| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Syn-X Pharma Inc., 6354 Viscount Rd., Mississauga, Ontario, L4V 1H3 Canada
aauthor for correspondence: fax 905-677-1674, e-mail empty{at}ICA.net
Glutamine synthetase (GS) is a ubiquitous enzyme present at high concentrations in liver, muscle, kidney, and brain (1). GS in the brain is astrocyte specific, is involved in the detoxification of ammonia and glutamate, and is overexpressed after brain injury (2). Monomeric GS protein was found in 38 of 39 cerebrospinal fluid samples obtained from Alzheimer disease (AD) patients (3), and the concentration of GS in the lumbar cerebral spinal fluid of AD patients was increased significantly but nonspecifically (4). GS has also been reported in serum (4).
Markers for AD have poor sensitivity and specificity for overt or preclinical AD (5)(6). We have developed an ELISA that is rapid, highly sensitive, and GS specific. This ELISA, together with the S100B and neuron-specific enolase (NSE) ELISAs, was used to retrospectively assess serum GS in healthy controls, AD patients, and other dementia patients.
Recombinant human GS (rhGS) cDNA was purchased from ATCC (ATCC No. 409071). The full-length rhGS open reading frame was obtained by PCR and subcloning in pET28a (Ndel/XhoI). The construct included a polyhistidine tag at the N-terminal domain of the rhGS open reading frame and no extra sequence at the C-terminal domains. The protein was expressed in E. coli BL21 (DE3) (7). We prepared extracts and solubilized products as described (8). Affinity purification was performed by nickel-nitrilotriacetic acid chromatography.
Female Balb/c mice (7–8 weeks of age) were immunized by subcutaneous injection of 50 µg of rhGS emulsified in complete Freund’s adjuvant. An intraperitoneal booster injection was given 3 weeks later (50 µg in incomplete Freund’s adjuvant) and then at 2- to 3-week intervals (three times or more) with a final injection of 50 µg of antigen in buffer via intraperitoneal route 3 days before cell fusion. Hybridoma cultures were screened by ELISA, and positive cultures were cloned at least twice with limiting dilutions as described (9). Monoclonal antibodies (MAbs) were purified by protein G chromatography (Phamacia Biotech), and subclass was determined (Mouse Typer®; Bio-Rad).
Serum S100B and NSE
were analyzed by two ELISAs (SMART S100B and SMART NSE; Syn-X Pharma, Inc.). For GS, MaxiSorpTM plates (NUNC) were coated for 16–24 h at 4–8 °C with 10 mg/L MAb 1G3 in 0.125 mL/well of coating buffer (20 mmol/L NaHPO4 buffer, pH 7.4). The plates were washed once with wash buffer (phosphate-buffered saline containing 0.5 g/L Tween 20, pH 7.4), followed by another wash with phosphate-buffered saline, pH 7.4, and the free binding sites were blocked at room temperature for 1 h with 0.25 mL of blocking buffer (phosphate-buffered saline containing 5 g/L bovine serum albumin), after which time the plates were washed three times. rhGS was used as the calibrator. The purified rhGS was quantified by the Bradford method (Bio-Rad). The rhGS stock solution was diluted to 100 µg/L in the calibrator diluent (human serum from a pool of 15 apparently healthy donors) and kept frozen in 100-µL aliquots at -75 °C until use. After thawing, the rhGS solution was diluted further to 0–2 µg/L in the diluent supplemented immediately before use with 4 mmol/L Pefa-blocSC and PSC-protector (Boehringer Mannheim) and 1 mL/L ProClin-300 (Supelco) followed by filtration through a sterile 0.2-µm membrane.
Incubation steps of the ELISA were at room temperature on an orbital shaker unless otherwise stated. Calibrators and serum samples were pipetted in duplicate at 50 µL/well, followed by 50 µL of incubation buffer (50 mmol/L NaHPO4, 450 mmol/L NaCl, 0.5 mL/L Tween 20, pH 8.0) per well. After 30 min, the plate was washed three times, and 100 µL of detector antibody (biotinylated MAb 5G4) was added and incubated for 30 min. The plate was again washed three times, followed by an addition of peroxidase-conjugated streptavidin (Boehringer Mannheim). The plate was incubated for 15 min. After washing, the bound enzyme activity was visualized by the addition of 100 µL of 3,3',5,5'-tetramethylbenzidine substrate solution (MOSS). Substrate hydrolysis was continued for 5 min at room temperature in the dark. The reaction was stopped with 100 µL of 0.5 mol/L H2SO4. Absorbance was measured in a microplate reader (Bio-Rad) at 450 nm. The Bio-Rad microplate reader software was used to calculate the GS concentrations of the unknown samples by reference to a calibration curve fitted to a quadratic function.
From three cell fusions, 13 clones were established: 1 produced IgM, k; 1 IgA, k; 4 IgG2b, k; and 7 IgG1, k. None cross-reacted with any isoform of S100 protein, NSE, NSE
, myelin basic protein, or glial fibrillary acidic protein at 10 mg/L in the ELISA. Pairing analysis was performed on all clones, using a BIAcore biosensor, and two IgG1 clones, 1G3 and 5G4, were chosen for ELISA assay development. To examine the specificity for brain GS, cross-reactivities of the selected capture MAb (1G3) were tested by Western blotting against brain, liver, and muscle tissue extracts with 5 µg of protein. MAb 1G3 reacted with only the brain GS (data not shown). The total GS assay time was 80 min with a detection limit of 0.017 µg/L GS (zero calibrator, + 2 SD; n = 20). The calibration curve was 0–2 µg/L. Intraassay (n = 32) and interassay (n = 20) imprecision (CV) was assessed with a pool of four serum samples with 0.15 and 1 µg/L added rhGS. The maximum CV was 7.25%. No hook effect was observed at a GS concentration of 500 µg/L, although the ELISA signal reached its maximum at 100 µg/L. Recovery of rhGS (0.15 and 0.4 µg/L) added to three sera was 90–103% (mean, 97%). To assess linearity, four patient sera with high GS, diluted with the zero calibrator (to 2-, 4-, 8-, and 16-fold dilutions), were measured in triplicates. The measured values were 87–112% of those expected. In serum and plasma samples from six blood donors with added rhGS (0.5 µg/L), heparin plasma results were 84–97% of serum values, citrated plasma results were 90–108% of serum, and EDTA plasma results were 120–146% of serum values.
We obtained serum from 153 healthy donors (18–87 years; mean, 71 ± 10 years): 80 were from employees of Syn-X Pharma Inc.; 40 were purchased from Intergen (donors with no known medical conditions); and 33 samples were age-matched controls from the clinics where samples from dementia patients were obtained. For reference values, we used the mean +2 SD: GS = 0.022 µg/L, S100B = 0.02 µg/L, and NSE = 8.34 µg/L.
Serum samples were collected prospectively between November 2000 and March 2001 from dementia patients who were a subgroup of the controls from a previous study "Investigating the Clinical Values of Serum Biochemical Markers in Patients with Acute Cerebrovascular Disease" (St. Joseph’s Hospital, Hamilton, Ontario, and St. Joseph’s Health Centre, London, Ontario). The study evaluated 38 consecutive patients, male or female of any age with possible or probable AD as assessed by the National Institute of Neurological Disorders and Stroke–Alzheimer’s Disease and Related Disorders Association diagnostic criteria. Information on disease severity was collected with the Mental State Examination and other routine tests. Patients (n = 24) were diagnosed with AD and 14 with various types of dementia other than AD. The diagnoses of AD and other dementia were made by clinicians blinded to the biochemical marker results. Likewise, the technologists performing biochemical marker assays were blinded to the diagnosis of patients. Exclusion criteria (part of the Institutional Review Board protocol) were as follows: cerebral tumor, hypothyroidism, vitamin B12 deficiency, dementia by chronic substance abuse, and HIV-related dementia. The mean age of the patients with dementia was 74 ± 11 years (range, 53–91 years). Blood samples were obtained and centrifuged after clotting; aliquots of sera were stored frozen at -75 °C until analysis. Institutional Review Board approval was obtained to collect blood samples, and each study participant or their power of attorney signed an informed consent form.
Healthy individuals in the sixth decade had higher mean concentrations of GS than those in other decades (Fig. 1
; P <0.001, Tukey–Kramer honestly significant difference test). Those with high GS in the sixth decade may have had undiagnosed AD, whereas AD patients in the seventh decade had developed clinical signs, which would have excluded them from the healthy population.
|
Serum GS did not differ between sexes. Results in patients are shown in Table 1
. GS correlated with the severity of AD, i.e., Mini-Mental State Examination score (r = 0.65), whereas no significant correlation was observed with NSE. The positivity rate of S100B was very low, but when this marker is increased, it may be an indication of ongoing destruction of astrocytes in the brain attributable to acute events as observed in patients AD-10 and AD-18.
|
Of the 14 non-AD dementia samples, only 1 had increased concentrations of both GS and S100B, whereas 7 samples had increased NSE (Table 1
). The GS-positive patient (DC-008) with dementia with Lewy bodies may have had undetected AD; many patients with Lewy body dementia have concomitant AD pathology (10). The area under the ROC curve for GS was 0.950 (SE, 0.021; 95% confidence interval, 0.911–0.975).
Although the increase of NSE is not specific for AD, NSE may contribute as an indicator of the ongoing destruction of neurons that occurs in AD (11). GS is a sensitive and specific marker for AD when the protein is analyzed in blood. The serum concentration of S100B is not increased in most dementia patients, but it may indicate acute events (e.g., transient ishemic attack, stroke, and hypoxia leading to ischemia), which are common among older individuals who are the target population of AD. Our preliminary studies of the GS ELISA indicate an encouraging sensitivity and specificity for AD. Studies on a large number of serum samples from patients with various conditions (AD, non-AD dementia, and diseases involving liver and muscle) are in progress.
Acknowledgments
We thank Drs. M. Gagnon and C. Hobbs of St. Joseph’s Hospital, Hamilton, Canada, and Drs. J. Howard and B. Gaw of St. Joseph’s Health Centre, London, Canada, for the supply of serum samples. We also thank Kathy Ciebiera for performing the NSE and S100B ELISA analyses.
References
-aminobutyric acid A receptor in the rat brain. J Comp Neurol 1994;350:260-271.[ISI][Medline]
[Order article via Infotrieve]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
