QJM Advance Access originally published online on January 30, 2008
QJM 2008 101(3):225-230; doi:10.1093/qjmed/hcm146
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The value of serial cerebrospinal fluid 14-3-3 protein levels in adult community-acquired bacterial meningitis
1From the Department of Neurology, 2Department of Emergency Medicine, Chang Gung Memorial Hospital-Kaohsiung Medical Center, Chang Gung University College of Medicine, Kaohsiung, 3Department of Biological Science, National Sun Yat-Sen University, Kaohsiung, 4Section of Infectious Diseases, Department of Internal Medicine, Kaohsiung Veterans General Hospital and National Yang-Ming University, Taipei, and 5Department of Neurosurgery Chang Gung Memorial Hospital-Kaohsiung Medical Center, Chang Gung University College of Medicine, Kaohsiung, Taiwan
Address correspondence to Dr C.-H. Lu, MD, MSc, Department of Neurology, Chang Gung Memorial Hospital, 123, Ta Pei Road, Niao Sung Hsiang, Kaohsiung Hsien, Taiwan. email: chlu99{at}pchome.com.tw
Received 5 August 2007 and in revised form 3 November 2007
 |
Summary |
|---|
 Top Summary IntroductionPatients and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Background: Increased levels of cerebrospinal fluid (CSF) 14-3-3 proteins have been reported in acute bacterial meningitis. We tested the hypothesis that CSF 14-3-3 protein levels are substantially increased in acute bacterial meningitis and decreased after anti-microbial therapy, and that CSF 14-3-3 protein levels can predict treatment outcomes.
Methods: We examined serial pan-CSF 14-3-3 (14-3-3-P) protein and five major isoform (β, 
, 
, 
, 
) levels in 29 adult community-acquired bacterial meningitis (ACABM) patients. The CSF 14-3-3 protein levels were also evaluated in 12 aseptic meningitis patients during the study period.
Results: All of the meningitis patients had a positive result on admission. Levels of CSF 14-3-3 protein in ACABM cases were significantly increased initially, and substantially decreased thereafter. Most of those who survived (survivors = 25 and non-survivors = 4) had nearly cleared their 14-3-3 protein from the CSF before discharge. Conversely, patients who died never cleared their CSF 14-3-3 protein. The median value of CSF 14-3-3-P and 14-3-3 , 14-3-3 and 14-3-3 isoforms on admission in the bacterial meningitis group were 173.7, 137.7, 42.2 and 9.1, respectively, which were statistically significant than those of the aseptic meningitis group (48.4, 39.6, 2.5 and 0, respectively). Stepwise logistic regression analysis showed only CSF 14-3-3 isoform on admission was independently associated with outcome (P = 0.05, OR = 0.991).
Conclusions: Serial 14-3-3 protein isoform actually meets the major requirements for outcome prediction in the treatment of ACABM patients. Assay of the 14-3-3 protein isoform should be added as a neuro-pathologic marker among the panel of conventional CSF parameters.
|
Introduction |
|---|
|
TopSummaryIntroductionPatients and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Despite the advent of new anti-microbial drugs, mortality and morbidity from adult community-acquired bacterial meningitis (ACABM) remains high.1 The 14-3-3 proteins are conserved abundantly in the cytoplasmic fraction of mammalian neurons2 that are in high concentrations within neuronal cells and are involved in signal transduction, cell cycle regulation and apoptosis.3 Seven distinct isoforms (β,
, , , , 
and 
) and the phosphorylated forms of β and have been reported in mammalian tissues.4 Although only the isoform seems to be brain-specific, the five major isoforms (β, , , , ) are found in neurons of the human cerebral cortex.5 The presence of 14-3-3 protein and its isoform patterns in cerebrospinal fluid (CSF) results from neuronal disruption and leakage of brain proteins into the CSF. This protein is found in CSF specimens obtained from patients with several acute or sub-acute neurological disorders, such as Creutzfeldt-Jacob disease, herpes simplex encephalitis, multiple sclerosis and para-neoplastic neurologic disorders have been reported.6–9 To our knowledge, only one clinical research has focused specifically on the relationship between bacterial meningitis and CSF 14-3-3 protein level.10 Through this prospectively designed study, we aimed to report on the relationship between serial CSF 14-3-3 protein level and therapeutic outcome in ACABM patients.
|
Patients and methods |
|---|
|
TopSummaryIntroductionPatients and methodsResultsDiscussionAcknowledgementsReferences |
|---|
This prospective study on the time course of CSF 14-3-3 protein levels in ACABM patients was conducted in a 2-year period (July 2004–June 2006). During this period, 29 adult (
16 years of age) ACABM patients at the Chang Gung Memorial Hospital (CGMH)-Kaohsiung were enrolled. CGMH-Kaohsiung, a 2482-bed acute-care teaching hospital, is the largest medical center in Southern Taiwan that provides both primary and tertiary referral care. In this study, the diagnostic criteria for ACABM included a positive culture result of CSF in patients with clinical presentations of bacterial meningitis, with at least one of the following parameters of purulent CSF inflammation: (i) leukocytosis with leukocyte counts >250 x 106/l and predominant polymorphonuclear (PMN) cells; (ii) CSF glucose ratio (CSF glucose/blood glucose) <0.4; or 3) CSF glucose <2.5 mmol/l if without blood glucose.11,12 In patients with negative culture results, a diagnosis of acute bacterial meningitis is based on compatible clinical features and pleocytosis of at least 100 x 106 PMN/l.11
For comparison, 12 aseptic meningitis patients were included as controls. The diagnostic criteria for aseptic meningitis included clinical evidence of acute meningitis such as fever, headache, or other signs of meningeal irritation, absence of any microorganism on Gram stain of CSF, negative routine bacterial culture of CSF in the absence of antibiotic treatment before obtaining the first CSF sample, and CSF parameters of leukocytosis with leukocyte counts >15 x 106/l. A small and variable increase in protein and glucose content of the CSF is normal.13
Bacterial meningitis was characterized as either being nosocomial or community-acquired.1,14 ‘Community-acquired bacterial meningitis’ was defined as bacterial meningitis contracted outside a hospital environment. Patients were enrolled in this study only if full written informed consent was obtained from the patients or their families. Informed consent was obtained from all of the study subjects. The study protocol was approved by Chang Gung Memorial Hospital's Institutional Review Committee on Human Research. A Total of 28 patients, including 7 with a history of head trauma and skull fracture and 21 with neuro-surgical procedures were excluded.
In every patient, lumbar puncture was performed at the time of admission and CSF specimens were analyzed for pathogen culture, cell count and levels of protein, lactate, glucose and 14-3-3 protein isoforms (β, , , and ). Follow-up CSF studies in most patients were fixed for Day 4, Week 2 (Days 10–14) and Week 3 (Days 20–24) after admission. To detect 14-3-3 protein in CSF, we initially analyzed CSF using polyclonal rabbit antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA), which react to the common epitopes of all of the 14-3-3 isoforms (pan 14-3-3). Commercially available antibodies for 14-3-3 β, 14-3-3 , 14-3-3 , 14-3-3 and 14-3-3 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) were also evaluated. Procedural details have been described previously.10
An aliquot of 20 µl of CSF specimens were loaded onto a 10–15% polyacrylamide gel and transferred to polyvinylidene difluoride membranes. The membranes were reacted with pan 14-3-3 antibodies and the isoform-specific antibodies (0.2 µg/ml for SC-629 pan 14-3-3 and 0.4 µg/ml for 14-3-3 β, , , and ) at 4°C overnight. The membranes were incubated with 1:5000 dilutions of horseradish peroxidase (HRP)-conjugated anti-rabbit or anti-goat immunoglobulin G (Santa Cruz Biotechnology). The blots were developed using an enhanced chemi-luminescent system. Densitometric values of each sample were obtained with a computer-assisted laser scanner, after correction for the background, and the content was assessed by densitometry using ImageQuant 3.3 (molecular Dynamic, Sunnydale, CA). The total amount of 14-3-3 protein as quantified from each CSF sample was expressed in arbitrary units.
Three separate statistical analyses were performed. First, data for CSF WBC count and levels of protein, lactate, glucose and pan-CSF 14-3-3 (14-3-3-P) protein and isoform (β, , , and ) levels on admission were logarithmically transformed to improve normality and comparison with the aseptic meningitis group was made using the independent t-test. Second, data of the bacterial meningitis patients for age, CSF lactate, protein and glucose levels, WBC counts and pan-CSF 14-3-3 (14-3-3-P) protein and isoform (β, , , and ) levels at the time of admission between the fatal and non-fatal groups were logarithmically transformed to improve normality and were compared using Student's t-test. Data of clinical manifestations between the fatal and non-fatal groups were analyzed by means of Chi-square test or Fisher's exact test. Third, stepwise logistic regression was used to evaluate the relationship between CSF parameter and therapeutic outcomes, with adjustments made for other potential confounding factors. Variables with a zero cell count in a 2-by-2 table were eliminated from logistic analysis, while only variables with a strong association with fatality rate (P < 0.05) were included in the final model.
Receiver operating characteristic (ROC) curves were generated for CSF 14-3-3 isoform, CSF Glucose/blood glucose ratio and CSF White cell count. The areas under the ROC curves were calculated for each parameter and compared. All of the statistical analyses were conducted using the SAS software package, version 9.1 (2002, SAS Statistical Institute, Cary, North Carolina).
|
Results |
|---|
|
TopSummaryIntroductionPatients and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Baseline characteristics of the study patients
The baseline characteristics of the 29 adult ACABM cases and 12 aseptic meningitis patients are listed in Table 1. Of the ACABM cases, 22 had one or more underlying disease. Regarding causative pathogens, Gram-negative bacilli were the most prevalent, followed by Streptococcus and Staphylococcus species.
|
Time course of CSF 14-3-3 proteins
Both bacterial and aseptic meningitis patients had positive results on admission. The time course of CSF white cell count and levels of protein, lactate, glucose, white cell count and pan-14-3-3 proteins and isoforms (β,
, , and ) in patients with community-acquired bacterial meningitis are listed in Table 2. The median value [interquartile range (IQR)] of pan-CSF 14-3-3 (14-3-3-P) proteins and CSF 14-3-3 β, CSF 14-3-3 , CSF 14-3-3 , CSF 14-3-3 and CSF 14-3-3 isoforms on admission of the bacterial meningitis group were 173.7 (83.3, 342.4) densitometric units (DU), 102.6 (28.6, 163.1) DU, 137.7 (29.4, 331.9) DU, 9.1 (0, 46.8) DU, 42.2 (10.1, 139.0) DU and 193.2 (26.7, 307.8) DU, respectively. The median value (IQR) of pan-CSF 14-3-3 proteins and CSF 14-3-3 β, CSF 14-3-3 , CSF 14-3-3 , CSF 14-3-3 and CSF 14-3-3 isoforms in the aseptic meningitis group were 48.5 (16.8, 134.4) DU, 51.9 (9.1, 132.0) DU, 39.6 (0, 83.4) DU, 0 (0, 7.4) DU, 18.7 (2.5, 50.1) DU and 94.7 (55.2, 164.3) DU, respectively.
|
The median value of pan-CSF 14-3-3 (14-3-3-P) proteins and its five isoforms on admission in the bacterial meningitis and aseptic meningitis groups are shown as a box-plot in Figure 1. Levels of CSF 14-3-3 protein on admission between the bacterial and aseptic meningitis groups revealed the following significant findings: CSF 14-3-3-P proteins (P = 0.049), CSF 14-3-3
isoform (P = 0.032), CSF 14-3-3 isoform (P = 0.028) and CSF 14-3-3 isoform (P = 0.04) (independent t-test, two-sided, data were logarithmically transformed to improve normality). Furthermore, the levels of CSF 14-3-3 protein in ACABM cases were significantly increased initially, and substantially decreased thereafter (Table 2).
|
P = 0.049, bacterial meningitis vs. aseptic meningitis groups; 
P = 0.032, bacterial meningitis vs. aseptic meningitis groups; 
P = 0.028, bacterial meningitis vs. aseptic meningitis groups; 
P = 0.04, bacterial meningitis vs. aseptic meningitis groups.Outcome and CSF 14-3-3 proteins
Regarding therapeutic outcome, 25 cases survived and four died of meningitis. At the time of final CSF study before discharge, the median value (IQR) of CSF 14-3-3-P protein and CSF 14-3-3 β, CSF 14-3-3
, CSF 14-3-3 , CSF 14-3-3 and CSF 14-3-3 isoforms among survivors were 0 (0, 25.9) DU, 0 (0, 5.9) DU, 0 (0, 0) DU, 0 (0, 0) DU, 0 (0, 4) DU and 0 (0, 23.6) DU, respectively. Among those who did not survive, the median values (IQR) were 52.8 (43.2, 521.7) DU, 26.2 (8.8, 275.5) DU, 157.3 (65.3, 431.3) DU, 0 (0, 297.1) DU, 19.0 (4.0, 271.9) DU and 43.1 (8.9, 681.8) DU, respectively.
Furthermore, before discharge, we compared the last CSF 14-3-3 protein levels between the fatal and non-fatal groups of the 29 adult ACABM cases. The statistical results revealed that the CSF 14-3-3 isoform in the fatal group was significantly higher than that in the non-fatal group (P = 0.004, independent t-test, two-side, data were logarithmically transformed to improve normality).
Outcome and prognostic factors
The potential prognostic factors of the 29 ACABM patients at the 3-month end-point are listed in Table 3. Statistical analysis of the clinical manifestations, laboratory data and neuro-imaging studies between the fatal and non-fatal groups revealed the following significant findings: CSF Glucose/blood glucose ratio (P = 0.04), CSF White cell count (P = 0.007) and CSF 14-3-3 isoform (P = 0.03). The variables used in the logistic regression included CSF Glucose/blood glucose ratio, CSF White cell count and CSF 14-3-3 isoform. After analysis of the above-mentioned variables, only CSF 14-3-3 isoform at the time of admission (P = 0.05, OR = 0.991, 95% CI = 0.982–1.00) was independently associated with outcome.
|
Furthermore, the area under the ROC curve for CSF 14-3-3
isoform, CSF Glucose/blood glucose ratio and CSF White cell count were 0.85 (P = 0.027, 95% CI = 0.68–1.02), 0.78 (P = 0.076, 95% CI = 0.51–1.06) and 0.83 (P = 0.037, 95% CI = 0.63–1.03), respectively. Thus, this model discriminated well between patients who survived and those who did not. |
Discussion |
|---|
|
TopSummaryIntroductionPatients and methodsResultsDiscussionAcknowledgementsReferences |
|---|
The 14-3-3 proteins have been shown to interact with various protein kinases, receptor proteins, enzymes, structural and cytoskeletal proteins, proteins involved in cell cycle and transcriptional control and proteins involved in apoptosis.15 In addition to their possible role in neuronal function, 14-3-3 proteins have recently attracted much interest due to their possible involvement in the pathophysiology of various neurologic disorders.6–9 One study indicates that the presence of 14-3-3 proteins in CSF may be due to the extensive destruction of brain tissue and leakage of normal cellular proteins into the CSF.16 Other studies have confirmed this hypothesis by detecting 14-3-3 proteins in patients with various neurologic disorders involving massive neural disruption and leakage of normal cellular proteins.6–9
In the present study, we examined the time course of CSF 14-3-3 proteins levels and its isoform in ACABM patients, and produced four major findings. First, all of the meningitis patients investigated had a positive result on admission and the levels of 14-3-3-P and 14-3-3 isoform, CSF 14-3-3 isoform and CSF 14-3-3 isoform in the bacterial meningitis group was significantly higher than those of the aseptic meningitis group. Second, the level of CSF 14-3-3 protein in the ACABM cases was significantly increased and then substantially decreased thereafter. Third, most of those who survived nearly cleared the 14-3-3 protein from the CSF before hospital discharge. Finally, 14-3-3 protein isoform levels on admission and before hospital discharge between those who survived and those who did not were statistically significant.
Although the sample size was not large, the numbers of variables considered for multiple logistic regression analysis was small. Furthermore, based on stepwise procedures, only one variable was selected as the important variable predicting outcome. Nonetheless, the maximum likelihood estimates of the coefficients were valid in the analysis.
In the patients who recovered, clearance of the 14-3-3 protein occurred before the other CSF parameters returned to normal levels, which suggests that 14-3-3 protein clearance may actually be the earliest CSF parameter to provide a distinct, reliable and favorable prognostic sign. Accordingly, in terms of pathophysiology, these findings also suggest that neuronal loss may actually cease before inflammatory markers decrease.
There are two main limitations in this study. First, inadequate isoform specificity of the commercially available antibodies used may have caused differences in isoform patterns. Improvement in terms of isoform specificity of the antibodies may provide more reliable results.17 Second, levels of CSF 14-3-3 protein isoform measured in this study were semi-quantitatively using western immuno-blotting. This justifies establishing alternative methods, like enzyme-linked immunosorbent assay (ELISA) procedures that will allow high-capacity screening and precise quantification. The use of isoform-specific ELISA should further improve the diagnostic specificity of 14-3-3 in CSF as a surrogate marker for neurologic disorders.18
On the basis of our results, serial 14-3-3 protein isoform seems to meet the major requirement for outcome prediction in the treatment of ACABM. The 14-3-3 protein assay can be considered as a neuro-pathologic marker that should be added to the panel of conventional CSF parameters.
|
Acknowledgements |
|---|
|
TopSummaryIntroductionPatients and methodsResultsDiscussionAcknowledgementsReferences |
|---|
This study was supported by grants from the National Science Council of Taiwan (Grants NSC 93-2314-B-182A-161 and NSC94-2314-B-182A-014).
Conflict of interest: None declared.
|
References |
|---|
|
TopSummaryIntroductionPatients and methodsResultsDiscussionAcknowledgementsReferences |
|---|
1. Lu CH, Huang CR, Chang WN, Chang CJ, Cheng BC, Lee PY, et al. Community-acquired bacterial meningitis in adults: the epidemiology, timing of appropriate antimicrobial therapy, and prognostic factors. Clin Neurol Neurosurg (2002) 104:352–8.[CrossRef][Web of Science][Medline]
2. Boston PF, Jackson P, Thompson RJ. Human 14-3-3 protein: radioimmunoassay, tissue distribution, and cerebrospinal fluid levels in patients with neurological disorders. J Neurochem (1982) 38:1475–82.[Web of Science][Medline]
3. Aitken A, Jones D, Soneji Y, Howell S. 14-3-3 proteins: biological function and domain structure. Biochem Soc Trans (1995) 23:605–11.[Web of Science][Medline]
4. Ichimura T, Isobe T, Okuyama T, Takahashi N, Araki K, Kuwano R, et al. Molecular cloning of cDNA coding for brain-specific 14-3-3 protein, a protein kinase-dependent activator of tyrosine and tryptophan hydroxylases. Proc Natl Acad Sci U S A (1988) 85:7084–8.
5. Isobe T, Ichimura T, Sunaya T, Okuyama T, Takahashi N, Kuwano R, et al. Distinct forms of the protein kinase-dependent activator of tyrosine and tryptophan hydroxylases. J Mol Biol (1991) 217:125–32.[CrossRef][Web of Science][Medline]
6. Sanchez-Juan P, Green A, Ladogana A, Cuadrado-Corrales N, Saanchez-Valle R, Mitrovaa E, et al. CSF tests in the differential diagnosis of Creutzfeldt-Jakob disease. Neurology (2006) 22; 67:637–43.
7. Wiltfang J, Otto M, Baxter HC, Bodemer M, Steinacker P, Bahn E, et al. Isoform pattern of 14-3-3 proteins in the cerebrospinal fluid of patients with Creutzfeldt-Jakob disease. J Neurochem (1999) 73:2485–90.[CrossRef][Web of Science][Medline]
8. Satoh J, Yukitake M, Kurohara K, Takashima H, Kuroda Y. Detection of the 14-3-3 protein in the cerebrospinal fluid of Japanese multiple sclerosis patients presenting with severe myelitis. J Neurol Sci (2003) 212:11–20.[CrossRef][Web of Science][Medline]
9. Saiz A, Graus F, Dalmau J, Pifarre A, Marin C, Tolosa E. Detection of 14-3-3 brain protein in the cerebrospinal fluid of patients with paraneoplastic neurological disorders. Ann Neurol (1999) 46:774–7.[CrossRef][Web of Science][Medline]
10. Bonora S, Zanusso G, Raiteri R, Monaco S, Rossati A, Ferrari S, et al. Clearance of 14-3-3 protein from cerebrospinal fluid heralds the resolution of bacterial meningitis. Clin Infect Dis (2003) 36:1492–5.[CrossRef][Web of Science][Medline]
11. Durand ML, Calderwood SB, Weber DJ, Miller SI, Southwick FS, Caviness VS Jr, et al. Acute bacterial meningitis in adults: a review of 493 episodes. N Engl J Med (1993) 328:21–8.
12. Hussein AS, Shafran SD. Acute bacterial meningitis in adults. A 12-year review. Medicine (2000) 79:360–8.[Medline]
13. Bottner A, Daneschnejad S, Handrick W, Schuster V, Liebert UG, Kiess W. A season of aseptic meningitis in Germany: epidemiologic, clinical and diagnostic aspects. Pediatr Infect Dis J (2002) 21:1126–32.[CrossRef][Web of Science][Medline]
14. Chang WN, Lu CH, Huang CR, Chuang YC. Mixed infections in adult bacterial meningitis. Infection (2000) 28:8–12.[CrossRef][Web of Science][Medline]
15. Yaffe MB. How do 14-3-3 proteins work? — Gatekeeper phosphorylation and the molecular anvil hypothesis. FEBS Lett (2002) 513:53–7.[CrossRef][Web of Science][Medline]
16. Hsich G, Kenney K, Gibbs CJ, Lee KH, Harrington MG. The 14-3-3 brain protein in cerebrospinal fluid as a marker for transmissible spongiform encephalopathies. N Engl J Med (1996) 335:24–30.
17. Shiga Y, Wakabayashi H, Miyazawa K, Kido H, Itoyama Y. 14-3-3 protein levels and isoform patterns in the cerebrospinal fluid of Creutzfeldt-Jakob disease patients in the progressive and terminal stages. J Clin Neurosci (2006) 13:661–5.[CrossRef][Web of Science][Medline]
18. Baxter HC, Fraser JR, Liu WG, Forster JL, Clokie S, Steinacker P, et al. Specific 14-3-3 isoform detection and immunolocalization in prion diseases. Biochem Soc Trans (2002) 30:387–91.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||