Increased Reactivity to HTLV-I

in Inflammatory Nervous System Diseases

Francesco Lolli, MD, Sten Fredrikson, MD, Slavenka Kam-Hansen, MD, and Hans Link, MD

The presence of IgG antibodies reacting with purified and disrupted human T-lymphotropic virus type I (HTLV-I)
was examined by an indirect enzyme-linked immunosorbent assay (ELISA) i n sera from 49 patients with multiple
sclerosis (MS),2 1 patients with aseptic meningoencephalitis (AM), 12 patients with Guillain-Barre syndrome (GB), and
30 patients with tension headache (TH). This was also assessed in the concentrated cerebrospinal fluid (CSF) of most of
these patients, as well as in sera of 60 blood donors (BD). Standardized amounts of serum IgG and CSF IgG were used
in ELISA. For sera, higher reactivity with HTLV-I was found in all four patient groups compared with the BD group,
but no significant differences were observed among the four groups. There was higher reactivity with HTLV-I in the
CSF of patients with MS, AM, and GB compared to findings in patients with TH. Ten serum (2 MS, 3 GB, 3 TH, 2 BD)
and 3 CSF (1 MS, 1 GB, 1 TH) specimens considered positive by ELISA for HTLV-I were found negative on
confirmatory Western blot analysis. We extended this study to analyze the in vitro production of anti-HTLV-I-IgG
antibodies by the 24-hour cultivation of unstimulated lymphocytes from peripheral blood and CSF of 6 additional
patients with MS directly in HTLV-I antigen-coated wells of microtiter plates. This was followed by determination of
specific antibodies by ELISA in the same wells. No antibody production was measurable. Our data do not favor the
hypothesis of an HTLV-I-related human retrovirus in the etiology of MS.
Lolli F, Fredrikson S, Kam-Hansen S, Link H : Increased reactivity to HTLV-I in inflammatory
nervous system diseases. Ann Neurol 22:67-71, 1987

A human retrovirus may be a logical candidate in the

etiology of multiple sclerosis (MS) because such agents
have been found to (1) be capable of altering the immune response; (2) be neurotropic; and ( 3 ) resemble,
both biologically and genetically, visna virus which
causes a chronic demyelinating disease of sheep that in
many ways resembles MS (for review see {18]). Visna
virus is a retrovirus belonging to the same subgroup,
so-called lentiviridae, as the third discovered human
retrovirus, the human immunodeficiency virus (HIV)
[Z, 81 which causes acquired immune deficiency syndrome. HIV is also lymphotropic and neurotropic 1.11,
161. Human T-lymphotropic virus type I (HTLV-I),
the first discovered human retrovirus which is the
cause of a rare form of human T-cell leukemia [201,
has also been associated with neurological disorders,
first with tropical spastic paraparesis [9] and then with
MS {lS]. Its association with MS, however, has been
the subject of considerable controversy [lo, 141.
We have examined serum and cerebrospinal fluid
(CSF) for the presence of antibodies reactive with
HTLV-I in patients with MS and other inflammatory
as well as noninflammatory neurological diseases. Since
antibodies might be bound to the target or metab-

From the Department of Neurology, Karolinska Institute, Huddinge University Hospital, S-141 86 Huddinge, Stockholm,

Sweden.

olized and their absence in body ffuids does not rule

out the possibility of an immune response [7, 197, we
have also studied the in vitro production of antibodies against HTLV-I by lymphocytes isolated from
peripheral blood and CSF. Our results, even though
they agreed with those reported by Koprowski and
associates {15], do not show an association between
HTLV-I and MS because the use of proper controls
results in a different interpretation.

Materials and Methods

Patients
Serum was obtained from the following groups. Forty-nine
patients (32 females) had clinically definite MS according to
the criteria of Schumacher and colleagues {23}. Their age
varied between 19 and 70 years (median, 40.5 years). Thirtythree of the MS patients had no o r slight disability, that is,
they were in all respects able to manage on their own; the
remaining 16 patients had moderate to severe disability and
required assistance in professional and social activities. The
duration of MS varied from 1 month to 24 years (median, 5
years). Of the 49 patients, 22 were examined during an
exacerbation, consisting of the sudden appearance of new
symptoms and signs or a sudden reappearance or worsening
of previous findings within one month of examination. Eigh-

Received Oct 10, 1986, and in revised form Dec 4. Accepted for
publication Dec 5, 1986.
Address correspondence to Prof. Link.

67

teen of the MS patients were in remission and 9 had a

chronic progressive form of MS. All except 3 patients had
oligoclonal IgG bands in CSF, demonstrated by agarose
isoelectric focusing 1171, and 43 of the 49 patients displayed
intrathecal IgG production, as reflected by an elevated CSF
IgG index (higher than 0.7) [24}. Seven of the patients had a
slight disturbance of the blood-brain barrier (BBB), as determined by the CSF/serum albumin ratio [24]. None of
the MS patients had been treated with immunomodulatory
drugs.
Twenty-one patients (8 females) had aseptic meningoencephalitis (AM; age range, 16 to 60 years; median, 36 years).
Thirteen had AM of unknown cause and showed the usual
benign clinical course; 5 had positive serology for infection
with BoweIia bnrgdot$mi and 3 for herpes simplex.
Twelve patients (4 females) fulfilled the criteria of Asbury
[l} for Guillain-Barre syndrome (GB; age range, 18 to 81
years; median, 48 years). All had moderate to severe BBB
damage, as indicated by the CSF1serum albumin ratio.
Control subjects consisted of 30 patients (2 1 females) with
tension headache (TH) and 60 healthy blood donors (BD; 22
females). Their age varied from 16 to 67 years (median, 40.5
years) and 23 to 61 years (median, 34 years), respectively.
All the TH patients had normal CSF findings regarding cell
count, CSF/serum albumin ratio, CSF IgG index, and agarose
isoelectric focusing patterns.
Corresponding CSF was available in sufficient quantity
from 34 ( 2 3 females) of the patients with MS (age range, 21
to 70 years; median, 37.5 years), 10 (4females) with AM
(age range, 19 to 58 years), 6 (3 females) with G B (age range,
18 to 55 years), and 13 (9 females) withTH (age range, 21 to
62 years).
For in vitro assay of HTLV-I antibody production, peripheral blood lymphocytes and CSF lymphocytes were obtained
from 6 additional patients with clinically definite MS. Their
age range was 16 to 45 years (median, 42 years). Of these
patients, 4 were suffering an exacerbation, one was in remission, and the remaining one had a chronic progressive form
of MS. Four of these patients had moderate to severe disability. Numbers of mononuclear cells in the CSF varied between 8 and 29 per microliter (median, l l ) . The CSFherum
albumin ratio was normal in all 6 patients, and all had an
elevated CSF IgG index and oligoclonal bands in the CSF.
As control, peripheral blood lymphocytes were obtained
from 10 healthy subjects who were members of the hospital
staff.

Determination of Reactivity with HTLV-1

in Serum and CSF Using ELISA
The presence of antibodies reactive with purified and disrupted HTLV-I was studied with an indirect enzyme-linked
immunosorbent assay (ELISA), according to the method of
Saxinger and Gallo (22). To avoid the influence of different
IgG concentrations [21}, the samples were adjusted to the
same IgG level (500 mg/L). Commercially available microtiter plates coated with purified and disrupted HTLV-I
(NEA-985 1; DuPont) were used. The wells were filled with
100 pl of 2% bovine serum albumin (Sigma A-4503) in
0.01 M phosphate-buffered saline (PBS) containing 0.15 M
NaCl and 0.025% NaN3 (pH, 7.4). The plates were then
incubated for 1 hour at room temperature, rinsed four times

68 Annals of Neurology Vol 22 N o 1 July 1987

with PBS containing 0.05% Tween 80 (PBS-T), and incubated in duplicate with 10O-kl samples diluted in PBS
containing 10% normal rabbit serum, to reach an IgG concentration of 500 m&. The CSF samples were concentrated
in macrocollodion bags (Sari:orius Membranfilter, Gottingen,
F.R.G.), and the IgG in the concentrated specimens was
measured by a sandwich ELISA. The IgG concentration optimal (500 mdL) for the detection of anti-HTLV-I antibodies
was determined by titration using positive serum from a patient with T-cell leukemia and negative serum (both from
DuPont). O n each plate were included in duplicate the positive and negative reference sera diluted 1120 in PBS containing 10% normal rabbit serim, plus blank and background
control wells (221. After 2 hours of incubation at room temperature, the wells were washed five times with PBS-T and
incubated for 1 hour at room temperature with 100 kI of
alkaline phosphatase-conjugated (type VII-S; Sigma) rabbit
antiserum against human y-chains (code A090; Dakopatts,
Copenhagen, Denmark), diluted 114,000 in PBS containing
1% normal rabbit serum. The conjugate was obtained according to the method of Voller and associates [25J After
washing, in each well was applied 100 kl of a solution containing 1 mg/ml of the substrate p-nitrophenylphosphate
(Sigma, 104-105) in 10% d~~ethanolamine
buffer containing
0.02% MgClz (pH, 9.8). After 30 minutes of incubation at
room temperature, the plates were read at 405 nm in a
Titertek Multiscan plate reader (Eflab Oy, Helsinki, Finland).
Results are expressed as the so-called ELISA ratio, which is
calculated by: (absorbance sample - absorbance blank): (absorbance reference negative sera - absorbance blank). The
system showed good reproducibility, with an interday coefficient of variation (CV) for positive and negative controls of
<lo% and a CV between pairs of observations of <5%. All
the samples with an ELISA ratio higher than 2 in repeated
observations were considered positive and were tested by
Western blot analysis using commercial strips (NEA-8005,
DuPont), according to the companys instruction.

Determination of HTLV-1 Antibody

Production in Vitro
The production of anti-HTLV-I IgG antibodies by peripheral blood lymphocytes and CSF lymphocytes was examined
by in vitro cultivation of the cells directly in antigen-coated
wells, with subsequent determination of the specific antibodies in the same culture wells by ELISA, so-called cellELISA as described recently [7}. Fifteen milliliters of CSF
was obtained by lumbar puncture and kept at 4C. Total and
differential cell counts were performed within 30 minutes
using phase contrast microscopy. CSF was centrifuged at 200
g for 20 minutes at 4C. The CSF pellet was resuspended in
tissue culture media (TCM, to be discussed) and kept at 4C
for at most 30 minutes before cultivation. Peripheral blood
lymphocytes were isolated by Ficoll-Hypaque density gradient centrifugation 141. The cells were washed three times
in TCM, consisting of RPMI. 1640 (Flow Lab), 1% glutamine, 1% penicillin-streptomycin-neomycin, and 20% fetal
calf serum. The cells were nsuspended to achieve a final
concentration of 1 X lo6 lymphocytes per milliliter (0.5 X
lo6 for three CSF preparations) in TCM. Cell viability was
checked by the trypan blue dye exclusion test before cultivation and was always higher th,an 95%.

Two hundred microliters of the cell suspension was applied in triplicate (duplicate for CSF lymphocytes) to HTLVI-coated wells. The following controls were included on
each microtiter plate: the positive and negative reference
sera diluted 1/20 in TCM; the patient's serum diluted 1/20 in
TCM and the undiluted CSF to detect the possible presence
of free antibodies; and the fluid from the third peripheral
blood lymphocyte washing (undiluted) for control of possible
contamination of the cell preparation by antibodies present
in serum and peripheral blood lymphocytes obtained from a
healthy subject at the same cell concentration in TCM. Cultivation was performed at 37C in 5% COZ in air and 100%
humidity for 24 hours (24 and 48 hours in 2 cases). Thereafter, the plates were washed five times with PBS-T and the
presence of specific antibodies detected, as described.

Results
The Figure shows reactivity with HTLV-I evaluated by
ELISA and presented as values of the ELISA ratio for
serum and CSF from each of the patients as well as
values for serum from BDs. The ELISA results for
HTLV-I IgG antibodies in serum and CSF showed
normal logarithmic distributions for the ELISA ratio.
In the BD group, HTLV-I ELISA revealed a median
ELISA ratio of 0.5, a geometric mean of 0.5, and an
upper reference limit (antilogarithm of the geometric
mean + 2 standard deviations in logarithmic values)
for the ELISA ratio of 2. No age- or sex-related variation was observed. Based on these data, a cutoff value
of 2 was chosen for the ELISA ratio. Values higher
than 2 were thus considered as positive for the pres-

I
I

ER

serum

.
...

2
1

.
. .
..
..:. ...f .
.* .
"& .:.
..,...P.... +.: 4. 1
:
... ... +..
..
.... ....
....
at.

0.5

+@

8qj

0.1

CSF

ence of HTLV-I antibodies. In similar systems, this

value has previously shown the highest sensitivity { 5 ,
261. The sera of MS patients reacted significantly
higher with HTLV-I (median ELISA ratio, 0.84) compared with BDs (median ELISA ratio, 0.5; Wilcoxon
nonpaired test, p < 0.0001). Similarly, the reactivity of
the sera from patients with AM (median ELISA ratio,
0.92) was higher in comparison with BDs Cp <
O.OOOl), as was the reactivity of the sera from the
patients with GB (median ELISA ratio, 0.95; p <
0.001) and TH (median ELISA ratio, 0.82; p < 0.01).
Conversely, no significant differences were found
among the groups of subjects with MS, AM, GB, and
TH. In the CSF studies, patients with MS (median
ELISA ratio, 0.52), AM (0.5), and GB (0.52) had
significantly higher reactivity with HTLV-I compared
with TH subjects (median ELISA ratio, 0.21; p <
0.004, p < 0.05, p < 0.02, respectively). In MS patients, no differences in reactivity with HTLV-I were
demonstrable either in serum or in CSF when related
to disease activity or degree of disability.
As is apparent from the Figure, an ELISA ratio
higher than 2, considered as positive for the presence
of HTLV-I antibodies, was found among the sera in 2
of the 49 patients with MS, 3 of the 12 with GB, 3 of
the 30 with TH, and also in 2 of the 60 BDs, but was
found in none of the 2 1 patients with AM. The frequency among the G B patients versus BDs was
significantly higher Cp < 0.05 by Fisher's exact test),
while the differences in frequencies were otherwise
not significant. Looked at in another way, the elevation
of the ELISA ratio was very slight in 2 of the 3 GB
patients. In the CSF, an ELISA ratio higher than 2 was
demonstrated in 1 patient in each of the MS, GB, and
TH groups of patients. The ELISA ratios in the corresponding serum of these 3 patients were also the highest observed. Among these positive serum and CSF
specimens, only one serum-from a patient with
GB-revealed an ELISA ratio of greater than 5.
As a confirmatory test, Western blot analysis was
carried out on all positive (ELISA ratio, > 2) serum
and CSF samples. However, there were no detectable
antibodies that were reactive with HTLV-I proteins
p24 or p19, or with any other protein.
With the use of cell-ELISA for demonstrating
HTLV-I IgG antibodies in short-term cultures, no production of such antibodies could be detected in any of
the 6 MS patients examined, whether CSF lymphocytes or peripheral blood lymphocytes were cultivated.

Discussion
Based on the analogy between our observation that
CSF lymphocytes from patients with MS respond
poorly or not at all to stimulation with T-cell mitogens
C131 and the low mitogen responses of virus-infected
lymphocytes reported in other diseases, we proLolli et ab HTLV-I in Multiple Sclerosis 69

posed-at a time when human retroviruses were not

yet discovered-that a viral infection which primarily
affected lymphocytes might be the cause of MS {12}.
With the present knowledge about the capacity of human retroviruses, such an agent is considered a serious
candidate as an etiological factor in MS. When employing an ELJSA similar if not identical to the one
used by Koprowski and associates 1151, the results we
obtained for serum and CSF from our MS patients
agreed with those reported by these authors. However, when we extended our study to include adequate
numbers of controls, sera from patients with AM, GB,
and TH also displayed significantly increased reactions
with HTLV-I, as did CSF from patients with AM and
GB. The patients with MS in these two studies are
comparable regarding geographic origin and clinical
characteristics, and none of our patients was treated
with immunomodulatory drugs.
A confirmatory test, e.g., Western blot analysis, is
necessary to prove that the reactions obtained with
HTLV-I ELISA are specific { S , 263. Ten of our sera
and 3 CSF specimens positive in HTLV-I ELISA were
all negative by Western blot. These findings indicate
that positive results obtained by ELISA and the differences observed for sera and CSF values among
groups did not reflect the presence of HTLV-Ispecific antibodies. It has previously been shown that
false positive results for HTLV-I IgG antibodies frequently occur among bload donors when employing
the ELISA used in the present study {3,21,22). Technical factors such as the presence of antibodies reactive
with cellular components like HLA antigens in the
antigen preparation, or serum factors such as acutephase reactants or immune complexes might interfere
in the ELISA {3, 6). When testing for HIV, the situation is quite similar IS, 267.
It is questionable whether the use of other HTLV-I
antigen preparations would change the results presented in our study, since we were able to demonstrate
with our ELISA system variations in accordance and
within the same range as those reported by Koprowski
and associates { 151. Furthermore, overloading of the
plastic may increase the frequency of false positive
reactions 131.
Since the absence of specific antibodies in body
fluids does not rule out the presence of a specific immune response {19], we employed the cell-ELISA on
peripheral blood lymphocytes and CSF lymphocytes
from patients with MS to detect production of antibodies against HTLV-I. Cell-ELISA has recently been
shown to be more sensitive in demonstrating an intraBBB antibody response than determining antibodies
in CSF and serum by ELISA alone {71. However, even
cell-ELISA yielded negative results for JgG antibodies
reactive with HTLV-I.
Taken together, these data indicate that a human
70 Annals of Neurology

Vol 22

No 1 July 1987

retrovirus, if relevant in the etiology of MS, has no

demonstrable relationship to HTLV-I. Nevertheless,
the hypothesis of a lymphotropic viral infection in the
etiology of MS 112) is atitractive, and studies to isolate
a possible but unknown human retrovirus from mononuclear cells in MS are of great relevance.
This study was supported by grants from the Swedish Medical Research Council (project No. 338 1) and the Swedish Medical Society.
Dr Lolli received partial support from the AISM (Associazione
Italiana Sclerosi Multipla, sezione Toscana).

References
1. Asbury AK: Diagnostic considerations in Guillain-Bar6 syndrome. Ann Neurol 9(suppl):1-5, 1981
2. Barr6-Sinoussi F, Chermann JC, Ruy F, et al: Isolation of a Tlymphotropic retrovirus from a patient at risk for acquired immune deficiency syndrome (AIDS). Science 220:868-87 1,
1983
3. Biggard JB, Saxinger C, Gardiner C, et al: Type-I HTLV antibody in urban and rural Ghana, West Africa. Int J Cancer
34:215-219, 1984
4. Boyum A: Isolation of monoriuclear cells and granulocytes from
human blood. Scand J Clin Iab Invest 2l(suppl 97):77-79,
1968
5 . Carlson JR, Bryant ML, Hinrichs SH, et al: AIDS serology in
low and high risk groups. JAIMA 23:3405-3408, 1985
6. Essex M, McLane MF, Lee T H , et al: Antibodies to cell membrane antigens associated with human T-cell leukemia virus in
patients with AIDS. Science 220:859-862, 1983
7. Forsberg P, Kam-Hansen S, Fryden A: Production of specific
antibodies by CSF lymphocyaes in patients with herpes zoster,
mumps meningitis and HSV encephalitis. Scand J Immunol
24~261-271, 1986
8. Gallo RC, Salahuddin SZ, Popovic M, et al: Frequent detection
and isolation of cytopathic retrovirus (HTLV-111) from patients
with AIDS and at risk for AIDS. Science 224:500-503, 1984
9. Gessain A, Vernant JC, Maws L, et al: Antibodies to human
T-lymphotropic virus type I in patients with tropical spastic
paraparesis. Lancet 2:407-410, 1985
10. Hauser SL, Aubert C, Burks JS, et al: Analysis of human Tlymphotropic virus sequences in multiple sclerosis tissue. Nature 332:176-177, 1986
11. H o DD, Rota TR, Schooley R J , et al: Isolation of HTLV-111
from cerebrospinal fluid and neural tissue of patients wirh
neurologic syndromes related to the acquired immunodeficiency syndrome. N Engl J Med 24:1498-1503, 1985
12. Kam-Hansen S: Distribution and function of lymphocytes from
the cerebrospinal fluid and blood in patients with multiple sclerosis. Acta Neurol Scand 62(suppl 75 ): 1-8 1, 1980
13 Kam-Hansen S, Lnk H, Fryden A, Moller E: Reduced in vitro
response of CSF lymphocytes to mitogen in multiple sclerosis.
Scand J Immunol 10:161-169, 1979
14. Karpas S, Kampf U, Siden hi, et ;d:Lack of evidence for involvement of known human retroviruses in multiple sclerosis. Nature
332:177-178, 1986
15 Koprowski H, DeFreitas EC, Harper ME, et al: Multiple sclerosis and human T-cell lymphotrophic retroviruses. Nature 3 18:
154-160, 1985
16. Levy JA, Hollander H, Shimabumro J. et al: Isolation of AIDS
associated retroviruses from cerebrospinal fluid and brain of
patients with neurological symptoms. lancet 2:586-588, 1985
17. Link H , Kostulas V: Utility of isoelectric focusing of cerebrospi-

18.

19.

20.

21.

nal fluid and serum on agarose evaluated for neurological patients. Clin Chem 2992310-815, 1983
Narayan 0, Cork LC: Lentiviral diseases of sheep and goats:
chronic pneumonia leukoencephalomyetitis and arthritis. Rev
Infect Dis 7:89-98, 1985
Olsson T, Henriksson A, Link H: In vitro synthesis of immunoglobulins and autoantibodies by lymphocytes from various
body compartments during chronic relapsing experimental allergic encephalomyelitis. J Neuroimmunol 9:293-305, 1985
Poiesz BS, Ruschetti FW,Gatdar AF, 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 77:7415-7419, 1980
Saxinger CW, Blattner WA, Levine PH, et al: Human T-cell
leukemia virus (HTLV-I) antibodies in Africa. Science 225:
1473-1476, 1984

22. Saxinger CW, Gallo RC. Application of the indirect enzyme

linked immunosorbent assay microtest to the detection and sutveillance of human T cell leukemia-lymphoma virus. Lab Invest
3:361-377, 1983
23. Schumacher GA, Beebe G, Kibler RE, et al: Problems of experimental trials of therapy in multiple sclerosis. Ann NY Acad
Sci 122:552-568, 1965
24. Tibbling G, Link H, Ohman S : Principles of albumin and IgG
analyses in neurological disorders. I. Establishment of reference
values. Scand J Clin Lab Invest 37:385-390, 1977
25. Voller A, Bidwell DE, Bartlett A: Enzyme immunoassay in
diagnostic medicine. Theory and practice. Bull WHO 53:5565, 1976
26. Weiss SH, Goedert JJ, Sarngadharan MG, et al: Screening test
for HTLV-111 (AIDS agent) antibodies. JAMA 23:221-225,
1985

Lolli et al: HTLV-I in Multiple Sclerosis 71