Use of Horseradish Peroxidase- and ... - Clinical Chemistry

3 downloads 153 Views 444KB Size Report
Use of Horseradish Peroxidase- and. Fluorescein-modified Cisplatin Derivatives for Simultaneous Labeling of. Nucleic Acids and Proteins. Rob P.M. van ...
Clinical Chemistry 48:8 1352–1359 (2002)

Cancer Diagnostics: Discovery and Clinical Applications

Use of Horseradish Peroxidase- and Fluorescein-modified Cisplatin Derivatives for Simultaneous Labeling of Nucleic Acids and Proteins Rob P.M. van Gijlswijk,1,2* Eduard G. Talman,1,3* Inge Peekel,1 Judith Bloem,1 Marcel A. van Velzen,1 Rob J. Heetebrij,1,3 and Hans J. Tanke2†

Background: Microarray platforms will change immunochemical and nucleic acid-based analysis of cell homogenates and body fluids compared with classic analyses. Microarrays use labeled target and immobilized probes, rather than fixed targets and labeled probes. We describe a method for simultaneous labeling of nucleic acids and proteins. Methods: Horseradish peroxidase- and fluoresceinmodified cisplatin derivatives were used for labeling of nucleic acids and proteins. These reagents, called the Universal Linkage System (ULS), bind to sulfur- and nitrogen-donor ligands present in amino acids and nucleotides. For automated screening of proteins and nucleic acids on microarrays, it is advantageous to label these biomolecules without pre- or postpurification procedures. The labeling of antibodies and nucleic acids in whole serum was therefore pursued. Results: Immunoglobulins in nonpurified serum were labeled efficiently enough to be used for immunochemistry. To investigate whether protein-adapted labeling allowed nucleic acid labeling as well, 1 ␮g of plasmid DNA was added to 1 ␮L of serum. DNA and serum proteins were simultaneously labeled, and this labeled DNA could be used as a probe for direct fluorescence in situ hybridization.

1 Kreatech Biotechnology, Vlierweg 20, 1032 LG Amsterdam, The Netherlands. 2 Laboratory for Cytochemistry and Cytometry, Department of Molecular Cell Biology, Leiden University Medical Center, Wassenaarseweg 72, 2333 AL Leiden, The Netherlands. 3 Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden University, PO Box 9502, 2300 RA Leiden, The Netherlands. *Both authors contributed equally to this work. †Author for correspondence. E-mail [email protected]. Received March 26, 2002; accepted May 23, 2002.

Conclusion: ULS provides a direct labeling tool for the (simultaneous) modification of proteins and nucleic acids even in unpurified samples. © 2002 American Association for Clinical Chemistry

The specific interaction of biomolecules underlies cytochemical methods such as immunocytochemistry (antigens and antibodies) and in situ hybridization (complementary nucleic acid sequences). These methods share a common factor: a specific reagent is labeled with a reporter molecule to investigate target molecules present in cells, tissues, or chromosomes. Frequently used reporter molecules are fluorophores, enzymes, or electron dense particles. Technical developments of the last few years, mainly in the area of genomic research, have led to the application of a reversed approach in which the target molecule in extracts, homogenates, or body fluids is labeled and then applied to platforms that contain the capture molecule. One of the early examples is comparative genomic hybridization (1 ). With the introduction of array technology, the use of this approach has increased. Thus, target labeling is increasingly applied in research and is foreseen in diagnostic applications. Because the latter may be based on the identification of nucleic acids (DNA and RNA) and proteins in the same specimen, there is a need for a procedure that labels biomolecules in a sample, in a single step, irrespective of their nature. Methods for target labeling must meet additional demands (2 ) for routine clinical analysis. Because thousands of samples per analysis are required, the number of labeling reactions will increase as the amount of target material decreases. Furthermore, purification steps should be avoided to maintain the biomolecule ratio in the sample and to decrease assay times as well as prevent sample contamination.

1352

Clinical Chemistry 48, No. 8, 2002

We describe the use of cisplatin derivatives for the target labeling of nucleic acids and proteins. The reagents that were used are called the Universal Linkage System (ULS)4 (3 ), and they have been used for the labeling of nucleic acids in various hybridization applications with applications in molecular biology, cell biology, pathology, and cytogenetics (3– 6 ). In nucleic acids, ULS reagents link to guanine and, less frequently, to adenine (7 ). In proteins they link to free thiol groups (cysteine), methionine, and histidine (8, 9 ). The reporter molecules that are used for ULS labeling are either fluorophores or haptens and thus do restrict ULS-based target labeling to nonenzymatic fluorescent methods only. Horseradish peroxidase (HRP) is frequently used as a label for nucleic acids in combination with sensitive chemiluminescence detection in case of filter hybridization (10, 11 ). Larger DNA probes, such as PCR products and plasmids suitable for filter hybridizations, can be labeled according to the method of Renz and Kurz (12 ). When these labeled probes were combined with luminol as the chemiluminescent substrate, the sensitivities of Southern and Northern blotting applications were reported to be comparable to those radioactive methods (13 ). Although this method would suggest that other nucleic acid probes, such as RNA and oligonucleotides, can also be labeled, it appears to be restricted to DNA probes. Different methods to modify oligonucleotides with HRP have been described (14, 15 ). Fluorescence detection allows the use of HRP-labeled oligonucleotides for fluorescence in situ hybridization (FISH) (15, 16 ). HRP-labeled oligonucleotides were used in combination with tyramide signal amplification (TSA) (17–19 ) and allowed sensitive FISH applications, such as the detection of highly and moderately repetitive DNA sequences in metaphase preparations (15 ) and the detection of lowabundance mRNA in cells (16 ). We describe here the synthesis of HRP-ULS®, a new label for the ULS system. Whole serum was used to investigate the use of ULS for protein target labeling. To investigate simultaneous labeling of nucleic acids and proteins, a plasmid probe was added to serum. ELISAs were used to visualize serum-labeled antibodies. In addition, the principle of simultaneous labeling of nucleic acids and proteins was pursued because this would allow for the same-labeled mixture to be analyzed for genomics as well as proteomics (20 ) without the need of different labeling procedures or purification steps.

4 Nonstandard abbreviations: ULS, Universal Linkage System; HRP, horseradish peroxidase; FISH, fluorescence in situ hybridization; TSA, tyramide signal amplification; NMR, nuclear magnetic resonance; ECL, enhanced chemiluminescence; HPV-6, human papillomavirus type 6; PBS, phosphate-buffered saline; BSA, bovine serum albumin; and TMB, 3,3⬘,5,5⬘tetramethylbenzidine.

1353

Materials and Methods preparation of hrp-modified platination reagents The general outline of the synthesis of HRP-ULS is shown in Fig. 1. The first phase of the synthesis involved the preparation of hexanediamine-modified deoxyguanosine (1): 10.5 mg of 2⬘-deoxyguanosine 5⬘-triphosphate (sodium salt; cat no. G-8877; Sigma) was dissolved in 1.0 mL of Millipore-quality water; 13.4 mg of 1,6-diaminohexane (cat. no. H-2381; Sigma) in 220 ␮L of dimethylformamide (cat. no. 7032; JT Baker) and 2.0 mL of 100 mmol/L imidazole (cat. no. 12202; Acros; dissolved in MilliQ water, pH adjusted to 6.0 with HCl) were added and mixed. Finally, 55.6 mg of 1-(3-dimethylaminopropyl-3ethylcarbodiimide hydrochloride (cat. no 16146-2; Aldrich) dissolved in MilliQ water was added, mixed, and reacted overnight at 40 °C. Hexanediamine-modified deoxyguanosine was purified on a Mono Q column (Mono Q HR 5/5; cat. no. 17-0546-01; APBiotech) in a 1.0 mol/L ammonium hydrogen carbonate gradient in water (flow, 1.0 mL/min; linear 0 –100% gradient; main product at 37% ammonium hydrogen carbonate). Identity of the product was confirmed by proton nuclear magnetic resonance (1H-NMR): (D2O, ppm) 8.10 H8 (s), 6.31 H1⬘ (dd, 3J ⫽ 6.9 Hz), 4.74 H3⬘ (m), 4.22 H4⬘ (m), 4.17 H5⬘/5⬙ (m), 3.19 H2⬘ (m), 2.93 ␥CH2 (dd, 3J ⫽ 7.4 Hz), 2.83 ␣CH2 (m), 2.51 H2⬙ (m), 1.56 ⑀CH2 (m), 1.36 ␤CH2 (m), 1.24 ␥CH2/␦CH2 (m), where s is a singlet, dd is a doublet-doublet, and m is a multiplet; and 31P-NMR (D2O, ␦ ppm): 0.15 ␥P (broad), ⫺10.47 ␣P (sharp), ⫺21.61 ␤P (broad). Both NMR spectra were recorded on a 300-MHz Bruker DPX-300 spectrometer. The second step of the synthesis was the preparation of HRP-guanosine conjugates (2): 21.9 mg of HRP (cat. no. 31491; Pierce) was dissolved in 1.0 mL of a 50 mmol/L sodium acetate buffer, pH 5.5, and 137 ␮L of a freshly prepared 100 mmol/L solution of sodium periodate (cat. no. 30200; BDH Chemicals Ltd.) in MilliQ water was then added. The color rapidly changed from brown to green. This solution was incubated for 2 h on ice. Thirty microliters of diethylene glycol (cat. no. 803131; Merck) was added to stop the reaction. After an additional incubation for 15 min on ice, the reaction mixture was purified on a 10 ⫻ 1 cm Sephadex-G50 column (PD-10 Prepacked Disposable Column; product code 17-0851-01; APBiotech) with 50 mmol/L sodium phosphate, pH 7.5, as the eluent. The colored fractions were pooled (⬃2.5 mL), and 6.5 mg of hexanediamine-modified deoxyguanosine in 1.0 mL of MilliQ water was added. This mixture was allowed to react for 2 h at room temperature. The remaining aldehyde moieties were reduced by the addition of 137 ␮L of a 60 g/L solution of sodium cyanoborohydride (cat. no. 16855; Acros) in MilliQ water. After a 2-h incubation on ice, the product was purified and concentrated to 75 ␮L with Centricon 30 devices (cat. no. 4208; Amicon) according to the supplied manual. The HRP concentration was 70 g/L, and the activity was 28 500 kilounits/L.

1354

van Gijlswijk et al.: Simultaneous Labeling of Proteins and Nucleic Acids

Fig. 1. Outline of the synthesis of HRP-ULS. EDC, 1-(3-dimethylaminopropyl-3-ethylcarbodiimide hydrochloride; G, guanosine; Pt(en)-Cl, [chloro(ethylenediamine)platinum(II)]⫹.

Clinical Chemistry 48, No. 8, 2002

The third step was the preparation of HRP-ULS reagent (3): to 1.0 mg of guanosine-modified HRP (15 ␮L of product 2) in 50 mmol/L sodium phosphate, pH7.5, 50 ␮L of a 57 mmol/L solution of cisplatin-Cl-NO3 (4) in MilliQ water was added. The mixture was allowed to react for 15 min at 42 °C, and then 60 ␮L of 10 mmol/L Tris-HCl, pH 9.0, was added, mixed, and incubated for 15 min at 42 °C, followed by a 20-h incubation at 25 °C. HRP-ULS was purified from unreacted cisplatin and concentrated to 100 ␮L using Centricon 30 devices according to the supplied manual. The HRP concentration was 10 g/L, and the activity was 3900 kilounits/L. HRP-ULS reagents could be stored at 4 °C for 6 months without loss of labeling performance.

hrp labeling of nucleic acids For comparison, double-stranded DNA probes were labeled with HRP, using HRP-ULS reagents as well as with the enhanced chemiluminescence (ECL) labeling and detection reagent set (cat. no. RPN3000; APBiotech). The probes used were specific for human papillomavirus type 6 (HPV-6) and human chromosome 1q12 (4 ). Before labeling with HRP, the double-stranded DNA probes (100 ng in 10 ␮L of Tris-EDTA) were denatured by incubation for 5 min at 100 °C, followed by incubation on ice for 5 min. For ULS labeling, 40 ␮g (4 ␮L) of HRP-ULS was added to 100 ng of denatured DNA probe. Reaction times varied, and a reaction time of 15 min at 37 °C was sufficient. The probes were immediately diluted in hybridization mixture and used for hybridization within 24 h. ECL labeling was done according to the instructions provided by the manufacturer.

filter hybridizations and fish For filter hybridization assays, a threefold dilution series of unlabeled HPV-6 DNA was spotted onto Hybond-N⫹ (cat. no. RPN303B; APBiotech) and hybridized with HRPlabeled HPV-6 probes. Filter hybridizations were performed with reagents from the ECL reagent set and according to the supplied instructions. For the HRP-labeled 1q12 probe on human metaphase chromosomes, pretreatment was performed as described by Cooke and Hindley (21 ). HRP-labeled probes were hybridized at concentrations of 2–5 mg/L in ECL-gold hybridization mixture (cat. no. RPN3006; APBiotech). Hybridization times varied from 1 to 16 h at 37 °C. After hybridization, the slides were washed twice (5 min each time) in 2⫻ standard saline citrate containing 1 mL/L Tween 20 at 37 °C, followed by two stringent washes (3 min each) in 2⫻ standard saline citrate containing 500 mL/L formamide at 42 °C. After a short wash in phosphate-buffered saline (PBS) at room temperature, slides were subjected to fluorescent TSA detection as described by van Gijlswijk et al. (15 ).

1355

hrp labeling of avidin Avidin-D (Vector Laboratories) was labeled with HRPULS by incubation of 100 ␮g of avidin in 100 ␮L of PBS with 25 ␮g of HRP-ULS for 16 h at 37 °C. The remaining reactive platinum sites were blocked with 44 mmol/L sodium diethyldithiocarbamate (cat. no. 21161; Acros), and the conjugate was used without further purification. Avidin-HRP conjugates were tested for their performance in ELISA. Biotinylated bovine serum albumin [BSA; 10 mg/L in PBS; 1 mg of BSA (Sigma type V) was labeled with biotin-lc-succinimidyl ester (Pierce) to ⬃4 biotin molecules/BSA molecule] was coated overnight at room temperature in 96-well plates (high binding; Greiner). As negative controls, some wells were coated with unlabeled BSA (Sigma type V). After four washes with PBS, the plates were blocked with 30 g/L BSA in PBS. The wells were emptied and incubated for 1 h at 37 °C with avidin-HRP conjugates at various concentrations. A commercially available avidin-HRP conjugate (Dako) was used as a control. A solution of 20 g/L casein (cat. no. C-8654; Sigma), 10 g/L BSA, and 0.5 g/L thimerosal (cat. no. 817043; Merck) in PBS was used as incubation buffer for the ELISA. The wells were washed four times with PBS and incubated 30 min at room temperature with 0.4 g/L 3,3⬘,5,5⬘-tetramethylbenzidine (TMB; Kreatech Biotechnology) in a citrate–phosphate buffer, pH 5.5. The addition of one volume of 0.5 mol/L sulfuric acid stopped the reaction. Absorbance was measured at 450 nm.

uls labeling of whole serum Normal goat serum (Sigma) as well as mouse IgG-immunized goat serum (Sigma) were used for target labeling with ULS. One microliter of serum was diluted in 50 ␮L of PBS, and various amounts of fluorescein- or HRP-ULS were added. The addition of 15 ␮g of fluorescein-ULS or 10 ␮L of HRP-ULS was optimal. After an overnight incubation at 37 °C, the reaction was quenched with 5 ␮L of 44 mmol/L sodium diethyldithiocarbamate (Acros 21161). Mouse IgG (Sigma; 4 mg/L in PBS) was coated to 96-well plates and blocked as described above. The wells were incubated with different dilutions of the labeled sera. Fluorescein-labeled sera were incubated with a HRPlabeled rabbit anti-fluorescein antibody (Kreatech Diagnostics). Incubation solutions, washes, and detection with TMB were performed as described for avidin-HRP conjugates. Plasmids containing the 1.77-kb fragment that hybridizes to human chromosome 1q12 (a generous gift of Dr. H. Cooke, UCLA, Los Angeles, CA) were sonicated to a fragment length of 200 –300 bp. This DNA was added to human serum in various amounts, and this mixture was labeled as described above. Part of the labeled serum was used for hybridization and part for ELISA. Without any purification serum was diluted in hybridization mixture to a DNA concentration of 1 mg/L. FISH was performed on human chromosome preparations as described previ-

1356

van Gijlswijk et al.: Simultaneous Labeling of Proteins and Nucleic Acids

ously (15 ). The labeling of proteins in this mixture was determined by analyzing the labeling of human IgM in serum. Rabbit anti-human IgM antibodies (Dako) were coated onto the plates (10 mg/L in PBS for 2 h at 37 °C). After four washes with PBS, the plates were blocked with 30 g/L BSA in PBS. The wells were emptied and incubated for 1 h at 37 °C with fluorescein-labeled serum, followed by an anti-fluorescein-HRP incubation step and TMB detection as described above. A solution containing 20 g/L casein, 10 g/L BSA, and 0.5 g/L thimerosal in PBS was used as incubation buffer for the ELISA.

Results Several synthetic approaches were undertaken to modify HRP by platinum-based reactive groups. Because cisplatin binds to proteins, the straightforward synthesis of HRP-ULS involves incubating HRP with this bifunctional platinum reagent. A HRP-bound monofunctional platinum group was obtained. Determination of the platinum concentration by flameless atomic absorbance spectroscopy showed platination of HRP at ⬃1 platinum/HRP molecule and, thus, one reactive group per HRP. However, for DNA labeling, relatively high concentrations of HRP-ULS had to be used, which combined with prolonged incubation times decreased HRP activity to 20% and consequently yielded suboptimal results for filter hybridizations. To increase binding of a platinum unit to HRP, guanosine, which is known to be very reactive toward platinum derivatives, was conjugated to the protein. dGTP was conjugated to 1,6-diaminohexane to yield amino-dGTP. Amino-dGTP was coupled to HRP, and this conjugate was subsequently modified with a platinum species. By varying the chemistry, we obtained products with different degrees of platinum modification. These spaced HRP-ULS reagents allowed much lower platinum concentrations as well as milder reaction conditions for the last step of the preparation of HRP-ULS, better preserving HRP activity. The product described in Materials and Methods contains 5.5 ⫾ 0.5 platinum molecules/HRP molecule. When applied for filter hybridization, HPV-6 probes prepared with HRP-ULS performed at least as well as the control, which was prepared with the ECL method (Fig. 2). Fig. 3 shows that HRP-labeled probes could be used for TSA-based FISH. Variations in chemistry included the use of carbohydrazide as spacer in place of 1,6-diaminohexane, which eliminated the need for a final borohydride reduction step. In addition, GTP modification of HRP was done in reverse order: polyamino-HRP was prepared by periodate oxidation and coupling with a diamino reagent, followed purification by gel filtration and incubation with GTP, 1-(3-dimethylaminopropyl-3-ethylcarbodiimide hydrochloride, and imidazole. After reaction with dichloro(ethylenediamine)platinum(II) [Pt(en)Cl2], each method yielded platinum-to-HRP ratios comparable to the method described in Materials and Methods.

Fig. 2. HPV-6 DNA labeled with HRP-ULS, detected by ECL. The probe was hybridized on nylon membranes containing spots of HPV-6 DNA ranging from 1000 to 0.1 pg. An ECL signal could be detected for concentrations up to 1 pg.

Avidin was used to investigate protein labeling with HRP-ULS. As shown in Fig. 4, biotin-reactive HRPlabeled avidin was obtained after incubation with HRPULS. Furthermore, labeling was at least as efficient as with standard HRP-labeled avidin. Optimum ratios of avidin to HRP-ULS were empirically determined by labeling small quantities of avidin (10 –50 ␮g) with various concentrations of HRP-ULS, which were then tested for their biotin-binding capacity in ELISA. In addition, goat serum was used as a model for protein labeling in highly complex mixtures. Whole goat serum containing antibodies against mouse IgG was labeled with ULS, and an ELISA was used to monitor labeled anti-mouse antibodies in the serum. Serum from a nonimmunized goat was used as a negative control. As shown in Fig. 5, labeled mouse IgG-reactive antibodies

Fig. 3. Fluorescence detection of plasmid probe pUC1.77 labeled with HRP-ULS. The probe was hybridized on metaphase spreads and detected using the Cyanine 3 tyramide detection system (Perkin Elmer Life Sciences). Fluorescence signals are visible on the centromere of chromosome 1 (1q1.2).

1357

Clinical Chemistry 48, No. 8, 2002

Fig. 4. Comparison of avidin-HRP reagents in ELISA. Avidin-HRP was incubated on BSA-coated microtiter plates (dotted line) or on biotin-labeled BSA-coated microtiter plates (solid lines). HRP-ULS-labeled avidin (ⴱ) was compared with commercially available avidin-HRP (E).

were obtained with both fluorescein-ULS and HRP-ULS in the immunized serum but not in the control. Similar to avidin, the optimum ULS-to-protein ratios were empirically determined. These optimum conditions were subsequently used for the simultaneous labeling of nucleic acids and proteins within the same mixture. For this purpose, 1 ␮g of the DNA probe was added to 1 ␮L of human serum, and this mixture was allowed to react with fluorescein-ULS. The mixture was then tested by direct fluorescence hybridization and ELISA with detection by labeled human IgM. As shown in Fig. 6B, the pUC1.77 DNA in the mixture was labeled sufficiently to show specific hybridization signals on human chromosome 1q12 (21 ). FISH results were easily detected by direct fluorescence and appeared as bright as the control probe labeled in the absence of protein. The efficiency of protein labeling in this mixture is shown in Fig. 6A. A capture-ELISA for human IgM was used to monitor fluorescein labeling. Detection limits were ⬍50 ␮g/L IgM.

Fig. 5. Labeling of mouse IgG-immunized serum (F and Œ) or nonimmunized normal goat serum (E and ‚) with either fluorescein-ULS (FLU; circles) or HRP-ULS (triangles).

Fig. 6. Simultaneous labeling of proteins (A) and DNA (B) in human serum with fluorescein-ULS. (A), human IgM detected by ELISA; (B), pUC1.77 plasmid detected by FISH.

Discussion The results described here show the application of a method to synthesize a reactive HRP-labeled ULS reagent that was used to label both nucleic acids and proteins based on coordination chemistry. Both fluorescein-ULS and HRP-ULS were used for target labeling of immunoglobulins and DNA in whole serum. HRP-ULS-labeled probes were used in different hybridization formats, and labeled immunochemical reagents were tested in an ELISA. Introduction of a spacer between HRP and the nucleobase or amino-acid-specific reactive group as well as the introduction of multiple reactive groups on a single HRP allowed for mild reaction conditions for HRP labeling of biomolecules. Mild conditions are important for preservation of enzyme activity and probe stability. Because double-stranded DNA must be denatured before labeling with a heat- or pH-sensitive enzyme, minimal incubation times would allow minimal renaturation of the probe. Furthermore, nucleic acids that

1358

van Gijlswijk et al.: Simultaneous Labeling of Proteins and Nucleic Acids

are sensitive to degradation, such as RNA, would also benefit from short incubation times. In filter hybridization applications, HRP-ULS-labeled probes were as sensitive as the ECL method, which is known for sensitive filter applications such as Southern and Northern blotting (13 ). Furthermore, we showed that HRP-ULS-labeled probes combined with a TSA reaction could be used in FISH applications. Avidin was used to investigate protein labeling with HRP-ULS. Small quantities were labeled with various concentrations of HRP-ULS. The reaction could easily be scaled up for optimum ratios. Avidin retained its biotinbinding capacity, and performance was comparable to that for commercially available avidin-HRP conjugates. Different ULS reagents were used for the target labeling of nucleic acids and proteins. Fluorescent labeling of different biomolecules and subsequent detection using nucleic acid- and/or protein-array-bound probes on microarrays would allow quick analyses of thousands of nucleic acids and/or proteomes (20 ). The applications of this method could be numerous because many different kinds of targets in cell lysates and body fluids can be labeled and analyzed by a single detection step. For routine screening of samples, it would be beneficial to minimize or exclude purification steps for the lysates or fluids. We investigated whether ULS chemistry allows labeling of serum without purification before or after labeling. The targets used for labeling were IgG, IgM, and DNA in serum. As shown in Fig. 5, immunoreactive IgG was obtained after labeling with fluorescein-ULS as well as with HRP-ULS. The procedure involved only the dilution of serum, addition of the ULS reagent, and incubation. After the remaining reactive ULS was inactivated, the mixture was used for ELISA without any purification. Simultaneous labeling of proteins and nucleic acids in, for example, cell lysates would allow for more direct comparison of genomes and proteomes. As shown in Fig. 6, ULS chemistry also allows for combined labeling of nucleic acids and proteins without any pre- or postlabeling purification step. As a model system, a plasmid probe was added to serum to investigate whether nucleic acids and proteins can both be labeled simultaneously and efficiently enough to allow detection in immunoassays as well as hybridization assays. The labeled mixture could be diluted directly into hybridization buffer, denatured, and used for a fluorescence hybridization assay. The same mixture was used for ELISA. The labeling efficiencies of IgG, IgM, and DNA allowed for sensitive detection in various assays. In the ELISA, fluorescein-ULS enabled the use of serum dilutions ranging from 1:5000 for IgM to 1:25 000 for IgG. For example, in-serum-labeled antibodies of tuberculosis patients were detected on 96-well plates coated with a Mycobacterium tuberculoses antigen mixture with a sensitivity and specificity similar to indirectly detected immunoglobulins (data not shown). The labeling procedure

required only 1 ␮L and was reproducible. The DNA probe labeled in serum performed as well as control probes labeled in buffer only. It has already been shown that ULS-labeled probes perform as well as probes labeled by standard enzymatic procedures (3 ). In conclusion, we showed that fluorescent labeling of protein targets as well as nucleic acid targets in serum is possible (target labeling). It has already been shown that nucleic acids can be successfully labeled in whole cell lysates (22 ); here we show that ULS has the same ability for protein labeling in relatively nonpure environments. The simultaneous labeling of antibodies and DNA in nonpurified whole serum was sufficient to allow their use in immunochemical and hybridization assays. Because the volume of serum can be scaled down and the immunochemical components can be used without any purification step, the number of labeling reactions could easily be scaled up. This could create an automated two-step method for serum analysis on protein arrays. With the addition of HRP as a ULS marker reagent, sensitive detection with fluorescent substrates would further expand the sensitivity of these arrays. Because nucleic acids can also be targeted, labeling of cell lysates and subsequent analysis by “DNA chips” as well as “protein chips” has become feasible.

We thank Prof. Dr. J. Reedijk (Leiden Institute of Chemistry, Leiden University, The Netherlands) for helpful discussions and for critical reading of the manuscript.

References 1. Kallioniemi A, Kallioniemi OP, Sudar D, Rutovitz D, Gray JW, Waldman F, et al. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 1992;258:818 – 21. 2. Figeys F, Pinto D. Lab-on-a-chip: a revolution in biological and medical sciences. Anal Chem 2000;72:330 –5. 3. Van Gijlswijk RPM, Talman EG, Janssen PJA, Snoeijers SS, Killian J, Tanke HJ, et al. Universal Linkage System: versatile nucleic acid labeling. Expert Rev Mol Diagn 2001;1:81–91. 4. Alers JC, Rochat J, Krijtenburg PJ, van Dekken H, Raap AK, Rosenberg C. Universal Linkage System: an improved method for labeling archival DNA for comparative genomic hybridization. Genes Chromosomes Cancer 1999;25:301–5. 5. Hoevel T, Holz H, Kubbies M. Cisplatin-digoxigenin mRNA labeling for nonradioactive detection of mRNA hybridized onto nucleic acid cDNA arrays. Biotechniques 1999;27:1064 –7. 6. Wiegant JCAG, van Gijlswijk RPM, Heetebrij RJ, Bezrookove V, Raap AK, Tanke HJ. ULS: a versatile method of labeling nucleic acids for FISH based on a monofunctional reaction of cisplatin derivatives with guanine moieties. Cytogenet Cell Genet 1999;87: 47–52. 7. van den Berg FM, Lempers ELM, Bloemink MJ, Reedijk J, inventors. Pt-containing compound, process for its preparation, and application of such compounds. US Patent 5,580,990, 1996. 8. Cox MC, Barnham KJ, Frenkiel TA, Hoeschele JD, Mason AB, He Q-Y, et al. Identification of platination sites on human serum

Clinical Chemistry 48, No. 8, 2002

9.

10. 11.

12. 13.

14.

15.

16.

transferring using 13C and 15N NMR spectroscopy. J Biol Inorg Chem 1999;4:621–31. Trynda-Lemiesz L, Kozlowski H, Keppler BK. Effect of cis-, transdiamminedichloroplatinum(II) and DBP on human serum albumin. J Inorg Biochem 1999;77:141– 6. Kricka LJ. Chemiluminescent and bioluminescent techniques. Clin Chem 1991;37:1472– 81. Dent AH, Aslam M. Horseradish peroxidase. In: Aslam M, Dent AH, eds. Bioconjugation—protein coupling techniques for the biomedical sciences. London: Macmillan Reference Ltd., 1998: 118 –32. Renz M, Kurz C. A colorimetric method for DNA hybridization. Nucleic Acids Res 1984;12:3435– 44. Pollard-Knight D, Read CA, Downes MJ, Howard LA, Leadbetter MR, Pheby SA, et al. Nonradioactive nucleic acid detection by enhanced chemiluminescence using probes directly labeled with horseradish peroxidase. Anal Biochem 1990;185:84 –9. Agrawal S, Christodoulou C, Gait MJ. Efficient methods for attaching non-radioactive labels to the 5⬘ ends of synthetic oligodeoxyribonucleotides. Nucleic Acids Res 1986;14:6227– 45. van Gijlswijk RPM, van de Corput MPC, Bezrookove V, Wiegant J, Tanke HJ, Raap AK. Synthesis and purification of using horseradish peroxidase-labeled oligodeoxynucleotides for tyramide-based fluorescence in situ hybridization. Histochem Cell Biol 2000;113: 175– 80. van de Corput MPC, Dirks RW, van Gijlswijk RPM, van Binnendijk

17.

18.

19.

20. 21.

22.

1359

E, Hattinger CM, de Paus RA, et al. Sensitive mRNA detection by fluorescence in situ hybridization using horseradish peroxidaselabeled oligodeoxynucleotides and tyramide signal amplification. J Histochem Cytochem 1998;46:1249 –59. Bobrow MN, Thomas D, Harris D, Shaughnessy KJ, Litt GJ. Catalyzed reporter deposition, a novel method of signal amplification. Application to immunoassays. J Immunol Methods 1989; 124:279 – 85. Bobrow MN, Shaughnessy KJ, Litt GJ. Catalyzed reporter deposition, a novel method of signal amplification. II. Application to membrane immunoassays J Immunol Methods 1991;137:103– 12. Bobrow MN, Litt GJ, Shaughnessy KJ, Mayer PC, Conlon J. The use of catalyzed reporter deposition as a means of signal amplification in a variety of formats. J Immunol Methods 1992;150:145–9. Borrebaeck CAK. Antibodies in diagnostics—from immunoassays to protein chips. Immunol Today 2001;21:379 – 82. Cooke HJ, Hindley J. Cloning of human satellite III DNA: different components are on different chromosomes. Nucleic Acids Res 1979;6:3177–97. Van Leeuwen W, Schalk M, Veuskens J, Libregts C, Verbrugh H, van Belkum A. Binary typing of Staphylococcus aureus through reversed hybridization using digoxigenin-Universal Linkage System-labeled bacterial genomic DNA. J Clin Microbiol 2001;39: 328 –31.