Bioconjugate Chem. 1994, 5, 463-467
463
Use of Psoralens for Covalent Immobilization of Biomolecules in Solid Phase Assays Henrik I. E l m e r * a n d Saren Mouritsen M&E, Lersa, Parka116 40, 2100 Copenhagen, Denmark. Received April 7, 1994@
The ability of compounds to adsorb passively to hydrophobic polymer surfaces composed of, e.g., polystyrene generally is restricted to limited types of molecules such as proteins. Some proteins, many peptides, polysaccharides, oligonucleotides, and small molecules as well as pro- and eucaryotic cells cannot adsorb directly to such surfaces. Also, solid phase adsorbed antigens, antibodies, or gene probes may not be recognized by its corresponding ligand due to denaturation or steric hindrance of the molecular tertiary structure. Covalent binding, on the other hand, orientates all immobilized compounds in a defined way on the solid phase, thereby exposing the interacting sites on the enzymes, antibodies, gene probes, etc. Here we describe a method for modifying a polymer surface by contacting the polymer with derivatives of psoralen under irradiation with long-wavelength UV light. The psoralen derivatives were immobilized covalently on the polymer surface by this process. The psoralen molecules was conjugated to appropriate chemical linkers, incubated in aqueous solutions, and irradiated with UV light. This resulted in solid phase introduction of functional groups such as, e.g., amino groups on the polystyrene surface. The functional groups could subsequently be used for immobilization of biomolecules using conventional cross-linker technology. The method only involved premodification of the psoralens to be immobilized whereas no pretreatment of the polymer was required. Psoralen modified microtiter plates seems to have future application for the development of solid phase hybridization and immunoassays.
INTRODUCTION
The usual way to immobilize biomolecules such as antibodies or protein antigens in solid phase assays is by passive adsorption on, e.g., polystyrene or poly(viny1 chloride) surfaces (1). Many molecules, microorganisms, and cells cannot, however, be immobilized by this method. Polysaccharides, peptides, gene probes, small organic molecules, viruses, and pro- and eukaryotic cells generally have to be immobilized by other techniques. Furthermore, passive adsorption is not an irreversible process (2), which may affect the reproducibility of the immunoassays, especially when the antigedantibody coated solid phase is stored in dry form. The nature of passive adsorption predominantly involves multiple hydrophobic interactions between the solid phase and the biomolecule. Passive adsorption may therefore interfere with the structure and function of adsorbed antigens and antibodies ( 3 , 4 ) .Also, solid phase adsorbed antigen may not be recognized by its corresponding antibody due to denaturation of the antigen tertiary structure (5, 6 ) . Epitopes of, e.g., peptides may be hidden and prevented from recognition by antibodies (7), and the biological activity of passively adsorbed antibodies or enzymes may decline with time (8). Covalent binding, in contrast to passive adsorption, orientates all immobilized compounds in defined ways on the solid phase, thereby exposing defined areas on, e.g., antigens, antibodies, or enzyme catalytic sites to the fluid phase. The antigen epitopes or active sites on these compounds will therefore probably be more conserved. Irreversible immobilization of molecules may furthermore be advantageous in relation to storge of the immobilized compounds. In order to immobilize biomolecules covalently on microtiter wells, these surfaces must posses some kind of functional groups. Methods to introduce amino groups @
Abstract published in Advance ACS Abstracts, August 1,
1994.
1043-1802/94/2905-0463$04.50/0
have been described previously (9), but large scale production of modified microtiter plates using these methods generally has been impractical. Here we describe a method for covalent immobilization of biomolecules on polystyrene microtiter wells, which can be used for immobilization of many kinds of molecules. Functional groups on the polymer surface are introduced by contacting the polymer with derivatives of psoralen under irradiation with long-wavelength UV light. The herein described method for introducing secondary amino groups on polystyrene surfaces is currently being used by Nunc M S (Roskilde, Denmark), and microtiter plates are being marketed under the name CovaLink. Several papers reporting the usefulness of these plates have been published. EXPERIMENTAL PROCEDURES
Materials and General Procedures. All chemicals for preparation of buffers were of analytical grade unless noted, and chemicals for synthesis were standard commercial. Elemental analyses were performed at the Microanalytical Laboratory of The H. C. 0rsted Institute a t The University of Copenhagen. Thin-layer chromatography was performed on silica gel 60 Fz54 precoated aluminum sheets (layer thickness, 0 . 2 mm) from E. Merck, Darmstadt, Germany. The plates were visualized by W light (254 nm). IH-NMR were recorded a t 90 MHz on a Jeol FX 90 Q spectrometer. Chromatography was performed on a 60 x 1.5 cm column using silica gel grade 60 from E. Merck, Darmstadt, Germany. Polystyrene microtiter plates (Maxisorp) were from Nunc M S , Roskilde, Denmark. All experiments have been performed a t least three times in duplicate, and the values shown in Figures 1-5 are mean values. NJV-Dimethyl-N-[3-( psoralen-8-yloxy)propyl] -N'(tert-butoxycarbony1)hexanediamine(I). 3-Bromol-(psoralen-8-yloxy)propane(10,11), (2.3 g, 8.2 mmol) and N-( tert-butoxycarbony1)-NjV-dimethylhexanediamine (12) (1.75 g, 8.2 mmol) were mixed in acetone (100 mL) 0 1994 American Chemical Society
464 Bioconjugate Chem., Vol. 5, No. 5, 1994
with potassium carbonate (2.7 g) and refluxed for 72 h, whereupon the mixture was filtered and evaporated in vacuum. The residue was purified on a silica gel column using methanol in methylene chloride as eluent giving rise to I(1.7 g, 3.5 mmol, 43%). lH NMR (CDCl3): 7.736.17 (5H, m, psoralen), 4.44 (2H, t, OCHZ),3.54 (2H, q, CH2), 3.07 (4H, q, CHz), 2.70 (3H, s, CH3), 2.53 (4H, m, CHd, 2.25 (2H, m, CHd, 2.13 (3H, s, CHd, 1.90 (4H, m, CHd, 1.35 (9H, s, Boc). Anal. Calcd for C27H38NzOs: C, 66.66; H, 7.82; N, 5.76. Found: C, 66.57; H, 7.88; N, 5.77. TLC: Rf = 0.25,10% CH30H in CH2C12. N,W-Dimethyl-N-[3-(psoralen-8-yloxy)propyll hexanediamine (11). I(1.7 g, 3.5 mmol) was suspended in hydrochloric acid (35 mL, 4 M), and after 1h the solution was evaporated in vacuum, and the main product (11) was purified on a silica gel column using triethylamine and ethanol as eluent (1.4 g, 28 mmol, 77%). IH NMR (DzO, DC1): 7.73-6.09 (5H, m, psoralen), 4.32 (2H, t, OCHd, 3.45-2.97 (6H, m, CH2),2.86 (3H, s, CH3), 2.65 (3H, s, CH3), 1.99-1.41 (lOH, q, CHZ). Anal. Calcd for Cz~H30N204:C, 68.39; H, 7.78; N, 7.26. Found: C, 68.16; H, 7.82; N, 7.04. TLC: Rf = 0.30, 50% CH30H, 50% N(CH3CHd3. N,W-Dimethyl-N-[3(psoralen-8-yloxy)propyl] -Wbiotinylhexanediamine (111). I1 (0.62 g, 1.34 mmol) was solubilized in DMF whereupon triethylamine (1mL, 7.2 mmol) and N-hydroxysuccinimide-biotin (0.46 g, 1.34 mmol) were added under stirring. The next day the solution was evaporated in vacuum, and the main product (111)was purified on silica gel as described above (0.4 g, 0.60 mmol, 45%). 'H NMR (DMSO): 8.24-6.40 (5H, m, psoralen), 4.60 (2H, t, OCHZ),4.55-4.01 (2H, m), 3.50-2.93 (9H, m), 2.81 (3H, s, CH3), 2.77 (3H, s, CH3), 2.19-1.01 (18H, m). Anal. Calcd C32HaN406S: C, 66.73; H, 7.19; N, 9.15. Found: C, 65.97; H, 7.43; N, 9.06. TLC: Rf= 0.90, 50% CHBOH,50%N(CHsCH&. Photochemical Modification of Polystyrene. (a) Binding of NJ7-Dimethyl-N-[3-(psoralen-8-yloxy)propyl]N-biotinylhexanediamine (HI) to Microtiter Wells. A stock solution of I1 (10 mg/mL in DMSO) could be stored for months at -4 "C. Ten-fold dilutions in distilled water were made from this stock in concentrations ranging from 0.1 to 1000 pg/mL, and 100 &/well of each dilution was added to the wells of polystyrene microtiter wells. The wells were then irradiated for 2 h a t room temperature with long-wavelength (A > 350 nm) W light from a Phillips TL 20W/09N lamp placed 20 cm above the microtiter wells. Subsequently, the wells were washed three times with 200 pL of washing buffer (pH, 7.2, 0.1 M phosphate, 0.5 M NaCl (PBS), 1%Triton X-100). The same experiment was done without irradiation. For detection of immobilized biotin on the solid phase, a solution of 100 &/well of horse radish peroxidase conjugated avidin (HRP-avidin, Sigma, St. Louis, MO) was added to the wells (25 pg of HRP-avidin, 100 mg of BSA, 10 mL of washing buffer) and incubated a t 37 "C for 1h. The wells were then washed three times with washing buffer and 100 $ / w e l l of a solution of the chromogenic substrate 0-phenyldiamine (OPD, 10 mg) and hydrogen peroxide (1pL, 35%) in citrate/phosphate buffer (0.1 M, 10 mL, pH 5.0) was added. After approximately 3 min the colored reaction was stopped with sulfuric acid (100 pL, 1 M), and the optical density (OD) was read on a Titertek Multiscan ELISA-photometer. (b) The Influence of Irradiation Times on the SolidPhase Binding of N,"-Dimethyl-N-[3-(psoralen-8-yloxy)propyl/-N'-biotinylhexanediamine(III). A stock solution of I11 (10 mg/mL, DMSO) was diluted in water to a concentration of 10 pg/mL and 100 pg/mL, respectively. One hundred pL/well of each dilution was transferred to
Elmer and Mouritsen
microtiter wells, W-irradiated for different time periods, and subsequently treated as described above. (c) Binding of NN-Dimethyl-N-[3-(psoralen-8-yloxy)propyllhexanediamine (I& to Microtiter Wells. A solution of I1 (500 pglmL) in phosphate-buffered saline (0.1 M, pH 8 . 2 , 2 M NaC1) was made. One hundred pL per well was added, and the microtiter plates were irradiated for 1 h as described above. Each well was washed three times with demineralized water, and solutions of N-hydroxysuccinimide-biotin (NHS-biotin, Hoechst, lot no. 410074, Calbiochem) in carbonate buffer (pH 9.6, 0.005 M) was then added in 2-fold dilution series starting a t 125 pglmL. Subsequently, the microtiter wells were incubated for 2 h at room temperature, and each well was washed three times with washing buffer. The coupling efficiency of biotin was visualized using HRPavidin as described above. (e) Binding of 5-[(Trimethylammoni0)[~Hlmethyl1-8methoxypsoralen Bromide of Microtiter Wells. Aqueous solutions of 5-[(trimethylamm0nio)[~H]methyl]-8-methoxypsoralen bromide (13) were added to polystyrene microtiter wells (100 pL, 1 mg/mL, 3 800 000 cpm) and diluted 10-fold in distilled water to a final concentration of 0.01 pg/mL. The wells were then irradiated for 2 h a t room temperature and washed eight times with 200 pL of washing buffer. The same experiment was performed without U V irradiation. The wells were emptied, separated mechanically, and transferred to scintillation vials. Ten mL of scintillation fluid (Instagel, Pachard) was added, and the samples were analyzed in a scintillation counter (Beckman LS7000). (d) Treatment of Microtiter Wells with N-(4-Azido-2nitrophenyl)-N'-[3-(biotinylamino)propyll-N'-methyl-1,3propanediamine (Photobiotin). It was also attempted to biotinylate polystyrene using N-(4-azido-2-nitrophenyl)N'-[3-(biotinylamino)propyl]-N'-methyl- 1,3-propanediamine) (photobiotin, Sigma cat. no. A 7667). This reagent contains a photoreactive aryl azide group connected t o a chemical linker similar to the one used for 111. The Stability of the Psoralen-Modified Surface. One hundred pL of the following aqueous solutions were added to microtiter wells modified with I1 as described above: 1M NaOH; 1M HC1; 1%Triton X-100; 10% acetic acid; 0.1 M citrate, pH 5.5;0.1 M PBS pH 7.2, 0.1 M Carbonate buffer pH 9.6, 10% ethanol; 10% methanol; 1%DMSO, 1%DMF, water, washing buffer, and absolute ethanol. After 2 h the wells were emptied and washed five times with water. Detection of secondary amino groups on the surface was performed as described above. RESULTS
Preparation of Derivatives of S-(Propyloxy)psoralen. Three psoralen derivatives were synthesized starting from 3-bromo-l-(psoralen-8-yloxy)propane and N-(tert-butoxycarbonyl)-NJV'-dimethylhexanediamine, one derivative with a Boc-group (I),one with a secondary amine (II), and one conjugated to biotin (111)(Scheme 1). I1 and I11 were used for photomodification of polystyrene surfaces. All three compounds were characterized by elemental analysis, TLC, and H1 NMR. Photobinding of N,"-Dimethyl-N-[3-(psoralen8-yloxy)propyl]-W-biotinylhexanediamine (111) to Microtiter Wells. By using HRP-avidin it was shown that I11 bound to polystyrene under irradiation of W light. When no W light was used no significant binding of I11 could be observed (Figure 1). The amount of 111 that could be detected depended on the added amount, and this relationship was almost linear in concentrations of ranging from 0 t o 1 mg/mL. The influence of the UV irradiation time was most pronounced in a time range
Covalent Immobilization of Biomolecules
Roconjugate Chem., Vol. 5, No. 5, 1994 465
Scheme 1. Synthesis of NJ”-Dimethyl-N-[3-(psoralen-8-yloxy)propyll-N’-(tert-butoxycarbonyl)hexanediamine (I), NJV’-Dimethyl-N-[3-(psoralen-8-yloxy)propyllhexanediamine(111, and NJ”-Dimethyl-N-[3-(psoralen-8-yloxy)propyll- N’-biotinylhexanediamine(111) I
qD0
w n-
0
AoY I
0
0-Er
>
0
I 1) HCI,
2) E t , N
>
0
111 3*0 2.5
1
2.0
1.5
2.0 N
N
m
m
+ n 1.5
d 0
0
1.0
1 .o
0.5 0.5
0.0
0.0
io-‘
loo
10’
IO*
lo3
propyl]-N-biotinylhexanediamine(111)to microtiter wells with ( 0 )and without irradiation (0) of U V light.
from 0 to 90 min (Figure 2). Longer irradiation periods did not increase the amount of immobilized biotin, although the used amount of I11 influenced the maximal amount of biotin which could be detected. Photobinding of N,iV’-Dimethyl-N-(3-psoralen-8y1oxy)hexanediamine (111) to Microtiter Wells. A clear correlation could be demonstrated between the added amounts of N-hydroxysuccinimide-biotin and the signal level obtained with HRP-avidin on microtiter wells modified with this psoralen derivative (Figure 3). Addition of N-hydroxysuccinimide-biotin (125 pglmL) to nonmodified wells gave an average signal level of only 0.077 OD units. Photobinding of 5-[(TrimethyIammonio)[’H3methyI]-8-methoxypsoralenBromide to Microtiter Wells. Using a solution of 1 mg/mL of 5-[(trimethylammoni0)[~H]methyl1-8-meth0q~p~0ralen bromide, it could be demonstrated that 87 ng were photochemically immobilized on the polymer surface, whereas only 17 ng were immobilized in nonirradiated wells. When using a concentration of 5-[(trimethylammoni0)[~HImethyl]-8methoxypsoralen bromide of 100 ,uglmL, 48 ng were immobilized to the solid phase whereas nonirradiated
l
l
l
l
I
I
L
Min.
Pg/ml
Figure 1. Binding of N,”-dimethyl-N-[(3-psoralen-8-yloxy)-
l
0 20 40 60 80100120140
Figure 2. Influence of irradiation time on the binding of N,”dimethyl-N-[3-(psoralen-8-yloxy)propyll-~-biotinylhexanediamine (111)to microtiter wells using concentrations of 100 ,ug/ mL ( 0 )and 10 pg/mL (01, respectively.
wells showed near-background values (12 ng) (Figure 4). At concentrations lower than 100 pglmL no significant immobilization of 5-[(trimethylammonio)[1Hlmethyll-8methoxypsoralen bromide could be detected (Figure 4). Binding of Photobiotin. The same experiments were performed as for 111, but no biotinylation of the polystyrene solid phase could be detected. The Stability of the Psoralen-ModifiedSurface. Treatment of microtiter wells modified with I1 using different buffers and solvents had no significant effect on the amount of secondary amino groups which subsequently could be detected on the surface (Figure 5). DISCUSSION
We describe here a n easy method for introduction of functional groups on polymer surfaces. Such groups were introduced using the psoralen derivative as the surface reactive compound. Secondary amino groups were immobilized on the surface through an appropriate linkage to the psoralen moiety, and these functional groups could subsequently be used for chemical coupling of, e.g., activated biotin.
Elsner and Mouritsen
466 Bioconjugate Chem., Vol. 5, No. 5, 1994 3.0 2.5
2.0 N
m
*
cs
0
1.5 1 .o 0.5
90 -
I
I
I
I
1
100
80 70 60
-
213 50 40 30 4 4
-
20 10 0 -
10‘ i o 2 io3 CLg/ml Figure 4. Binding of 5-[(trimethylammonio)[’H]methyl1-8methoxypsoralen bromide to microtiter wells with ( 0 ) and without irradiation (0) of UV light.
io-’
t I
l1e.o 5
ioo
I
Figure 5. Treatment of psoralen-modified surfaces with 1 M NaOH (11, 1M HCl(2), 1%Triton X-100 (31, 10% acetic acid in water (41, 0.1 M citric acid buffer pH 5.5 (51, 0.1 M PBS pH 7.2 (6),0.1 M carbonate buffer pH 9.6 (71, 10% ethanol in water (81, 10% methanol in water (9), 1%DMSO in water (101, 1% DMF in water (ll),water (la), washing buffer (131, absolute ethanol (141, no treatment (15).
A clear correlation between the added amounts of I11 and the detectable amounts of solid phase immobilized biotin was seen after irradiation with W light (Figure 1). Without irradiation only a small amount of biotin could be detected, showing that the surface modification of polystyrene was coursed by a photochemical reaction. The process depended on the amounts of psoralen derivative used and on the irradiation time. When using concentrations of 100 and 10 pglmL, respectively, the
amounts of immobilized psoralen derivative were in linear proportion to the irradiation time in a n interval from 0 to 90 min (Figure 2). A psoralen derivative containing a secondary amine I1 was synthesized for purpose of introduction of secondary amines on the polystyrene surface using the same protocol as for photobinding of 111. These amino groups could be confirmed using a n active ester of biotin and HRP-avidin. A dose-dependent relation was seen between the added amounts of NHS-biotin and the immobilized peroxidase activity, and as expected NHSbiotin itself was not able to react with nonmodified wells (Figure 4). The amount of psoralen derivative immobilized by the method was estimated using radioactively labeled 5[(trimethylammonio)[1H]methyll-8-methoxypsoralen bromide. The maximum amount, which could be immobilized of this compound-although different from I1 and 111-was 87 ng/well (0.28 nmovwell) when using a solution of 1000 ,ug/mL (Figure 5). At concentrations less than 100 ,ug/mL no significant immobilization of 5-[(trimethylammonio)[1H]methyl]-8-methoxysoralenbromide could be detected. The amount of secondary groups introduced on polystyrene with I1 has also been determined using a colorimetric method (14). Approximately the same result was obtained corresponding to about 1014 amino groups per well. The psoralen-modified surfaces were treated with different strong aqueous eluents in order to examine the nature of binding to the surface. Neither detergents, strong acid, strong base, buffers, high salt concentrations, alcohols, 1%DMSO, nor 1%DMF in water were able to remove the introduced secondary amino groups (Figure 5). This suggest that the mechanism of binding involves covalent bonds between the psoralen derivative and the polystyrene surface. It is well known that psoralens are able to photoreact with DNA through a cycloaddition between the doublebond of the coumaridfuran in the psoralen molecule and the 5,6 double bond in thymine (15). It has recently been shown that psoralens are capable of making photocycloaddition to pyrimidine (16). These reactions take place under irradiation of long-wavelength UV light, and since we previously have shown that this wavelength interval also is optimal for the herein described method (data not shown), it seems likely that the observed photobinding of the psoralen derivatives to polystyrene was established through a similar mechanism. Perhaps a photocyclization takes place between the coumaridfuran part of a psoralen molecule and a styrene group on the polymer surface. Two psoralen molecules are furthermore able to react with each other under formation of dimers (17) resulting in further modification of the surface. Others have premodified polystyrene surfaces with W light (18, 19) and with W light in combination with cerium ammonium nitrate (20). The nature of the introduced functional groups was, however, not characterized, and the mechanisms seemed to be entirely different from the herein described method, where the introduced functional groups were known to be secondary amino groups. This enables the use of crosslinking reagents for subsequent, well-defined immobilization of biomolecules. A known photoreactive derivative of biotin, in which the active compound was arylazide (21),was also examined. This compound did, however, not bind to polystyrene, which may be due to a n immediate reaction between the photoreactive azide and the aqueous solvent. Psoralens, on the other hand, do not react with water.
Covalent Immobilization of Biomolecules
Psoralen-modified surfaces containing secondary amines were shown to be able to react with a succinimide ester of biotin, but other active compounds have been shown by others to react with this surface. Crosslinking reagents such as l-[3-(dimethylamino)propyll-3-ethylcarbodiimide (EDC) and succinimidyl4-(N-maleimidomethyl)cyclohexane-1-carboxylate (SMCC) have been shown to react selectively with the modified surface (22). Oligonucleotides (23),peptides (22,24),and biotin (22)have been immobilized on I1 modified surfaces using EDC as the coupling reagent. The immobilized oligonucleotides could subsequently be used in solid phase hybridization assays where it was possible to detect very low amounts of hybridized DNA (23). Peptides were immobilized with EDC on the same surface for the development of an ELISA for detection of peptide antibodies (24),and an even smaller molecule (a steroid) could be immobilized on I1 modified microtiter wells (25) and subsequently detected with antibodies. Photomodification of microtiter wells using derivatives of psoralen may solve many of the problems concerning immobilization and detection of small molecules. These surfaces may furthermore have many future interesting applications within the field of automated solid phase DNA hybridization. LITERATURE CITED (1) Engvall, E., and Perlmann, P. (1971) Enzyme-linked immunosorbentassay (ELISA). Quantitative assay of immunoglobulin G. Immunochemistry 8, 871-874. (2) Lehtonen, 0. P., and Viljanen, M. K. (1980) Antigen attachment in ELISA. J . Zmmunol. Methods 34, 61-70. (3) Darst, A. S., Robertson, C. R., and Berzofsky, J. A. (1988) Adsorption of the protein antigen myoglobin affects the binding of conformation-specific monoclonal antibodies. Biophys. J . 53, 533-535. (4) Dierks, S.E., Butler, J. E., and Richerson, H. B. (1986) Altered recognition of surface-adsorbed compared to antigenbound antibodies in the ELISA. Molec. Zmmunol. 23, 403411. (5) Kurki, P. and Virtanen, I. (1984) The detection of human antibodies against cytoskeletal components. J . Immunol. Methods 67, 209-23. (6) Butler, J. E., Nessler, Ni, L., Nessler, R., Joshi, K. S., Suter, M., Rosenberg, Chang, J., Brown, W. R., and Cantarero, L. A. (1992) The physical and functional behaviour of capture antibodies adsorbed on polystyrene. J . Zmmunol. Methods 150, 77-90. (7) Wood, W. G., and Gadow, A. (1983) Immobilisation of Antibodies and Antigens on Macro Solid Phases-A Comparison Between Adsorptive and Covalent Binding. J . Clin. Chem. Clin. Biochem. 21, 789-797. (8) Ansari, A. A., Hattikudur, N. S., Joshi, S. R., and Medeira, M. A. (1985) ELISA Solid Phase: Stability and Binding Characteristics. J . Zmmunol. Methods 84, 117-124. (9) Neurath, A. R., and Strick, N. (1981) Enzyme-linked fluorescence immuno assays using beta-galactosidase and antibodies covalently bound to polystyrene plates. J . Virol. Methods 3 , 155-65.
Bioconjugate Chem., Vol. 5, No. 5, 1994 467 (10) Elsner, H. I., Buchardt, O., Moeller, J., and Nielsen, P. E. Photochemical crosslinking of protein and DNA in chromatin. Synthesis and application of psoralen-cystamine-arylazido photocrosslinking reagents. Anal. Biochem. 149, 575-81. (11) Henriksen, U., Buchardt, O., and Nielsen, P. E. (1991) Azidobenzoyl-, azidoacridinyl-, diazocyclopentadienylcarbonyl- and 8-propyloxypsoralen photobiotinylation reagents. Syntheses and photoreactions with DNA and protein. J . Photochem. Photobiol. 57, 331-42. (12) Hansen, J. B., Nielsen, M. C., Ehrbar, U., and Buchardt, 0. (1985) Partially Protected Polyamines. Synthesis 5,404405. (13) Buchardt, O., Ebbesen, P., Kantrup, A., Karup, G., Knudsen, P. H., Nielsen, P., Hansen, J. B., Bjerring, P. E., Nielsen, M. C., Norden, B., and Ygge, B. (1985) Psoralenamines synthesis, pharmacological behavior and DNA binding of 5-(aminomethyl)-8-methoxy-, 5-([[3-aminopropylloxylmethylland -[(3-aminopropyl)oxylpsoralenderivatives. J . Med. Chem. 28, 1001-10. (14) Kakabakos, S. E., Tyllianakis, P. E., Evangelatos, G. P., and Ithakissios, D. S. (1993) Colorimetric Determination of Amino Groups of Covalink NH Microwells. Nunc Bull. 11, 1-2. (15) Parsons, B. J. (1980) Psoralen photochemistry. Photochem. Photobiol. 32, 813-21. (16) Bisagni, E. (1992) Synthesis of psoralens and analogs. J . Photochem. Photobiol. 14, 23-46. (17) Shim, S. C., Lee, S. S., and Choi, S. J. (1990) The C4-photocyclodimers of 4,5’,84rimethylpsoralen (TMP). J . Photochem. Photobiol. 51, 1-7. (18) Zouali, M., and Stollar, B. D. (1986) A rapid ELISA for measurement of antibodies to nucleic acid antigens using UVtreated polystyrene microplates. J . Zmmunol. Methods 90, 105-10. (19) Boudet, F., Theze, J., and Zouali, M. (1991) W - t r e a t e d polystyrene micro-titer plates for use in a n ELISA to measure antibodies against synthetic peptides. J . Zmmunol. Methods 142, 73-82. (20) Buchardt, O., Jergensen, W. A., Henriksen, U., Rasmussen, M., Lohse, C., Lgvborg, U., Bjerrum, 0. J., and Nielsen, P. E. (1993) Photochemical surface modification of polystyrene in the presence of cerium(IV) ammonium nitrate: improved binding of proteins, amines and mercaptans in the presence of detergent. Biotechnol. Appl. Biochem. 17, 223-237. (21) Forster, A. C., McInnes, J. L., Skingle, D. C., and Symons, R. H. (1985) Nonradioactive hybridization probes prepared by the chemical labeling of DNA and RNA with a novel reagent, photobiotin. Nucleic Acids Res. 13, 745-61. (22) Rasmussen, S. E. (1990) Covalent immobilization of biomolecules onto polystyrene MicroWells for use in biospecific assays. Ann. Biol. Clin. 48, 647-650. (23) Rasmussen, S. R., Larsen, M. R., and Rasmussen, S. E. (1991) Covalent Immobilization of DNA onto Polystyrene Microwells: The Molecules Are Only Bound at the 5’End. Anal. Biochem. 198, 138-142. (24) Sendergbrd-Andersen, J., Lauritzen, E., Lind, K., and Holm, A. (1990) Covalently linked peptides for enzyme-linked immunosorbent assay. J . Immunol. Methods 131, 99-104. (25) Yonezawa, S., Kambegawa, A., and Tokudome, S. (1993) Covalent coupling of steroid to microwell plates for use in a competitive enzyme-linked immunosorbent assay. J . Zmmunol. Methods 166, 55-61.
