Clinical Chemistry
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 47: 1811-1820, 2001;
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (36)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kema, I. P.
Right arrow Articles by de Vries, E. G.E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kema, I. P.
Right arrow Articles by de Vries, E. G.E.
Related Collections
Right arrow Endocrinology and Metabolism
(Clinical Chemistry. 2001;47:1811-1820.)
© 2001 American Association for Clinical Chemistry, Inc.

Articles

Profiling of Tryptophan-related Plasma Indoles in Patients with Carcinoid Tumors by Automated, On-Line, Solid-Phase Extraction and HPLC with Fluorescence Detection

Ido P. Kema1a, Wim G. Meijer2, Gert Meiborg1, Bert Ooms3, Pax H.B. Willemse2 and Elisabeth G.E. de Vries2

Departments of
1 Pathology and Laboratory Medicine, and
2 Medical Oncology, University Hospital Groningen, 9700 RB Groningen, The Netherlands.
3 Spark Holland B.V., 7800 AJ Emmen, The Netherlands.

aAddress correspondence to this author at: Department of Pathology and Laboratory Medicine, University Hospital Groningen, PO Box 30001, 9700 RB Groningen, The Netherlands. Fax 31-50-6612290; e-mail i.p.kema{at}lab.azg.nl.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: Profiling of the plasma indoles tryptophan, 5-hydroxytryptophan (5-HTP), serotonin, and 5-hydroxyindoleacetic acid (5-HIAA) is useful in the diagnosis and follow-up of patients with carcinoid tumors. We describe an automated method for the profiling of these indoles in protein-containing matrices as well as the plasma indole concentrations in healthy controls and patients with carcinoid tumors.

Methods: Plasma, cerebrospinal fluid, and tissue homogenates were prepurified by automated on-line solid-phase extraction (SPE) in Hysphere Resin SH SPE cartridges containing strong hydrophobic polystyrene resin. Analytes were eluted from the SPE cartridge by column switching. Subsequent separation and detection were performed by reversed-phase HPLC combined with fluorometric detection in a total cycle time of 20 min. We obtained samples from 14 healthy controls and 17 patients with metastasized midgut carcinoid tumors for plasma indole analysis. In the patient group, urinary excretion of 5-HIAA and serotonin was compared with concentrations of plasma indoles.

Results: Within- and between-series CVs for indoles in platelet-rich plasma were 0.6–6.2% and 3.7–12%, respectively. Results for platelet-rich plasma serotonin compared favorably with those obtained by single-component analysis. Plasma 5-HIAA, but not 5-HTP was detectable in 8 of 17 patients with carcinoid tumors. In the patient group, platelet-rich plasma total tryptophan correlated negatively with platelet-rich plasma serotonin (P = 0.021; r = -0.56), urinary 5-HIAA (P = 0.003; r = -0.68), and urinary serotonin (P <0.0001; r = -0.80).

Conclusions: The present chromatographic approach reduces analytical variation and time needed for analysis and gives more detailed information about metabolic deviations in indole metabolism than do manual, single-component analyses.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
5-Hydroxytryptamine (serotonin) is a tryptophan-derived biogenic amine that is involved in a variety of physiologic processes in the human body, including smooth muscle contraction, blood pressure regulation, and both peripheral and central nervous system neurotransmission (1)(2). Serotonin is synthesized from the essential amino acid tryptophan in a process in which ~2% of dietary tryptophan is converted into serotonin. The major part of tryptophan is used for protein synthesis, in which its major catabolic route is via kynurenine and 3-hydroxyanthranilic acid (3)(4). Serotonin is synthesized and stored predominantly in the enterochromaffin cells of the gastrointestinal tract, where it accounts for ~80% of total body serotonin content (5). Other quantitatively important stores are found in the dense granules of platelets (storage only) and in the serotonergic neurons of the central nervous system (1).

Serotonin is implicated in several pathologic conditions, such as schizophrenia, autism, migraine, Parkinson disease, and carcinoid syndrome. The most pronounced aberrations in peripheral serotonin production and metabolism are encountered in carcinoid tumors. These tumors are classified according to their site of origin into fore-, mid-, and hindgut tumors (6). Notably, midgut tumors may give rise to the carcinoid syndrome, of which flushing, diarrhea, valvular lesions of the right heart compartment, and bronchoconstriction are the major features (7). Depletion of the free tryptophan body pool because of continuous augmentation of the serotonin biosynthetic pathway can produce symptoms similar to those seen in pellagra, such as dermatitis, diarrhea, and dementia (8).

Depending on the site of origin and mass, carcinoids are known to give rise to excessive synthesis, storage, and release of serotonin; its precursor, 5-hydroxytryptophan (5-HTP); 1 and its major metabolite, 5-hydroxyindole-3-acetic acid (5-HIAA). For the clinical chemical diagnosis and follow-up of patients with carcinoid tumors, measurements of serotonin and its metabolites in body fluids and tissues are used. The assay of 5-HIAA in urine is the most commonly used, whereas assessment of urinary and platelet serotonin may be useful in the diagnosis of specific forms of carcinoid tumors (9)(10). In previous studies, we showed that platelet serotonin is the most sensitive indole marker for the diagnosis and follow-up of carcinoids that are characterized by production of only small amounts of serotonin (10)(11). Moreover, in contrast to urinary 5-HIAA, platelet serotonin is not influenced by short-term dietary intake of serotonin-containing foods (12).

Several analytical techniques have been developed for the measurement of either single indoles or a selection of tryptophan-derived indoles (profiling methods). These include thin-layer chromatography, enzyme immunoassay, gas chromatography, gas chromatography-mass spectrometry, and HPLC with ultraviolet, fluorometric, or electrochemical detection (13). HPLC-based techniques have gradually replaced less-specific colorimetric methods in clinical chemical laboratories (14)(15). Developments in reversed-phase HPLC have enabled the separation of several metabolically related indoles. Together with fluorometric detection and injection of minimally prepurified biological samples, it offers the possibility to combine selectivity, sensitivity, and high precision. Automation of HPLC analysis, using direct injection procedures in combination with column switching, is an emerging technique that offers the possibility of combining sample prepurification, concentration, and analysis in one run. Automated methods thus enable the reduction of analytical variation attributable to differences in manual sample pretreatment and, moreover, decrease analysis turnaround time (16)(17)(18)(19)(20)(21).

Using on-line solid-phase extraction (SPE) and gradient HPLC with fluorometric detection, we developed an automated indole-profiling method that enables the simultaneous analysis of tryptophan, 5-HTP, serotonin, and 5-HIAA in protein-containing matrices, such as platelet-rich plasma, cerebrospinal fluid (CSF), and tissue homogenates. This method permits the direct injection of bio-fluids without sample pretreatment to remove proteins. This is accomplished through a stream-switching design that connects a SPE column (cartridge), packed with a nonselective strong hydrophobic sorbent, with the analytical column. Results of analysis of plasma, tissue, and CSF are shown.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
reagents
HPLC-grade acetonitrile was obtained from Rathburn; serotonin creatine sulfate, ascorbic acid, and potassium dihydrogen phosphate were from Merck; 5-HTP, tryptophan, and 5-HIAA were from Sigma Chemical Co.; 5-methoxytryptophan (5-MTP) was from Acros; and EDTA and sodium metabisulfite (Na2S2O5) were from BDH. All reagents were at least of analytical reagent grade. Reagent-grade water, obtained from a Millipore MilliQ-system, was used throughout.

stock solutions and samples
Stock solutions containing 0.5 mmol/L 5-MTP, 1 mmol/L 5-HTP or serotonin, or 4 mmol/L tryptophan or 5-HIAA were prepared in 0.01 mol/L acetic acid containing 0.5 g/L Na2S2O5. Stock solutions were stored protected from light at -20 °C for up to a maximum of 6 months. Saturated stock solutions containing ascorbic acid (300 g/L), dipotassium EDTA (750 g/L), and Na2S2O5 (750 g/L) were prepared in water and stored at room temperature for 1 month.

Venous blood samples from 14 healthy controls who had given informed consent (10 females and 4 males; median age, 45 years; range, 40–63 years) and 17 patients with liver-metastasized midgut carcinoid tumors (9 females and 8 males; median age, 62 years; range, 47–77 years) were collected in 10-mL Vacutainer Tubes (Becton Dickinson) containing 0.12 mL of (0.34 mol/L) EDTA solution. The samples were immediately placed on ice. Platelet-rich plasma was prepared from the EDTA blood within 1 h of sampling by centrifugation for 30 min at 120g and 4 °C. Platelet concentrations were measured in platelet-rich plasma with a Coulter counter Model S plus 4 (Coulter Electronics). Na2S2O5 and EDTA were added as preservatives to a final concentration of ~10 g/L each. Samples were stored at -20 °C until analysis.

Urine samples (24-h) were collected from the patients with carcinoid tumors into brown polypropylene bottles (Sarstedt) containing ~250 mg each of Na2S2O5 and EDTA as preservatives. Urine samples were acidified to pH 4 with acetic acid and stored at -20 °C.

A carcinoid tumor tissue sample was obtained from a patient with a midgut carcinoid who underwent tumor resection and histopathologic diagnosis. After weighing, the sample was transferred to a 0.01 mol/L acetic acid solution containing 10 g/L each of Na2S2O5 and EDTA. Approximately 100 mg of carcinoid tumor tissue sample was homogenized at 4 °C in 10 mL of a 0.01 mol/L acetic acid solution containing Na2S2O5 and EDTA as above, using a potter S homogenizer (Braun). After homogenization, the sample was centrifuged at 12 000g for 15 min. The supernatant was stored at -20 °C.

Midstream lumbar CSF was obtained from a 2-year-old child for biochemical assessment of a neurologic developmental disorder. After sample collection, the CSF was centrifuged for 10 min at 1500g at 4 °C to remove cells and stored at -80 °C.

Before analysis of samples, aliquots of thawed platelet-rich plasma samples (50 µL), tissue homogenate (50 µL), or CSF (400 µL) were mixed with 200 µL of antioxidant solution (150 µL of ascorbic acid stock solution, 25 µL of dipotassium EDTA stock solution, 25 µL of Na2S2O5 stock solution) and 50 µL of internal standard solution (6 µmol/L 5-MTP). The plasma and tissue homogenate were diluted with 1.2 mL of water. All samples were subsequently centrifuged at 25 000g for 5 min in brown Eppendorf vials, and 20 µL of each sample was injected into the HPLC as described below. This volume was equivalent to ~0.7 µL of plasma, 12 µL of CSF, or 7 µg of tissue homogenate.

analysis and quantification
Instrumentation.
The on-line SPE unit used was a Prospekt (Spark Holland, Emmen, The Netherlands) consisting of three modules: a SPE controller unit, a solvent delivery unit, and an autosampler. The SPE unit contains two connectable 6-way valves and a SPE cartridge-exchange module. The solvent delivery unit delivers solvents for conditioning, sample application, and clean-up of SPE cartridges. Samples were injected with a Midas autosampler (Spark Holland BV). The HPLC pump used was a Gynkotek Series P580A binary high-pressure gradient pump (Softron GmbH); detection was performed with a Jasco 920 FP spectrofluorometer (excitation wavelength, 285 nm; emission wavelength, 340 nm). For chromatographic separation of analytes, we used a 250 x 3.0 (i.d.) mm Inertsil ODS-3 analytical column filled with 5-µm spherical particles (GL Sciences Inc.). The analytical column was preceded by a 10 x 4.0 (i.d.) mm ODS-2 reversed-phase guard column filled with 5-µm spherical particles (Hichrom Limited). Hysphere Resin SH SPE cartridges (Spark Holland BV) were used for clean-up of samples and concentrating analytes. Detector output was integrated using ChromPerfect (Ver. 3.51) integration software (Nelson).

Gradient.
The binary gradient system consisted of eluent A (50 mmol/L potassium dihydrogen phosphate in water adjusted to pH 3.3 with phosphoric acid) and eluent B (700 mL/L eluent A–300 mL/L acetonitrile). Before use, all eluents were filtered through a 0.45 µm membrane filter (Schleicher & Schuell) and degassed at reduced pressure for 5 min. Gradient elution was performed according to the following elution program: 0–2 min, 80% A, 20% B; 2–10 min, 25% A, 75% B; 10–10.1 min, 0% A, 100% B; 10.1–15 min, 0% A, 100% B; 15–15.1 min, 80% A, 20% B; 15.1–20 min, 80 A, 20% B. The gradients applied were linear; the flow rate was 0.6 mL/min. Chromatography was performed at ambient temperature

On-line SPE.
On-line SPE was performed according to the timetable specified in Table 1 and as schematically depicted in Fig. 1 in the following way. The system was designed to proceed automatically through a series of programmable routines in which, in broad outlines, the SPE cartridge was loaded, purged for clean-up, eluted to the analytical column, and regenerated or exchanged. During the "loading" routine, the SPE cartridge was conditioned sequentially with acetonitrile, water, 5 g/L dipotassium EDTA in water, and water, all provided by the solvent delivery unit at a flow rate of 3.0 mL/min. After this, the autosampler injected the sample, thereby loading the cartridge with the sample in the forward flush mode. The solvent delivery unit then continued to pump water for ~1 min, thus cleaning up the cartridge. In the following "elute" step, valve 1 was switched to the elute position, and analytes were eluted from the cartridge with eluent B in the direction opposite the direction in which the sample was loaded (backward-flush flow). The gradient was started simultaneously, and the eluted components were directed toward the analytical column. In front of the analytical column and after the T-union connecting the lines transporting eluents A and B, 80 cm of coiled line was mounted to facilitate the mixing of eluents A and B before they reached the analytical column. During the "regenerate" step (after switching of both valves), the cartridge was flushed, independent of the ongoing chromatographic separation, in the "backward-flow" direction, successively with eluent B, water, and acetonitrile, all provided by the solvent delivery unit at a flow rate of 3.0 mL/min. After the chromatographic separation on the analytical column was completed (t = 20 min), the system was ready for the next injection. If necessary, a cartridge could be changed automatically before the next cycle.


View this table:
[in this window]
[in a new window]
  		Table 1. Timetable summarizing the sequence of events used for on-line SPE, as described in Materials and Methods.



View larger version (26K):
[in this window]
[in a new window]
  		Figure 1. Schematic representation of the on-line SPE system coupled to HPLC with fluorescence detection.

In the top panel (SPE), the system is depicted during conditioning of the cartridge and, after sample injection, SPE of the sample (e.g., platelet-rich plasma). The middle panel (Elute) represents the system during backward-flush elution of the cartridge with eluent B, subsequent mixing of the eluent with eluent A in the helical coil, and transportation of the combined eluent toward the analytical column and detector. The bottom panel (Regenerate) shows the system during regeneration of the cartridge. Instead of regenerating the cartridge, the system can be programmed to automatically replace the cartridge at this point. For further details, see Table 1Up and Materials and Methods. HPG, high-pressure gradient. F1 and F2 denote valve positions.

Tryptophan, 5-HTP, serotonin, and 5-HIAA were quantified by calculation of the peak-area ratios relative to that of the internal standard (5-MTP) in samples. Sample peak-area ratios were compared with the peak-area ratios obtained for the calibration solutions at five different concentrations. Calibrators were prepared fresh by diluting stock solutions to concentrations of 5–20 µmol/L with 0.01 mol/L acetic acid. Peak identity was verified by the response of the retention time to different chromatographic conditions and the addition of calibrators.

Urinary 5-HIAA concentrations were determined in ether extracts by HPLC with fluorometric detection, essentially as described by Rosano et al. (22). Urinary serotonin concentrations were determined as described previously (23) by clean-up, using cation-exchange chromatography, and separation by HPLC with fluorometric detection. Urinary creatinine was measured by a picric acid method on a Merck Mega analyzer (Merck). Single-component analyses of serotonin in platelet-rich plasma, used for validation of the plasma-profiling method, were performed by HPLC with fluorometric detection as described previously (23).

analytical characteristics
In on-line SPE, detection limits depend on the degree of sample preconcentration on the SPE cartridge. With the format described for plasma, detection limits were determined by injecting calibration solutions with decreasing concentrations of tryptophan, 5-HTP, serotonin, and 5-HIAA. The detection limit was defined as the injected amount that produced a signal-to-noise ratio of 3. We estimated the percentage of carryover between sequential analyses performed on a single SPE cartridge by alternating injections of water and plasma samples with high concentrations of indoles.

quality control and validation of the plasma indole-profiling method
Recoveries were estimated by the addition of indoles (tryptophan, 5-HTP, serotonin, and 5-HIAA) in three different concentrations to a platelet-rich plasma sample. Within- and between-series precision was determined from six measurements of each concentration of added indoles and of the nonsupplemented platelet-rich plasma. Between-series precision was assessed on 6 different days over a 1-month period. Experiments were performed using either one SPE cartridge per series or one cartridge per sample, as indicated.

The plasma-profiling method was validated by comparison of its results with those obtained from single-component analyses of serotonin that were routinely used in our laboratory. For this, 64 platelet-rich plasma samples obtained from clinically suspected patients and patients with carcinoid tumors screened for routine diagnostic procedures were used.

statistics
The Cusum test was used for testing the linearity of the profiling and the routinely applied methods for platelet-rich plasma serotonin. For method comparisons, we used Deming regression. With this regression, the slope and intercept and their 95% confidence intervals were calculated as well as the standard error of prediction (Sy|x). Differences between plasma concentrations of indoles between healthy controls and patients were assessed by means of the Mann-Whitney U-test. The relationships between concentrations of plasma and urinary indoles from controls and patients with carcinoid tumors were examined by means of the Spearman rank test. P <=0.05 was considered significant (24).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
chromatography
A chromatogram of five calibrators obtained by automated on-line SPE and HPLC with fluorometric detection (as depicted schematically in Fig. 1Up ) is shown in Fig. 2A . Complete separation of all components was achieved within 13 min. Total cycle time was 20 min (Table 1Up ). Fig. 2 , B–D, shows indole profiles of platelet-rich plasma samples obtained from a healthy adult and from two patients with liver-metastasized carcinoid tumors. Comparison of the plasma indole profiles shows grossly increased serotonin concentrations in the samples of the two patients with carcinoid tumors. Plasma total tryptophan in these patients was decreased, notably in the patient depicted in Fig. 2D . Both patients with carcinoid tumors showed increased plasma 5-HIAA concentrations, whereas 5-HIAA was not detectable in the healthy control. 5-HTP was not detectable in any of the plasma samples. Fig. 2E depicts the chromatogram obtained after on-line SPE extraction of carcinoid tissue homogenate obtained from the primary ileum tumor of a patient with a liver-metastasized midgut carcinoid tumor. Both serotonin and 5-HIAA are present in substantial amounts, whereas the tryptophan concentration is low. Fig. 2F illustrates the indole profile of a CSF sample.



View larger version (19K):
[in this window]
[in a new window]
  		Figure 2. Chromatograms of body fluids and tissues from healthy persons and patients with carcinoid tumors, as obtained by on-line SPE coupled to HPLC with fluorescence detection.

Excitation wavelength, 285 nm; emission wavelength, 340 nm. (A), separation of five indole calibrators [injected amounts, tryptophan (TRP), 39.5 µmol/L; 5-HTP (5HTP), 9.9 µmol/L; serotonin (5HT), 10.3 µmol/L; 5-HIAA (5HIAA), 42.5 µmol/L; 5-MTP (5MTP), 7.2 µmol/L]. (B–D), indole profiles in platelet-rich plasma from a healthy adult (B) and patients with liver-metastasized carcinoid tumors (C and D) obtained by on-line SPE of 20-µL aliquots of platelet-rich plasma. Calculated concentrations in the plasma samples were (for B, C, and D, respectively) as follows: tryptophan, 59.5, 60.6, and 23.6 µmol/L; 5-HTP, not detectable; serotonin, 2000, 25 826, and 16 296 nmol/L; 5-HIAA, not detectable, 4999, and 17 362 nmol/L. (E), chromatogram of SPE-prepurified tissue homogenate from carcinoid of midgut (ileum) origin. On a wet weight basis, the calculated concentrations of compounds were as follows: 5-HTP, not detectable; tryptophan, 103 nmol/g; 5-HIAA, 476 nmol/g; and serotonin, 1273 nmol/g. (F), lumbar CSF. Calculated concentrations were as follows: tryptophan, 2266 nmol/L; 5-HTP, not detectable; serotonin, not detectable; 5-HIAA, 263 nmol/L. Peak identity was verified by standard addition. *, compounds that could not be identified.

analytical characteristics
With on-line SPE, detection limits depend on the degree of sample preconcentration on the SPE cartridge. Using the format described for plasma, we achieved a signal-to-noise ratio of 3 for tryptophan, 5-HTP, serotonin, and 5-HIAA at concentrations of 19, 17, 67, and 61 nmol/L, respectively. Analysis of increasing amounts of calibrators added to a plasma sample showed that the assay is linear to injected amounts of at least 150 µmol/L tryptophan, 15 µmol/L 5-HTP, 20 µmol/L serotonin, and 60 µmol/L 5-HIAA. The percentage of carryover between sequential analyses performed on one single SPE cartridge was <1% for all five indoles tested.

quality control and validation of the plasma indole-profiling method
The results for the within- and between-series recoveries and precision for platelet-rich plasma samples determined in six replicates are shown in Table 2 . The within-series data represent recoveries and CVs for six samples analyzed in one series, using one SPE cartridge per series. The between-series data represent recoveries and CVs for six samples analyzed in six series, using one SPE cartridge per series. Within-series recoveries for the four analytes were 88.7–102.8%. Between-series recoveries were 91.9–110.4%. Analysis of six replicates of a nonsupplemented platelet-rich plasma sample from a patient with carcinoid tumors in one series (precision), using one SPE cartridge per series, gave the following results (mean ± SD): tryptophan, 92.8 ± 1.3 µmol/L; 5-HTP, not detectable; serotonin, 4275 ± 64 nmol/L; and 5-HIAA, 7350 ± 470 nmol/L. The use of one cartridge per sample reduced CVs from 1.4% to 0.8% for tryptophan, from 1.9% to 1.5% for 5-HT, and from 6.4% to 1.9% for 5-HIAA. The between-series reproducibility tested in a nonsupplemented platelet-rich plasma sample from a patient with carcinoid tumors was as follows (mean ± SD): tryptophan, 43.9 ± 2.3 µmol/L; 5-HTP, not detectable; serotonin, 7225.0 ± 440 nmol/L; and 5-HIAA, 46 000 ± 2714 nmol/L.


View this table:
[in this window]
[in a new window]
  		Table 2. Within- and between-series1 recovery and precision of tryptophan, 5-HTP, serotonin, and 5-HIAA added to a platelet-rich plasma sample.2

Serotonin values obtained by the profiling method for platelet-rich plasma were correlated to the results obtained with the single-component method. For this, 64 platelet-rich plasma samples obtained from patients with clinically suspected carcinoid tumors and patients with carcinoid tumors screened for routine diagnostic procedures were analyzed. Using the Deming regression, we obtained a slope of 1.054 (range, 1.004–1.089) and a y-intercept of -75 (-119.2 to 3.0) nmol/L. The correlation coefficient was 0.9936 (Sy|x = 292 nmol/L). A graphical representation of these data is given in Fig. 3 . Using the Cusum test, we found no deviations from linearity.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 3. Comparison of results for serotonin (5-HT) in 64 platelet-rich plasma samples as performed by the indole profiling method (y axis) and the manual single-component method (x axis).

Results of Deming regression: slope, 1.054 (range, 1.004–1.089); y-intercept, -75 nmol/L (range, -119.2 to 3.0 nmol/L); r = 0.9936; Sy|x = 292 nmol/L. The solid line represents the Deming regression line.

indole concentrations in platelet-rich plasma and urine in healthy controls and patients with carcinoid tumors
The results of the indole analyses in platelet-rich plasma from 14 healthy controls and 17 patients with liver-metastasized midgut carcinoid tumors as obtained by on-line SPE are shown in Table 3 . For the patient group, additional data for the urinary excretion of 5-HIAA and serotonin are given. For both groups, total tryptophan and serotonin were the most abundant indoles in platelet-rich plasma. In the patient group, 5-HIAA was detectable in 8 of 17 platelet-rich plasma samples, whereas it was not detectable in samples from the healthy control group. 5-HTP was not detectable in any of the analyzed platelet-rich plasma samples. Comparison of indoles in platelet-rich plasma in the two groups showed that both serotonin and 5-HIAA concentrations were higher in patients with carcinoid tumors than in the control group (P <0.0001 and P = 0.04, respectively). The median total tryptophan concentration was lower in the patient group (45.7 vs 50.4 µmol/L), but the range in the patient group was wide (21.0–68.0 µmol/L), perhaps explaining the lack of a statistically significant difference. In the patient group, total tryptophan in platelet-rich plasma was negatively correlated with serotonin in platelet-rich plasma (P = 0.021), urinary 5-HIAA (P = 0.003), and urinary serotonin [P <0.0001; Spearman’s rho (r) = -0.56, -0.68, and -0.80, respectively]. The serotonin concentration in platelet-rich plasma correlated positively with urinary 5-HIAA (P <0.0001; r = 0.79) and urinary serotonin (P = 0.001; r = 0.73). The excretions of 5-HIAA and serotonin in urine also showed a positive correlation (P <0.0001; r = 0.88). The correlation between 5-HIAA in platelet-rich plasma and urinary 5-HIAA excretion did not reach significance (P = 0.058). In the healthy control group, no correlation was found between serotonin and the total tryptophan concentration in platelet-rich plasma.


View this table:
[in this window]
[in a new window]
  		Table 3. Indole concentrations in platelet-rich plasma and urine samples of 14 healthy adults and 17 patients with liver metastases of a midgut carcinoid tumor.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Several methods have been described for the determination of serotonin, including automated HPLC-based methods (13). We developed the present method for the automated analysis of four clinically important indoles in protein-containing matrices to reduce factors such as hands-on time, turnaround time, and analytical variation attributable to differences in manual sample pretreatment. The on-line SPE extraction and clean-up system that we used, which consisted of two connectable 6-way valves combined with a SPE cartridge-exchange module, has previously been described for several other, notably single-component, applications (17)(18)(19)(20).

Because tryptophan-related indoles differ considerably with respect to functional groups, a rather nonselective prepurification method had to be found to enable their profiling in biological materials. For this, a few SPE procedures, using porous polystyrene polymer, Sep-Pak C18, or protein-coated ODS, have been described (25)(26)(27). Other methods include protein precipitation and filtration or centrifugation (28)(29). To allow on-line SPE, we were restricted to those resins available in cartridge format. Of the tested resins, only Hysphere Resin SH (a strong hydrophobic polystyrene resin) sufficiently retained all components of interest. A side effect of the high capacity factor of the resin was that, notably, serotonin was retained strongly on the cartridge. To elute all components from the cartridge simultaneously, a high concentration of organic modifier (300 mL/L acetonitrile) had to be used. To circumvent differences in elution patterns between serotonin and the other indoles, compounds were eluted from the cartridge in the back-flush mode (opposite to the direction in which extraction was performed). Elution of the compounds of interest from the SPE cartridge with a high concentration of organic modifier would typically interfere with the subsequent reversed-phase chromatographic separation of the compounds. In fact, this limitation in the choice of SPE eluent can be a major drawback of on-line SPE when SPE sorbents are used that have a higher capacity factor for the compounds of interest than does the stationary phase of the analytical column. To circumvent this problem, we used a binary high-pressure HPLC pump and changed the tubing in such a way that the cartridge eluent was mixed with the other, more polar eluent of the binary system in a helical coil (Fig. 1Up ). The combined eluent has a lower organic modifier content, which does not interfere with subsequent chromatographic separation on the analytical column. Coil length and diameter influenced the efficacy of eluent mixing (results not shown). This adaptation of the system, which to our knowledge is described here for the first time, allows on-line elution of SPE cartridges almost independent of the eluent used for chromatographic separation.

As shown in Table 2Up , the presented method allows reproducible quantification of several indoles in biological matrices. A feature of the system is that it not only enables automated extraction and clean-up but also permits on-line concentration (trace enrichment), as was shown for CSF. Concentration can be accomplished by increasing the ratio between the sample volume and the elution volume. Tryptophan and serotonin are present in comparatively high concentrations in platelet-rich plasma. Consequently, no trace enrichment is required for these two compounds, and only a relatively low volume needs to be extracted for their reproducible quantification. The plasma concentrations of 5-HTP and 5-HIAA, however, are substantially lower in healthy individuals (9)(30)(31). Simultaneous quantification of these compounds together with the more abundant plasma indoles, however, not only requires that the detector used have a wide dynamic linear range, but also puts high demands on the chromatographic method used to separate the compounds. Because our focus is primarily on tryptophan and serotonin in the diagnosis and follow-up of carcinoid tumors, the injected plasma volume was optimized for these two compounds. In case of lower indole concentrations, such as in CSF, the applied volume can be increased >500-fold because the cartridges are designed to extract up to 1 mL of plasma. The reported detection limits for plasma indoles thus should be considered in combination with the extracted volumes: increasing the extracted volume will lower the detection limits proportionally.

The on-line SPE system can be programmed to automatically replace the SPE cartridges after each plasma sample. In fact, most of the applications described for on-line SPE make use of this feature. We found, however, that the extraction of small volumes of plasma enables regeneration of the SPE cartridge up to at least 40 times. Increased pressure over the cartridge or clear changes in the chromatographic pattern signaled cartridge deterioration. The time between the elution step and end of the chromatography (Table 1Up ) allowed thorough and reproducible regeneration of the SPE cartridge, thus substantially reducing costs. Conditioning of cartridges with an EDTA-containing solution before their first use was incorporated in the conditioning routine (Table 1Up ) because it is essential in preventing changes in extraction recovery during the first injections. As was shown, the degree of cross-contamination between analyses performed on a single cartridge was negligible under the conditions described. Although the use of one cartridge per analysis reduces the analytical variation, we routinely reuse the SPE cartridges up to 30 times because the effect on the analytical precision is considered acceptable. The analytical variation for plasma serotonin has been reduced by a factor of 2 since the automated SPE method replaced the manual weak cation-exchange extraction method we used on our laboratory 1 year ago.

In a recent study, we compared the diagnostic accuracy of three indole markers for the diagnosis of carcinoid tumors (11). We showed that platelet serotonin has a higher discriminating capacity than does urinary 5-HIAA or urinary serotonin. For this reason, we routinely use platelet serotonin as the primary marker for screening patients with clinically suspected carcinoid tumors. Because at higher serotonin production rates the saturable platelet serotonin does not hold pace with urinary 5-HIAA, the latter is used for estimation of total serotonin production and follow-up of patients with grossly increased serotonin production. Profiling of tryptophan-related indoles in (platelet-enriched) plasma of patients with carcinoid tumors can provide additional information on, e.g., the occurrence of an enzyme deficiency in foregut carcinoids. Foregut carcinoids usually secrete both serotonin (although to a lesser extent) and 5-HTP. In foregut carcinoids with aromatic L-amino acid decarboxylase (EC 4.1.1.28) deficiencies, tumor 5-HTP accumulation and secretion may prevail (32)(33)(34). Decarboxylation of 5-HTP in kidneys and other tissues increases urinary and platelet serotonin (9)(10)(34). In our patient group, 11 of 17 patients exhibited increased urinary excretion of serotonin, which could indicate increased production of 5-HTP. We did not find plasma 5-HTP concentrations above the detection limit in any of the patients with carcinoid tumors tested. Because no quantitative data on plasma 5-HTP have been reported, we can only assume that its concentration is comparable to that of urinary serotonin. Our observations suggest that either the metabolic pressure on circulating 5-HTP is high, leading to low plasma 5-HTP concentrations, or that increased urinary serotonin excretion in our (midgut) patient group is a result of renal clearance of circulating free serotonin.

We found plasma 5-HIAA concentrations above the detection limit in 8 of 17 patients with carcinoid tumors. The correlation between urine and plasma concentrations of this metabolite did not reach significance (P = 0.058) probably because plasma 5-HIAA was lower than the detection limit in 9 of the 17 patients tested in the present setting. Although increased urinary 5-HIAA is a known hallmark of advanced carcinoid disease, relatively little is known about plasma 5-HIAA concentrations in these patients. Lee et al. (30) found increased plasma 5-HIAA in patients with tumors of the gastrointestinal tract, but did not investigate patients with carcinoid tumors. More recently, Degg et al. (31) compared plasma and urinary 5-HIAA in 11 patients with carcinoids of both fore- and midgut origin. Although the diagnostic sensitivity of both markers was comparable in this patient group, the authors noted that, overall, 5-HIAA showed a much greater relative increase in plasma than in urine, suggesting that the former might be a more sensitive indicator of carcinoid disease. Degg et al. (31) also observed that ingestion of serotonin-containing foods, such as pineapple and walnuts, causes an increase in plasma 5-HIAA concentration, as has been noted previously for urinary 5-HIAA (12)(35). Comparison of the concentrations of plasma 5-HIAA in our patient group with those reported by Degg et al. (31) showed similar concentrations.

The total plasma tryptophan data reported for controls in our study are comparable to those reported previously (36). We did not find differences between the plasma tryptophan concentrations in controls and patients with carcinoid tumors. Plasma samples were obtained from both controls and patients in an undefined metabolic state, at different times of the day. Because plasma total tryptophan is influenced by food intake and exhibits a circadian rhythm (36), potential differences may have become obscured. A significant negative correlation was found between plasma tryptophan concentrations and concentrations of other plasma and urine indoles in the patient group. Long-term augmentation of the serotonin biosynthetic pathway in these patients leads to seriously reduced free tryptophan body pool and slow development of symptoms comparable to those found in pellagra (8). Simultaneous determination of tryptophan concentrations provides a means of early detection of deficiencies and enables timely supplementation with tryptophan or niacin (37).

In conclusion, automated on-line SPE in combination with reversed-phase HPLC and fluorometric detection enables reproducible automated profiling of tryptophan and related indoles in protein-containing samples, such as platelet-rich plasma and CSF. The present chromatographic approach reduces turnaround time for analysis and gives more detailed information about metabolic deviations in indole metabolism than do manual single-component analyses. Profiling of tryptophan-related indoles can especially be of use in patients with carcinoid tumors, in whom indole marker quantification is performed routinely, or in patients in whom aberrations in tryptophan metabolism are suspected.


   Acknowledgments
 
We are grateful to Drs. R.C.J. Verschueren and H. de Vries (Department of Surgery) for their cooperation in the collection of samples from patients with carcinoid tumors. We thank M. Volmer for statistical advice, and E.E.A. Venekamp-Hoolsema and A. Oosterloo-Duinkerken for skillful technical assistance and preparation of the figures. Spark Holland loaned the PROSPEKT instrument during the start of the study.


   Footnotes
 
1 Nonstandard abbreviations: 5-HTP, 5-hydroxytryptophan; 5-HIAA, 5-hydroxyindole-3-acetic acid; SPE, solid-phase extraction; CSF, cerebrospinal fluid; and 5-MTP, 5-methoxytryptophan.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Sirek A, Sirek OV. Serotonin: a review. Can Med Assoc J 1970;102:864-869.
  2. Grahame-Smith DG. Serotonin (5-hydroxytryptamine, 5-HT). Q J Med 1988;67:459-466.[Free Full Text]
  3. Ruthven CRJ, Sandler M. Neurogenic amines and secreting tumors. Brown SS Mitchell FL Young DS eds. Chemical diagnosis of disease 1979:1217-1291 Elsevier Amsterdam. .
  4. Tyce GM. Biochemistry of serotonin. VanHoutte PM eds. Serotonin and the cardiovascular system 1985:1-13 Raven Press New York. .
  5. Bertaccini G, Chieppa S. Urinary excretion of 5-hydroxyindoleacetic acid after removal of the large intestine in man. Lancet 1960;i:881.
  6. Williams ED, Sandler M. The classification of carcinoid tumours. Lancet 1963;i:238-239.
  7. Maton PN. The carcinoid syndrome. JAMA 1988;260:1602-1605.[Abstract/Free Full Text]
  8. Roberts LJ, II, Oates JA. Disorders of vasodilator hormones: the carcinoid syndrome and mastocytosis. Wilson JD Foster DW eds. Williams textbook of endocrinology 1985:1363-1378 WB Saunders Philadelphia. .
  9. Feldman JM. Urinary serotonin in the diagnosis of carcinoid tumors. Clin Chem 1986;32:840-844.[Abstract/Free Full Text]
  10. Kema IP, de Vries EGE, Schellings AMJ, Postmus PE, Muskiet FAJ. Improved diagnosis of carcinoid tumors by measurement of platelet serotonin. Clin Chem 1992;38:534-540.[Abstract/Free Full Text]
  11. Meijer WG, Kema IP, Volmer M, Willemse PHB, de Vries EGE. Discriminating capacity of indole markers in the diagnosis of carcinoid tumors. Clin Chem 2000;46:1588-1596.[Abstract/Free Full Text]
  12. Kema IP, Schellings AMJ, Meiborg G, Hoppenbrouwers CJM, Muskiet FAJ. Influence of a serotonin and dopamine rich diet on platelet serotonin content and urinary excretion of biogenic amines and their metabolites. Clin Chem 1992;38:1730-1736.[Abstract/Free Full Text]
  13. Kema IP, de Vries EGE, Muskiet FAJ. Clinical chemistry of serotonin and metabolites. J Chromatogr B 2000;747:33-48.
  14. Van Haard PM. Chromatography of urinary indole derivatives. J Chromatogr 1988;429:59-94.[Web of Science][Medline] [Order article via Infotrieve]
  15. Deacon AC. The measurement of 5-hydroxyindoleacetic acid in urine. Ann Clin Biochem 1994;31:215-232.
  16. Anderson DJ. High performance liquid chromatography in clinical analysis. Anal Chem 1999;71:314R-327R.[Medline] [Order article via Infotrieve]
  17. Koch DD, Kissinger PT. Determination of serotonin in serum and plasma by liquid chromatography with precolumn sample enrichment and electrochemical detection. Anal Chem 1980;52:27-29.[Medline] [Order article via Infotrieve]
  18. Palmerini CA, Cantelmi MG, Minelli A, Fini C, Zampino M, Floridi A. Determination of plasma serotonin by high-performance liquid chromatography with pre-column sample enrichment and fluorimetric detection. J Chromatogr 1987;417:378-384.[Web of Science][Medline] [Order article via Infotrieve]
  19. Slingerland RJ, Van Kuilenburg AB, Bodlaender JM, Overmars H, Voute PA, Van Gennip AH. High-performance liquid chromatographic analysis of biogenic amines in cells and in culture media using on-line dialysis and trace enrichment. J Chromatogr B 1998;716:65-75.
  20. Takkenberg B, Endert E, van Ingen HE, Ackermans M. Improved method for the determination of serotonin in plasma by high-performance liquid chromatography using on-line sample pre-treatment. J Chromatogr 1991;565:430-435.[Web of Science][Medline] [Order article via Infotrieve]
  21. Panholzer TJ, Beyer J, Lichtwald K. Coupled-column liquid chromatographic analysis of catecholamines, serotonin, and metabolites in human urine. Clin Chem 1999;45:262-268.[Abstract/Free Full Text]
  22. Rosano TG, Meola JM, Swift TA. Liquid-chromatographic determination of urinary 5-hydroxy-3-indoleacetic acid, with fluorescence detection. Clin Chem 1982;28:207-208.[Abstract/Free Full Text]
  23. Kwarts E, Kwarts J, Rutgers H. A simple paired-ion liquid chromatography assay for serotonin in cerebrospinal fluid, platelet-rich plasma, serum and urine. Ann Clin Biochem 1984;21:425-429.
  24. Siegel SN, Castellan NJ, Jr. Nonparametric statistics for the behavioral sciences, 2nd ed. 1988:58-64 McGraw-Hill New York. .
  25. Morita I, Masujima T, Yoshida H, Imay H. Enrichment and high-performance liquid chromatography analysis of tryptophan metabolites in plasma. Anal Biochem 1981;118:142-146.[Web of Science][Medline] [Order article via Infotrieve]
  26. Chou PP, Jaynes PK. Determination of urinary 5-hydroxyindole-3-acetic acid using solid-phase extraction and reversed-phase high-performance liquid chromatography with electrochemical detection. J Chromatogr 1985;341:167-171.[Web of Science][Medline] [Order article via Infotrieve]
  27. Morita I, Kawamoto M, Hattori M, Eguchi K, Sekiba K, Yoshida H. Determination of tryptophan and its metabolites in human plasma and serum by high-performance liquid chromatography with automated sample clean-up system. J Chromatogr 1990;526:367-374.[Web of Science][Medline] [Order article via Infotrieve]
  28. Raynaud F, Pevet P. Determination of 5-methoxyindoles in pineal gland and plasma samples by high performance liquid chromatography with electrochemical detection. J Chromatogr 1991;564:103-113.[Web of Science][Medline] [Order article via Infotrieve]
  29. Chin JR. determination of six indolic compounds, including melatonin, in rat pineal using high-performance liquid chromatography with serial fluorimetric-electrochemical detection. J Chromatogr 1990;528:111-121.[Web of Science][Medline] [Order article via Infotrieve]
  30. Lee MS, Cheng FC, Yeh Z, Liou TY, Liu JH. Determination of plasma serotonin and 5-hydroxyindoleacetic acid in healthy subjects and cancer patients. Clin Chem 2000;46:422-423.[Free Full Text]
  31. Degg TJ, Allen KR, Barth JH. Measurement of plasma 5-hydroxyindoleacetic acid in plasma in carcinoid disease: an alternative to 24-h urine collections?. Ann Clin Biochem 2000;37:724-726.
  32. Campbell ACP, Gowenlock AH, Platt DS, Snow PJD. A 5-hydroxytryptophan-secreting tumor. Gut 1962;3:61-67.
  33. Oates JA, Sjoerdsma A. A unique syndrome associated with secretion of 5-hydroxytryptophan by metastatic gastric carcinoids. Am J Med 1962;32:333-342.
  34. Feldman JM. Serotonin metabolism in patients with carcinoid tumors: incidence of 5-hydroxytryptophan-secreting tumors. Gastroenterology 1978;75:1109-1114.[Web of Science][Medline] [Order article via Infotrieve]
  35. Feldman JM, Lee EM. Serotonin content of foods: effect on urinary excretion of 5-hydroxyindoleacetic acid. Am J Clin Nutr 1985;42:639-643.[Abstract/Free Full Text]
  36. Eynard N, Flachaire E, Lestra C, Broyer M, Zaidan R, Claustrat B, Quincy C. Platelet serotonin content and free and total plasma tryptophan in healthy volunteers during 24 h. Clin Chem 1993;39:2337-2340.[Abstract]
  37. Swain CP, Tavill AS, Neale G. Studies of tryptophan and albumin metabolism in a patient with carcinoid syndrome, pellagra, and hypoproteinemia. Gastroenterology 1976;71:484-489.[Web of Science][Medline] [Order article via Infotrieve]



The following articles in journals at HighWire Press have cited this article:


Home page
 Cancer Res.Home page
O. C. Neels, K. P. Koopmans, P. L. Jager, L. Vercauteren, A. van Waarde, J. Doorduin, H. Timmer-Bosscha, A. H. Brouwers, E. G.E. de Vries, R. A.J.O. Dierckx, et al.
Manipulation of [11C]-5-Hydroxytryptophan and 6-[18F]Fluoro-3,4-Dihydroxy-L-Phenylalanine Accumulation in Neuroendocrine Tumor Cells
Cancer Res., September 1, 2008; 68(17): 7183 - 7190.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
K. P. Koopmans, O. C. Neels, I. P. Kema, P. H. Elsinga, W. J. Sluiter, K. Vanghillewe, A. H. Brouwers, P. L. Jager, and E. G.E. de Vries
Improved Staging of Patients With Carcinoid and Islet Cell Tumors With 18F-Dihydroxy-Phenyl-Alanine and 11C-5-Hydroxy-Tryptophan Positron Emission Tomography
J. Clin. Oncol., March 20, 2008; 26(9): 1489 - 1495.
[Abstract] [Full Text] [PDF]

Home page
 J. Pharmacol. Exp. Ther.Home page
D. Zolkowska, M. H. Baumann, and R. B. Rothman
Chronic Fenfluramine Administration Increases Plasma Serotonin (5-Hydroxytryptamine) to Nontoxic Levels
J. Pharmacol. Exp. Ther., February 1, 2008; 324(2): 791 - 797.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
W. H.A. de Jong, K. S. Graham, J. C. van der Molen, T. P. Links, M. R. Morris, H. A. Ross, E. G.E. de Vries, and I. P. Kema
Plasma Free Metanephrine Measurement Using Automated Online Solid-Phase Extraction HPLC Tandem Mass Spectrometry
Clin. Chem., September 1, 2007; 53(9): 1684 - 1693.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
S. Russo, I. P. Kema, E. B. Haagsma, J. C. Boon, P. H.B. Willemse, J. A. den Boer, E. G.E. de Vries, and J. Korf
Irritability Rather Than Depression During Interferon Treatment Is Linked to Increased Tryptophan Catabolism
Psychosom Med, September 1, 2005; 67(5): 773 - 777.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
E. J. Mulder, A. Oosterloo-Duinkerken, G. M. Anderson, E. G.E. De Vries, R. B. Minderaa, and I. P. Kema
Automated On-Line Solid-Phase Extraction Coupled with HPLC for Measurement of 5-Hydroxyindole-3-acetic Acid in Urine
Clin. Chem., September 1, 2005; 51(9): 1698 - 1703.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
K. H. Schulpis, G. A. Karikas, J. Tjamouranis, H. Michelakakis, and S. Tsakiris
Acetylcholinesterase Activity and Biogenic Amines in Phenylketonuria
Clin. Chem., October 1, 2002; 48(10): 1794 - 1796.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (36)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kema, I. P.
Right arrow Articles by de Vries, E. G.E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kema, I. P.
Right arrow Articles by de Vries, E. G.E.
Related Collections
Right arrow Endocrinology and Metabolism

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS