- Email: [email protected]
HPLC analysis of carbohydrate deficient transferrin isoforms isolated by the Axis-Shield %CDT method
Clinica Chimica Acta 356 (2005) 143 – 146 www.elsevier.com/locate/clinchim
HPLC analysis of carbohydrate deficient transferrin isoforms isolated by the Axis-Shield %CDT method Anna Alde´na, Sten Ohlsona, Peter P3hlssonb, Ingvar Ryde´nc,* a
Department of Chemistry and Biomedical Sciences, University of Kalmar, SE-391 82 Kalmar, Sweden b Department of Biomedicine and Surgery, Linko¨ping University, SE-581 85 Linko¨ping, Sweden c Department of Clinical Chemistry, Kalmar County Hospital, SE-391 85 Kalmar, Sweden Received 19 November 2004; received in revised form 12 January 2005; accepted 12 January 2005
Abstract Background: Carbohydrate-deficient transferrin (CDT) is elevated during prolonged overconsumption of alcohol and CDT is considered to be the most specific biochemical marker for alcohol overconsumption. However, an accurate method for analysing CDT is necessary because the test is frequently used for example in legal matters. Methods: Patient serum samples were analysed with the Axis-Shield %CDT and eluates were pooled together. Transferrin was purified from the pool by affinity chromatography and further analysed with HPLC to determine the ratios of different transferrin isoforms. Results: In the eluates using the Axis-Shield %CDT method, a substantial amount of trisialo transferrin was found, which is generally not considered a CDT isoform. Conclusions: The fact that trisialo transferrin is present may generate falsely elevated CDT results and it could at least partly explain the discrepancy between results of the Axis-Shield %CDT assay and HPLC in routine analysis. D 2005 Elsevier B.V. All rights reserved. Keywords: Carbohydrate-deficient transferrin; Ion-exchange chromatography; Affinity chromatography; HPLC
1. Introduction Carbohydrate-deficient transferrin (CDT) generally refers to the asialo, monosialo and disialo isoforms of serum transferrin. These isoforms in-
T Corresponding author. Tel.: +46 480 81049; fax: +46 480 81025. E-mail address: [email protected] (I. Ryde´n). 0009-8981/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cccn.2005.01.021
crease as a result of alcohol overconsumption and CDT is considered to be the most specific biochemical marker of chronic alcohol abuse [1–5]. High-performance liquid chromatography (HPLC) based on anion-exchange chromatography is commonly used to measure the amount of CDT in relation to the total amount of transferrin and HPLC is the preferred method for analysis of CDT in legal matters and for reanalyzing positive results from immunological methods [6]. Recently a new version
144
A. Alde´n et al. / Clinica Chimica Acta 356 (2005) 143–146
of an immunological method for analysis of CDT was introduced, Axis-Shield %CDT (Axis-Shield A/ S, Oslo, Norway), in which disposable columns are used to separate CDT fractions by ion-exchange chromatography. The eluted CDT isoforms are then quantified by turbidimetry or nephelometry, using rabbit anti-human transferrin antibodies. The total transferrin concentration in the sample is determined separately by using the same antibodies and the fraction of CDT isoforms is calculated as percent of total transferrin. According to the manufacturer of the Axis-Shield %CDT method, the assay only includes asialo, monosialo and disialo transferrin isoforms. Whether the trisialo isoform should be included in CDT analysis has been debated [7,8]. However, recent studies have shown that trisialo transferrin has no applicable value in measuring CDT for detection of alcohol abuse and high levels of monosialo and trisialo transferrin have been seen in patients without alcohol abuse [7–9]. In a previous study, CDT results from the Axis-Shield %CDT method were compared with HPLC (asialo, monosialo and disialo isoforms) in 130 clinical samples [10]. An acceptable correlation (r 2=0.92) was found between the two methods. However, the Axis-Shield %CDT method showed higher CDT levels in a majority of the studied patients. Ionexchange chromatography, followed by turbidimetric immunoassays has shown to give higher CDT levels compared to HPLC also in other studies [11]. In the present study, we analysed CDT fractions isolated by the Axis-Shield %CDT method with HPLC, in order to find a reason for the discrepancy between the two methods.
2.2. Purification of transferrin Transferrin from the pool was purified using a HiTrap N-hydroxysuccinimide-activated affinity gel (5 mL, 1.62.5 cm, Amersham Biosciences, Uppsala, Sweden) with immobilized polyclonal rabbit antihuman transferrin antibodies (A0061, Dako, Copenhagen, Denmark) according to the manufacturer’s instructions. Briefly the anti-transferrin column was equilibrated with starting buffer (0.1 M sodium phosphate, 0.15 M NaCl; pH 7.0) and absorbance was measured at 280 nm. A volume of 250 mL of the same pool was applied at 2 mL/min (20 8C). After washing with starting buffer, elution was performed using 0.2 M glycine-HCl; pH 2.0. Fractions (1 mL) were collected and they were immediately neutralized with 0.2 M phosphate buffer; pH 8.9 to a pH of 6.5. 2.3. Transferrin quantification Fractions containing transferrin were pooled, concentrated with Vivacell Concentrator (VS 6001, 10 000 MWCO, 70 and 4 mL, Vivascience, Hanover, Germany) and sodium azide (0.08%) was added. Fractions were stored at 4 8C. The recovery of transferrin was determined according to a standard double-antibody sandwich ELISA method. As coating antibody rabbit anti-human transferrin (A0061, Dako) was used and as secondary antibody the same antibody was used but coupled to peroxidase by reductive amination according to standard procedures. 2.4. Determination of CDT isoforms
2. Materials and methods 2.1. The samples Eluates from routine analysis, performed with the Axis-Shield %CDT method [12] on a Beckman Coulter ImmageR (Beckman Coulter, Fullerton, USA), were pooled and stored at 70 8C. The investigated pool contained 430 serum samples and had a volume of 860 mL. Some of these samples were from repeated testing of the same patients in a follow-up period of treatment programs.
The CDT isoforms were analysed by HPLC using an anion-exchange column (Resource Q, 1 mL, 6.4 mm i.d.30 mm, Amersham Biosciences). Elution was performed with an increasing NaCl gradient (Fig. 1) in Bis-Tris (Bis(2-hydroxyethyl)amino-tris(hydroxymethyl)methane) buffer essentially as previously described [6]. Purified CDT from the pool (33 Ag) was saturated with FeCl3 and injected (100 Al). Analysis was performed at a flow rate of 1.5 mL/min at 25 8C using a 2690 Alliance pumpinjector-oven system and a SPD-10A spectrophotometric detector recording at 460 nm (Waters, Milford,
A. Alde´n et al. / Clinica Chimica Acta 356 (2005) 143–146
Fig. 1. HPLC analysis of purified CDT isoforms from the pool (solid line), and a reference pool (broken line). The reference is pooled serum from patients with high levels of CDT. pI 5.9 is asialo, 5.7 disialo, 5.6 trisialo, 5.4 tetrasialo and 5.2 pentasialo transferrin. The procedure was performed as described earlier [6]. Elution buffers were (A) Bis-Tris 20 mM; pH 6.2 and (B) BisTris 20 mM, NaCl 175 mM; pH 6.2. The salt gradient in this chromatogram is linear and starts at 2 min with 100% buffer A and ends at 20 min with 50% buffer A and 50% buffer B.
145
purified and concentrated transferrin fractions contained a low amount (~5%) of bands reacting with anti-transferrin antibody but having a molecular weight below the molecular weight of native transferrin. These bands probably reflect degraded transferrin and may not be detected in the immunological analysis of ELISA. In addition, the fraction also contained minor amounts of IgG (b1%). The acidic conditions used during affinity chromatography have in previous experiments been shown not to affect glycoprotein sialylation (data not shown). An HPLC chromatogram was obtained which showed separation of the CDT isoforms in the pool (Fig. 1). The ratios of the different isoforms in the pool were as follows; asialo transferrin (pI 5.9) 1.5%, disialo transferrin (pI 5.7) 70.7% and trisialo transferrin (pI 5.6) 27.8%. The small peak corresponding to pI 5.9 was calculated using valley– valley integration and the large double peak (pI 5.7 and 5.6) was calculated using forced baseline, since these two peaks were not fully separated. The most abundant transferrin isoform was disialo transferrin (pI 5.7), which would be expected, since it is the CDT isoform mainly related to overconsumption of alcohol. In addition, trisialo transferrin (pI 5.6) could also be detected. The identity of the isoforms
USA). Evaluation software was the Millenium chromatography data system. A reference sample of pooled serum obtained from Linkfping University Hospital with high CDT concentration was included as a positive control. Molecular mass analysis was performed with a Voyager-DEk Pro Bio-SpectrometrykWorkstation Perceptive Biosystem (Applied Biosystems, Foster City, USA), operating in the positive mode of detection with an acceleration voltage of 25 kV. As a matrix 3.5-dimetoxy-4-hydroxycinnamic acid (sinapinic acid, Sigma) was used in a saturated solution of 30% acetonitrile (ACN) and 0.15% trifluoro acetic acid (TFA).
3. Results and discussion The recovery of transferrin after affinity chromatography was 73% as measured by ELISA. According to SDS–PAGE and Western blot, the
Fig. 2. MALDI TOF mass spectrum of purified CDT isoforms from the CDT pool. Three peaks are clearly visible with the molecular weights of 75442, 77479 and 79294 Da. These molecular weights correspond to the transferrin isoforms of asialo, disialo and trisialo transferrin respectively.
146
A. Alde´n et al. / Clinica Chimica Acta 356 (2005) 143–146
was confirmed by MALDI TOF mass spectroscopy (Fig. 2). The calculated molecular mass for transferrin containing two biantennary glycans is 79570 Da. The spectrum shows two major peaks with molecular masses of 77479 and 79294 which corresponds well with the reported molecular weights of the disialo and trisialo transferrin isoforms, respectively [4]. In addition, a small peak with a molecular weight of 75523 is observed in the spectrum corresponding to the asialo transferrin isoform. The trisialo isoform of the rare allelic transferrin variant D elutes as the normal disialo transferrin C isoform in the HPLC analysis. Inclusion of some patients carrying the transferrin D variant could potentially give an overestimation of the amount of disialo transferrin in the HPLC analysis [2]. However, in the mass spectrometry analysis, the trisialo transferrin D would have the same molecular weight as trisialo transferrin C. According to the manufacturer of the Axis-Shield %CDT [12], this method should measure only the asialo, monosialo and disialo transferrin isoforms. However, the data presented here show that trisialo transferrin is also present which may partly account for the discrepancy found between the Axis-Shield %CDT and HPLC assays.
Acknowledgements Thanks are due to Inger Gustafsson at Kalmar County Hospital department for providing the patient pools and to Martin Johansson at the Clinical Biochemistry department at Linkfping University Hospital for the technical support. This work was funded by grants from the Knowledge Foundation and the University of Kalmar.
References [1] Arndt T. Carbohydrate-deficient transferrin as a marker of chronic alcohol abuse: a critical review of preanalysis, analysis, and interpretation. Clin Chem 2001;47(1):13 – 27. [2] Helander A, Eriksson G, Stibler H, Jeppsson J-O. Interference of transferrin isoform types with carbohydrate-deficient transferrin quantification in the identification of alcohol abuse. Clin Chem 2001;47(7):1225 – 33. [3] Landberg E, P3hlsson P, Lundblad A, Arnetorp A, Jeppsson JO. Carbohydrate composition of serum transferrin isoforms from patients with high alcohol consumption. Biochem Biophys Res Commun 1995;210(2):267 – 74. [4] Jochen P, Unverzagt C, Engel W-D, Renauer D, Seidel C, Hfsel W. Identification of carbohydrate deficient transferrin forms by MALDI-TOF mass spectrometry and lectin ELISA. Biochim Biophys Acta 1998;1380:93 – 101. [5] Stibler H, Borg S, Joustra M. Micro anion exchange chromatography of carbohydrate-deficient transferrin in serum in relation to alcohol consumption (Swedish patent 84005875). Alcohol: Clin Exp Res 1986;10(5):535 – 44. [6] Jeppsson J-O, Kristensson H, Fimiani C. Carbohydratedeficient transferrin quantified by HPLC to determine heavy consumption of alcohol. Clin Chem 1993;39(10):2115 – 20. [7] Dibbelt L. Does trisialo-transferrin provide valuable information for the laboratory diagnosis of chronically increased alcohol consumption by determination of carbohydrate-deficient transferrin? Clin Chem 2000;46(8 Pt1):1203 – 5. [8] Arndt T, Korzec A, B7r M, Kropf J. Further arguments against including trisialo-Fe2-transferrin in carbohydrate-deficient transferrin (CDT): a study on male alcoholics and hazardous drinkers. Med Sci Monit 2002;8(6):411 – 8. [9] Legros FJ, Nuyens V, Baudoux M, et al. Use of capillary zone electrophoresis for differentiating excessive from moderate alcohol consumption. Clin Chem 2003;49(3):440 – 9. [10] Ryden I, Gustafsson I, Svensson M, Jeppsson J-O. A new method for CDT analysis—a comparison of Axis-Shield %CDT with HPLC. Swedish Society of Clinical Chemistry Annual Spring Meeting, vol. 3. Uppsala, Sweden: Fyris-Tryck; 2001. [11] Arndt T, Keller T. Forensic analysis of carbohydrate-deficient tranferrin (CDT): implementation of a screening and confirmatory analysis concept is hampered by the lack of CDT isoform standards. For Sci Int 2004;146:9 – 16. [12] Instruction Manual, Axis-Shield %CDT, Axis-Shield A/S. P.O. Box 206 akern, N-0510, Oslo, Norway.
HPLC analysis of carbohydrate deficient transferrin isoforms isolated by the Axis-Shield %CDT method Anna Alde´na, Sten Ohlsona, Peter P3hlssonb, Ingvar Ryde´nc,* a
Department of Chemistry and Biomedical Sciences, University of Kalmar, SE-391 82 Kalmar, Sweden b Department of Biomedicine and Surgery, Linko¨ping University, SE-581 85 Linko¨ping, Sweden c Department of Clinical Chemistry, Kalmar County Hospital, SE-391 85 Kalmar, Sweden Received 19 November 2004; received in revised form 12 January 2005; accepted 12 January 2005
Abstract Background: Carbohydrate-deficient transferrin (CDT) is elevated during prolonged overconsumption of alcohol and CDT is considered to be the most specific biochemical marker for alcohol overconsumption. However, an accurate method for analysing CDT is necessary because the test is frequently used for example in legal matters. Methods: Patient serum samples were analysed with the Axis-Shield %CDT and eluates were pooled together. Transferrin was purified from the pool by affinity chromatography and further analysed with HPLC to determine the ratios of different transferrin isoforms. Results: In the eluates using the Axis-Shield %CDT method, a substantial amount of trisialo transferrin was found, which is generally not considered a CDT isoform. Conclusions: The fact that trisialo transferrin is present may generate falsely elevated CDT results and it could at least partly explain the discrepancy between results of the Axis-Shield %CDT assay and HPLC in routine analysis. D 2005 Elsevier B.V. All rights reserved. Keywords: Carbohydrate-deficient transferrin; Ion-exchange chromatography; Affinity chromatography; HPLC
1. Introduction Carbohydrate-deficient transferrin (CDT) generally refers to the asialo, monosialo and disialo isoforms of serum transferrin. These isoforms in-
T Corresponding author. Tel.: +46 480 81049; fax: +46 480 81025. E-mail address: [email protected] (I. Ryde´n). 0009-8981/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cccn.2005.01.021
crease as a result of alcohol overconsumption and CDT is considered to be the most specific biochemical marker of chronic alcohol abuse [1–5]. High-performance liquid chromatography (HPLC) based on anion-exchange chromatography is commonly used to measure the amount of CDT in relation to the total amount of transferrin and HPLC is the preferred method for analysis of CDT in legal matters and for reanalyzing positive results from immunological methods [6]. Recently a new version
144
A. Alde´n et al. / Clinica Chimica Acta 356 (2005) 143–146
of an immunological method for analysis of CDT was introduced, Axis-Shield %CDT (Axis-Shield A/ S, Oslo, Norway), in which disposable columns are used to separate CDT fractions by ion-exchange chromatography. The eluted CDT isoforms are then quantified by turbidimetry or nephelometry, using rabbit anti-human transferrin antibodies. The total transferrin concentration in the sample is determined separately by using the same antibodies and the fraction of CDT isoforms is calculated as percent of total transferrin. According to the manufacturer of the Axis-Shield %CDT method, the assay only includes asialo, monosialo and disialo transferrin isoforms. Whether the trisialo isoform should be included in CDT analysis has been debated [7,8]. However, recent studies have shown that trisialo transferrin has no applicable value in measuring CDT for detection of alcohol abuse and high levels of monosialo and trisialo transferrin have been seen in patients without alcohol abuse [7–9]. In a previous study, CDT results from the Axis-Shield %CDT method were compared with HPLC (asialo, monosialo and disialo isoforms) in 130 clinical samples [10]. An acceptable correlation (r 2=0.92) was found between the two methods. However, the Axis-Shield %CDT method showed higher CDT levels in a majority of the studied patients. Ionexchange chromatography, followed by turbidimetric immunoassays has shown to give higher CDT levels compared to HPLC also in other studies [11]. In the present study, we analysed CDT fractions isolated by the Axis-Shield %CDT method with HPLC, in order to find a reason for the discrepancy between the two methods.
2.2. Purification of transferrin Transferrin from the pool was purified using a HiTrap N-hydroxysuccinimide-activated affinity gel (5 mL, 1.62.5 cm, Amersham Biosciences, Uppsala, Sweden) with immobilized polyclonal rabbit antihuman transferrin antibodies (A0061, Dako, Copenhagen, Denmark) according to the manufacturer’s instructions. Briefly the anti-transferrin column was equilibrated with starting buffer (0.1 M sodium phosphate, 0.15 M NaCl; pH 7.0) and absorbance was measured at 280 nm. A volume of 250 mL of the same pool was applied at 2 mL/min (20 8C). After washing with starting buffer, elution was performed using 0.2 M glycine-HCl; pH 2.0. Fractions (1 mL) were collected and they were immediately neutralized with 0.2 M phosphate buffer; pH 8.9 to a pH of 6.5. 2.3. Transferrin quantification Fractions containing transferrin were pooled, concentrated with Vivacell Concentrator (VS 6001, 10 000 MWCO, 70 and 4 mL, Vivascience, Hanover, Germany) and sodium azide (0.08%) was added. Fractions were stored at 4 8C. The recovery of transferrin was determined according to a standard double-antibody sandwich ELISA method. As coating antibody rabbit anti-human transferrin (A0061, Dako) was used and as secondary antibody the same antibody was used but coupled to peroxidase by reductive amination according to standard procedures. 2.4. Determination of CDT isoforms
2. Materials and methods 2.1. The samples Eluates from routine analysis, performed with the Axis-Shield %CDT method [12] on a Beckman Coulter ImmageR (Beckman Coulter, Fullerton, USA), were pooled and stored at 70 8C. The investigated pool contained 430 serum samples and had a volume of 860 mL. Some of these samples were from repeated testing of the same patients in a follow-up period of treatment programs.
The CDT isoforms were analysed by HPLC using an anion-exchange column (Resource Q, 1 mL, 6.4 mm i.d.30 mm, Amersham Biosciences). Elution was performed with an increasing NaCl gradient (Fig. 1) in Bis-Tris (Bis(2-hydroxyethyl)amino-tris(hydroxymethyl)methane) buffer essentially as previously described [6]. Purified CDT from the pool (33 Ag) was saturated with FeCl3 and injected (100 Al). Analysis was performed at a flow rate of 1.5 mL/min at 25 8C using a 2690 Alliance pumpinjector-oven system and a SPD-10A spectrophotometric detector recording at 460 nm (Waters, Milford,
A. Alde´n et al. / Clinica Chimica Acta 356 (2005) 143–146
Fig. 1. HPLC analysis of purified CDT isoforms from the pool (solid line), and a reference pool (broken line). The reference is pooled serum from patients with high levels of CDT. pI 5.9 is asialo, 5.7 disialo, 5.6 trisialo, 5.4 tetrasialo and 5.2 pentasialo transferrin. The procedure was performed as described earlier [6]. Elution buffers were (A) Bis-Tris 20 mM; pH 6.2 and (B) BisTris 20 mM, NaCl 175 mM; pH 6.2. The salt gradient in this chromatogram is linear and starts at 2 min with 100% buffer A and ends at 20 min with 50% buffer A and 50% buffer B.
145
purified and concentrated transferrin fractions contained a low amount (~5%) of bands reacting with anti-transferrin antibody but having a molecular weight below the molecular weight of native transferrin. These bands probably reflect degraded transferrin and may not be detected in the immunological analysis of ELISA. In addition, the fraction also contained minor amounts of IgG (b1%). The acidic conditions used during affinity chromatography have in previous experiments been shown not to affect glycoprotein sialylation (data not shown). An HPLC chromatogram was obtained which showed separation of the CDT isoforms in the pool (Fig. 1). The ratios of the different isoforms in the pool were as follows; asialo transferrin (pI 5.9) 1.5%, disialo transferrin (pI 5.7) 70.7% and trisialo transferrin (pI 5.6) 27.8%. The small peak corresponding to pI 5.9 was calculated using valley– valley integration and the large double peak (pI 5.7 and 5.6) was calculated using forced baseline, since these two peaks were not fully separated. The most abundant transferrin isoform was disialo transferrin (pI 5.7), which would be expected, since it is the CDT isoform mainly related to overconsumption of alcohol. In addition, trisialo transferrin (pI 5.6) could also be detected. The identity of the isoforms
USA). Evaluation software was the Millenium chromatography data system. A reference sample of pooled serum obtained from Linkfping University Hospital with high CDT concentration was included as a positive control. Molecular mass analysis was performed with a Voyager-DEk Pro Bio-SpectrometrykWorkstation Perceptive Biosystem (Applied Biosystems, Foster City, USA), operating in the positive mode of detection with an acceleration voltage of 25 kV. As a matrix 3.5-dimetoxy-4-hydroxycinnamic acid (sinapinic acid, Sigma) was used in a saturated solution of 30% acetonitrile (ACN) and 0.15% trifluoro acetic acid (TFA).
3. Results and discussion The recovery of transferrin after affinity chromatography was 73% as measured by ELISA. According to SDS–PAGE and Western blot, the
Fig. 2. MALDI TOF mass spectrum of purified CDT isoforms from the CDT pool. Three peaks are clearly visible with the molecular weights of 75442, 77479 and 79294 Da. These molecular weights correspond to the transferrin isoforms of asialo, disialo and trisialo transferrin respectively.
146
A. Alde´n et al. / Clinica Chimica Acta 356 (2005) 143–146
was confirmed by MALDI TOF mass spectroscopy (Fig. 2). The calculated molecular mass for transferrin containing two biantennary glycans is 79570 Da. The spectrum shows two major peaks with molecular masses of 77479 and 79294 which corresponds well with the reported molecular weights of the disialo and trisialo transferrin isoforms, respectively [4]. In addition, a small peak with a molecular weight of 75523 is observed in the spectrum corresponding to the asialo transferrin isoform. The trisialo isoform of the rare allelic transferrin variant D elutes as the normal disialo transferrin C isoform in the HPLC analysis. Inclusion of some patients carrying the transferrin D variant could potentially give an overestimation of the amount of disialo transferrin in the HPLC analysis [2]. However, in the mass spectrometry analysis, the trisialo transferrin D would have the same molecular weight as trisialo transferrin C. According to the manufacturer of the Axis-Shield %CDT [12], this method should measure only the asialo, monosialo and disialo transferrin isoforms. However, the data presented here show that trisialo transferrin is also present which may partly account for the discrepancy found between the Axis-Shield %CDT and HPLC assays.
Acknowledgements Thanks are due to Inger Gustafsson at Kalmar County Hospital department for providing the patient pools and to Martin Johansson at the Clinical Biochemistry department at Linkfping University Hospital for the technical support. This work was funded by grants from the Knowledge Foundation and the University of Kalmar.
References [1] Arndt T. Carbohydrate-deficient transferrin as a marker of chronic alcohol abuse: a critical review of preanalysis, analysis, and interpretation. Clin Chem 2001;47(1):13 – 27. [2] Helander A, Eriksson G, Stibler H, Jeppsson J-O. Interference of transferrin isoform types with carbohydrate-deficient transferrin quantification in the identification of alcohol abuse. Clin Chem 2001;47(7):1225 – 33. [3] Landberg E, P3hlsson P, Lundblad A, Arnetorp A, Jeppsson JO. Carbohydrate composition of serum transferrin isoforms from patients with high alcohol consumption. Biochem Biophys Res Commun 1995;210(2):267 – 74. [4] Jochen P, Unverzagt C, Engel W-D, Renauer D, Seidel C, Hfsel W. Identification of carbohydrate deficient transferrin forms by MALDI-TOF mass spectrometry and lectin ELISA. Biochim Biophys Acta 1998;1380:93 – 101. [5] Stibler H, Borg S, Joustra M. Micro anion exchange chromatography of carbohydrate-deficient transferrin in serum in relation to alcohol consumption (Swedish patent 84005875). Alcohol: Clin Exp Res 1986;10(5):535 – 44. [6] Jeppsson J-O, Kristensson H, Fimiani C. Carbohydratedeficient transferrin quantified by HPLC to determine heavy consumption of alcohol. Clin Chem 1993;39(10):2115 – 20. [7] Dibbelt L. Does trisialo-transferrin provide valuable information for the laboratory diagnosis of chronically increased alcohol consumption by determination of carbohydrate-deficient transferrin? Clin Chem 2000;46(8 Pt1):1203 – 5. [8] Arndt T, Korzec A, B7r M, Kropf J. Further arguments against including trisialo-Fe2-transferrin in carbohydrate-deficient transferrin (CDT): a study on male alcoholics and hazardous drinkers. Med Sci Monit 2002;8(6):411 – 8. [9] Legros FJ, Nuyens V, Baudoux M, et al. Use of capillary zone electrophoresis for differentiating excessive from moderate alcohol consumption. Clin Chem 2003;49(3):440 – 9. [10] Ryden I, Gustafsson I, Svensson M, Jeppsson J-O. A new method for CDT analysis—a comparison of Axis-Shield %CDT with HPLC. Swedish Society of Clinical Chemistry Annual Spring Meeting, vol. 3. Uppsala, Sweden: Fyris-Tryck; 2001. [11] Arndt T, Keller T. Forensic analysis of carbohydrate-deficient tranferrin (CDT): implementation of a screening and confirmatory analysis concept is hampered by the lack of CDT isoform standards. For Sci Int 2004;146:9 – 16. [12] Instruction Manual, Axis-Shield %CDT, Axis-Shield A/S. P.O. Box 206 akern, N-0510, Oslo, Norway.