- Email: [email protected]
A sensitive assay of transthyretin (prealbumin) in human cerebrospinal fluid in nanogram amounts by ELISA
Clinica Chimica Acta, 197 (1991) 19-26 Elsevier Science Publishers B.V. ADONIS 000989819100077W
19
CCA 04936
A sensitive assay of transthyretin ( prealbumin) in human cerebrospinal fluid in nanogram amounts by ELISA G.T. Vatassery, H.T. Quach, W.E. Smith, B.A. Benson and J.H. Eckfeldt Research, Neurology and GRECC Services, VA Medical Center, Minneapolis, MN and the Departments of Neurology and Laboratory Medicine and Pathology, Medical School, University of Minnesota, Minneapolis, MN (USA) (Received 17 September 1990; accepted 7 November 1990)
Key words: Transthyretin; Prealbumin; Cerebrospinal fluid; ELISA
A sensitive ELISA method for determining transthyretin (prealbumin) in human cerebrospinal fluid (CSF) is described. The method utilizes goat antihuman transthyretin antibody (IgG fraction) for capture and peroxidase conjugated antibody for color development. The assay has a linear range of l-4 ng transthyretin added per well. The within-day and between-day coefficients of variation are 5.1 and 6.18, respectively. The concentration of transthyretin in CSF (ranging from 5 to 20 mg per L) correlated significantly with the corresponding serum concentrations (range 170-420 mg/l). This suggests that synthesis of transthyretin in the brain and peripheral tissues is under similar biological control in normal subjects. The transthyretin concentrations in CSF did not correlate with total CSF protein concentration or age of the subject.
Introduction
Human transthyretin, previously termed prealbumin, is a protein with four identical subunits and a molecular weight of approximately 55 000. This protein is known to be actively involved in the transport of thyroxine and retinol [1,2]. Correspondence to: G.T. Vatassery, Ph.D., Research Service (151), VA Medical Center, One Veterans Drive, Minneapolis, MN 55417, USA.
20
Transthyretin is not known to have any specific function in the central nervous system even though it seems to be synthesized specifically by the choroid plexus and secreted into the cerebrospinal fluid (CSF) [l]. Therefore, concentration of transthyretin in CSF may serve as a useful marker of the function of choroid plexus in man and thus provide an index of the status of the blood-CSF barrier. A sensitive routine method for determination of transthyretin in CSF would also have other practical diagnostic applications (see Discussion). Transthyretin in serum has been analyzed by rate nephelometry, radial immunodiffusion, immunoturbidimetry and radioimmunoassay [3-81. A rapid, reliable and routine method which does not require expensive instruments or involve the use of radioactivity for determination of transthyretin would make this assay more available in smaller clinical laboratories. Therefore, we have developed a sensitive ELISA for transthyretin. The method can be used for determination of transthyretin in CSF, serum or other physiological fluids.
Materials and methods
Chemicals Reagent chemicals of high purity were purchased from standard commercial sources. Tween 20 was purchased as a 10 g/d1 peroxide-free aqueous solution from Pierce Chemical Company (Rockford, IL, USA). Thimerosal and transthyretin were from Sigma Chemical Company (St. Louis, MO, USA). The immunochemicals were purchased from the following sources: goat antihuman transthyretin antibody as an IgG fraction from Atlantic Antibodies (Scarborough, ME, USA); sheep serum from Colorado Serum Company (Denver, CO, USA); Immulon I plates from Dynatech Laboratories (Chantilly, VA, USA); antihuman prealbumin peroxidase conjugate from Serotec (Kidington, Oxford, UK).
Subjects All sample collections were done under the guidelines of the Institutional Review Board for human studies after the project was approved by this board. The CSF samples which were used to compare the two techniques of analysis for transthyretin were collected at random from specimens submitted for analysis to a university clinical chemistry laboratory. Samples of CSF for all the other studies were obtained from 30 males aged 24 to 70 years. These subjects were being seen in a neuroradiology clinic, primarily for evaluation of degenerative disc disease. Subjects with neurological diseases (such as nervous system tumors, cerebrovascular disease and multiple sclerosis) as well as systemic metabolic diseases like diabetes were excluded. The CSF samples were collected, centrifuged and stored at - 30 o C before analysis. If any trace of a red pellet was observed after centrifugation (suggesting contamination with blood), the CSF was not used in the study. The total CSF protein concentrations were all below 600 mg/l.
21
Biochemical assays
Total protein concentrations in CSF were determined by the method of Markwell et al. [9]. The ELISA technique for transthyretin reported in this paper was checked against a standard kinetic nephelometric method using an automated nephelometer (Beckman Array, Beckman Instruments Inc., Fullerton, CA, USA). Undiluted or six-fold diluted samples of CSF were used in contrast to the 36-fold dilutions used for serum assays. Solutions for the transythretin assay
A. Coating buffer. 50 mmol/l sodium carbonate/bicarbonate buffer at pH 9.6 with 0.1% thimerosal (sodium ethylmercurithiosalicylate). This solution is stable at room temperature for up to 2 wk. B. Wash buffer. Phosphate-buffered saline at pH 7.4 containing 0.05% Tween 20. C. Backcoat buffer. Sheep serum was heated at 55°C for 30 minutes and filtered through 0.45 pm Millipore filters. The heated sheep serum (0.9 ml) was then mixed with 29.1 ml of phosphate-buffered saline. D. Assay buffer. The assay buffer was prepared by mixing 0.5 ml heated sheep serum, 0.5 ml 10% Tween 20 and 49 ml of phosphate-buffered saline. E. Enzyme-antibody buffer. This solution was prepared by adding 2.5 ml of heated sheep serum to 47.5 ml phosphate-buffered saline. F. Color reagent. Ortho phenylenediamine (OPD) reagent was purchased from Abbot Laboratories, North Chicago, IL. Two tablets of the reagent were dissolved in 10 ml diluent as recommended by the manufacturer. G. Standard solution. Prealbumin solutions in the range of lo-30 ng per ml were made by diluting a stock solution of prealbumin in assay buffer (concentration of stock = 220 mg/l). Procedure
Aliquots of the capture antibody (0.2 pg per well) in the coating buffer were pipetted into all the wells of an Immulon I plate. After incubating for 2 h at room temperature (20-22”(Z) the plates were covered with parafilm and placed in a refrigerator at - 4” C for at least 48 h. On the day of assay the plates were washed with 250 ~1 of wash solution and then incubated at room temperature for 45 minutes after the addition of 280 ~1 of back-coat buffer. The solution was removed and the wells were washed twice. Aliquots (100 ~1 each) of diluted CSF (dilution 1:750) or prealbumin standard solutions (1 to 4 ng per well) were pipetted into the wells and incubated at room temperature for 90 min. After washing 4 times, the peroxidase-conjugated antibody solution (100 ~1) in the enzyme-antibody buffer was added and incubated for 60 minutes at room temperature. The wells were washed 4 times with wash buffer and 3 times with phosphate-buffered saline. To each well 100 ~1 of the OPD color reagent was added. Color was developed for 30 min and the absorbance read at 405 nm for nonstopped and 492 nm for acid-stopped reactions using a microplate reader. The results were calculated using a standard curve.
22
Standard curves with the logarithm of transthyretin on the Y axis and absorbance on the X axis are linear within the limited range of l-4 ng of transthyretin added per well. The correlation coefficients are usually more than 0.95. Samples of controls, sera and CSF have to be diluted appropriately with assay buffer in order to be within range of the standards used. Some of the experimental parameters such as the amount of capture antibody and antibody-peroxidase conjugate can be altered within a range of 0.5 to 2 times the recommended amount. The final amounts to be used are chosen based on obtaining lower blanks and higher absorbance values for the samples and standards. Techniques involving the use of OPD color reagent can be done with or without stopping the reaction with sulfuric acid. Comparison of standard curves obtained under these two conditions showed that the two procedures were nearly identical with respect to the linear range of concentrations and the coefficients of variation of the experimental data. It is recommended that the plates be read at a fixed time interval after the addition of the OPD reagent since this eliminates one additional step (stopping the reaction) in the procedure. The amount of capture antibody needed per well was determined by adding varying amounts of the antibody and measuring the absorbance produced after the OPD reaction with a constant level of transthyretin. It was found that 0.2 pg capture antibody was sufficient to get maximum color with additional amounts of up to 3 pg per well producing no enhancement in the final color. Many procedures also use bovine serum albumin as a blocking agent instead of heated sheep serum. The two blocking agents were very similar with regard to the development of low and reproducible blanks and therefore the sheep serum was used in the final procedure.
60
1
Transthyretin
concentrations WWL)
from
ELISA
Fig. 1. Comparison of the results of analysis of transthyretin concentrations fluids by ELISA and nephelometric methods.
in human cerebrospinal
23
5
8!
0
20
I 30
I 40 Age
Fig. 2. Correlation
between
I 50
I 60
I 70
I 80
(years)
age and transthyretin concentrations human, male subjects.
in the cexebrospinal
fluid of adult,
As expected, the development of color with the OPD reagent is dependent on the time cf incubation. The standard curves are linear between 30 and 60 min with the absorbance at 60 min being about 20% higher. The final recommended procedure calls for a color development time of 30 min. The coefficients of variation for the recommended assay were: within-day 5.1% and between-day 6.1%. The ELISA technique for transthyretin was then tested against a standard automated nephelometric assay. Samples of CSF collected at random from a clinical chemistry laboratory were analyzed by both methods and the results in Fig. 1 show that the two methods are in close agreement. Results of assays of transthyretin in serum also showed similar good agreement.
500 l
400 -
l
l
300 -
200 -
100 -
l
Concentration
of cerebrospinal
Fig. 3. Relationship
between
transthyretin
transthyretin fluid
in
(mg/L)
concentrations in serum and cereborspinal human, male subjects.
fluid from adult,
24
Samples of CSF from 30 human, male, control subjects were analyzed for transthyretin by the recommended technique. The age of the subject did not correlate significantly with transthyretin concentrations in CSF (Fig. 2). Among these subjects there was no significant correlation between total protein and transthyretin concentrations within the CSF. This is to be expected since it is well-known that the bulk of total protein in CSF originates from plasma whereas most of the transthyretin is synthesized within the choroid plexus. The concentrations of transthyretin in serum and CSF showed a significant correlation as shown in Fig. 3 (r2 = 0.626, P -C0.01). Discussion
The procedure for transthyretin assay is relatively simple and reproducible. The variability of results of ELISA is generally more than that of assays using chromatography or spectrophotometry. Therefore, our laboratory routinely analyzes samples in quadruplicate. Transthyretin concentrations in CSF could be used to assess the blood-CSF barrier component within the blood-brain-CSF barriers. Similarly, total protein concentrations in CSF can be a measure of the permeability of the blood-brain barrier. The lack of correlation between CSF total protein and transthyretin concentrations is due to the fact that most of the transthyretin is synthesized and secreted by choroid plexus. The observation that there was no significant correlation between CSF transthyretin concentrations and age in this population suggests that choroid plexus function is not altered with age. Even though the bulk of transthyretin in CSF is produced locally by the choroid plexus, there was a significant relationship between serum and CSF transthyretin levels among these ‘normal subjects’ (Fig. 3). Whether this relationship holds good under pathological conditions should be determined separately. It is entirely conceivable that since the majority of transthyretin in CSF is locally synthesized, its concentration in CSF may not reflect peripheral pathological conditions such as liver disease and protein-caloric malnutrition where serum transthyretin concentrations are of diagnostic value [lO,ll]. In any case, it is important to measure transthyretin concentrations in both CSF and serum when the data is used for clinical diagnosis. A few reports in the literature indicate that CSF transthyretin levels may be of diagnostic significance in neurologic disorders. Weisner and Roethig [12] have suggested that the concentration of transthyretin in CSF is a useful indicator of CSF circulation disorders. Studies have shown that transthyretin is a constituent of senile amyloid deposits [13] and that transthyretin is reduced in CSF from patients with Alzheimer’s disease [14]. An increase in total CSF protein accompanied by a decrease in CSF transthyretin have been reported in pyogenic meningitis and primary and metastatic tumors [15]. The sensitive ELISA method for determination of transthyretin in CSF reported in this paper could be useful in the diagnosis of the disorders mentioned above.
Acknowledgements The investigations were supported by grant AGO6309 stitute of Aging and by research funds from the Department
from the National Inof Veterans Affairs.
References DW. Transthyretin in familial amyloidotic polyneuropathy: 1 Saraiva MJM, Costa JP, Goodman genetic and functional aspects. Adv Nemo1 1989;48:189-200. 2 Herbert J, Wilcox JN, Pham KTC, Fremeau RT, Zeviani M, Dwork A, Soprano DR, Makover A, Goodman DS, Zimmerman EA, Roberts JL, Schon EA. Transthyretin: a choroid plexus-specific transport protein in human brain. Neurology 1986;36:900-911. DA. Turbidimetric determination of transthyretin (prealbumin) with a 3 Hamlin CR, Pankowsky centrifugal analyzer. Clin Chem 1987;33:144-146. BM, Munson S. Rate nephelometry and radial immunodiffusion compared for measuring 4 Goldsmith serum prealbumin. Clin Chem 1987;33:161-163. assay of transthyretin by random-access 5 Pressac M, Vignoli L, Aymard P. Immunoturbidimetric analysis. Clin Chem 1988;34:176. of transthyretin (prealbumin) in 6 Ledue TB, Rifai N, Irish GR, Silverman LM. Inununoturbidimetry human serum. Clin Chem 1987;33:1260. RC, Fetterhoff TJ, Luckey DW. Three immunoassay methods evaluated for quantifying I McCarthy prealbumin (transthyretin) in serum. Clin Chem 1987;33:1430-1431. of human prealbumin in body fluids. Clin 8 Cowley DM, Davey JF, Hjelm NM. Radioimmunoassay Chim Acta 1983;134:69-76. of the Lowry procedure to 9 Markwell MAK, Haas SM, Bieber LL, Tolbert NE. A modification simplify protein determinations in membrane and lipoprotein samples. Anal Biochem 1978;87:206210. DR, Halliwell RP, Smith MG, Parke DV. Serum prealbumin as an index of liver function 10 Hutchinson in human hepatobiliary disease. Clin Chim Acta 1981;114:69-74. malnutrition. Clin Chem 1984;30:1286-1299. 11 Haider M, Haider SQ. Assessment of protein-calorie of prealbumin in cerebrospinal fluid (CSF), indicator of 12 Weisner B, Roethig HJ. The concentration CSF circulation disorders. Eur Neurol 1983;22:96-105. P, Rubinow A, Cohen AS, Brun A, Kemper TL. Senile cerebral 13 Shirahama T, Skinner M, Westennark amyloid: prealbumin as a common constituent in the neuritic plaque, in the neurofibrillary tangle and in the microangiopathic lesion. Am J Path01 1982;107:41-50. (transthyretin) in CSF of severely demented patients with Alxheimer’s 14 Riisoen H. Reduced prealbumin disease. Acta Neurol Stand. 1988;78:455-459. levels of cerebrospinal fluid in neurological disorders. Ind 15 Bimanpalli MV, Ghaswala PS. Prealbumin J Med Res 1988;88:26-164.
19
CCA 04936
A sensitive assay of transthyretin ( prealbumin) in human cerebrospinal fluid in nanogram amounts by ELISA G.T. Vatassery, H.T. Quach, W.E. Smith, B.A. Benson and J.H. Eckfeldt Research, Neurology and GRECC Services, VA Medical Center, Minneapolis, MN and the Departments of Neurology and Laboratory Medicine and Pathology, Medical School, University of Minnesota, Minneapolis, MN (USA) (Received 17 September 1990; accepted 7 November 1990)
Key words: Transthyretin; Prealbumin; Cerebrospinal fluid; ELISA
A sensitive ELISA method for determining transthyretin (prealbumin) in human cerebrospinal fluid (CSF) is described. The method utilizes goat antihuman transthyretin antibody (IgG fraction) for capture and peroxidase conjugated antibody for color development. The assay has a linear range of l-4 ng transthyretin added per well. The within-day and between-day coefficients of variation are 5.1 and 6.18, respectively. The concentration of transthyretin in CSF (ranging from 5 to 20 mg per L) correlated significantly with the corresponding serum concentrations (range 170-420 mg/l). This suggests that synthesis of transthyretin in the brain and peripheral tissues is under similar biological control in normal subjects. The transthyretin concentrations in CSF did not correlate with total CSF protein concentration or age of the subject.
Introduction
Human transthyretin, previously termed prealbumin, is a protein with four identical subunits and a molecular weight of approximately 55 000. This protein is known to be actively involved in the transport of thyroxine and retinol [1,2]. Correspondence to: G.T. Vatassery, Ph.D., Research Service (151), VA Medical Center, One Veterans Drive, Minneapolis, MN 55417, USA.
20
Transthyretin is not known to have any specific function in the central nervous system even though it seems to be synthesized specifically by the choroid plexus and secreted into the cerebrospinal fluid (CSF) [l]. Therefore, concentration of transthyretin in CSF may serve as a useful marker of the function of choroid plexus in man and thus provide an index of the status of the blood-CSF barrier. A sensitive routine method for determination of transthyretin in CSF would also have other practical diagnostic applications (see Discussion). Transthyretin in serum has been analyzed by rate nephelometry, radial immunodiffusion, immunoturbidimetry and radioimmunoassay [3-81. A rapid, reliable and routine method which does not require expensive instruments or involve the use of radioactivity for determination of transthyretin would make this assay more available in smaller clinical laboratories. Therefore, we have developed a sensitive ELISA for transthyretin. The method can be used for determination of transthyretin in CSF, serum or other physiological fluids.
Materials and methods
Chemicals Reagent chemicals of high purity were purchased from standard commercial sources. Tween 20 was purchased as a 10 g/d1 peroxide-free aqueous solution from Pierce Chemical Company (Rockford, IL, USA). Thimerosal and transthyretin were from Sigma Chemical Company (St. Louis, MO, USA). The immunochemicals were purchased from the following sources: goat antihuman transthyretin antibody as an IgG fraction from Atlantic Antibodies (Scarborough, ME, USA); sheep serum from Colorado Serum Company (Denver, CO, USA); Immulon I plates from Dynatech Laboratories (Chantilly, VA, USA); antihuman prealbumin peroxidase conjugate from Serotec (Kidington, Oxford, UK).
Subjects All sample collections were done under the guidelines of the Institutional Review Board for human studies after the project was approved by this board. The CSF samples which were used to compare the two techniques of analysis for transthyretin were collected at random from specimens submitted for analysis to a university clinical chemistry laboratory. Samples of CSF for all the other studies were obtained from 30 males aged 24 to 70 years. These subjects were being seen in a neuroradiology clinic, primarily for evaluation of degenerative disc disease. Subjects with neurological diseases (such as nervous system tumors, cerebrovascular disease and multiple sclerosis) as well as systemic metabolic diseases like diabetes were excluded. The CSF samples were collected, centrifuged and stored at - 30 o C before analysis. If any trace of a red pellet was observed after centrifugation (suggesting contamination with blood), the CSF was not used in the study. The total CSF protein concentrations were all below 600 mg/l.
21
Biochemical assays
Total protein concentrations in CSF were determined by the method of Markwell et al. [9]. The ELISA technique for transthyretin reported in this paper was checked against a standard kinetic nephelometric method using an automated nephelometer (Beckman Array, Beckman Instruments Inc., Fullerton, CA, USA). Undiluted or six-fold diluted samples of CSF were used in contrast to the 36-fold dilutions used for serum assays. Solutions for the transythretin assay
A. Coating buffer. 50 mmol/l sodium carbonate/bicarbonate buffer at pH 9.6 with 0.1% thimerosal (sodium ethylmercurithiosalicylate). This solution is stable at room temperature for up to 2 wk. B. Wash buffer. Phosphate-buffered saline at pH 7.4 containing 0.05% Tween 20. C. Backcoat buffer. Sheep serum was heated at 55°C for 30 minutes and filtered through 0.45 pm Millipore filters. The heated sheep serum (0.9 ml) was then mixed with 29.1 ml of phosphate-buffered saline. D. Assay buffer. The assay buffer was prepared by mixing 0.5 ml heated sheep serum, 0.5 ml 10% Tween 20 and 49 ml of phosphate-buffered saline. E. Enzyme-antibody buffer. This solution was prepared by adding 2.5 ml of heated sheep serum to 47.5 ml phosphate-buffered saline. F. Color reagent. Ortho phenylenediamine (OPD) reagent was purchased from Abbot Laboratories, North Chicago, IL. Two tablets of the reagent were dissolved in 10 ml diluent as recommended by the manufacturer. G. Standard solution. Prealbumin solutions in the range of lo-30 ng per ml were made by diluting a stock solution of prealbumin in assay buffer (concentration of stock = 220 mg/l). Procedure
Aliquots of the capture antibody (0.2 pg per well) in the coating buffer were pipetted into all the wells of an Immulon I plate. After incubating for 2 h at room temperature (20-22”(Z) the plates were covered with parafilm and placed in a refrigerator at - 4” C for at least 48 h. On the day of assay the plates were washed with 250 ~1 of wash solution and then incubated at room temperature for 45 minutes after the addition of 280 ~1 of back-coat buffer. The solution was removed and the wells were washed twice. Aliquots (100 ~1 each) of diluted CSF (dilution 1:750) or prealbumin standard solutions (1 to 4 ng per well) were pipetted into the wells and incubated at room temperature for 90 min. After washing 4 times, the peroxidase-conjugated antibody solution (100 ~1) in the enzyme-antibody buffer was added and incubated for 60 minutes at room temperature. The wells were washed 4 times with wash buffer and 3 times with phosphate-buffered saline. To each well 100 ~1 of the OPD color reagent was added. Color was developed for 30 min and the absorbance read at 405 nm for nonstopped and 492 nm for acid-stopped reactions using a microplate reader. The results were calculated using a standard curve.
22
Standard curves with the logarithm of transthyretin on the Y axis and absorbance on the X axis are linear within the limited range of l-4 ng of transthyretin added per well. The correlation coefficients are usually more than 0.95. Samples of controls, sera and CSF have to be diluted appropriately with assay buffer in order to be within range of the standards used. Some of the experimental parameters such as the amount of capture antibody and antibody-peroxidase conjugate can be altered within a range of 0.5 to 2 times the recommended amount. The final amounts to be used are chosen based on obtaining lower blanks and higher absorbance values for the samples and standards. Techniques involving the use of OPD color reagent can be done with or without stopping the reaction with sulfuric acid. Comparison of standard curves obtained under these two conditions showed that the two procedures were nearly identical with respect to the linear range of concentrations and the coefficients of variation of the experimental data. It is recommended that the plates be read at a fixed time interval after the addition of the OPD reagent since this eliminates one additional step (stopping the reaction) in the procedure. The amount of capture antibody needed per well was determined by adding varying amounts of the antibody and measuring the absorbance produced after the OPD reaction with a constant level of transthyretin. It was found that 0.2 pg capture antibody was sufficient to get maximum color with additional amounts of up to 3 pg per well producing no enhancement in the final color. Many procedures also use bovine serum albumin as a blocking agent instead of heated sheep serum. The two blocking agents were very similar with regard to the development of low and reproducible blanks and therefore the sheep serum was used in the final procedure.
60
1
Transthyretin
concentrations WWL)
from
ELISA
Fig. 1. Comparison of the results of analysis of transthyretin concentrations fluids by ELISA and nephelometric methods.
in human cerebrospinal
23
5
8!
0
20
I 30
I 40 Age
Fig. 2. Correlation
between
I 50
I 60
I 70
I 80
(years)
age and transthyretin concentrations human, male subjects.
in the cexebrospinal
fluid of adult,
As expected, the development of color with the OPD reagent is dependent on the time cf incubation. The standard curves are linear between 30 and 60 min with the absorbance at 60 min being about 20% higher. The final recommended procedure calls for a color development time of 30 min. The coefficients of variation for the recommended assay were: within-day 5.1% and between-day 6.1%. The ELISA technique for transthyretin was then tested against a standard automated nephelometric assay. Samples of CSF collected at random from a clinical chemistry laboratory were analyzed by both methods and the results in Fig. 1 show that the two methods are in close agreement. Results of assays of transthyretin in serum also showed similar good agreement.
500 l
400 -
l
l
300 -
200 -
100 -
l
Concentration
of cerebrospinal
Fig. 3. Relationship
between
transthyretin
transthyretin fluid
in
(mg/L)
concentrations in serum and cereborspinal human, male subjects.
fluid from adult,
24
Samples of CSF from 30 human, male, control subjects were analyzed for transthyretin by the recommended technique. The age of the subject did not correlate significantly with transthyretin concentrations in CSF (Fig. 2). Among these subjects there was no significant correlation between total protein and transthyretin concentrations within the CSF. This is to be expected since it is well-known that the bulk of total protein in CSF originates from plasma whereas most of the transthyretin is synthesized within the choroid plexus. The concentrations of transthyretin in serum and CSF showed a significant correlation as shown in Fig. 3 (r2 = 0.626, P -C0.01). Discussion
The procedure for transthyretin assay is relatively simple and reproducible. The variability of results of ELISA is generally more than that of assays using chromatography or spectrophotometry. Therefore, our laboratory routinely analyzes samples in quadruplicate. Transthyretin concentrations in CSF could be used to assess the blood-CSF barrier component within the blood-brain-CSF barriers. Similarly, total protein concentrations in CSF can be a measure of the permeability of the blood-brain barrier. The lack of correlation between CSF total protein and transthyretin concentrations is due to the fact that most of the transthyretin is synthesized and secreted by choroid plexus. The observation that there was no significant correlation between CSF transthyretin concentrations and age in this population suggests that choroid plexus function is not altered with age. Even though the bulk of transthyretin in CSF is produced locally by the choroid plexus, there was a significant relationship between serum and CSF transthyretin levels among these ‘normal subjects’ (Fig. 3). Whether this relationship holds good under pathological conditions should be determined separately. It is entirely conceivable that since the majority of transthyretin in CSF is locally synthesized, its concentration in CSF may not reflect peripheral pathological conditions such as liver disease and protein-caloric malnutrition where serum transthyretin concentrations are of diagnostic value [lO,ll]. In any case, it is important to measure transthyretin concentrations in both CSF and serum when the data is used for clinical diagnosis. A few reports in the literature indicate that CSF transthyretin levels may be of diagnostic significance in neurologic disorders. Weisner and Roethig [12] have suggested that the concentration of transthyretin in CSF is a useful indicator of CSF circulation disorders. Studies have shown that transthyretin is a constituent of senile amyloid deposits [13] and that transthyretin is reduced in CSF from patients with Alzheimer’s disease [14]. An increase in total CSF protein accompanied by a decrease in CSF transthyretin have been reported in pyogenic meningitis and primary and metastatic tumors [15]. The sensitive ELISA method for determination of transthyretin in CSF reported in this paper could be useful in the diagnosis of the disorders mentioned above.
Acknowledgements The investigations were supported by grant AGO6309 stitute of Aging and by research funds from the Department
from the National Inof Veterans Affairs.
References DW. Transthyretin in familial amyloidotic polyneuropathy: 1 Saraiva MJM, Costa JP, Goodman genetic and functional aspects. Adv Nemo1 1989;48:189-200. 2 Herbert J, Wilcox JN, Pham KTC, Fremeau RT, Zeviani M, Dwork A, Soprano DR, Makover A, Goodman DS, Zimmerman EA, Roberts JL, Schon EA. Transthyretin: a choroid plexus-specific transport protein in human brain. Neurology 1986;36:900-911. DA. Turbidimetric determination of transthyretin (prealbumin) with a 3 Hamlin CR, Pankowsky centrifugal analyzer. Clin Chem 1987;33:144-146. BM, Munson S. Rate nephelometry and radial immunodiffusion compared for measuring 4 Goldsmith serum prealbumin. Clin Chem 1987;33:161-163. assay of transthyretin by random-access 5 Pressac M, Vignoli L, Aymard P. Immunoturbidimetric analysis. Clin Chem 1988;34:176. of transthyretin (prealbumin) in 6 Ledue TB, Rifai N, Irish GR, Silverman LM. Inununoturbidimetry human serum. Clin Chem 1987;33:1260. RC, Fetterhoff TJ, Luckey DW. Three immunoassay methods evaluated for quantifying I McCarthy prealbumin (transthyretin) in serum. Clin Chem 1987;33:1430-1431. of human prealbumin in body fluids. Clin 8 Cowley DM, Davey JF, Hjelm NM. Radioimmunoassay Chim Acta 1983;134:69-76. of the Lowry procedure to 9 Markwell MAK, Haas SM, Bieber LL, Tolbert NE. A modification simplify protein determinations in membrane and lipoprotein samples. Anal Biochem 1978;87:206210. DR, Halliwell RP, Smith MG, Parke DV. Serum prealbumin as an index of liver function 10 Hutchinson in human hepatobiliary disease. Clin Chim Acta 1981;114:69-74. malnutrition. Clin Chem 1984;30:1286-1299. 11 Haider M, Haider SQ. Assessment of protein-calorie of prealbumin in cerebrospinal fluid (CSF), indicator of 12 Weisner B, Roethig HJ. The concentration CSF circulation disorders. Eur Neurol 1983;22:96-105. P, Rubinow A, Cohen AS, Brun A, Kemper TL. Senile cerebral 13 Shirahama T, Skinner M, Westennark amyloid: prealbumin as a common constituent in the neuritic plaque, in the neurofibrillary tangle and in the microangiopathic lesion. Am J Path01 1982;107:41-50. (transthyretin) in CSF of severely demented patients with Alxheimer’s 14 Riisoen H. Reduced prealbumin disease. Acta Neurol Stand. 1988;78:455-459. levels of cerebrospinal fluid in neurological disorders. Ind 15 Bimanpalli MV, Ghaswala PS. Prealbumin J Med Res 1988;88:26-164.