Post on 26-Apr-2023
www.elsevier.com/locate/jns
Journal of the Neurological Sciences 217 (2004) 181–188
Intrathecal production of Chlamydia pneumoniae-specific high-affinity
antibodies is significantly associated to a subset of multiple
sclerosis patients with progressive forms
Enrico Fainardia,b,*, Massimiliano Castellazzia, Ilaria Casettaa, Rosario Cultrerac,Luca Vaghia, Enrico Granieria, Carlo Continic
aMultiple Sclerosis Center, Department of Neurology, University of Ferrara, Arcispedale S. Anna, Corso della Giovecca 203, Ferrara I-44100, ItalybSection of Neuroradiology, Department of Neurosciences, Azienda Ospedaliera-Universitaria, Arcispedale S. Anna, Corso della Giovecca 203,
Ferrara I-44100, ItalycSection of Infectious Diseases, Department of Clinical and Experimental Medicine, University of Ferrara, via Fossato di Mortara, 23, Ferrara I-44100, Italy
Received 15 July 2003; received in revised form 19 September 2003; accepted 24 September 2003
Abstract
The purpose of this study was to provide further insight into the effective relevance of the association between Chlamydia pneumoniae
and MS. We evaluated by ELISA technique cerebrospinal fluid (CSF) and serum levels of anti-C. pneumoniae IgG in 46 relapsing–
remitting (RR), 14 secondary progressive (SP) and 11 primary progressive (PP) MS patients grouped according to clinical and Magnetic
Resonance Imaging (MRI) evidence of disease activity. Fifty-one patients with other inflammatory neurological disorders (OIND) and 52
with non-inflammatory neurological disorders (NIND) were used as controls. A C. pneumoniae-specific intrathecal IgG synthesis as detected
by the relative specific index was present in a small proportion of MS (17%), OIND (22%) and NIND (2%) patients and was significantly
more frequent in MS and in OIND than in NIND ( p < 0.001) and in SP and PP MS than in RR MS patients ( p < 0.02). Among the patients
with C. pneumoniae-specific intratecally produced antibodies, CSF high-affinity anti-C. pneumoniae IgG were found in the majority of SP
or PP MS, occasionally in OIND, but not in RR MS and NIND patients. These findings confirm that the presence of a humoral immune
response to C. pneumoniae within the central nervous system (CNS) is not selectively restricted to MS, but is shared by several
inflammatory neurological conditions. In addition, our results suggest that an intrathecal production of C. pneumoniae-specific high-affinity
IgG can occur in a subset of patients with MS progressive forms in which a C. pneumoniae brain chronic persistent infection may play an
important pathogenetic role.
D 2003 Elsevier B.V. All rights reserved.
Keywords: Chlamydia pneumoniae; Multiple sclerosis; Cerebrospinal fluid; Intrathecal synthesis; Antibody affinity
1. Introduction Recently, a potential role as a causative agent of MS has
Multiple sclerosis (MS) is commonly considered as an
autoimmune chronic inflammatory demyelinating disease of
the central nervous system (CNS) [1]. The exposure to
environmental factors, such as a virus, in combination with
genetic predisposition could be implicated in MS pathogen-
esis [2]. However, direct evidence for an infectious etiology
in MS is still lacking [1,2].
0022-510X/$ - see front matter D 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.jns.2003.09.012
* Corresponding author. Department of Neurology, Section of
Neurology, University of Ferrara, Corso della Giovecca 203, Ferrara I-
44100, Italy. Tel.: +39-532-236304; fax: +39-532-205525.
E-mail address: henryfai@tin.it (E. Fainardi).
been suggested for Chlamydia pneumoniae, a Gram-nega-
tive, obligate intracellular bacterium that infects and repli-
cates within monocytes, macrophages and endothelial cells,
can induce a chronic infection, and presents an high
seroprevalence in adults [3]. However, the initial high rates
reported for molecular and culture demonstration of C.
pneumoniae in cerebrospinal fluid (CSF) of MS patients
[4] were not reproduced in the subsequent studies in which
CSF evidence of C. pneumoniae in MS was either less
frequent than previously described [5–9] or absent [10–
13]. In addition, whereas C. pneumoniae was not identified
in MS brain lesions [14,15], a C. pneumoniae-specific
epitope was able to induce an experimental autoimmune
E. Fainardi et al. / Journal of the Neurological Sciences 217 (2004) 181–188182
encephalomyelitis, the animal model of MS, in rats [16].
Contradictory findings have also been reported when the C.
pneumoniae-specific intrathecal humoral immune response
was evaluated. A quantitative production of anti-C. pneu-
moniae antibodies within the CNS was found in approxi-
mately 20–30% of MS patients and, to a lesser extent, in
patients with other neurological disorders [9,13,17]. CSF-
restricted oligoclonal IgG bands specific for C. pneumoniae
were detected in over 80% [18] or in none of MS patients
[13]. The assessment of the C. pneumoniae-specific anti-
body production in the brain is crucial to understand the
effective relevance of this pathogen in MS because it is
well known that, while in MS a polyspecific intrathecal
immune response composed by anti-viral low-affinity anti-
bodies occurs, in infectious diseases intrathecally synthe-
sized high-affinity antibodies, of which at least 30%
specifically directed against the causative agents, predom-
inate [19–22].
Thus, in order to clarify the significance of C. pneumo-
niae infection in MS, in the present study we addressed the
affinity distributions of intrathecally produced anti-C. pneu-
moniae antibodies in MS patients and controls.
2. Patients and methods
2.1. Patient characteristics
Seventy-one consecutive patients (51 women and 20
men; mean age = 40.3F 11.4) satisfying the criteria for
clinically or laboratory-supported definite MS according
to the classification of Poser et al. [23] and followed by
the MS Centre of the Department of Neurology, University
of Ferrara during the period from June 1997 to December
2001 were retrospectively selected for the study. Of these,
46 were relapsing–remitting (RR), 14 were secondary
progressive (SP) and 11 were primary progressive (PP)
in agreement with the criteria of Lublin and Reingold [24].
Relapse was defined as the onset of new or recurrent
symptoms or signs or the worsening of already present
neurological abnormalities persisting for at least 24 h in
the absence of fever and preceded by at least 1 month of
stable or improved neurological state [23]. Evidence of a
relapse at admission was considered as clinical disease
activity. Accordingly, at the time of enrolment in the study
39 (32 RR and 7 SP) MS were clinically active, while 21
(14 RR and 7 SP) MS were clinically stable. Given that PP
MS patients, by definition [24], were not experiencing
relapses these subjects were excluded by these two sub-
groups. Disease severity was scored in all MS patients at
the time of sample collection using Kurtzke’s Expanded
Disability Status Scale (EDSS) [25] (mean at entry =
3.2F 1.3; range from 0 to 6). The duration of the disease
was expressed in months (mean at entry = 73.1F 84.3;
range from 2 to 460). Patients treated with immunosup-
pressive (e.g. azathioprine or methylprednisolone) or im-
munomodulatory (e.g. interferon-beta or glatiramer acetate)
drugs within 6 months before entering the study were
excluded.
Fifty-one patients with other inflammatory neurological
disorders (OIND) and 52 subjects with other non-inflam-
matory neurological disorders (NIND) were selected as
neurological controls. Among the OIND patients (24 wom-
en and 27 men, mean age = 50.96F 15.99), 14 had chronic
inflammatory demyelinating polyneuropathy, 8 had viral
encephalomyelitis, 7 had acute inflammatory demyelinating
polyneuropathy, 5 had bacterial meningitis, 5 had optic
neuritis, 4 had HIV encephalopathy, 4 had neurolupus, 3
had neuroSjogren and 1 had neuroBechet. None of the five
patients with optic neuritis had CSF oligoclonal bands or
Magnetic Resonance Imaging (MRI) MS-like lesions sug-
gesting a future development of MS. Among the NIND
patients (25 women and 27 men, mean age = 47.96F 17.50)
8 had transient ischemic attack, 7 had epilepsy, 6 had
headache, 5 had cervical spondylosis, 5 had hereditary
ataxia, 5 had vascular dementia, 4 had migraine, 4 had
amyotrophic lateral sclerosis, 4 had compression neuropa-
thy, 2 had parestesias, and 2 had Alzheimer disease. All
OIND and NIND patients were not undergoing treatment
with immunosuppressant drugs, including steroids, at the
time of sampling.
2.2. Handling of CSF and serum samples
CSF samples were obtained by atraumatic lumbar punc-
ture performed with appropriate clinical indications. Both
CSF and serum samples were collected under sterile con-
ditions, coded, frozen and stored in aliquots at � 70 jCuntil the assay.
The study was approved by the Regional Committee for
Medical Ethics in Research. CSF and serum IgG and
albumin levels were measured by immunochemical nephe-
lometry with the Beckman Array Protein System (Beckman
Instruments, Fullerton, CA, USA) according to the proce-
dure of Salden et al. [26] (Salden et al., 1988). Blood-CSF-
barrier (B-CSF-B) dysfunction was determined by CSF/
serum albumin quotient (QAlb) [27].
2.3. MRI procedures
All MS patients underwent brain MRI scans at entry on a
1-Tesla MRI unit (GE Signa Horizon, General Electric
Medical Systems, Milwaukee, WI). Routinely used T1-
weighted axial spin echo images were obtained approxi-
mately 10 min after intravenous injection of 0.1 mmol/kg of
Gd-DTPA in each patient. Lesions showing Gd-enhance-
ment on T1-weighted scans were defined as active. MRI
evidence of disease activity was present in 30 (24 RR, 3 SP
and 3 PP) MS patients, whereas in 41 (22 RR, 11 SP and
8 PP) MS patients it was not detected. All brain MRI scans
were evaluated by one investigator (EF) blinded to clinical
and sample data.
Table 1
Serum and CSF detectable levels of anti-C. pneumoniae IgG in patients
with Multiple Sclerosis (MS), Other Inflammatory Neurological Disorders
(OIND) and Non-Inflammatory Neurological Disorders (NIND)
Serum CSF
Total MS (n= 71) 63/71 (88.7%)* 58/71 (81.7%)y
OIND (n= 51) 42/51 (82.3%)# 44/51 (86.3%)z
NIND (n= 52) 33/52 (63.5%) 36/52 (69.2%)
*Total MS vs. NIND ( p< 0.001).y Total MS vs. NIND ( p< 0.02).# OIND vs. NIND ( p< 0.05).z
E. Fainardi et al. / Journal of the Neurological Sciences 217 (2004) 181–188 183
2.4. Elisa assay for serum and CSF levels of anti-C.
pneumoniae IgG
As recently described [28], CSF and serum concentra-
tions of anti-C. pneumoniae specific IgG were measured by
enzyme-linked immunosorbent assay (ELISA) by using a
commercially available ELISA kit (C. pneumoniae IgG
ELISA, Bioclone Australia, Cat. No. 40CPG0096).
2.5. Calculation of C. pneumoniae-specific antibody index
According to Reiber’s formula [29] production of anti-C.
pneumoniae IgG was determined by Antibody Specific
Index (ASI). ASI means the ratio between CSF and serum
AU values (QSpec) and CSF/serum total IgG levels expressed
as mg/dl (QIgG) in accordance with the formula:
ASI ¼ QSpec=QIgG ð1Þ
with QSpec =AUCSF/AUserum and QIgG = IgGCSF/IgGserum.
In the case of intrathecal synthesis of total IgG, QIgG
appeared more elevated than Reiber’s hyperbolic discrimi-
nation line (QLim) [30]. QLim represented the completely
OIND vs. NIND ( p< 0.01).
Fig. 1. Distribution of Antibody Specific Index (ASI) abnormal values for anti-C
remitting (RR) MS, 14 secondary progressive (SP) MS, 11 primary progressive (P
Inflammatory Neurological Disorders (NIND) patients, with higher rates in Total M
(*p< 0.001), in PP MS than in RR MS (^p< 0.02) and in NIND (jp< 0.01) and
blood-derived CSF total IgG fraction calculated from the
individual QAlb of a single patient. Therefore, owing to the
introduction of this barrier-related correction of the QIgG,
the ratio became between QSpec and QLim in agreement with
the formula [31]:
ASI ¼ QSpec=QLim ð2Þ
with QLim ¼0:93
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiðQAlbÞ2 þ 6� 10�6 � 1:7� 10�3
q
Therefore, to avoid false-negative results Eq. (1) (QIgG>
QLim) was used when no significant intrathecal total IgG
synthesis occurred, while Eq. (2) (QIgG <QLim) was pre-
ferred in case of intense brain-derived total IgG production
in CSF.
C. pneumoniae-specific intrathecal IgG synthesis was
assumed for values of ASI greater than 1.5.
2.6. Determination of affinity distributions of CSF anti-C.
pneumoniae IgG
Affinity distributions of anti-C. pneumoniae IgG were
measured in CSF by performing an ELISA protocol based
on the employment of increasing concentrations of sodium
thioyanate (NaSCN). All the assays were carried out using
plates with 8 rows� 12 columns of the above-mentioned
ELISA kit (C. pneumoniae IgG ELISA, Bioclone Aus-
tralia, Cat. No. 40CPG0096). In agreement with the
previous C. pneumoniae-specific IgG measurements, the
same CSF dilutions were utilized. Briefly, 100 Al of eachCSF sample 1:2 diluted in 0.4% saline were added to all
12 wells of each the eight rows of the precoated ELISA
microplate. After 1 h incubation at 37 jC and three
washing cycles, 12 consecutive dilutions of NaSCN
(range: 0–5 M in 0.4% saline) were placed in each row
and incubated for 10 min at 37 jC. The different concen-
. pneumoniae IgG in 71 Total multiple sclerosis (Total MS), 46 relapsing
P) MS, 51 Other Inflammatory Neurological Disorders (OIND) and 52 Non
S than in NIND (jp< 0.01), in SP MS than in RR MS (^p< 0.02) and NIND
in OIND than in RRMS (#p< 0.05) and in NIND (*p< 0.001).
Table 2
Clinical features and C. pneumoniae-specific CSF affinity findings in
Multiple Sclerosis (MS) patients and controls with intrathecally synthesized
IgG anti-C. pneumoniae as indicated by the presence of antibody specific
index values >1.5
Total MS
(n= 12)
OINDa
(n= 11)
NIND
(n= 1)b
Sex (male/female) 3/9 8/3 0/1
Age, years (meanF SD) 42.17F 11.68 60F 11.87 23
Disease duration,
months (meanF SD)
108.33F 79.01
Disease severity,
EDSS (meanF SD)
4.0F 1.0
Clinical course
Relapsing–
Remitting (n)
3/12 (25%)
Secondary
progressive (n)
5/12 (45.4%)
Primary
progressive (n)
4/12 (33.3%)
Clinical activityc (n) 3/12 (25%)
MRI activityd (n) 4/12 (33.3%)
CSF RA at 5 M
NaSCNe (%)
Range 0–8.1 0–3.4 0
Median 3.15 0 0
CSF affinity ratiof
(meanF SD)
Range 0–198.2 0–77.7 0
Median 31.05 0 0
a OIND=Other Inflammatory Neurological Disorders.b NIND=Non Inflammatory Neurological Disorders.c Clinical activity = presence of relapse at entry.d MRI activity =MRI appearance of gadolium enhancing lesions.e CSF RA at 5 M NaSCN=CSF with Relative Affinity (RA) value at
the highest concentration (5.0 M) of sodium thioyanate expressed in
percentages.f CSF affinity ratio (AR) = a single numerical value which relates the
percentages of high affinity antibodies to the percentages of low affinity
antibodies in CSF containing anti-C. pneumoniae IgG.
E. Fainardi et al. / Journal of the Neurological Sciences 217 (2004) 181–188184
trations of NaSCN we applied into the 12 wells of each
row from the first to the twelfth well were: 0, 0.5, 0.75,
1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, and 5 M, where 0 M
corresponding to 0.4% saline buffer. After incubation,
NaSCN was removed and after three washing cycles wells
were blocked by adding 100 Al/wells of 3% bovine serum
albumin (BSA) for 30 min at 37 jC. After incubation,
BSA was discarded and detection antibody reaction and the
colour development were performed as described in
ELISA assay for serum and CSF levels of anti-C. pneumo-
niae IgG. Appropriate background wells (blank) were
included in each test and their optical density values were
subtracted from all readings. The affinity distribution
histograms for CSF C. pneumoniae-specific IgG were
obtained assuming that, for each CSF sample, the results
observed in the first well, in which NaSCN were absent,
represented the total antigen-specific IgG. In each row, we
divided the difference in OD values between two succes-
sive wells (e.g. well 1 and well 2) by OD values of the
total antigen-specific IgG, and multiplied by 100. In this
way, the variation in OD values between two subsequent
wells was expressed as percentage of OD values of the
total antigen-specific IgG measured in the first well. The
difference in OD values between two consecutive wells
reflected the proportion in which the bounded IgG mole-
cules were reduced by the higher concentration of NaSCN
dispensed in the two wells, and the corresponding percent-
age was considered as the Relative Affinity (RA) of
antibodies. When OD values measured in the well with
higher levels of NaSCN were slightly different compared
to those detected in the adjacent well with lower levels of
NaSCN, the RA was referred as zero. When positive OD
values were still measurable in the twelfth well, we
compared these OD values with those of a simulated
thirteenth well which was considered to be zero. Based
on this principle, we obtained an RA value showing the
proportion of antigen-specific IgG which was not removed
by the highest concentration (5.0 M) of NaSCN, and then
the highest antibody affinity. Thus, the affinity histogram
for each CSF sample was generated by plotting the relative
affinities on the x-axis and the corresponding proportion of
IgG expressed as percentages on y-axis. According to
Luxton et al. [21], CSF samples with RA value more than
2.5% from well number twelve containing the highest
concentration (5.0 M) of NaSCN were considered as
having high-affinity anti-C. pneumoniae IgG, whereas
CSF samples with RA value equal or less than 2.5% at
the highest concentration of 5 M NaSCN were regarded as
showing C. pneumoniae-specific low-affinity IgG. In ad-
dition, to increase the sensitivity in the detection of
antibody binding affinity, the affinity of CSF anti-C.
pneumoniae IgG was also defined by Affinity Ratio
(AR), a single numerical value which relates the percen-
tages of high affinity antibodies, expressed by the summa-
tion of the RA values of wells 11 and 12, to the
percentages of low affinity antibodies, represented by the
summation of the RA values of wells 1 through 4,
according to the formula [21]:
AR ¼
XRA11�12
XRA1�4
þ0:1
8>>>><>>>>:
9>>>>=>>>>;
� 100:
2.7. Statistics
Mann–Whitney U-test was used to compare mean values
among the various groups. The patient group percentages
were compared using the Log-Likelihood Ratio Test (G).
The Spearman rank correlation coefficient test was used to
identify possible relationships among the different variables
in the MS group. A value of p < 0.05 was accepted as
statistically significant.
E. Fainardi et al. / Journal of the Neurological Sciences 217 (2004) 181–188 185
3. Results
3.1. Serum and CSF levels and intrathecal synthesis of anti-
C. pneumoniae IgG in MS patients and controls
Detectable serum and CSF levels of anti-C. pneumoniae
IgG were statistically more frequent in total MS
( p < 0.001 and p < 0.02, respectively) and in OIND
( p < 0.05 and p < 0.01, respectively) than in NIND
patients, without any significant differences between total
MS and OIND (Table 1). In contrast, there were no
significant differences for serum and CSF anti-C. pneumo-
niae IgG mean concentrations among total MS, OIND and
NIND groups also when MS patients were evaluated
according to clinical course, and clinical and MRI activity
(data not shown).
Fig. 2. Panel A describes the distribution CSF high affinity IgG to C. pneumon
sclerosis (Total MS), 3 relapsing remitting (RR) MS, 5 secondary progressive
Neurological Disorders (OIND) with ASI values >1.5. CSF high affinity anti-C. pn
SP (*p< 0.02) and PP (jp< 0.05) MS than in RR MS. Panel B reports CSF mean
relates the percentages of high affinity antibodies to the percentages of low affinity
of MS patients and controls. CSF Affinity Ratio mean levels are higher (*p< 0.05)
represents the 25th–75th quartile divided by the median, the whiskers the range.
A C. pneumoniae-specific intrathecal IgG synthesis as
suggested by the presence of ASI abnormal values was
significantly more represented ( p< 0.001) in total MS (12/
71; 16.9%) and in OIND (11/51; 21.6%) than in NIND (1/
52; 1.9%) (Fig. 1). When MS patients were stratified
according to clinical types, and clinical and MRI activity,
ASI values greater than 1.5, indicating an intrathecal
production of anti-C. pneumoniae IgG were significantly
more frequent in SP and PP ( p < 0.02) than in RR MS, in
SP ( p < 0.001) and in PP MS ( p < 0.01) than in NIND and
in OIND ( p < 0.05) than RR MS, with no statistical differ-
ences between SP and PP MS (Fig. 1). No significant
discrepancies were observed for intrathecal synthesis of
anti-C. pneumoniae IgG between clinically active and
clinically stable MS as well as between MS with and
without MRI appearance of disease activity (data not
iae, as reflected by RA value at 5 M NaSCN>2.5%, in 12 Total multiple
(SP) MS, 4 primary progressive (PP) MS and 11 Other Inflammatory
eumoniae IgG is more frequent in Total MS than in OIND (jp< 0.05), and invalues of CSF Affinity Ratio, considered as a single numerical value which
antibodies in CSF containing anti-C. pneumoniae IgG, in the same subsets
in total MS than in OIND, and in SP and PP MS than in RR MS. The box
E. Fainardi et al. / Journal of the Neurolo186
shown). With regard to OIND and NIND, there were no
significant differences for serum and CSF anti-C. pneumo-
niae IgG detectable and mean concentrations as well as for
C. pneumoniae-specific intrathecal IgG synthesis among the
different clinical entities included in these two groups (data
not shown).
3.2. Affinity distributions of intrathecally synthesized anti-
C. pneumoniae IgG
To test the strength of binding of C. pneumoniae-specific
intrathecally produced antibodies, affinity distributions of
IgG directed against C. pneumoniae were studied in all 24
CSF samples (12 MS, 11 OIND and 1 NIND) with an
intrathecal production of anti-C. pneumoniae IgG (Table 2)
and in 12 CSF samples (9 MS, 2 OIND and 1 NIND)
without evidence of C. pneumoniae-specific intrathecal IgG
synthesis in which levels of anti-C. pneumoniae IgG were
detectable in CSF. We also tested, as controls, 3 CSF
samples (1 MS, 1 OIND and 1 NIND) without measurable
CSF levels of anti-C. pneumoniae IgG.
As illustrated in Fig. 2, panel A, in patients with ASI
values >1.5, CSF high-affinity IgG to C. pneumoniae, as
indicated by RA value at 5 M NaSCN>2.5%, was detected
more frequently in total MS (7/12) than in OIND (2/11) and
in NIND (0/1) patients, with a significant difference
( p < 0.05) between total MS and OIND. The two OIND
patients with CSF high-affinity anti-C. pneumoniae IgG
showed one a bacterial meningitis, and the other an acute
inflammatory demyelinating polyneuropathy. As affinity
measurements were performed in only one NIND patient,
this group was not statistically compared with total MS and
OIND. Among the different MS clinical courses, an RA
value at 5 M NaSCN>2.5% was present in the majority of
SP (4/5) and PP (3/4) MS and in none of RR (0/3) MS
patients. Consequently, CSF high-affinity anti-C. pneumo-
niae IgG were significantly more frequent in SP MS
( p < 0.02) and PP MS ( p < 0.05) than in RR MS. Similar
proportions of CSF high-affinity anti-C. pneumoniae IgG
were found in clinically active (2/3; 66.3%) and in clini-
cally stable (2/5; 40%) MS, in MS patients with Gd
enhancing lesions (2/4; 50%) and in those without MRI
focal activity (5/8; 62.5%). Accordingly, CSF Affinity Ratio
mean values were higher ( p < 0.05) in total MS than in
OIND, and in SP and PP MS than in RR MS (Fig. 2, panel
B), with no significant differences among the remaining
groups evaluated.
All patients with ASI values < 1.5 and with detectable C.
pneumoniae-specific IgG levels in CSF, showed CSF low-
affinity IgG to C. pneumoniae, as suggested by RA value at
5 M NaSCN< 2.5%, and comparable CSF Affinity Ratio
mean values (data not shown).
No antibodies with affinity for C. pneumoniae were
demonstrated in CSF of the three controls without measur-
able CSF levels of anti-C. pneumoniae IgG (data not
shown).
3.3. Correlation between CSF and serum C. pneumoniae-
specific humoral immune response and MS clinical features
In MS patients, CSF and serum levels and intrathecal
synthesis of anti-C. pneumoniae IgG, and affinity distribu-
tions of intrathecally synthesized anti-C. pneumoniae IgG
were not significantly correlated to severity, as expressed by
EDSS, and duration of the disease (data not shown).
gical Sciences 217 (2004) 181–188
4. Discussion
We investigated whether the appearance of an intrathecal
production of anti-C. pneumoniae IgG in MS represents a
manifestation of a C. pneumoniae-specific brain infection.
In our study, an intrathecal release of anti-C. pneumoniae
IgG was found in a small proportion of MS (17%), OIND
(22%) and NIND (2%) patients and, as opposed to previous
reports [10,13,17], it was significantly more prevalent in MS
and OIND than in NIND and in SP and PP MS than in RR
MS. In agreement with prior studies [9,13,17], MS patients
with and without relapse as well as with and without Gd
enhancing lesions produced intrathecally the same amount
of anti-C. pneumoniae IgG. In contrast with Sriram et al. [4]
but according to several other publications [5,9–13,17,32],
no significant differences were detected for CSF and serum
mean concentrations of anti-C. pneumoniae IgG among MS,
OIND and NIND. As previously reported [17], we identified
no statistical correlation between CSF and serum levels and
intrathecal synthesis of anti-C. pneumoniae IgG and severity
or duration of the disease. Discrepancies between our
findings and those obtained by other groups may be due
to the use of different procedures during the assay and to the
existence of distinct study designs.
Overall, these findings do not support a central role for
C. pneumoniae in the pathogenesis of all MS patients and
confirm that a C. pneumoniae-specific intrathecal humoral
immune response can be detected in a subgroup of MS and
OIND patients [9,13,17,32], strengthening the assumption
that the presence of C. pneumoniae within the CNS is not
selectively associated with MS, but may be linked to a large
number of inflammatory neurological conditions [8,9,32]. In
addition, the lack of differences observed in our report for
intrathecal synthesis of anti-C. pneumoniae IgG between
MS and OIND do not corroborate the recent hypothesis that
the clearance of C. pneumoniae from CSF could be closely
reduced in MS [8,9,32].
Of particular relevance was the observation that in MS a
C. pneumoniae-specific intrathecal IgG synthesis was pre-
ponderant in a subset of patients with MS progressive
forms. Although SP and PP MS differ in clinical, genetic,
immunological, pathological and radiological features
[33,34], the strong intrathecal production of anti-C. pneumo-
niae IgG we found in a part of both these clinical types of
MS was not unexpected since SP and PP MS share not only
some immunological patterns [34] but also similar mecha-
E. Fainardi et al. / Journal of the Neurological Sciences 217 (2004) 181–188 187
nisms leading to progression [33]. Furthermore, this result
was in agreement with the frequent culture demonstration of
C. pneumoniae previously described in CSF of progressive
MS [4]. Given that in progressive forms of MS there appears
a chronic immune activation sustained by a continual
antigen presentation within the CNS [35], an intrathecal
synthesis of C. pneumoniae-specific IgG may mirror a
persistent infection which promotes an autoimmune re-
sponse within the CNS [2,35].
In the present report we assessed, for the first time, the
affinity distributions of CSF anti-C. pneumoniae IgG in a
subgroup of patients with C. pneumoniae-specific intrathe-
cally produced antibodies. In the course of an infection, the
process of immune maturation requires a progressive in-
crease in antibody affinity that leads to an intensive pro-
duction of high-affinity antibodies specifically directed
against the causative pathogen [36]. In addition, neurolog-
ical infectious diseases are characterized by the presence of
CSF antibodies with high-affinity against the causative
agent [21]. Our principal finding was that, among the
patients with C. pneumoniae-specific intratecally produced
antibodies, CSF high-affinity anti-C. pneumoniae IgG were
found in 7 out of 12 (58.3%) MS patients, all belonging to
SP or PP MS clinical types, and in 2 out of 11 (18.2%)
OIND subjects. Accordingly, AR mean values were more
consistent in total MS than in OIND as well as in SP and PP
MS than in RR MS. These results raise the possibility that
an intrathecal production of C. pneumoniae-specific high-
affinity antibodies, suggesting a C. pneumoniae brain chron-
ic persistent infection, is present in a subset of patients with
MS progressive forms and, more occasionally, in neuro-
inflammatory diseases which have been related to C. pneu-
moniae infection [3].
A contribution of C. pneumoniae to progressive disabil-
ity in MS, however, is not supported by the lack of a
significant correlation between CSF high-affinity anti-C.
pneumoniae IgG and EDSS. This study does not clarify
whether the presence of intrathecal high-affinity anti-C.
pneumoniae antibodies reflects a continual antigenic stimu-
lation sustained by a specific brain chronic persistent
infection or whether it is due to the reactivation of C.
pneumoniae in macrophages and endothelial cells promoted
by the overactive immune response operating in progressive
forms of MS [35]. A causal association between CSF high-
affinity specific-antibodies and CNS infection has been
demonstrated only in acute and not in chronic encephalitis
[21]. In addition, we did not evaluate ASI and CSF affinity
distributions of specific-IgG for other infectious agents,
such as measles, rubella, varicella zoster, frequently
detected in CSF of MS patients, thus questioning the
specificity of C. pneumoniae antibodies we found in CSF
for progressive MS.
The hypothesis suggested in this study requires more
extensive research and a larger number of patients before
definite conclusions can be drawn on the role of C. pneumo-
niae in MS.
Acknowledgements
The study was supported by Fondazione Cassa di
Risparmio di Ferrara, Italy. We thank Dr. Tiziana Bellini
and Professor Franco Dallocchio of the Biochemistry
Department of University of Ferrara and Dr. Roberto Furlan
of Neuroimmunology Unit, Department of Neurology, San
Raffaele Scientific Institute of Milan for their insightful
discussions and thoughtful suggestions in the development
of ELISA method to measure antibody relative affinity in
CSF.
References
[1] Noseworthy JH, Lucchinetti C, Rodriguez M, Weinshenker BG. Mul-
tiple sclerosis. N Engl J Med 2000;343:938–52.
[2] Meinl E. Concepts of viral pathogenesis of multiple sclerosis. Curr
Opin Neurol 1999;12:303–7.
[3] Yucesan C, Sriram S. Chlamydia pneumoniae infection of the central
nervous system. Curr Opin Neurol 2001;14:355–9.
[4] Sriram S, Stratton CW, Yao SY, Tharp A, Ding L, Bannan JD, et al.
Chlamydia pneumoniae infection of the central nervous system in
multiple sclerosis. Ann Neurol 1999;46:6–14.
[5] Layh-Schmitt G, Bend C, Hildt U, Dong-Si T, Juttler E, Schnitzler P,
et al. Evidence for infection with Chlamydia pneumoniae in a sub-
group of patients with multiple sclerosis. Ann Neurol 2000;47:652–5.
[6] Treib J, Haaß A, Stille W, Maass M, Stephan C, Holzer G, et al.
Multiple sclerosis and Chlamydia pneumoniae. Ann Neurol 2000;
47:408.
[7] Ikejima H, Haranaga S, Takemura H, Kamo T, Takahashi Y, Friedman
H, et al. PCR-based method for isolation and detection of Chlamydia
pneumoniae DNA in cerebrospinal fluid. Clin Diagn Lab Immunol
2001;8:499–502.
[8] Gieffers J, Poh D, Treib J, Dittmann R, Stephan C, Klotz K, et al.
Presence of Chlamydia pneumoniae DNA in the cerebral spinal fluid
is a common phenomenon in a variety of neurological diseases and
not restricted to multiple sclerosis. Ann Neurol 2001;49:585–9.
[9] Hao Q, Miyashita N, Matsui M, Wang HY, Matsushima T, Saida T.
Chlamydia pneumoniae infection associated with enhanced MRI spi-
nal lesions in multiple sclerosis. Mult Scler 2002;8:436–40.
[10] Boman J, Roblin PM, Sundstrom P, Sandstrom M, Hammerschlag
MR. Failure to detect Chlamydia pneumoniae in the central nervous
system of patients with MS. Neurology 2000;54:265.
[11] Pucci E, Taus C, Carthechini E, Morelli M, Giuliani G, Clementi M,
et al. Lack of Chlamydia infection of the central nervous system in
multiple sclerosis. Ann Neurol 2000;48:340–99.
[12] Saiz A, Marcos MA, Graus F, Vidal J, Jimenez de Anta MT. No
evidence of CNS infection with Chlamydia pneumoniae in patients
with multiple sclerosis. J Neurol 2001;248:617–8.
[13] Derfuss T, Gurkov R, Bergh FT, Goebels N, Hartmann M, Barz C, et
al. Intrathecal antibody production against Chlamydia pneumoniae in
multiple sclerosis is part of a polyspecific immune response. Brain
2001;124:1325–35.
[14] Hammerschlag MR, Ke Z, Lu F, Roblin P, Boman J, Kalman B. Is
Chlamydia pneumoniae present in brain lesions of patients with multi-
ple sclerosis? J Clin Microbiol 2000;38:4274–6.
[15] Morre SA, De Groot CJA, Killestein J, Meijer CJLM, Polman CH,
Van der Valk P, et al. Is Chlamydia pneumoniae present in the central
nervous system of multiple sclerosis patients? Ann Neurol 2000;
48:399.
[16] Lenz DC, Lu L, Conant SB, Wolf NA, Gerard HC, Whittum-Hudson
JA, et al. A Chlamydia pneumoniae-specific peptide induces experi-
mental autoimmune encephalomyelitis in rats. J Immunol 2001;
167:1803–8.
E. Fainardi et al. / Journal of the Neurological Sciences 217 (2004) 181–188188
[17] Krametter D, Niederwieser G, Berghold A, Birnbaum G, Strasser-
Fuchs S, Hartung HP, et al. Chlamydia pneumoniae in multiple scle-
rosis: humoral immune responses in serum and cerebrospinal fluid
and correlation with disease activity marker. Mult Scler 2001;7:13–8.
[18] Yao SY, Stratton CW, Mitchell WM, Sriram S. CSF oligoclonal bands
in MS include antibodies against Chlamydiophila antigens. Neurol-
ogy 2001;56:1168–76.
[19] Sindic CJM, Monteyne P, Laterre EC. The intrathecal synthesis of
virus-specific oligoclonal IgG in multiple sclerosis. J Neuroimmunol
1994;54:75–80.
[20] Conrad AJ, Chiang EY, Andeen LE, Avolio C, Walker SM, Baum-
hefner RW, et al. Quantitation of intrathecal measles virus IgG anti-
body synthesis rate: subacute sclerosing panencephalitis and multiple
sclerosis. J Neuroimmunol 1994;54:99–108.
[21] Luxton RW, Zeman A, Holzel H, Harvey P, Wilson J, Kocen R, et al.
Affinity of antigen-specific IgG distinguishes multiple sclerosis from
encephalitis. J Neurol Sci 1995;132:11–9.
[22] Reiber H, Ungefehr S, Jacobi C. The intrathecal, polyspecific and
oligoclonal immune response in multiple sclerosis. Mult Scler 1998;
4:111–7.
[23] Poser CM, Paty DW, Scheimberger L, McDonald WI, Davis FA,
Ebers GC, et al. New diagnostic criteria for MS: guidelines for re-
search protocols. Ann Neurol 1983;13:227–31.
[24] Lublin DF, Reingold SC. Defining the clinical course of multiple
sclerosis: results of an international survey. Neurology 1996;46:
907–11.
[25] Kurtzke JF. Rating neurological impairment in multiple sclerosis: an
expanded disability scale (EDSS). Neurology 1983;33:1444–52.
[26] Salden HJM, Bas BM, Hermas JTH, Janson PCV. Analytical perform-
ance of three commercially available nephelomethers compared quan-
tifying protein in serum and cerebrospinal fluid. Clin Chem 1988;
34:1594–6.
[27] Tibbling G, Link H, Ohman S. Principles of albumin and IgG anal-
yses in neurological disorders: I. Establishment of reference value.
Scand J Clin Lab Invest 1977;37:385–90.
[28] Contini C, Fainardi E, Seraceni S, Granieri E, Castellazzi M, Cultrera
R. Molecular identification and antibody testing of Chlamydiophila
pneumoniae in a subgroup of patients with HIV-associated dementia
complex preliminary results. J Neuroimmunol 2003;136:172–7.
[29] Reiber H, Lange P. Quantification of virus-specific antibodies in cere-
brospinal fluid and serum: sensitive and specific detection of antibody
synthesis in brain. Clin Chem 1991;37:1153–60.
[30] Reiber H, Felgenhauer K. Protein transfer at the blood cerebrospinal
fluid barrier and the quantitation of the humoral immune response
within the central nervous system. Clin Chem Acta 1987;163:319–28.
[31] Reiber H. Evaluation of blood cerebrospinal fluid barrier function and
quantification of the humoral immune response within the central
nervous system. In: Thompson EJ, Trojano M, Livrea P, editors.
Cerebrospinal fluid analysis in multiple sclerosis. Milano: Springer;
1996. p. 51–72.
[32] Sotgiu S, Piana A, Pugliatti M, Sotgiu A, Deiana GA, Sgaramella E,
et al. Chlamydia pneumoniae in the cerebrospinal fluid of patients
with multiple sclerosis and neurological controls. Mult Scler 2001;
7:371–4.
[33] Thompson AJ, Polman CH, Miller DH, Mc Donald WI, Brochet B,
Filippi M, et al. Primary progressive multiple sclerosis. Brain 1997;
120:1085–96.
[34] McDonnell GV, Hawkins SA. Primary progressive multiple sclerosis:
increasing clarity but many unanswered questions. J Neurol Sci 2002;
199:1–15.
[35] Hafler DA. The distinction blurs between an autoimmune versus
microbial hypothesis in multiple sclerosis. J Clin Invest 1999;104:
527–9.
[36] Luxton RW, Thompson EJ. Affinity distribution of antigen-specific
IgG in patients with multiple sclerosis and in patients with viral
encephalitis. J Immunol Methods 1990;131:277–82.