Elevated Levels of Carbonyl Proteins in Cerebrospinal Fluid of Patients with Neurodegenerative Diseases

Abstract

The importance of oxidative stress in the pathogenesis of neuroimmunological and neurodegenerative diseases, such as multiple sclerosis (MS), has been discussed for a long time. However, markers for oxidative stress in cerebrospinal fluid are hardly detected. The aim of the present study is to assess whether carbonyl proteins as end products of metabolic processes may serve as a marker for oxidative stress in the cerebrospinal fluid (CSF) of patients with neuroimmunological and neurodegenerative diseases. Levels of carbonyl proteins in the CSF were assessed in 15 patients suffering from MS, four patients with neurodegenerative diseases, including one patient with dementia complicated by carcinomatous meningitis due to breast cancer, and four control subjects with no established neurological disease. Levels of carbonyl proteins were measured with a commercially available KIT. A significant difference (P = 0.025) was shown for mean values of various subgroups with highest levels for patients with neurodegenerative diseases (756.1 pmol/mg), followed by the MS (630.8 pmol/mg) and the control group (356.5 pmol/mg). Post-hoc pair wise comparisons showed significant differences between the MS group and healthy controls (P = 0.016) as well as for patients with neurodegenerative diseases and healthy controls (P = 0.02). This pilot trial showed that carbonyl proteins might serve as measure for oxidative stress in the CSF of relapsing as well as progressive MS patients and in patients with neurodegenerative diseases. Larger trials have to show whether they may serve as biomarkers and be helpful in monitoring patients with MS or neurodegenerative diseases.

Full text

Carbonyl Proteins in Neurological Diseases 313Tohoku J. Exp. Med., 2014, 234, 313-317
313
Received August 11, 2014; revised and accepted December 1, 2014. Published online December 18, 2014; doi: 10.1620/tjem.234.313.
Correspondence: Joachim Greilberger, Institute for Physiological Chemistry, Medical University Graz, Harrachgasse 21/II, 8010 Graz,
Austria.
e-mail: joachim.greilberger@medunigraz.at
Elevated Levels of Carbonyl Proteins in Cerebrospinal Fluid of
Patients with Neurodegenerative Diseases
Paulus S Rommer,1 Joachim Greilberger,2 Sabine Salhofer-Polanyi,1 Eduard Auff,1
Fritz Leutmezer1 and Ralf Herwig3
1Department of Neurology, Medical University of Vienna, Vienna, Austria
2Institute for Physiological Chemistry, Medical University Graz, Graz, Austria
3Department of Urology, Medical University of Vienna, Vienna, Austria
The importance of oxidative stress in the pathogenesis of neuroimmunological and neurodegenerative
diseases, such as multiple sclerosis (MS), has been discussed for a long time. However, markers for
oxidative stress in cerebrospinal fluid are hardly detected. The aim of the present study is to assess
whether carbonyl proteins as end products of metabolic processes may serve as a marker for oxidative
stress in the cerebrospinal fluid (CSF) of patients with neuroimmunological and neurodegenerative
diseases. Levels of carbonyl proteins in the CSF were assessed in 15 patients suffering from MS, four
patients with neurodegenerative diseases, including one patient with dementia complicated by
carcinomatous meningitis due to breast cancer, and four control subjects with no established neurological
disease. Levels of carbonyl proteins were measured with a commercially available KIT. A significant
difference (P = 0.025) was shown for mean values of various subgroups with highest levels for patients with
neurodegenerative diseases (756.1 pmol/mg), followed by the MS (630.8 pmol/mg) and the control group
(356.5 pmol/mg). Post-hoc pair wise comparisons showed signicant differences between the MS group
and healthy controls (P = 0.016) as well as for patients with neurodegenerative diseases and healthy
controls (P = 0.02). This pilot trial showed that carbonyl proteins might serve as measure for oxidative
stress in the CSF of relapsing as well as progressive MS patients and in patients with neurodegenerative
diseases. Larger trials have to show whether they may serve as biomarkers and be helpful in monitoring
patients with MS or neurodegenerative diseases.
Keywords: carbonyl proteins; cerebrospinal uid; multiple sclerosis; neurological disorders; oxidative stress
Tohoku J. Exp. Med., 2014 December, 234 (4), 313-317. © 2014 Tohoku University Medical Press
Introduction
Multiple sclerosis (MS) is the most common chronic
neuroimmunological disease in young adults. Hitherto the
pathogenesis of MS is not fully understood; an autoimmune
process seems to be the underlying cause (Weinshenker
1998; Weiner 2009). Diagnosis of clinically denite MS
(CDMS) is established after new symptoms or progression
of disability have occurred after a rst clinical event indica-
tive of MS (clinically isolated syndrome [CIS]) (Polman et
al. 2011). MS may proceed in attacks (relapses) with full or
limited remission (relapsing remitting MS [RRMS]), or it
may be characterized by steady worsening of neurological
functions either from disease onset (primary progressive
MS [PPMS]) or after an initial relapsing remitting course
(secondary progressive MS [SPMS]) (Lublin and Reingold
1996). At an early stage, the progression of disease is asso-
ciated with inammatory processes, whereas neurodegener-
ative processes become more important in the later stages
of MS (Lassmann et al. 2012).
The importance of oxidative stress (OS) in the etiology
of immunological and degenerative diseases has been dis-
cussed for a long time (Perry et al. 2002; Greilberger et al.
2008; Drechsel et al. 2012). In MS, for instance, the link-
age between inflammation and OS is well established.
Additionally, neurodegeneration seems to be driven by oxi-
dative injury and mitochondrial dysfunction (Lassmann et
al. 2012).
An imbalance between antioxidants and radical oxy-
gen species (ROS) leads to the development of OS
(Gonsette 2008). ROS cause damage to cells by oxidizing
proteins, lipids and DNA (Lassmann et al. 2012). Amino
acids are modied, resulting in the accumulation of car-
bonyl- and hydroxyl groups. Higher levels of nitric oxide
metabolites could be detected in patients suffering from MS
(Rejdak et al. 2004). Lipid peroxidation products and glu-

P. S Rommer et al.
314
tathione peroxidase activity were observed in the serum of
RRMS patients (Ortiz et al. 2009). Increased levels of sulf-
hydryl groups, transferrin, DJ-1 and transthyretin isoforms
mirror OS and were assessed in the serum of MS patients
(Calabrese et al. 1994, 1998; Zeman et al. 2000; Hirotani et
al. 2008; Poulsen et al. 2012).
Carbonyl proteins (CPs) are the end product of meta-
bolic processes (Greilberger et al. 2010; Pennisi et al. 2011)
and recent trials suggested CPs as a marker for OS in the
plasma of patients with degenerative diseases like
Alzheimer’s disease (Greilberger et al. 2010).
Markers for OS in the cerebrospinal uid (CSF) were
hardly detected and pose a substantial challenge. Nitrit
oxide metabolites were assessed in the CSF of MS patients
by Rejdak et al. (2008). Whereas CPs have recently been
assessed in the CSF of relapsing MS patients (Pennisi et al.
2011), the relevance in the CSF of patients with diseases
characterized by neurodegeneration remains unexplored.
The aim of this proof of concept trial is to explore CPs in
the CSF of neurodegenerative diseases, and whether they
may serve as a marker for OS in the CSF of patients with
neurodegenerative diseases.
Methods and Patients
The study was approved by the institutional ethics committee
(Medical University of Vienna, Vienna, Austria; EK-NR 588/2009).
All patients gave their written and informed consent.
Subgroups were analyzed by Kruskal-Wallis-Test. Kolmo-
gorov-Smirnov-test was performed for checking distribution in
various subgroups. In the case of signicant results post-hoc pair
wise comparisons were performed by Mann-Whitney-U test.
Statistics were performed with software IBM® SPSS® Statistics
Version 20 (Armonk, U.S.A.).
All lumbar punctures were performed after excluding increased
intracranial pressure and were performed for diagnostic purposes
only. In all cases atraumatic “Sprotte” cannulas (Pajunk Medizin-
technologie GmbH, Geisingen, Germany) were used. 200 µl of CSF
were collected for further analysis. CSF was immediately held on ice
and stored at −80°C. EDTA blood was collected; plasma was sepa-
rated by centrifugation and stored at −80°C. CP content was mea-
sured with the Carbonyl Protein ELISA Kit (Immundiagnostik AG,
Germany) as originally described (Greilberger et al. 2008, 2010).
Each sample (4 μL) was reacted with 80 μL di-nitro-phenylhydrazine
(DNPH) for 45 min at room temperature. DNPH-derivatised proteins
were separated by centrifugation from non-protein constituents and
unconjugated DNPH reagent using 5,000 Da molecular weight cut-
off lters (Amicon ultra-centrifugal lter, Merck, Vienna, Austria).
One part of the diluted proteins is adsorbed onto an ELISA plate and
incubated with anti-DNPH antibody followed by antibody-linked
horseradish peroxidase. The absorbance of the carbonylated protein
content of the samples is related to a standard curve prepared with
oxidized human serum albumin. The carbonyl protein content is cal-
culated from the estimated carbonyl concentration and the total pro-
tein content of the sample. In parallel, a protein determination of the
centrifuged samples was measured using a PIERCE 660 nm protein
assay (Pierce Biotechnology, Rockford, IL).
Results
In total, 23 patients were included in this study. EDTA
blood was collected from 16 patients, whilst CSF was
assessed in all 23 patients. List of patients see Table 1. Out
of the 23 enrolled patients, 15 patients suffered from MS
(either CDMS or CIS), four patients suffered from neurode-
generative diseases. In one of the latter four patients,
dementia was initially diagnosed, but after careful diagnos-
tic work-up in the same patient, carcinomatous meningitis
due to breast cancer was nally diagnosed. In remaining
four patients, lumbar puncture was performed for ruling out
neurological diseases like subarachnoid hemorrhage.
The mean value of CPs in EDTA blood was 395.1
pmol/mg in all patients. The mean blood levels of CPs did
not show any correlation with clinical parameters: 368.8
pmol/mg for CDMS patients, 278.6 pmol/mg for CIS
patients, 366.4 pmol/mg for patients with neurodegenera-
tive diseases, and 389.3 pmol/mg for those without estab-
lished neurological disease. Thus, only levels of CPs in the
CSF are presented.
Average levels of CPs in the CSF of patients with MS
(CIS or CDMS) were 596.1 pmol/mg; 756.1 pmol/mg in
patients with neurodegenerative diseases, and 356.5 pmol/
mg for the healthy controls (without established neurologi-
cal disease). Kolmogorov-Smirnov-test revealed not nor-
mally distributed samples in the various subgroups.
Kruskal-Wallis-test showed signicant differences between
subgroups (P = 0.025); post-hoc pair wise comparisons
showed signicant differences between the MS group and
the healthy controls (P = 0.016) as well as between the neu-
rodegenerative group and the healthy controls (P = 0.020).
Further analyses showed that 12 patients suffered from
CDMS according the revised McDonald’s criteria (Polman
et al. 2011). 8 patients suffered from RRMS, 3 from SPMS
and 1 one from PPMS; mean values of CPs were 630.8
pmol/mg (SD 245.1) for all patients, 623.2 pmol/mg (SD
276) for RRMS patients, and 646 pmol/mg (SD 203) for
progressive MS (SPMS and PPMS) patients with no signi-
cant differences between both groups (P = 0.495).
Interestingly, two patients received natalizumab prior to
lumbar puncture and showed lower levels (364.6 pmol/mg
and 475 pmol/mg, respectively) than the average. 3 patients
suffered from CIS: mean value was 457.2 pmol/mg. One of
those patients developed CDMS (657.1 pmol/mg), whereas
the other 2 patients did not develop MS within follow up of
two years (317.9 pmol/mg and 396.7 pmol/mg, respec-
tively).
When analyzing CDMS patients and CIS patients sep-
arately Kruskal-Wallis-test showed signicant differences
between various groups (P = 0.024). Post-hoc pair wise
comparisons between subgroups showed signicant differ-
ences between the CDMS group and healthy controls (P =
0.034), but not for CIS and healthy controls (P = 0.289).
Kolmogorov-Smirnov-test revealed no normal distribution
for the various samples. Fig. 1 shows the results in detail.

Carbonyl Proteins in Neurological Diseases 315
Table 1. List of patients.
patient number sex age Disease Level of CPs
pmol/mg
1 f 21 RRMS acute relapse 513.2
2 f 41 RRMS 384.4
3 f 30 healthy 379.2
4 f 49 progressive MS 478.3
5 m 27 RRMS 430.7
6 m 70 ALS 719.8
7 m 66 Lewy-Body dementia 641.5
8 f 69 dementia carcinomatous meningitis 1,064.2
9 m 32 CIS 657.1
10 m 34 healthy 384
11 f 48 progressive MS 636.3
12 f 27 CIS 317.9
13 f 32 RRMS 364.4
14 f 47 progressive MS 534.9
15 m 41 progressive MS 934.4
16 f 34 RRMS acute relapse 796.7
17 f 35 RRMS 1,083.5
18 f 64 ALS 599.1
19 f 63 healthy 296.7
20 f 25 CIS 396.7
21 f 31 healthy 366
22 m 27 RRMS 937.7
23 f 38 RRMS 475
CIS, clinically isolated syndrome; RRMS, relapsing remitting multiple sclerosis; PPMS, primary progressive
multiple sclerosis; SPMS, secondary progressive multiple sclerosis; ALS, amytrophic lateral sclerosis.
group of patients
Fig. 1. Levels of carbonyl proteins in the CSF of patient groups and controls.
Shown are the CSF levels of carbonyl proteins (pmol/mg) in patients with clinically de nite MS (CDMS), clinically
isolated syndrome (CIS) or other neurodegenerative diseases (NDD) and in healthy controls (HC). The levels of CPs in
the CSF were highest in NDD, followed by patients with CDMS, CIS, and HC. The respective mean values are 756.1
pmol/mg (standard deviation: SD 211.3), 630.8 pmol/mg (SD 245.1), 457.2 pmol/mg (SD 177.5), and 356.5 pmol/mg
(SD 40.6). The levels differed significantly (P = 0.025) between the subgroups. Post-hoc pair wise comparisons
revealed signi cant differences between the CDMS and healthy controls (P = 0.034), as well as for NDD and healthy
control (P = 0.020), but not for CIS and healthy control (P = 0.289). Levels of CPs in CDMS and NDD did not differ
signi cantly (P = 0.395).

P. S Rommer et al.
316
Four patients suffered from neurodegenerative dis-
eases (2 patients with amyotrophic lateral sclerosis [ALS],
1 patient with Lewy-Body dementia, 1 patient with unspeci-
fied dementia). The fourth patient also suffered from
meningeosis carcinomatosa due to breast cancer. All over
mean value was 756.1 pmol/mg (SD 211.3). In healthy
controls mean value of CPs was 356.5 /mg (SD 40.6).
Discussion
The results of this proof of concept study showed dif-
ferences in the levels of CPs in patients with CDMS
(relapsing and progressive MS) compared to healthy con-
trols. These results suggest that CPs may serve as marker
for OS in the CSF of relapsing and progressive MS patients.
The signicance of the results in MS patients are underlined
by the levels of CPs in patients suffering from neurodegen-
erative diseases (ALS and Lewy-Body dementia) and from
malignancies (meningeosis carcinomatosa) with signicant
higher levels of CPs in the CSF than in controls with no
established neuroimmunological or neurodegenerative dis-
orders.
High levels of CPs in immunological and degenerative
diseases are explained by the fact that inammation as well
as neurodegeneration leads to OS. This is in accordance
with the ndings of Petzold (2005) who found higher levels
of neurolament heavy chains phosphoforms (NfH) in neu-
rodegenerative and neuroimmunological diseases. In pro-
gressive MS NfH seem to be a predictor of axonal damage
(Petzold 2005). Whereas higher levels of CPs in the CSF
are associated with inammatory or degenerative diseases,
this does not hold for levels in the EDTA blood. This is in
contrast to the ndings of Pennisi et al. (2011) who pre-
sented signicant differences in the plasma of MS patients
when compared to controls. Whereas Pennisi et al. (2011)
investigated CPs only in RRMS patients, we assessed the
levels in relapsing and progressive MS patients as well as
in patients with neurodegenerative diseases. In relapsing
MS patients, inammation drives disease activity. In later
stages of MS neurodegeneration is driven by inammation
beyond the blood-brain-barriers as well as by mitochondrial
dysfunction (Lassmann et al. 2012). Consequently, oxida-
tive reactions may be less detectable in peripheral blood of
progressive MS patients. Furthermore, environmental fac-
tors – like smoking – will inuence plasma levels. 20 of
the subjects included in the study were non-smokers, data
was not available for three subjects.
A limitation of our trial may be the small number of
controls. Moreover, former studies showed that levels of
CPs in the CSF of patients with RRMS were higher than in
controls (Pennisi et al. 2011), and elevated levels of CPs
were detected in the plasma of patients with Alzheimer’s
disease when compared with controls without dementia
(Greilberger et al. 2010). Here we show for the rst time
that CPs are increased in the CSF of neurodegenerative dis-
eases, such as ALS and progressive MS.
The ultimate goal of biomarkers is the prediction of
disease outcome and/or therapy response. Sbardella et al.
(2013) demonstrated that isoprostanes might serve as bio-
markers for OS in the CSF of MS patients. Interestingly,
higher levels in patients with CIS predicted progression to
denitive MS (Sbardella et al. 2013). Furthermore, nitric
oxide metabolites in the CSF were associated with disease
activity and sustained disability progression in MS patients
(Rejdak et al. 2008).
The results of our study (detection of elevated CPs in
the CSF of patients with neurodegenerative and neuroim-
munological diseases compared to controls) may help to
design future conrmative trials. Future trials will have to
show if levels of CPs may be a useful marker for therapy
response or for prognosis.
Conict of Interest
FL received speaker/consulting honorary from Bayer
Schering, Biogen Idec, Genzyme, Merck Serono, Novartis
and Teva. Other authors declare no conict of interest.
References
Calabrese, V., Bella, R., Testa, D., Spadaro, F., Scrofani, A., Rizza,
V. & Pennisi, G. (1998) Increased cerebrospinal fluid and
plasma levels of ultraweak chemiluminescence are associated
with changes in the thiol pool and lipid-soluble uorescence
in multiple sclerosis: the pathogenic role of oxidative stress.
Drugs Exp. Clin. Res., 24, 125-131.
Calabrese, V., Raffaele, R., Cosentino, E. & Rizza, V. (1994)
Changes in cerebrospinal uid levels of malondialdehyde and
glutathione reductase activity in multiple sclerosis. Int. J.
Clin. Pharmacol. Res., 14, 119-123.
Drechsel, D.A., Estévez, A.G., Barbeito, L. & Beckman, J.S.
(2012) Nitric oxide-mediated oxidative damage and the
progressive demise of motor neurons in ALS. Neurotox. Res.,
22, 251-264.
Gonsette, R.E. (2008) Neurodegeneration in multiple sclerosis: the
role of oxidative stress and excitotoxicity. J. Neurol. Sci., 274,
48-53.
Greilberger, J., Fuchs, D., Leblhuber, F., Greilberger, M.,
Wintersteiger, R. & Tafeit, E. (2010) Carbonyl proteins as a
clinical marker in Alzheimer’s disease and its relation to tryp-
tophan degradation and immune activation. Clin. Lab., 56,
441-445.
Greilberger, J., Koidl, C., Greilberger, M., Lamprecht, M.,
Schroecksnadel, K., Leblhuber, F., Fuchs, D. & Oettl, K.
(2008) Malondialdehyde, carbonyl proteins and albumin-
disulphide as useful oxidative markers in mild cognitive
impairment and Alzheimer’s disease. Free Radic. Res., 42,
633-638.
Hirotani, M., Maita, C., Niino, M., Iguchi-Ariga, S., Hamada, S.,
Ariga, H. & Sasaki, H. (2008) Correlation between DJ-1
levels in the cerebrospinal uid and the progression of disabil-
ities in multiple sclerosis patients. Mult. Scler., 14, 1056-
1060.
Lassmann, H., van Horssen, J. & Mahad, D. (2012) Progressive
multiple sclerosis: pathology and pathogenesis. Nat. Rev.
Neurol., 8, 647-656.
Lublin, F.D. & Reingold, S.C. (1996) Dening the clinical course
of multiple sclerosis: results of an international survey.
National Multiple Sclerosis Society (USA) Advisory
Commit tee on Clinical Trials of New Agents in Multiple Scle-
rosis. Neurology, 46, 907-911.
Ortiz, G.G., Macías-Islas, M.A., Pacheco-Moisés, F.P., Cruz-

Carbonyl Proteins in Neurological Diseases 317
Ramos, J.A., Sustersik, S., Barba, E.A. & Aguayo, A. (2009)
Oxidative stress is increased in serum from Mexican patients
with relapsing-remitting multiple sclerosis. Dis. Markers, 26,
35-39.
Pennisi, G., Cornelius, C., Cavallaro, M.M., Salinaro, A.T.,
Cambria, M.T., Pennisi, M., Bella, R., Milone, P., Ventimiglia,
B., Migliore, M.R., Di Renzo, L., De Lorenzo, A. & Calabrese,
V. (2011) Redox regulation of cellular stress response in
multiple sclerosis. Biochem. Pharmacol., 82, 1490-1499.
Perry, G., Cash, A.D. & Smith, M.A. (2002) Alzheimer disease
and Oxidative stress. J. Biomed. Biotechnol., 2, 120-123.
Petzold, A. (2005) Neurofilament phosphoforms: surrogate
markers for axonal injury, degeneration and loss. J. Neurol.
Sci., 233, 183-198.
Polman, C.H., Reingold, S.C., Banwell, B., Clanet, M., Cohen,
J.A., Filippi, M., Fujihara, K., Havrdova, E., Hutchinson, M.,
Kappos, L., Lublin, F.D., Montalban, X., O’Connor, P.,
Sandberg-Wollheim, M., Thompson, A.J., et al. (2011) Diag-
nostic criteria for multiple sclerosis: 2010 revisions to the
McDonald criteria. Ann. Neurol., 69, 292-302.
Poulsen, K., Bahl, J.M., Tanassi, J.T., Simonsen, A.H. & Heegaard,
N.H. (2012) Characterization and stability of transthyretin
isoforms in cerebrospinal uid examined by immunoprecipita-
tion and high-resolution mass spectrometry of intact protein.
Methods, 56, 284-292.
Rejdak, K., Eikelenboom, M.J., Petzold, A., Thompson, E.J.,
Stelmasiak, Z., Lazeron, R.H., Barkhof, F., Polman, C.H.,
Uitdehaag, B.M. & Giovannoni, G. (2004) CSF nitric oxide
metabolites are associated with activity and progression of
multiple sclerosis. Neurology, 63, 1439-1445.
Rejdak, K., Petzold, A., Stelmasiak, Z. & Giovannoni, G. (2008)
Cerebrospinal uid brain specic proteins in relation to nitric
oxide metabolites during relapse of multiple sclerosis. Mult.
Scler., 14, 59-66.
Sbardella, E., Greco, A., Stromillo, M.L., Prosperini, L., Puopolo,
M., Cefaro, L.A., Pantano, P., De Stefano, N., Minghetti, L. &
Pozzilli, C. (2013) Isoprostanes in clinically isolated
syndrome and early multiple sclerosis as biomarkers of tissue
damage and predictors of clinical course. Mult. Scler., 19,
411-417.
Weiner, H.L. (2009) The challenge of multiple sclerosis: how do
we cure a chronic heterogeneous disease? Ann. Neurol., 65,
239-248.
Weinshenker, B.G. (1998) The natural history of multiple scle-
rosis: update 1998. Semin. Neurol., 18, 301-307.
Zeman, D., Adam, P., Kalistová, H., Sobek, O., Kelbich, P., Andel,
J. & Andel, M. (2000) Transferrin in patients with multiple
sclerosis: a comparison among various subgroups of multiple
sclerosis patients. Acta Neurol. Scand., 101, 89-94.