Journal of the Neurological Sciences, 111 (1992)65-72
65
© 1992 ElsevierSciencePublishersB.V. All rightsreserved 0022-510X/92/$05.00 JNS 03803
Significance of CSF total neopterin and biopterin in inflammatory neurological diseases Yoshiaki Furukawa, Katsunori Nishi, Tomoyoshi Kondo, Kazutaka Tanabe and Yoshikuni Mizuno Department of Neurology, Juntendo UniversitySchool of.Medicine, Tokyo, Japan
(Received 2 December, 1991) (Revised, received 12 February,1992) (Accepted 18 February,1992) Key words: Neopterin; Biopterin; Cerebrospinal fluid; Neuro-sarcoidosis; Neuro-Beh~et's disease; Meningitis; Parkinson's
disease; Cellular immune system
Summary Total neopterin (T-N), a by-product in the biopterin biosynthesis and an indicator of activation of the cellular immune system, and total biopterin (T-B) levels in cerebmspinal fluid (CSF), were measured in patients with various inflammatory neurological diseases and Parkinson's disease, and the followingresults were obtained. (1) In patients with neuro-sarcoidosis, neuro-Beh~et's disease and meningitis, CSF T-N levels were markedly elevated in the exacerbation or acute stages of their neurological symptoms and remarkably decreased in the remission or chronic stages. In the neuro-sarcoidosis and neum-Beh~et's disease patients, however, CSF T-B levels showed no substantial change. (2) There was a significant positive correlation between CSF T-N levels and CSF/serum albumin ratios only in the meningitis patients. However, increases of CSF T-N levels were not associated with those of plasma T-N levels. (3) In the Parkinson's disease patients, CSF T-N levels remained normal, although CSF T-B levels significantlydecreased. (4) A gradient for the CSF T-N value (lumbar > ventricular CSF), being reverse to the CSF T-B value, was observed. These results indicate that the significanceof CSF T-N is quite different from CSF T-B, and that CSF T-N appears to be a valuable biochemical marker for evaluating the activity of inflammation within the central nervous system. Its measurement seems useful for therapeutic monitoring, especiallyof patients showingthe chronic exacerbating-remitting course.
Introduction Neopterin (N) and dihydroneopterin (NH 2) are degradation products of dihydroneopterin triphosphate (NH2P3), which is the first intermediate in the biosynthesis of tetrahydrobiopterin (BH 4) (Wachter et al. 1989), the natural cofactor for tyrosine hydroxylase (TH) (Brenneman and Kanfman 1964) and tryptophan hydroxylase (Lovenberg et al. 1967) in the brain. Tetrahydrobiopterin is highly concentrated in the human striatum (Sawada et al. 1987), and its decreases in the brain (Nagatsu et al. 1981) and the lumbar as well as
Correspondence to: Dr. Yoshiaki Furukawa,Departmentof New rology, Juntendo University School of Medicine, 2-1-1 Hongo, Bunk'yo, Tokyo 113, Japan. Tel.: (81)3-3813-3111, ext. 3323; Fax: (81)3-3813-7440.
ventricular cerebrospinal fluid (CSF) (Lovenberg et al. 1979; Furukawa et aL 1991, 1992) have been reported in patients with Parkinson's disease (PD). The distribution of N and NH 2 in the human brain, however, shows no correlation with that of BH 4 (Sawada et al. 1987). Recently, interferon-gamma (IFN-~), a product of activated T cells, has been shown to induce the release of NH 2 and N from macrophages in vitro (Huber et al. 1984; Schoedon et al. 1986, 1987) (Fig. 1). Since human macrophages have no 6-pyruvoyl tetrahydropterin synthase activity which catalyzes the subsequent conversion of NH2P 3, accumulated NH2P 3 is immediately converted to NH 2 and N when stimulated by IFN-3,. In contrast, although activated T cells are able to synthesize BH 4, it is not released in detectable amounts into the cell supernatant due to the presence of a feedback inhibition by BH 4 itself (Dhondt and Farriaux 1982;
66 Macrophage
T-lymphocyte
GTP
GTP
o:
NHa c
NH:P: 6-PPH4 synthase J
a~
6-PPH4
l
1
t
1
N
(/FN-r)
BH4
BH4
Fig. 1. Biosynthesis and metabolism of pterins in human immune cells and the mechanism by which dihydroneopterin and neopterin are produced by macrophages in response to IFN-y. Abbreviations: BH4, tetrahydrobiopterin; IFN--y, interferon-gamma; N, neopterin; NH 2, dihydroneopterin; NH2P3, dihydroneopterin triphosphate; GTP, guanosine triphosphate; GTP-CH l, GTP-cyclohydrolase !; 6-PPH4, 6-pyruvoyl tetrahydropterin.
Schoedon et al. 1986, 1987). Neopterin has thus received much attention as an indicator f activation of the cellular immune system. The relationship between N and BH 4 in vivo may be even more complex, but has not been studied enough from the aspects of its relation to brain injury and to the state of immune response. Therefore, in this study, we measured total neopterin (T-N) and biopterin (T-B) levels in the CSF of patients with various inflammatory neurological diseases (neuro-sarcoidosis, neuro-Beh~et's disease and meningitis) and PD, and correlated their levels with clinical features of the patients in order to evaluate the significance of CSF T-N. Because more than 90% of T-N (NH 2 + N) in the CSF exists as NH 2 in patients with central nervous system infections (Howells et al. 1987), we analyzed the T-N content by oxidizing NH 2 to N. Moreover, we assessed the effects of treatment with steroids on CSF T-N and T-B levels in patients with neuro-sarcoidosis and neuro-Beh~et's disease, who often show the chronic exacerbating-remitting course, with the aim of determining whether serial measurement of these levels is useful for therapeutic monitoring.
Patients and methods
Patients and controls The clinical subjects consisted of 4 sarcoidosis (all female, mean age :l=SE: 36.1 + 6.8 years), 8 Beh~et's disease (5 male and 3 female, 40.8 _+2.3), 14 meningitis (10 male and 4 female, 35.7 =l: 2.5) and 33 PD (14 male and 19 female, 61.5 + 2.2) patients, and 21 neurological controls (15 male and 6 female, 57.5 + 3.9; cerebral infarction, cervical spondyrosis, psychiatric disorders, and others). Moreover, in 12 additional PD or tremor patients including essential tremor and post-traumatic
tremor cases, who underwent stereotaxic surgery, T-N levels in both the ventricular and lumbar CSF were measured. Sarcoidosis: Three of the 4 patients were diagnosed as having neuro-sarcoidosis (N-S). Cases S-1 and S-2 were in the exacerbation stages and case S-3 was in the chronic stage of their neurological symptoms on admission. Brain CT revealed a large granuloma in the hypothalamus in case S-1. After increasing the steroid dosages, cases S-1 and S-2 attained the incomplete and complete remission stages, respectively. Case S-4 had uveitis due to sarcoidosis, but no neurological symptoms. Beh~et's disease: All of the 8 patients were diagnosed as having neuro-Beh~et's disease (N-B). Three of these patients were in the exacerbation stages on admission and they were successfully treated by steroids leading them to the remission stages (described in more detail about a representative patient below). Other patients were in the chronic stages. Meningitis: This group consisted of patients with 11 viral, 2 tuberculous and 1 cryptococcal meningitis. All of them were in the acute stages on admission. The clinical course of the representative N-B patient (case B-l), who was followed more than 30 months, is described below. In January 1986, she was diagnosed as having systemic Beh~et's disease and was treated with steroids and colchicine thereafter. In November, however, she stopped taking these medications through her own decision. In November 1988, she developed diplopia and ataxia and was admitted to our hospital 3 days after the appearance of these neurological symptoms. Neurological examination revealed vertical gaze limitation, disturbance of convergence, down-beating nystagmus on downgaze, and bilateral pyramidal and cerebellar signs. She was diagnosed as having N-B mainly involving the midbrain and cerebellum. After the first lumbar puncture, which was performed on the day of admission, the steroids treatment including pulse therapy (intravenous administration of methylprednisolone sodium succinate 1 g/day for 3 consecutive days) was started again. Then, her neurological symptoms gradually improved and she attained the remission stage. In June 1991, after the tapering off of the steroid dosages, she developed oral aphthous ulcers and erythema nodosum and her ataxia was exacerbated.
CSF collection All lumbar punctures were performed between 9 and 10 A.M. after overnight bed rest and fasting. The 2nd-4th ml of the lumbar CSF were used to analyze T-N and T-B levels. The CSF samples were immediately frozen and stored at -70°C under dark conditions until assayed.
67 Sixty CSF samples of patients with sarcoidosis (8 samples), Beh~et's disease (17 samples) and meningitis (35 samples) were analyzed. Forty-two of these CSF samples were obtained in the exacerbation or acute stage (both hereafter referred to as the 'active' stage: sarcoidosis 5, Beh~et's disease 8, and meningitis 29 samples) of their symptoms and other samples were obtained in the remission or chronic stage (both hereafter referred to as the 'inactive' stage: sarcoidosis 3, Behget's disease 9, and meningitis 6 samples). Each of the 33 PD patients and the 21 neurological controls underwent a lumbar puncture for diagnostic purposes. In the additional PD or tremor patients who underwent stereotaxic surgery, ventricular CSF samples were obtained during ventriculography prior to the surgery being performed under local anesthesia. These patients also underwent a lumbar puncture within 2 weeks prior to the surgery. Then, the T-N concentration in the ventricular CSF was compared with that in the lumbar CSF in each patient. At the time of CSF collection, none of our patients or controls studied had renal dysfunction or malignant diseases.
Total neopterin and biopterin assay Total neopterin and biopterin levels were determined according to the method of Fukushima and Nixon (1980) with slight modifications (Furukawa et al. 1991). High-performance liquid chromatography (Yanaco L-2000) with a fluorometric detector (Shimadzu RF 530) and a biophase ODS column II (BAS) was used. The mobile phase was 0.1 M sodium phosphate buffer (pH 3.0) containing 2 mM l-octane sulfonic acid and 4% acetonitrile. The flow rate was 1.5 ml/min and excitation and emission wavelengths were set at 360 and 445 nm, respectively. D-(+)-Neopterin (Sigma) and L-erythro-biopterin (a generous gift from Dr. S. Matsuura, Department of Chemistry, Nagoya University College of General Education) were used as the standards. The retention time of T-N was 2.4 min and that of T-B was 3.6 min. The retention time of 6-methylpterin (Sigma), used as an internal standard, was 10.2 min.
Statistical analysis Data are expressed as the means :1: SEM. Statistical analysis was carried out by paired and unpaired t-tests and Mann-Whitney U-test when appropriate. The association between variables was computed by means of Spearman's correlation coefficient (p). The criterion of statistical significance was P < 0.05.
Results In the lumbar CSF, the mean level of T-N in the controls was 27.7 ± 1.5 pmol/ml (Fig. 2). Based on this
• 1463JI
• 1204.e • 1203.9
II~I] lll:T • •
600t 500
812.5 O~'8,l
%
400
..~ ~,oo
•
IOD
@@ Be.
o
: Controls
PD
i N-S
H-O
N
Fig. 2. Cx)mparison of the mean CSF total neopterin ( + SE) levels in patients with active inflammatory neurological diseases (N-S, neuro-
sarcoidosis; N-B, neuro-Oeh~et'sdisease; M, meningitis) and Parkinson's disease(PD) and controls. •, Each value of total neopterin in patients with inflammatory neurological diseases during their active stages; [], This line indicates the upper reference limit of total neopterin. mean T-N concentration, we defined the upper reference limit of T-N (mean + 2 SD) in the lumbar CSF as 41.1 pmol/ml. The mean CSF T-N concentrations of the patients with N-S, N-B and meningitis in their active stages were markedly higher than that of the controls (N-S: 530.0 ± 245.0 pmol/ml, P < 0.001; N-B: 154,0 ± 21.4, P < 0.0001; meningitis: 449.1 ± 74.1, P < 0.0001). In the PD patients, the mean CSF T-N level (23.4 ± 1.6) was slightly lower than the controls, but the difference between them did not reach the statistical significance. The mean CSF T-B concentration in the controls was 14.4 ± 0.9 pmol/ml (Fig. 3). The mean CSF T-B level of the meningitis patients in their active stages (30.0:1: 2.9) was significantly high as compared with that in the controls ( P < 0.0001). However, no significant differences of the mean CSF T-B levels were noted between the controls and either the N-S or N-B patients in the active stages (N-S: 18.6 ± 1.8; N-B: 17.3 :t: 1.1). The mean CSF T-B level in the PD patients (8.1 ± 0.4) was significantly lower than that in the controls ( P < 0.0001).
68 In cases S-1 and S-2, after increasing the steroid dosages, their CSF T-lq levels markedly decreased with clinical improvements, but the CSF T-B concentrations did not change much (Fig. 4). The CSF angiotensinconverting e n z ~ e and lysozyme levels also showed decreases. In case S-3, who was in the inactive stage, the CSF T-N level was near the upper reference limit. The CSF T-N concentration in case S-4, who suffered from only uveitis, was within the normal range. In the N-B patients, the mean CSF T-N level in the inactive stages (49.8 ± 6.7) was significantly lower than that in the active stages ( P < 0.0005). Figure 5 shows the clinical course and the CSF T-N and T-B levels of case B-1 during the steroids therapy. The T-N level was highest on the day of admission when the neurological manifestations were severe. Then, her T-N levels gradually decreased with the clinical improvement and reached the upper reference limit 3 weeks after the treatment was begun. The T-N concentration, however, again in~.reased with the exacerbation of her ataxia. In contrast, the T-B levels during the course showed no substantial change. In the meningitis patients, the mean CSF T-N level in the inactive stages (44.3 ± 5.1)was significantly lower
0 7~',11
8[
o o
50
•
o o o
8o
4O
,D
o
.E J 3o O0
0
8
II
ooo Oo 0~
o
i:
Controls PO l-I N-O M Fig. 3. Comparisonof the mean CSF total biopterin(+ SE) levelsin patients with activeinflammatoryneurologicaldiseases(N-S, neurosarcoidosis;N-B, neuro-Beh~et'sdisease;M, meningitis)and Parkinson's disease(PD) and controls, o, Each valueof total biopterin in patients with inflammatoryneurologicaldiseasesduring their active stages.
than that in the active stages ( P < 0.0001). In these patients, the mean CSF T-B concentration in the inactive stages (16.9 + 2.7) was also substantially low as compared with that in the active stages ( P < 0.05). When we analyzed the relations between the CSF T-N levels and CSF mononuclear cell counts, CSF protein levels or CSF/serum albumin ratios in the patients with the inflammatory neurological diseases, a significant positive correlation was found between the T-N and protein levels in the sarcoidosis patients (p = 0.857, P < 0.05), and between the T-N levels and the mononuclear cell counts in the Beh~et's disease patients (p = 0.803, P < 0.005). Among the meningitis patients, significant positive correlations were observed between the T-N concentrations and the mononuclear cell counts, the protein levels and the CSF/serum albumin ratios (p = 0.485, P < 0.005; p = 0.630, P < 0.0005; p ffi 0.802, P < 0.005). As for the relationship between the CSF T-N and T-B levels, there was a significant positive correlation in the meningitis patients (p = 0.617, P < 0.0005), but not in the sarcoidosis and Beh~et's disease patients. Because there was a significant positive correlation between the CSF T-N levels and the CSF/serum albumin ratios in the meningitis patients, we additionally measured both the plasma and CSF T-N concentrations in 4 controls and 3 patients with meningitis in the active stages, and obtained the following results. In the controls, the mean T.N level in the CSF (24.4 + 2,8 pmol/ml) did not significantly differ from that in the plasma (21.9 + 5.7 pmol/ml). By contrast, in the patients with active meningitis, the CSF T.N levels (148.6, 828.1 and 262.3 pmol/ml) were markedly higher than the plasma T-N levels (17.3, 22.9 and 54.2 pmol/ml, respectively). In the 12 patients, who underwent stereotaxic surgery, the mean T-N level in the ventricular CSF (13.4 ± 1.7) was significantly lower than that in the lumbar CSF (23.8 _+3.8, P < 0.005) (Fig. 6).
Discussion
Elevated N concentrations in the CSF have been reported in patients with chronic inflammatory diseases; i.e., acquired immunodeficienoy syndrome (AIDS) (Fuchs et al. 1989; S6nnerborg et al. 1989; Brew et aL 1990; Griffin et al. 1991) and human T-lymphotropic virus type l-associated myelopathy (Nomoto et al. 1991). Although the etiologies of N-S and N-B are still unknown, they may be classified into the category of chronic inflammatory neurological disease and it has been reported that activated T cells may play an important role in these diseases. Current concept of the pathogenesis of sarcoidosis includes local stimulation and replication of activated
69
(Case S-I)
(Case S-2)
(Cases S-3 & S'-4)
1483.0 6oo
i
I0
\
|
i"~
°
\
|
~,. --. , Mononuclear cells(/ms) 7 Protein (ng/dl) 732
1 76
! ~
121
,~°
,2 °
E
R2
S-3
S"4
2 38
0 22
4 35
1
65
© 1992 ElsevierSciencePublishersB.V. All rightsreserved 0022-510X/92/$05.00 JNS 03803
Significance of CSF total neopterin and biopterin in inflammatory neurological diseases Yoshiaki Furukawa, Katsunori Nishi, Tomoyoshi Kondo, Kazutaka Tanabe and Yoshikuni Mizuno Department of Neurology, Juntendo UniversitySchool of.Medicine, Tokyo, Japan
(Received 2 December, 1991) (Revised, received 12 February,1992) (Accepted 18 February,1992) Key words: Neopterin; Biopterin; Cerebrospinal fluid; Neuro-sarcoidosis; Neuro-Beh~et's disease; Meningitis; Parkinson's
disease; Cellular immune system
Summary Total neopterin (T-N), a by-product in the biopterin biosynthesis and an indicator of activation of the cellular immune system, and total biopterin (T-B) levels in cerebmspinal fluid (CSF), were measured in patients with various inflammatory neurological diseases and Parkinson's disease, and the followingresults were obtained. (1) In patients with neuro-sarcoidosis, neuro-Beh~et's disease and meningitis, CSF T-N levels were markedly elevated in the exacerbation or acute stages of their neurological symptoms and remarkably decreased in the remission or chronic stages. In the neuro-sarcoidosis and neum-Beh~et's disease patients, however, CSF T-B levels showed no substantial change. (2) There was a significant positive correlation between CSF T-N levels and CSF/serum albumin ratios only in the meningitis patients. However, increases of CSF T-N levels were not associated with those of plasma T-N levels. (3) In the Parkinson's disease patients, CSF T-N levels remained normal, although CSF T-B levels significantlydecreased. (4) A gradient for the CSF T-N value (lumbar > ventricular CSF), being reverse to the CSF T-B value, was observed. These results indicate that the significanceof CSF T-N is quite different from CSF T-B, and that CSF T-N appears to be a valuable biochemical marker for evaluating the activity of inflammation within the central nervous system. Its measurement seems useful for therapeutic monitoring, especiallyof patients showingthe chronic exacerbating-remitting course.
Introduction Neopterin (N) and dihydroneopterin (NH 2) are degradation products of dihydroneopterin triphosphate (NH2P3), which is the first intermediate in the biosynthesis of tetrahydrobiopterin (BH 4) (Wachter et al. 1989), the natural cofactor for tyrosine hydroxylase (TH) (Brenneman and Kanfman 1964) and tryptophan hydroxylase (Lovenberg et al. 1967) in the brain. Tetrahydrobiopterin is highly concentrated in the human striatum (Sawada et al. 1987), and its decreases in the brain (Nagatsu et al. 1981) and the lumbar as well as
Correspondence to: Dr. Yoshiaki Furukawa,Departmentof New rology, Juntendo University School of Medicine, 2-1-1 Hongo, Bunk'yo, Tokyo 113, Japan. Tel.: (81)3-3813-3111, ext. 3323; Fax: (81)3-3813-7440.
ventricular cerebrospinal fluid (CSF) (Lovenberg et al. 1979; Furukawa et aL 1991, 1992) have been reported in patients with Parkinson's disease (PD). The distribution of N and NH 2 in the human brain, however, shows no correlation with that of BH 4 (Sawada et al. 1987). Recently, interferon-gamma (IFN-~), a product of activated T cells, has been shown to induce the release of NH 2 and N from macrophages in vitro (Huber et al. 1984; Schoedon et al. 1986, 1987) (Fig. 1). Since human macrophages have no 6-pyruvoyl tetrahydropterin synthase activity which catalyzes the subsequent conversion of NH2P 3, accumulated NH2P 3 is immediately converted to NH 2 and N when stimulated by IFN-3,. In contrast, although activated T cells are able to synthesize BH 4, it is not released in detectable amounts into the cell supernatant due to the presence of a feedback inhibition by BH 4 itself (Dhondt and Farriaux 1982;
66 Macrophage
T-lymphocyte
GTP
GTP
o:
NHa c
NH:P: 6-PPH4 synthase J
a~
6-PPH4
l
1
t
1
N
(/FN-r)
BH4
BH4
Fig. 1. Biosynthesis and metabolism of pterins in human immune cells and the mechanism by which dihydroneopterin and neopterin are produced by macrophages in response to IFN-y. Abbreviations: BH4, tetrahydrobiopterin; IFN--y, interferon-gamma; N, neopterin; NH 2, dihydroneopterin; NH2P3, dihydroneopterin triphosphate; GTP, guanosine triphosphate; GTP-CH l, GTP-cyclohydrolase !; 6-PPH4, 6-pyruvoyl tetrahydropterin.
Schoedon et al. 1986, 1987). Neopterin has thus received much attention as an indicator f activation of the cellular immune system. The relationship between N and BH 4 in vivo may be even more complex, but has not been studied enough from the aspects of its relation to brain injury and to the state of immune response. Therefore, in this study, we measured total neopterin (T-N) and biopterin (T-B) levels in the CSF of patients with various inflammatory neurological diseases (neuro-sarcoidosis, neuro-Beh~et's disease and meningitis) and PD, and correlated their levels with clinical features of the patients in order to evaluate the significance of CSF T-N. Because more than 90% of T-N (NH 2 + N) in the CSF exists as NH 2 in patients with central nervous system infections (Howells et al. 1987), we analyzed the T-N content by oxidizing NH 2 to N. Moreover, we assessed the effects of treatment with steroids on CSF T-N and T-B levels in patients with neuro-sarcoidosis and neuro-Beh~et's disease, who often show the chronic exacerbating-remitting course, with the aim of determining whether serial measurement of these levels is useful for therapeutic monitoring.
Patients and methods
Patients and controls The clinical subjects consisted of 4 sarcoidosis (all female, mean age :l=SE: 36.1 + 6.8 years), 8 Beh~et's disease (5 male and 3 female, 40.8 _+2.3), 14 meningitis (10 male and 4 female, 35.7 =l: 2.5) and 33 PD (14 male and 19 female, 61.5 + 2.2) patients, and 21 neurological controls (15 male and 6 female, 57.5 + 3.9; cerebral infarction, cervical spondyrosis, psychiatric disorders, and others). Moreover, in 12 additional PD or tremor patients including essential tremor and post-traumatic
tremor cases, who underwent stereotaxic surgery, T-N levels in both the ventricular and lumbar CSF were measured. Sarcoidosis: Three of the 4 patients were diagnosed as having neuro-sarcoidosis (N-S). Cases S-1 and S-2 were in the exacerbation stages and case S-3 was in the chronic stage of their neurological symptoms on admission. Brain CT revealed a large granuloma in the hypothalamus in case S-1. After increasing the steroid dosages, cases S-1 and S-2 attained the incomplete and complete remission stages, respectively. Case S-4 had uveitis due to sarcoidosis, but no neurological symptoms. Beh~et's disease: All of the 8 patients were diagnosed as having neuro-Beh~et's disease (N-B). Three of these patients were in the exacerbation stages on admission and they were successfully treated by steroids leading them to the remission stages (described in more detail about a representative patient below). Other patients were in the chronic stages. Meningitis: This group consisted of patients with 11 viral, 2 tuberculous and 1 cryptococcal meningitis. All of them were in the acute stages on admission. The clinical course of the representative N-B patient (case B-l), who was followed more than 30 months, is described below. In January 1986, she was diagnosed as having systemic Beh~et's disease and was treated with steroids and colchicine thereafter. In November, however, she stopped taking these medications through her own decision. In November 1988, she developed diplopia and ataxia and was admitted to our hospital 3 days after the appearance of these neurological symptoms. Neurological examination revealed vertical gaze limitation, disturbance of convergence, down-beating nystagmus on downgaze, and bilateral pyramidal and cerebellar signs. She was diagnosed as having N-B mainly involving the midbrain and cerebellum. After the first lumbar puncture, which was performed on the day of admission, the steroids treatment including pulse therapy (intravenous administration of methylprednisolone sodium succinate 1 g/day for 3 consecutive days) was started again. Then, her neurological symptoms gradually improved and she attained the remission stage. In June 1991, after the tapering off of the steroid dosages, she developed oral aphthous ulcers and erythema nodosum and her ataxia was exacerbated.
CSF collection All lumbar punctures were performed between 9 and 10 A.M. after overnight bed rest and fasting. The 2nd-4th ml of the lumbar CSF were used to analyze T-N and T-B levels. The CSF samples were immediately frozen and stored at -70°C under dark conditions until assayed.
67 Sixty CSF samples of patients with sarcoidosis (8 samples), Beh~et's disease (17 samples) and meningitis (35 samples) were analyzed. Forty-two of these CSF samples were obtained in the exacerbation or acute stage (both hereafter referred to as the 'active' stage: sarcoidosis 5, Beh~et's disease 8, and meningitis 29 samples) of their symptoms and other samples were obtained in the remission or chronic stage (both hereafter referred to as the 'inactive' stage: sarcoidosis 3, Behget's disease 9, and meningitis 6 samples). Each of the 33 PD patients and the 21 neurological controls underwent a lumbar puncture for diagnostic purposes. In the additional PD or tremor patients who underwent stereotaxic surgery, ventricular CSF samples were obtained during ventriculography prior to the surgery being performed under local anesthesia. These patients also underwent a lumbar puncture within 2 weeks prior to the surgery. Then, the T-N concentration in the ventricular CSF was compared with that in the lumbar CSF in each patient. At the time of CSF collection, none of our patients or controls studied had renal dysfunction or malignant diseases.
Total neopterin and biopterin assay Total neopterin and biopterin levels were determined according to the method of Fukushima and Nixon (1980) with slight modifications (Furukawa et al. 1991). High-performance liquid chromatography (Yanaco L-2000) with a fluorometric detector (Shimadzu RF 530) and a biophase ODS column II (BAS) was used. The mobile phase was 0.1 M sodium phosphate buffer (pH 3.0) containing 2 mM l-octane sulfonic acid and 4% acetonitrile. The flow rate was 1.5 ml/min and excitation and emission wavelengths were set at 360 and 445 nm, respectively. D-(+)-Neopterin (Sigma) and L-erythro-biopterin (a generous gift from Dr. S. Matsuura, Department of Chemistry, Nagoya University College of General Education) were used as the standards. The retention time of T-N was 2.4 min and that of T-B was 3.6 min. The retention time of 6-methylpterin (Sigma), used as an internal standard, was 10.2 min.
Statistical analysis Data are expressed as the means :1: SEM. Statistical analysis was carried out by paired and unpaired t-tests and Mann-Whitney U-test when appropriate. The association between variables was computed by means of Spearman's correlation coefficient (p). The criterion of statistical significance was P < 0.05.
Results In the lumbar CSF, the mean level of T-N in the controls was 27.7 ± 1.5 pmol/ml (Fig. 2). Based on this
• 1463JI
• 1204.e • 1203.9
II~I] lll:T • •
600t 500
812.5 O~'8,l
%
400
..~ ~,oo
•
IOD
@@ Be.
o
: Controls
PD
i N-S
H-O
N
Fig. 2. Cx)mparison of the mean CSF total neopterin ( + SE) levels in patients with active inflammatory neurological diseases (N-S, neuro-
sarcoidosis; N-B, neuro-Oeh~et'sdisease; M, meningitis) and Parkinson's disease(PD) and controls. •, Each value of total neopterin in patients with inflammatory neurological diseases during their active stages; [], This line indicates the upper reference limit of total neopterin. mean T-N concentration, we defined the upper reference limit of T-N (mean + 2 SD) in the lumbar CSF as 41.1 pmol/ml. The mean CSF T-N concentrations of the patients with N-S, N-B and meningitis in their active stages were markedly higher than that of the controls (N-S: 530.0 ± 245.0 pmol/ml, P < 0.001; N-B: 154,0 ± 21.4, P < 0.0001; meningitis: 449.1 ± 74.1, P < 0.0001). In the PD patients, the mean CSF T-N level (23.4 ± 1.6) was slightly lower than the controls, but the difference between them did not reach the statistical significance. The mean CSF T-B concentration in the controls was 14.4 ± 0.9 pmol/ml (Fig. 3). The mean CSF T-B level of the meningitis patients in their active stages (30.0:1: 2.9) was significantly high as compared with that in the controls ( P < 0.0001). However, no significant differences of the mean CSF T-B levels were noted between the controls and either the N-S or N-B patients in the active stages (N-S: 18.6 ± 1.8; N-B: 17.3 :t: 1.1). The mean CSF T-B level in the PD patients (8.1 ± 0.4) was significantly lower than that in the controls ( P < 0.0001).
68 In cases S-1 and S-2, after increasing the steroid dosages, their CSF T-lq levels markedly decreased with clinical improvements, but the CSF T-B concentrations did not change much (Fig. 4). The CSF angiotensinconverting e n z ~ e and lysozyme levels also showed decreases. In case S-3, who was in the inactive stage, the CSF T-N level was near the upper reference limit. The CSF T-N concentration in case S-4, who suffered from only uveitis, was within the normal range. In the N-B patients, the mean CSF T-N level in the inactive stages (49.8 ± 6.7) was significantly lower than that in the active stages ( P < 0.0005). Figure 5 shows the clinical course and the CSF T-N and T-B levels of case B-1 during the steroids therapy. The T-N level was highest on the day of admission when the neurological manifestations were severe. Then, her T-N levels gradually decreased with the clinical improvement and reached the upper reference limit 3 weeks after the treatment was begun. The T-N concentration, however, again in~.reased with the exacerbation of her ataxia. In contrast, the T-B levels during the course showed no substantial change. In the meningitis patients, the mean CSF T-N level in the inactive stages (44.3 ± 5.1)was significantly lower
0 7~',11
8[
o o
50
•
o o o
8o
4O
,D
o
.E J 3o O0
0
8
II
ooo Oo 0~
o
i:
Controls PO l-I N-O M Fig. 3. Comparisonof the mean CSF total biopterin(+ SE) levelsin patients with activeinflammatoryneurologicaldiseases(N-S, neurosarcoidosis;N-B, neuro-Beh~et'sdisease;M, meningitis)and Parkinson's disease(PD) and controls, o, Each valueof total biopterin in patients with inflammatoryneurologicaldiseasesduring their active stages.
than that in the active stages ( P < 0.0001). In these patients, the mean CSF T-B concentration in the inactive stages (16.9 + 2.7) was also substantially low as compared with that in the active stages ( P < 0.05). When we analyzed the relations between the CSF T-N levels and CSF mononuclear cell counts, CSF protein levels or CSF/serum albumin ratios in the patients with the inflammatory neurological diseases, a significant positive correlation was found between the T-N and protein levels in the sarcoidosis patients (p = 0.857, P < 0.05), and between the T-N levels and the mononuclear cell counts in the Beh~et's disease patients (p = 0.803, P < 0.005). Among the meningitis patients, significant positive correlations were observed between the T-N concentrations and the mononuclear cell counts, the protein levels and the CSF/serum albumin ratios (p = 0.485, P < 0.005; p = 0.630, P < 0.0005; p ffi 0.802, P < 0.005). As for the relationship between the CSF T-N and T-B levels, there was a significant positive correlation in the meningitis patients (p = 0.617, P < 0.0005), but not in the sarcoidosis and Beh~et's disease patients. Because there was a significant positive correlation between the CSF T-N levels and the CSF/serum albumin ratios in the meningitis patients, we additionally measured both the plasma and CSF T-N concentrations in 4 controls and 3 patients with meningitis in the active stages, and obtained the following results. In the controls, the mean T.N level in the CSF (24.4 + 2,8 pmol/ml) did not significantly differ from that in the plasma (21.9 + 5.7 pmol/ml). By contrast, in the patients with active meningitis, the CSF T.N levels (148.6, 828.1 and 262.3 pmol/ml) were markedly higher than the plasma T-N levels (17.3, 22.9 and 54.2 pmol/ml, respectively). In the 12 patients, who underwent stereotaxic surgery, the mean T-N level in the ventricular CSF (13.4 ± 1.7) was significantly lower than that in the lumbar CSF (23.8 _+3.8, P < 0.005) (Fig. 6).
Discussion
Elevated N concentrations in the CSF have been reported in patients with chronic inflammatory diseases; i.e., acquired immunodeficienoy syndrome (AIDS) (Fuchs et al. 1989; S6nnerborg et al. 1989; Brew et aL 1990; Griffin et al. 1991) and human T-lymphotropic virus type l-associated myelopathy (Nomoto et al. 1991). Although the etiologies of N-S and N-B are still unknown, they may be classified into the category of chronic inflammatory neurological disease and it has been reported that activated T cells may play an important role in these diseases. Current concept of the pathogenesis of sarcoidosis includes local stimulation and replication of activated
69
(Case S-I)
(Case S-2)
(Cases S-3 & S'-4)
1483.0 6oo
i
I0
\
|
i"~
°
\
|
~,. --. , Mononuclear cells(/ms) 7 Protein (ng/dl) 732
1 76
! ~
121
,~°
,2 °
E
R2
S-3
S"4
2 38
0 22
4 35
1
