a radiolabeled HTLV-I tax RNA probe
we have shown that HTLV-I R N A could be detected
in the spinal cord and cerebellum of HAM/TSP autopsy material. HTLV-I RNA was located within spinal
cord white matter and immunohistochemical analysis
indicated that some of these HTLV-I-infected cells
were colocalized within astrocytes. These results provide evidence for the in situ expression of HTLV-I tax
RNA in the CNS of HAM/TSP patients.
the pathogenesis of HAM/TSP. This is consistent with
observations that have also demonstrated an increase
in immunologic activity in the peripheral blood and
cerebrospinal fluid (CSF) of HAM/TSP patients. Studies have indicated that in HAM/TSP there is an activated T-cell state with increased levels of interleukin-2
receptor alpha (IL-2Rcx) on lymphocytes and soluble
IL-2R in patient serum 1151. This is probably related
to the phenomenon of in vitro spontaneous lymphoproliferation, which is elevated in HTLV-I-infected
individuals with neurologic disease Cl6, 17). In addition, high levels of circulating CDS , HTLV-I-specific
cytotoxic T-lymphocytes (,CTLs)restricted to immunodominant nonapeptides of the HTLV-I tax protein
have been demonstrated in the CSF and peripheral
blood of HAM/TSP patients [ 18-20], which is either
lower or absent in HTLV-I-seropositive, asymptomatic individuals. This suggests that HTLV-I-specific
CTLs may contribute to the pathogenesis of HTLV-Iassociated neurologic disease. If the high prevalence of
CD8 lymphocytes in the perivascular infiltrates in the
CNS of HAM/TSP patients is related to the CD8'
HTLV-I-specific CTL detected in peripheral blood
lymphocytes (PBLs) and CSF lymphocytes, then immune-mediated destruction of resident CNS cells may
lead co pathology. Since it has been difficult to demonstrate the presence of HTLV-I in nervous system tissue
[21), the target for these CTLs is unknown.
To determine if HTLV-I could be detected in
HAM/TSP CNS material, autopsy specimens from 3
patients with HAM/TSP were examined for the presence of HTLV-I by in situ hybridization techniques.
In previous studies, D N A from spinal cord and brain
sections of these 3 HAM/TSP patients were amplified
by the PCR with primers for the HTLV-I pol and tax
regions. Liquid hybridization with an HTLV-I polspecific probe demonstrated the presence of HTLV-I
provirus within the spinal cord and to a lesser extent
in the cerebrum [ 131. Since PCR could not distinguish
whether the signals obtained were localized within infiltrating mononuclear cells or neural cells, it was of
limited value in demonstrating which CNS cells ex+
Materials and Methods
A2Ltnp.r.y Specimens
CNS specimens were obtained from 3 patients with HAM/
TSP (Table 1). Patient 1 was a 68-year-old Hispanic woman
from the United States with a 25-year history of HAMITSP
whose pathological and immunological findings have been
described previously 1131. Spinal cord, cerebrum, brainstem,
and cerebellar specimens were obtained and analyzed by in
situ hybridization and PCR. Only spinal cord specimens were
available from Patients 2 and 3, a 7l-year-old Japanese
woman with a 4.5-year history of HAMiTSP and a 73-yea.rold Japanese man with a 10-year history of HAMITSP, respectively. Immunohistochemical studies from the latter two
specimens have also been described previously [9, 141. Control CNS material was obtained from 1 patient who died
of nonneurologic causes and a second patient with multiple
sclerosis who died of stroke. C N S tissue specimens were
fixed in 10% formalin and embedded into paraffin. An
HTLV-I-infected T-cell line, Hut- 102, and normal PBLs
were centrifuged (1,000 rpm) into a pellet, resuspended in a
1 : I mixture of thrombin and A plasma to form a thrombin
clot, fixed in 1 0 p paraformaldehyde, and embedded into
paraffin as described previously 1221.
Situ Hybridizutioiz
Six-micrometer-thick paraffin sections were placed onto sialinized slides (American Histolab, Gaithersburg, MD).In situ
hybridization was performed as described previously [23]
with a 4- to 6-day incubation for emulsion autoradiography.
Slides were counterstained with hematoxylin and eosin. The
radioactive R N A probe was a 1.8-kb fragment from the pS
T u & I. Hziman T-Ly?tiphotropic Virus Type I-associated Myelopathy /Tropical Spastic Paraparesis Autopsy Speiiiti~ti~
HAMiTSP Patient 1
Spinal cord
HAMiTSP Patient 2
Spinal cord
HAM/TSP Patient 3
Spinal cord
Hispanic, USA
CNS Pathology
Cause of Death
Inflammation, loss of myelin
and axons in the spinal cord
9, 14
Inflammation, loss of myelin
and axons
Inflammation, loss of myelin
and avons
Hepatic failure
9, 14
human T-lvmphotropicvirus tvpe I-associated myeiopathyitropicd spastic paraparesis, C N S
168 Annals of Neurology
Vol 37
No 2
February 1995
= central
nervous system
region of the HTLV-1 genome [24f. Each specimen was
tested with the antisense and sense HTLV-1 R N A probes,
the positive and negative probes, respectively. Control spinal
cord and cerebrum autopsy specimens as well as uninfected
PBLs were used as negative tissue controls. The Hut-102
was used as a positive control. A p-actin RNA probe (Lofstrand Inc, Gaithersburg, MD) was used to demonstrate the
presence of R N A in all samples. A human immunodeficiency
virus (HIV) R N A probe (Lofstrand) was used as a negative
control for the probe.
Phenotypic Analysis
Immunocytochemistry of the C N S tissue was performed prior
to in situ hybridization according to methods previously described [25]. Primary antibodies included: OPD4 (1 : 10 dilution) (Dako Corp, Carpinteria, CA) for CD4 T lymphocytes,
CD68 (1 : 50 dilution) (Zymed, South San Francisco, CA)
for macrophage/microglial cells, myelin basic protein (1 : 50
dilution) (Zymed) for oligodendrocytes, and glial fibrillary
acidic protein (GFAP) (1 : 100 dilution) (Chemicon, Temecula, CA) for astrocytes. The secondary antibodies were goat
anti-mouse IgG-biotin conjugate (1 : 100 dilution) (Vector,
Burlingame, CA) for the OPD4 and CD68 primary antibodies and goat anti-rabbit IgG-biotin conjugate (1 : 300 dilution)
(Vector) for the GFAP primary antibody. The antigenantibody-biotin complex was conjugated to avidin (ABC solution, Vector) and bound to diaminobenzidine (DAB; Isopac, Sigma, St Louis, MO), which results in a brown
pigmentation in positive cells. An RNase inhibitor (placental
RNase I N H , Boehringer-Mannheim Inc, Indianapolis, I N )
was added with the antibodies at a concentration of 280 Ug/
ml. Diethyl pyrocarbonate (DEPC)-treated deionized H,O
was used as a diluent and for washes.
Polymerase Chain Reaction
D N A was extracted from paraffin-embedded cell and tissue
sections as described [26}. In brief, specimens were cut into
5-micrometer-thick sections and five sections from each specimen were placed in a 1.5-ml microcentrifuge tube. These
sections were deparaffinized in 1 ml of xylene with gentle
continuous agitation for 30 minutes, at room temperature.
The samples were spun in a microcentrifuge at 14,000 rpm
for 5 minutes and the xylene removed. Deparaffinization
with xylene was repeated and followed by two washes with
0.5 ml of 100% ethanol. After removal of the ethanol, the
remaining pellets were dried in a Savant Speedvac concentrator, resuspended in proteinase digestion buffer (200 pg/ml
proteinase K, 50 mM Tris (pH 8.5), 1 EDTA, 0.001% Triton
X-100, and 0.00001% sodium dodecyl sulfate), and incubated at 55°C for 3 hours. Tubes were centrifuged briefly and
incubated at 95°C for 10 minutes to inactivate the protease.
PCR amplification of the HTLV-I pol region was performed using 10 pi of extracted genomic D N A from each
sample. Generic primer pairs SK110 (position 4757-4778)
and SK111 (position 4942-4919) were used for the amplification. PCR amplification product was analyzed by an enzyme oligonucleotide assay kit (GENE-detective for HTLV-I
pol EOA, Cellular Products Inc, Buffalo, NY) and the final
colorimetric reaction measured by optical density (OD)with
a microplate reader. To determine if a specimen was positive
a reactive threshold value (RTV) was calculated by adding
0.075 to the mean OD of the negative control. Positive samples were equal or above the RTV. Cell preparations [22)
made from Hut-102 cells and uninfected PBLs were used as
positive and negative controls for HTLV-I DNA, respectively. T o test for the presence of D N A after paraffin extraction, each sample was tested using PCR amplification with
human p-actin primers 127) and 10 pl of the amplified D N A
product was run on a 1% agarose gel. The tissue samples
were compared with cell preparations made from freshly
lysed Hut-102 cells 1281 and paraffin-embedded Hut-I02
cells 1221.
In situ hybridization was performed using radioactive
"S-labeled sense and antisense HTLV-I tax RNA
probes on autopsy specimens and cell preparations.
This region was chosen because immunologic studies
on HTLV-I-specific CTL from peripheral blood and
CSF lymphocytes of HAMlTSP patients indicated that
the HTLV-I tax protein was predominantly recognized
by these immune T cells [18j. The specificity of the
antisense HTLV-I tax R N A probe was demonstrated
by the accumulation of silver grains in an HTLV-Iinfected cell line, Hut-102 (Fig la). Silver grains indicate the presence of HTLV-I tax mRNA hybridized
to the radiolabeled antisense RNA probe. No such
grains were observed when a sense HTLV-I tax R N A
probe was used (Fig lb). To confirm the specificity of
the probe for HTLV-I-infected tissues, CNS autopsy
material from HTLV-I-uninfected donors (Fig lc) and
uninfected PBLs (data not shown) were tested with the
HTLV-I antisense probe with no evidence of positive
signal observed. In addition, positive signals could not
be demonstrated on any samples tested using an antisense HIV R N A probe (data not shown).
Once the specificity of the HTLV-I tax RNA probes
had been established, in situ hybridization was applied
to CNS material from the three HAM/TSP autopsies.
To maintain histoanatomical structures for comparison
purposes in each experiment, samples used for the
antisense and sense configurations of the HTLV-I
probe were taken from adjacent sections. As shown in
Figure Id, HTLV-I tax RNA could be detected in the
cerebellum of HAMiTSP Patient 1 with an antisense
HTLV-I RNA probe. Silver grains were present in
cells surrounding a vessel (Fig Id). N o such reactivity
was observed in an adjacent section with a sense
HTLV-I probe (Fig le), again indicating the specificity
of the probe in these tissue specimens. Spinal cord
specimens from the 3 HAM/TSP patients were probed
with the antisense HTLV-I R N A probe (Fig 2a-c).
Mononuclear cells containing a high density of silver
grains were detected with in situ hybridization. These
cells were located in the anterior funiculus (Fig 2a, b)
and lateral funiculus (Fig 2c). In addition, several cells
containing a high density of silver grains were detected
Lehky et al: HTLV-I In Situ Hybridization in HAMiTSP
F i x I . I n .situ hybridization using ail atrti3-etisrand .ienre ' S
/wniatz T-fyinphotropic i,irus type I (HTLV-I)R N A probe.
Thr hybridized R N A is detected as sili'er grains (black dots) iti
HTLV-I-inhi-ted cells when the antiretisf probe is used. The
senje probe i r inhowiologoi~sto niRNA and the e.xperiniental coiiditiom do Not fai,or RNA:DNA hybridizatioti. Therdire.
hzbriclizatiotr udth the j e m e probe shoirld not br detected.
(a, Hut-1 02 i-elh, iising the antisense HTLV-I R N A probe. uith
a high density of silwr graitis orvrlying the c-ells. ( 6 , Hut-102
i-elfl-. mirig the seme I-ITLV-I R N A probe. uith ar2 ab.rencr of
Jilirr grainJ oi'er the i.el'/s. (1-i S p l l i A f mrd ~pecitnenfroin miiftiple srlerosis patlent. iising the antiieme HTLV-I R N A probe.
1i.hic.h did not detrcl HTLV-I-poJ-itiw t-efl~.(dl HTLV-Iasrotiuted niyelopothyltropical spastic-paraparesis (HrlhlITSP)
c-erebellur specirneri (Putirnt 1 ) from a prriz,ascular area. using
thr antisense HTLV-I R N A probe, u'ith sei~erulHTLV-Ipo3itir.e cellj tiear the w ~ s e l !e)
. HAMITSP cerebellar .ipei.inien.
at2 adjacent .\ection to Figure 2e. using Jense HTLV-I R N A
probe. which did not detec-t HTLV-I-positiw cefh. (a+. light
held, x 300 magnijfiiltion bpfore 2 1'Y redidon.)
in the deep white matter of the cerebellum from
HAM/TSP Patient 1 (see Figs Id and I d ) . In both the
spinal cord and cerebellum, some of the positive cells
were observed adjacent to vessels (Fig I d ) but did not
appear t o be within infiltrating cells. O n average, only
two to three positive cells were seen per tissue section
suggesting that cells containing HTLV-I tax R N A
within the CNS may be a rare event using this technique. No significant difference in the number of
HTLV-I-infected cells per spinal cord section was ob-
170 Annals of Neurology
No 2 February 1995
served among these 3 HAMiTSP autopsy cases. T h e
presence of p-actin m R N A was demonstrated by in
situ hybridization in all brain and spinal cord autopsy
specimens indicating that R N A was still present in
these tissues. In addition, positive signals could not be
detected in any of these samples when an antisense
H I V R N A probe was used for in situ hybridization
(data not shown).
Immunohistochemical techniques were coupled to
in situ hybridization to attempt to colocalize HTLV-I
expression to a particular cell type. Positive reactivity is
indicated by brown-staining cells. Phenotypic analysis
using lymphocytic and neuroglial markers indicated
that CD4' cells could be demonstrated in all the C N S
material as represented in the spinal cord in Patient 2
(Fig ia). However, these C D 4 + cells did not contain
HTLV-I R N A (see Fig 3a). Conversely, cells that contained HTLV-I tax R N A were not CD4+ (Fig 3b).
Immunohistochemical staining with antibodies for the
GFAP, which defines glial cells of astrocytic origin,
indicated that some G F A P + cells did colocalize with
cells expressing HTLV-I R N A by in situ hybridization
(Fig 41, although not all infected cells were GFAP'.
Macrophage (anti-CD68) and oligodendrocyte markers
(anti-myelin basic protein) did not colocalize to the
HTLV-I-infected cells. Because of the decreased sensitivity of combined immunohistochemistry with in situ
hybridization, the majority of the HTLV-I-expressing
cells could not be phenotyped. Therefore, it is possible
F i g 2. I n situ hybridization of HTLV-I-associated tnyelopa-
tbyitropical spastic paraparesis (HAMITSPI spinal cord speciniem using an antisense "S hunian T-lyniphotropic uirus type I
(HTLV-I)R N A . The hybridized HTLV-I nzRNA is detected
as silwr grains (black dots) in HTLV-I cells. (a) HTLV-Ipositiw cell near anterior conzmisszire of spinal cord section (Patierzt 1 ) . (Light field. x 200 magmj5cation.1 (Insert: Same cell
photographed in light field with epipolarization, x 400 magnification.) (6) HTLV-I-positirie cell in the anteriorfuniculus of
spinal cord section (Patient 31. (Light field. x 400 nzagnification.) (Insert: Same cell photographed in dark field, x 400 niagmjfcation.) (c) HTLV-I-positiue cell in the lateralfuniculus of
spinal cord section (Patient 21. (Light field, x 400 magnifcation., (Insert: Same cell photographed in dark field, x 400 magnifiration.t (dt HTLV-l-positir,e cells in the deep ichite matter
of the cerebellar peduncle (Patient 1 )from another periz~ascular
region (see Fig Id. el. (Lightfield with epipolarization. x 400
niagnzjfcatiorz.t (All magtiif cations are before 28%) reduction.)
that HTLV-I infection of other neural cell types exists
in the CNS and is not yet identified.
To support the presence of HTLV-I in the CNS
specimens, D N A extraction from the paraffinembedded samples and PCR amplification for HTLV-I
pol was performed (Table 2). The PCR amplification
products were analyzed with an enzyme oligonucleotide assay (EOA). The EOA detected HTLV-I D N A
in all of the HAM/TSP spinal cord specimens and Hut102 cells but not in an uninfected spinal cord speci-
men. Uninfected cortex and PBLs did not contain detectable HTLV-I provirus. All of the reported samples
contained p-actin D N A when analyzed by PCR amplification.
The results of this study provide evidence for the in
situ expression of HTLV-I RNA in cells within the
white matter of the spinal cord and cerebellum of
HAM/TSP patients. HTLV-I-specific signals were not
detected within the perivascular inflammatory areas,
known to consist of predominantly lymphocytes and
macrophages [S, 171; nor could HTLV-I RNA be colocalized with CD4 cells. Combined immunohistochemistry/in situ hybridization experiments have demonstrated that some of these infected cells were
astrocytes. Presence of HTLV-I in the HAM/TSP
specimens was supported by the detection of proviral
HTLV-I with PCR amplification. Collectively, these
results suggest that in HAM/TSP, expression of
HTLV-I in the CNS may reside in neuroglial cells.
While in vitro studies have demonstrated that HTLV-I
is capable of infecting neuroglial cells, including neurons [as], astrocytes [271, and microglial C301, it has
been previously difficult to confirm this within the
CNS of patients with HAM/TSP. Therefore, the in
Lehky et al: HTLV-I In Situ Hybridization in H A M I T S P
172 Annals of Neurology
Vol 37 No 2
February 1995
Fig 3. Immiriioi3i~-toi~~e)tiist~
for CD4+ markers p h r biuriari
T-lymphphotvopii-i,iriis type I (HTLV-I) in sitir hybridization 115ing "S HTLV-I probe. The CD4' 1-ells are pigniented with dia~iinobrnzidjtie( b r o w n ) and the HTLV-I
ttiRNA is deteired ar sifter grains (black dots). ( a ) Uwinfected
CD4' cells (arrows) in spinal iord of Patietit 2. (Lightfield.
x 1.000 magii,ifi-atioribefove 857 rediic-tion.)(b) CD4 cell
containing HTLV-I mRNA from another region of same specinien us ab0z.e. (lightfield, x 1,000 niagnrfiiation before 8 7
i x 3. Ini~1.u~iohi~~toihe?n~st~
/or glial fibrillaty acidic protein+
I marker plr~ihiman T-lymphotropic
type I
(HTLV-11iii sitii hybridization ilsing "S HTLV-I antise)ise
are pigmented witb diatniprvbr. The GFAP ~aJ~troc>~tic/
iidemidine ( b r o w n ) and the HTLV-I mRNA i J detected aj. silw r grains. I n i.etrterfield, GFAP+ cell ( b r o w n ) iotitaining
HTLV-I mRNA iignal (black dots) is present siirroiitided by
GFAP cells withoiit HTLV-I mRNA signal. (Ligbt field.
x 1,000 rtiagnifcatiori.~(Insert: Same cell photographed in
dark ,field. x 1,000 magnifcation.)
situ demonstration of HTLV-I-infected CNS cells is
important in broadening our understanding of the
pathogenesis of this neurologic disorder.
As with other inflammatory diseases of the CNS,
such as multiple sclerosis, an early event in the CNS
disease appears to be blood-brain barrier breakdown
with perivascular influx of mononuclear cells. Susceptibility to this occurrence may involve a combination of
hemodynamic 13 11, host genetic, and environmental
factors [32-341. Similar events may also be operative
in HAM/TSP [ 3 5 ] , although, unlike multiple sclerosis, HAM/TSP involves a known etiologic agent. Potential CNS candidates for HTLV-I infection include
astrocytes, microglia, oligodendrocytes, and neurons.
While CNS HAM/TSP pathology involves predominantly white matter, the corticospinal tract axis tends
to be more affected than the more heavily myelinated
posterior columns. The demonstration of HTLV-I retrovirus in the white matter of HAM/TSP patients by in
situ hybridization, particularly the anterior and lateral
columns, is compatible with these pathologic observations. Furthermore, the in situ observation that at least
some of the HTLV-I-infected cells in the CNS are
astrocytes suggests that mechanisms other than direct
virus-induced demyelination may be involved in the
pathogenesis of HAM/TSP.
In addition to direct HTLV-I infection of glial elements in the CNS of HAM/TSP patients, immunologic factors appear to play an important role in the
immunopathogenesis of HAM/TSP. This is based on
the observations that high levels of circulating HTLVI-specific CD8+ cytotoxic T cells, restricted to immunodominant nonapeptides of the HTLV-I tax protein,
are present in the peripheral blood of patients with
HTLV-I-associated neurologic disease and are lower
or absent in HTLV-I-seropositive, asymptomatic
individuals [18). An estimate of the precursor frequency of these HTLV-I tax-specific CTLs (pCTLs) indicates that a frequency as high as 1 in 75 to 1 in 280
CDS- cells is present in the peripheral blood [20].
This suggests that HTLV-I-specific CTLs may contribute to the pathogenesis of HAMITSP. If these CTLs
play a role in the pathogenesis of this disorder, then
these cells would be expected to be present in the
CNS. This is supported by the demonstration of
HTLV-I-specific CTLs directly from lymphocytes
present in the CSF of affected individuals [ 191. Moreover, CTL precursor frequency analysis indicated a
comparable pCTL frequency of cytotoxic CD8' CSF
lymphocytes specific for HTLV-I to that found within
the peripheral blood C18-20). It is of interest that as
the duration of disease progresses in HAM/TSP, there
is an increase in the number of CD8+ cells infiltrating
the CNS [6, 8). This observation, coupled with the
presence of HTLV-I genomic sequences 10- 131,
upregulation of HLA class I and HLA class I1 molecules [ 131, increased expression of lymphokines [ 13,
141, and this present description of I-ITLV-I tax
mRNA expression in CNS material detected by in situ
hybridization techniques, further supports the role of
HTLV-I-specific immune responses in the pathogenesis of HAM/TSP. Other possible causative factors include the existence of neurotropic viral strains [ ,361,
genetic predisposition [37), and/or additional host immune responses 118, 38).
Table 2. Szimniavy of Human T-Lymphotropic Virzrs
Type I RNA In Sitii H-ybridization and Polymerase Chain
Reaction Ampllfication O N Parafjn-embedded Speiimens
HAMiTSP Patient 1
Spinal cord
HAMiTSP Patient 2
Spinal cord
HAMiTSP Patient 3
Spinal cord
Uninfected patient
Spinal cord
.'In situ hybridization (ISH) using a radiolabeled IITLV-I RNA
probe to the tax region performed on paraffin-embedded specimens
of CNS tissue. Positive samples contained individual cells with high
density of silver grains indicating sites of hybridization to the radiolabeled probe.
bPCR amplification of HTLV-I pol region performed on puaffinembedded specimens. Gene product was detected by an enzyme
oligonucleotide assay (EOA). Criteria for positive value described in
Materials and Methods.
'Not determined.
HTLV-I = human T-lymphotropic virus type I; PCR = polymerase
chain reaction; HAM/TSP = HTLV-I-associated myclopathy/tropical spastic paraparesis.
Lehky e t al: HTLV-I I n Situ Hybridization i n HAMITSP
In this report, it was of interest that while HTLV-Ipositive cells were detected in the C N S of HAM/TSP
patients by in situ hybridization, these represented
only a small number of cells. In general, 1 to 2 positive
cells were detected per tissue section. This may be due
to either a low number of productively infected cells
in the C N S or more likely reflect the insensitivity of
this technique to detect below 10 to 20 copy numbers
of HTLV-I R N A per cell 1371. This is consistent with
the difficulty of detecting HTLV-I R N A by in situ
hybridization in PBLs of HAM/TSP patients in which
the viral copy number fluctuates berween 2 and 2 0
copies per 100 cells [40]. Alternatively, it is possible
that the C N S cells are latently infected with nonexpressing proviral HTLV-I D N A . Proviral D N A is not
readily detected by this in situ hybridization technique,
because experimental conditions d o not favor R N A :
D N A hybridization to nondenatured D N A 14 11. Because of the paucity of HTLV-I-positive cells, it
was difficult to colocalize these signals with immunohistochemical markers for specific C N S cells such
as astrocytes, oligodendrocytes, neurons, o r microglia.
However, we did observe some GFAP cells that were
reactive with the HTLV-I R N A probe. Further studies
that can potentially PCR amplify, in situ, low copy
numbers of HTLV-I D N A and R N A (in situ PCR)
will be required to determine more definitively which
cell type expresses HTLV-I in the C N S of HAM/TSP
patients. While w e have successfully developed in situ
PCR for the detection of HTLV-1 D N A in HAM/TSP
PBLs [ 4 2 ] , it has been technically more difficult to
apply this technique to C N S material. This has also
been the experience of other investigators (A. Haase,
personal communication).
Given the above limitations, the present study has
succcssfully and reproducibly detected the presence of
HTLV-I R N A signals in C N S material from three different HAM/TSP patient autopsies. T h e detection of
HTLV-1 m R N A by in situ hybridization in C N S tissue,
the high levels of CD8’ HTLV-I-specific CTL in
both the peripheral blood and CSF coupled with the
immunohistochemical demonstration of C D 8 cells in
C N S lesions, strengthens the hypothesis for an immunopathological mechanism involved in HTLV-I-associated neurologic disease. In examining the broader
scope of HTLV-I-associated diseases such as uveitis
[43], arthropathy
Sjogren’s syndrome [45}, and
T-lymphocyte alveolitis 1461, our results suggest that
affected organs in these disorders should also be
screened for the in situ presence of HTLV-I to determine if similar immunopathogenic mechanisms may be
Dr Michael Levin is supported by a National Multiple Sclerosis Socitellowship.
174 Annals of Neurology
Vol 37
No 2
February 1995
I . Poiesz BJ, Ruscetri FW,Gazdar AR, et al. Detection and isolation of type C retrovirus particles from fresh and cultured lymphocytes of a patient with cutaneous T-cell lymphoma. Proc
Natl Acad Sci USA 1980;77:7415-7419
2. Gessain A, Barin F, Vernant JC, et al. Antibodies to human
T-lymphotropic virus type-I in patients with tropical spastic
paraparesis. Lancet 1985;2:407-4 10
3. Osame M, Usuku K, Izumo S. et al. HTLV-I associated myelopathy, a new clinical entity. Lancet 1985;1:397-398
4. &plan JE, Osame M, Kubota H, et al. The risk of development
of HTLV-I associated myelopathy/rropical spastic paraparesis. J
Acquir Immune Defic Syndr 1990,3:1096-1101
5. Piccardo P, Ceroni M. RodgemJohnson P. et al. Pathological
and immunological observations on tropical spastic paraparesis
in patients from Jamaica. Ann Neurol 1988;23isuppl):S156S160
6. Moore GRW, Traugott U, Scheinberg LC, Raine CS. Tropical
spastic paraparesis: a model of virus-induced cytotoxic T-cell
mediated demyelinariix~! Ann Neurol 1989;20.523-530
7. lwasaki I. Parhology o f chronic myelopathy associated with
HTLV-I infection (HAMITSP). J Neurol Sci 1990;96:103-123
8. Yoshioka A, tfirose G. Ueda Y ,et al. Neuropdthologicdk studies of the spinal cord in early stage HTLV-I-associated myelopathy (HAM). J Neurol Neurosurg Psychiatry 1993;56:10041007
9. Umehara F, lzumo S, Nakagd &I, et al. Immunocytochemical
analysis of the cellular infiltrates in the spinal cord lesions i n
HTLV-I associated myelopathy. J Neuropathol Exp Neurol
10. Bhigjee AI, Wiley CA, Wachsman W. et al. HTLV-I associated
myelopathy: clinicopathologic correlation with localization of
provirus to the spinal cord. Neurology 190 l ; 4 I : 1990-1992
1 I . KiraJ, Itoyama Y , Koyanagi Y. et al. Presence of HTLV-I proviral D N A in central nervous system o f patients with HTLV-I
associated mydopathy Ann Neurol 1992;3 1 :.59-45
I.?. Ohara Y. lwasaki Y. lzumo 5 , et al. Scarch tor human T-cell
leukemia virus type 1 (HTLV-I)proviral sequences by polymerase chain reaction in the central nervous system of HTLV-I
associated myr:loparhy. Arch Viral 1972;124.3L-43
13. Wu E, Dickson DW. Jacobson S, Raine CS. Neuroaxonal dystrophy in tropical spastic puaparesislHTLV-l associated myelopdthy. Acta Neuropathol 1991;86:221-2 35
14. Uinehara F, Izumo S, Ronquillo A, et al. Cytokine rxprrsslon
in the spinal cord lesions i n HTLV-I associared mythpathy. J
Neuropathol Exp Neurol 1994;5 i:72-’7
15. Tendler CL, Greenberg SJ, Blattner KIA, et al. Transactivation
of interleukin 2 and its receptor induces immune activation in
human T-cell lymphotropic virus type I-associated myelopathy:
pathogenic implications and a rationale for immunotherapy.
Proc Natl Acad Sci USA 1990;87:5218-5222
16. ltoyama Y, Minato S, Kira J . ct al. Spontaneous proliferation
of peripheral blood lymphocytes increased in patients with
HTLV-I-associated myek>pdthy. Neurology 198S;28:1302I307
I 7 Wucherpfenni,g KW. Hollsberg P, Richardson J H . et al. T-cell
activation by autologous human T-cell leukemia virus type Iinfected T-cell clones. I’roc Natl Acad Sci USA 1992;89:2 1102114
18 Jacobson S, Shida H. McFarlin DE, et al. Circulating C D 8 +
cytotoxic T-lymphocytes specific for HTLV-I p X in patients
with HTLV-1 ssociated neurological disease. Nature 1990;348:
1‘) Jacobson S, McFarlin DE, Robinson S. et al. HTLV-I specific
cytotoxic T-lymphocytes in the cerebrospinal Huid of patients
with HTLV-I associated neurological disease. Ann Neurol 1992;
32:(15 1-657
20. Elovaara I, Koenig S, Brewah AY, er al. High human T-cell
lymphotropic virus type 1 (HTLV-1)-specitic precursor cytotoxic
T lymphocyte frequencies in patients with HTLV-I-associated
neurologic disease. J Exp Med 1993;177:1567-1573
21. Liberski PP, Rodgers-Johnson P, Char P, et al. HTLV-I-like
particles in spinal cord cells in Jamaican tropical spastic paraparesis. Ann Neurol 1988;23(,suppl):S185-S187
22. Cottler-Fox M, Fox CH. Examining cells for infectious agents:
d novel approach. J Infect Dis 1991;164:1239-1240
23. Fox C, Coder-Fox M. In situ hybridization for detection of
HIV RNA. In: Coligan JE, Kruisbeek AM, Margulies D H , et
211, eds. Current protocols in immunology, supplement 6 . New
York: Greene Publishing & Wiley Interscience, 1993:1-21
24. Semmes OJ, Jeang KT. Mutational analysis of human T-cell
leukemia virus type 1 tax: regions necessary for function determined with 47 mutant proteins. J Virol 1992;66:7183-7192
25. Lehky TJ, Cowan EP, Lampson LA, Jacobson S. Induction of
HLA class I and I I expression in HTLV-I infected neuroblastoma cells. J Virol 1094;68:1854-1863
26. Wright DK, Manos MM. Sample preparation from paraffinembedded tissues. In: Innis MA, Gelfand D H , Sninsky JJ,
White TJ, eds. PCR protocols-a guide t o methods and applications. San Diego, CA: Academic Press, 1090:153-158
27. Ponte P, Ng SY, Engel J , e t al. Evolutionary conservation in
the untrdnslated regions of actin mRNAs: DNA sequence of a
human beta-actin cDNA. Nucleic Acids Res 198<;12:1687-
28. Kawasaki ES. Sample preparation from blood, cells, and other
fluids. In: Innis MA, Gelfand D H , Sninsky JJ, White TJ, eds.
PCR protocols-a guide to methods and applications. San
Diego, CA: Academic Press, 1990:146-152
29. Yamada M, Watabe K, Saida T, ffim SU. Increased susceptibility of human fetal astrocytes to human T-lymphotropic virus
type-l in culture. J Neuropathol Exp Neurol 1991;50:97-107
30. Hoffman PM, Dhib-Jalbut S, Mikovits JA. er al. Human T-cell
leukemia virus type-l infection of monocytes and microglial cells
in primary human cultures. Proc Natl Acad Sci USA 1992;89:
31. Ijichi S, lzumo S, Eiraku K, et al. An autoaggressive process
against bystander tissues in HTLV-I-infected individuals: a possible parhomechanism of HAM/TSP. Med Hypothesis 1993;41:
32. Poser CM. Pathogenesis of multiple sclerosis. Acta Neuropathol
l986;7 1: 1-10
33. McFarlin DE, McFarland HF. Multiple sclerosis. N Engl J Med
34. McFarland HF, Greenstein J , McFarlin DE, et al. Family and
twin studies in multiple sclerosis. Ann N Y Acad Sci 1984;436:
35. Gessain A, Gout 0. Chronic myeloparhy associated with human
T-lymphotropic virus type I (HTLV-I). Ann Intern Med 1992;
36. Haase A, Evangilista A, Minnegan. e t al. The issues of causation
and neurotropism in neurological diseases associated with infections by retroviruses. In: Blattner W, ed. Current issues in human rerrovirology. New York: Raven Press. 1990:15-25
37. Usuku K, Sonoda S, Osame M, et al. HLA haplotype-linked
high immune responsiveness against HTLV-I-associated myelopathy: comparison with adult T-cell leukemia/lymphoma
Ann Neurol 1988;23(suppl):S143-S150
38. Hollsberg P, Hafler DA. Pathogenesis of diseases induced by
human lymphotropic virus type I infection. N Engl J Med 1993;
39. Brahic M, Haase AT. Detection of viral sequences of low reiteration frequency by in situ hybridization. Proc Narl Acad Sci
USA 1978;75:6125-6129
40. Kubota R, Fujiyoshi T, Izumo S, et al. Fluctuation of HTLV-I
proviral D N A in peripheral blood mononuclear cells of HTLV-1
associated myelopathy. J Neuroimmunol 1993;42:147-1 54
41. Haase AT, Brahic M, Stowring L, Blum H. Detection of viral
nucleic acids by in situ hybridization. In: Maramorasch K, Koprowski H , eds. Methods in virology, vol vii. Orlando, FL: Academic Press, 1984:189-226
42. Walter MJ, Lehky TJ, Fox C H , Jacobson S. In situ PCR for
the detection of HTLV-I in HAMlTSP patients. NY Acad Sci
1994;724:404-4 13
43. Mochizuki IM,Watanabe T, Yamaguchi K, er al. HTLV-I uveitis:
a distinct clinical entity caused by HTLV-I. Jpn J Cancer Res
44. Nishioka K, Maruyama I, Sato K, et al. Chronic inflammatory
arthropathy associared with HTLV-I. Lancet 1989;1:441 (Letter)
45. Sumida T, Yonaha F, Maeda T , et al. Expression of sequences
homologous to HTLV-I tax gene in the labial salivary glands
of Japanese patients with Sjogren’s syndrome. Arthritis Rheum
46. Sugimoto M, Nakashima H , Watanabe S, e t al. T-Lymphocyte
alveolitis in HTLV-I associated myeloparhy. Lancet 1987;Z:1220
Lehky et
HTLV-I In Situ Hybridization in HAM/TSP
Без категории
Размер файла
985 Кб
spastic, patients, paraparesis, rna, detection, iцassociated, virus, typed, system, human, tax, central, lymphotropic, nervous, htlv, myelopathytropical, hybridization, situ
Пожаловаться на содержимое документа