Trityl tags for encoding in combinatorial synthesis

Article abstract of DOI:10.1016/S0040-4020(00)00223-4

New tags and an encoding strategy for combinatorial synthesis are described. Combinatorial libraries of short oligonucleotides attached to TentaGel beads were synthesised by a split-and-mix strategy using 5'-DMTr or 5'-Fmoc-protected nucleoside phosphoramidites. Trityl moieties with different masses were used to tag the nature and position of monomer units (bases) coupled at each step in the synthesis. Beads with a specific oligonucleotide were selected by hybridisation from combinatorial libraries. Tags orthogonal to the added nucleotides were produced by coupling amines of different molecular masses to an activated carboxyl group(s) on the trityl moiety. The tags may be released from the support by an acidic treatment or laser irradiation and then analysed by (MA)LDI-TOF. These properties make trityl- based tags promising for encoding in strategies not involving strong acids, such as oligonucleotide and peptide synthesis and small molecule combinatorial libraries. (C) 2000 Published by Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00223-4

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 2713±2724
Trityl Tags for Encoding in Combinatorial Synthesis
*
Mikhail S. Shchepinov, Rod Chalk and Edwin M. Southern
Department of Biochemistry, University of Oxford, South Parks Road, Oxford OX1 3QU, UK
Received 3 December 1999; revised 7 March 2000; accepted 9 March 2000
AbstractÐNew tags and an encoding strategy for combinatorial synthesis are described. Combinatorial libraries of short oligonucleotides
attached to TentaGel beads were synthesised by a split-and-mix strategy using 50-DMTr or 50-Fmoc-protected nucleoside phosphoramidites.
Trityl moieties with different masses were used to tag the nature and position of monomer units (bases) coupled at each step in the synthesis.
Beads with a speci®c oligonucleotide were selected by hybridisation from combinatorial libraries. Tags orthogonal to the added nucleotides
were produced by coupling amines of different molecular masses to an activated carboxyl group(s) on the trityl moiety. The tags may be
released from the support by an acidic treatment or laser irradiation and then analysed by (MA)LDI-TOF. These properties make trityl-based
tags promising for encoding in strategies not involving strong acids, such as oligonucleotide and peptide synthesis and small molecule
combinatorial libraries. q 2000 Published by Elsevier Science Ltd. All rights reserved.
Introduction
synthesis and assay, be cleaved speci®cally and orthogonally
to the tethered compound, and be readily identi®able in
pico- or femtomol quantities. To obtain orthogonal chemistry
oligonucleotides (which may be ampli®ed by polymerase
chain reaction (PCR)11), secondary amines12 and haloaryls13
(binary coding) have been used as encoding tags, with
varying degrees of success. Identi®cation has been carried
out by most spectroscopic and chromatographic methods,
including HPLC,12 GC,14,15 mass spectrometry,16±24 ¯uores-
cence,25 IR26,27 and NMR spectroscopy.28 Libraries of
limited size have been encoded by using physical sepa-
ration, either `tea-bags'29 or mini-capsules containing
radio-frequency microchips, which can then be identi®ed
using an electronic reader.30,31 Physical encoding by altering
the size, shape, density and colour of the beaded support has
also been investigated.32 However, none of these methods
has been particularly successful at encoding large libraries,
such as those formed by a complete set of all possible oligo-
nucleotides of a de®ned length. The sheer numbers involved
point to the use of bead libraries for their synthesis and
screening, but the limited sensitivity of gel-based sequenc-
ing methods rules out direct identi®cation.
The combinatorial approach to simultaneous synthesis of
large numbers of chemical compounds on solid supports,
®rst used for screening in electronic materials,1 has been
an important development in biological and medicinal
chemistry,2±4 allowing the use of rapid screening to speed
up lead identi®cation and optimisation. Two methods pre-
dominate: spatially addressable (or parallel) arrays, in which
synthesis steps are performed simultaneously on sets of
visibly separated starting materials;5,6 and bead libraries,
consisting of mixtures of microscopic resin beads each of
which carries a few pmol or mmol of a single compound,
usually synthesised using the `split-and-mix' method.7,8
Although resin libraries are ef®cient in their use of chemicals
and quickly screened, their application is limited if the
compound on a selected `hit' bead cannot be readily identi®ed.
One way around this is a deconvolution strategy, but these
are wasteful of both chemicals and time, and parallel
synthesis cannot always accommodate large numbers of
compounds.9 The alternative is to attach one or more tags
to the bead, which can be cleaved and identi®ed even at very
low concentration, and which will encode the synthesis
steps that bead has undergone. The problem of encoding
during `one-bead±one-compound' combinatorial synthesis
has been addressed by several groups and recently
reviewed.10 A successful encoding tag must survive the
Triarylmethyl (trityl)-type cations are stabilised by the reso-
nance effect of the phenyl rings, which makes their ethers
acid labile, and they are consequently a useful family of
protective groups, especially in nucleoside chemistry.33±36
Trityl groups generating cations of different colours have
been used to protect different nucleotides in oligonucleotide
synthesis.37 Modi®ed trityl groups have been used, for
example, to accelerate the formation of internucleotide
bonds in the phosphotriester approach,38,39 to reversibly
label synthetic oligonucleotides with biotin etc.40,41 to purify
them by immobilising on to a solid support after synthesis,42
Keywords: combinatorial chemistry; mass spectrometry; nucleic acids;
phosphoramidites.
* Corresponding author. Tel.: 144-1865-275-226; fax: 144-1865-275-
283; e-mail: misha@bioch.ox.ac.uk
0040±4020/00/$ - see front matter q 2000 Published by Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00223-4

2714
M. S. Shchepinov et al. / Tetrahedron 56 (2000) 2713±2724
Scheme 1. (a) SOCl2, re¯ux. (b) 2-Amino-2-methylpropan-1-ol, 2.5 equiv. (c) PhMgBr. (d) 80% AcOH, 48 h. (e) NHS, DCC. (f) AcCl, toluene, re¯ux. (g)
Grignard synthesis. Alternatively, 3 may be synthesised from benzophenone and 4-bromophenyl oxazoline.
to quantify the amount of amino groups on a solid support,43
and to controllably activate prodrug antibody conjugates.44±46
A modi®ed trityl group bearing a pyrenyl residue in
place of one of the aryl groups has been used for
more precise ¯uorescent detection (down to 10210 M)
of detritylation,47 and a 14C-labelled dimethoxytrityl
(DMTr) group was used for more sensitive monitoring
of coupling reactions on an aminated polypropylene
support.48 Apart from acidic treatment, some alternative
ways of removing the triarylmethyl group have been
explored, including treatment with anion-radicals49,50 and
ZnBr2.51 Importantly, for the tags described in this paper,
it has also been reported that the trityl cation may form
merely upon irradiation of the starting material by photo-
chemical ionisation.52
Results and Discussion
Synthesis of activated trityl blocks
A dimethoxytrityl group bearing an N-hydroxysuccinimide
(NHS)-activated carboxyl function (1) was previously used
for the reversible labelling of oligonucleotides and for other
applications.40±42,44±46 We used Koster's approach
to
40,41
È
synthesise trityl groups that are more stable to acid (both
with and without methoxy groups (Scheme 1)). To avoid the
synthesis of the Grignard reagent from 2-(4-bromophenyl)-
4,4-dimethyl-1,3-oxazoline, we took advantage of the
commercially available Grignard reagents (Aldrich) and
synthesised oxazolyl-protected 4-carboxybenzophenone
(4) starting from 4-carboxybenzophenone. Following the
Grignard reaction with phenylmagnesium bromide, subse-
quent steps were similar to those used for compounds 1 and
2.40,41
The characteristic signal of the DMTr1 cation (mono-
isotopic peak at 303.139 Da) is frequently present in
mass-spectra of DMTr-containing compounds, suggest-
ing that derivatives of trityl groups with different
masses could serve as the unique mass-tags in combi-
natorial synthesis. We describe here a new method of
encoding, which is based on the high desorption rate of
triphenylmethyl cation-based tags in the conditions
of laser desorption/ionisation TOF mass-spectrometry,
which makes detection simpler than in previously
described encoding systems.13±16 The trityl groups can
be released by acidic treatment and detected by laser
desorption/ionisation TOF analysis with or without matrix.
Alternatively, the cations can be generated directly by laser
irradiation.
Treating activated carboxyl group-containing trityl
synthons with different amines generates a variety of tags
with different masses. The masses of the majority of cheap
commercially available primary amines, which would with-
stand the conditions of oligonucleotide synthesis and depro-
tection (thus excluding, for example, all aromatic amines
unless phenoxyacetyl (PAC)- or similar `fast' phosphorami-
dites, are used for the combinatorial synthesis) lie mainly
in the range of 50±250 Da. For some applications it is
desirable to have several hundred mass-tags available. The
resolution of the tags in TOF mass-spectrometry was found
to be satisfactory with $2 Da difference between the masses

M. S. Shchepinov et al. / Tetrahedron 56 (2000) 2713±2724
2715
Table 1. Registered masses (LDI-TOF conditions, no matrix) of four different trityl-based pro-tags 1, 2, 3 and 6 treated with 80 amines. Dashed line indicates
no signal detected
#
MW of amines, Da
Chemical name
Tr(NHS)
(3)
MMTr(NHS)
(2)
DMTr(NHS)
(1)
MMTr(NHS)
(6)
1.
2.
3.
4.
5.
6.
7.
8.
17.03
31.06
45.09
59.11
73.14
74.09
85.15
87.17
87.17
89.14
89.14
89.14
97.12
99.18
Ammonia
Methylamine
Ethylamine
Propylamine
Butylamine
Glycinamide (xHCl)
Cyclopentylamine
Amylamine
2-Amino-3-methylbutane
2-Amino-1-methoxypropane
4-Amino-1-butanol
2-Amino-2-methyl-1-propanol
Furfurylamine
Cyclohexylamine
Hexylamine
5-Amino-1-pentanol
Thiomorpholine
2-(2-Aminoethoxy)-ethanol
Cycoheptylamine
1-(2-Aminoethyl)-pyrrolidine
Heptylamine
Phenethylamine
2-(2-Aminoethyl)-pyridine
1-(3-Aminopropyl)-imidazole
Cyclooctylamine
a-Amino-e-caprolactam
2-(2-Aminoethyl)-1-methylpyrrolidine
1-(2-Aminoethyl)-piperidine
Octylamine
4-(2-Aminoethyl)-morpholine
N,N-Diethyl-1,3-propanediamine
3-Phenylpropylamine
1-(4-Methoxyphenyl)-ethylamine
4-Methoxybenzylamine
2-Phenoxyethylamine
4-Fluoro-a-methyl-benzylamine
1-(3-Aminopropyl)-2-pyrrolidinone
3,5-Di¯uorobenzylamine
Nonylamine
4-(3-Aminopropyl)-morpholine
1,2,3,4-Tetrahydro-1-naphthylamine
2,2,3,3,3-Penta¯uoropropylamine
1-Methyl-3-phenylpropylamine
4-Phenylbutylamine
Norephedrin
2-(4-Methoxyphenyl)-ethylamine
1-Adamantylamine
4-tert-Butylcyclohexylamine
Menthylamine
286.27
300.33
314.38
328.39
342.47
±
316.25
330.32
344.35
358.44
372.46
372.44
384.47
386.60
386.42
388.34
388.39
388.43
396.42
398.52
400.55
402.60
402.46
404.53
412.49
413.45
414.48
420.48
421.54
424.58
426.61
427.54
427.67
427.62
428.57
429.49
429.55
434.51
434.55
436.55
436.55
438.58
411.67
442.55
442.65
443.52
446.49
448.46
448.61
448.61
450.63
450.66
450.62
454.68
454.63
455.63
456.49
456.64
457.67
459.64
461.70
466.60
470.76
475.66
±
346.31
360.43
374.45
388.53
402.51
403.60
414.53
416.52
416.50
418.14
418.41
418.51
426.54
428.58
430.69
432.63
432.55
434.56
442.58
443.59
444.55
450.48
451.63
454.73
456.77
457.65
457.71
457.69
458.70
459.60
459.66
464.66
464.65
466.63
466.56
468.63
471.77
472.67
472.74
473.58
476.59
478.58
478.58
478.73
480.77
480.71
480.75
480.75
484.77
485.79
486.55
486.70
487.81
489.72
491.75
496.65
500.88
504.52
505.74
509.71
±
359.36
387.45
415.54
443.64
471.71
474.55
495.71
499.73
499.65
503.57
503.65
503.66
519.64
523.78
527.76
531.78
531.72
535.66
551.86
553.68
555.81
567.63
569.84
576.34
580.00
581.91
±
581.91
583.92
585.94
585.85
595.72
595.89
599.85
599.91
603.91
610.02
611.85
612.10
±
619.82
623.78
623.95
623.95
627.98
627.95
628.06
636.06
635.82
±
639.69
640.02
642.12
645.97
649.99
659.94
668.21
676.03
677.93
685.94
±
354.42
356.43
356.38
358.35
358.40
358.39
366.43
368.47
370.56
372.52
372.47
374.45
382.50
383.43
384.49
390.52
391.62
394.24
396.62
397.56
397.64
±
398.64
399.56
399.53
404.54
404.59
406.57
406.61
408.62
411.63
412.56
412.67
413.57
416.55
418.50
418.65
418.64
420.62
420.63
420.69
424.73
424.68
±
426.52
426.68
427.75
428.67
431.69
436.63
440.85
444.65
445.72
449.60
±
452.60
454.80
459.73
460.80
462.75
466.75
468.90
469.59
477.71
476.53
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
101.19
103.17
103.19
105.14
113.20
114.19
115.22
121.18
122.17
125.18
127.23
128.18
128.22
128.22
129.25
130.19
130.23
135.17
135.21
137.18
137.18
139.17
142.20
143.14
143.27
144.22
147.22
149.06
149.24
149.24
151.21
151.21
151.25
155.29
155.29
156.27
157.22
157.30
158.29
160.22
162.24
167.21
171.33
175.15
176.26
180.23
181.25
183.25
185.36
190.29
191.27
193.15
197.28
199.38
200.26
206.75
207.28
1-(3-Aminopropyl)-2-pipecoline
1-Naphtalenemethylamine
Decylamine
2-Amino-5-diethylaminopentane
Tryptamine
1-Phenylpiperazine
2,6-Dimethoxybenzylamine
Undecylamine
4-(Tri¯uoromethyl)-1-benzylamine
1-Benzyl-3-aminopyrrolidine
1-(4-Fluorophenyl)-piperazine
Dicyclohexylamine
477.64
±
Aminodiphenylmethane
Dodecylamine
482.57
484.79
489.69
490.68
492.60
496.68
498.85
499.59
502.68
506.49
512.63
514.86
519.83
520.54
522.64
526.83
528.93
529.68
535.99
536.44
691.80
696.16
706.00
708.02
711.88
719.96
724.30
725.81
±
4-Amino-1-benzylpiperidene
2-Benzyloxycyclopentylamine
3-Fluoro-5-(tri¯uoromethyl)-benzylamine
1,2-Diphenylethylamine
Tridecylamine
4-(2-Aminoethyl)benzenesulfonamide
1-(2-Ethoxyphenyl)-piperazine (xHCl)
Amino-2,2-dimethyl-4-phenyl-1,3-dioxane
752.22

2716
M. S. Shchepinov et al. / Tetrahedron 56 (2000) 2713±2724
Table 1 (continued)
#
MW of amines, Da
Chemical name
Tr(NHS)
(3)
MMTr(NHS)
(2)
DMTr(NHS)
(1)
MMTr(NHS)
(6)
72.
73.
74.
75.
76.
77.
78.
79.
80.
213.41
221.31
227.44
231.30
241.46
243.15
269.50
297.57
339.61
Tetradecylamine
9-(Methylaminomethyl)-anthracene
Pentadecylamine
1-Pyrenemethylamine
Hexadecylamine
3,5-Bis(tri¯uoromethyl)-benzylamine
Octadecylamine
Didecylamine
482.81
490.77
496.88
509.83
512.74
±
538.99
566.97
609.00
512.79
520.83
526.90
540.93
542.71
±
568.92
596.99
638.99
542.81
±
557.03
571.08
572.78
±
599.15
627.02
±
752.30
±
780.21
808.36
812.17
860.46
864.52
920.55
±
Hexetidine
of tags. Therefore, the above range of amines can only yield
a limited number of tags. Secondary amines are not useful
because usually there are primary amines with the same
masses, and generally secondary amines are less reactive
owing to steric hindrance (cf. entries ##14 and 61, Table
1). Amines bearing other reactive groups, for example,
##11, 12 and 18 (Table 1), cannot be used as tags in the
conditions of oligonucleotide synthesis unless they are
introduced at the very last step or an additional capping
step is employed. Finally, some amines are unstable to the
deprotection conditions (e.g. #13). To increase the number
of mass-tags using readily available amines, we synthesised
an additional trityl-based reagent containing two NHS-
activated carboxyl groups (6), which allows attachment of
two amines to the same trityl moiety thus extending the series
of mass-tags into the higher mass range (9, Scheme 1).
products after quenching with sat. NaHCO3 (data not
shown). As expected, there was about one order of magni-
tude difference in stabilities between DMTr, MMTr and Tr.
The corresponding carboxy-derivatives were about two to
®ve times more stable, i.e. the stability was: DMTr,
DMTr(NHS) (1),MMTr,MMTr(NHS) (2),MMTr(2NHS)
(6),Tr,Tr(NHS) (3). The MMTr(NHS) group (2) is about
50 times more stable than the DMTr group. There was no
detectable difference between the acidic stabilities of NHS-
activated trityl groups and their corresponding amide
derivatives. Interestingly, the stability of the modi®ed trityls
correlated well with their relative signal intensities (Fig. 2).
No difference was detected for the signal intensity for trita-
nols as compared to the corresponding trityl ethers when
using laser ionisation instead of acidic treatment (data not
shown), suggesting photocleavage by the laser irradiation52
as a good alternative to acidic cleavage.
To evaluate these modi®ed trityl blocks as precursors for
mass-tags, compounds 1 and 2 were used to synthesise
50-protected thymidine.40±42 0.1 M solutions of these
nucleosides, and also of compounds 3 and 6 in the OH-form
(tritanols), in THF were reacted with 0.5±1 M solutions of
amines in THF or dioxane (5 equiv. of amine for mono-
NHS-based compounds 1±3 and 10 equiv. for 6), by mixing
200 ml of each of tritylated compounds with the correspond-
ing amount of an amine solution and allowing them to react
for 5 min. The reaction mixtures were then analysed by
mass spectrometry without matrix, to prevent the formation
of molecular ions, by applying 1 ml of these mixtures
directly on to a sample target plate and allowing them to
dry. Typical results are presented in Table 1. For all trityl
derivatives, the compounds that gave the strongest signal
were selected and analysed as a mixture by mixing all and
applying 1 ml of the mixture to the target plate. Typical
spectra are shown in Fig. 1(a)±(c). There is a certain
decrease in the signal intensities when the tags are analysed
as a mixture, which can partially be explained by the
decreased concentration of individual tags in the mixture.
But even in mixtures, the signal intensities are high enough
to distinguish the tags unambiguously. For applications in
which the tags can be removed prior to LDI-TOF analysis
(Scheme 2), the use of matrix may further increase the
signal intensities (Fig. 1(c)).
Encoding of combinatorial libraries
For the 300 mm Rapp beads used in our experiments, the
average mass of a single bead is about 10 mg, which, given
the capacity of the reactive hydroxyl groups on that polymer
as 0.21 mmol/g, gives a loading of ,2 nmol per bead; more
usually, for the supports used for combinatorial chemistry it
is ,200 pmol/bead.56 Assuming some loss of reactive
groups due to incomplete synthesis, we would still have
about 1 nmol of reactive groups per bead, a few percent of
which would be tagged by the scheme used in this study
(vide infra). Two-fold and ten-fold dilution experiments
(data not shown) showed that the lower limit of LDI-TOF
detection of trityl-based tags is around 10213 M concen-
tration level (per single tag). The diameter of the spot on
the sample well, which is covered by the laser beam, is
about 100±300 mm, which means that the actual amount
of sample necessary for detection is in the fmol range. With
about 5% of sites occupied by tags, one bead provides about
50 pmol of tags, which is more than enough for detection.
To be used as a tag in oligonucleotide synthesis, the trityl
group should give clean high intensity signal in (MA)LDI-
TOF analysis. It should also survive several steps of acidic
treatment used to remove the 50-DMTr group in oligo-
nucleotide synthesis, that is, be orthogonal to the other
protective groups involved. (The stability of the NHS-
group in conditions of oligonucleotide synthesis has been
proved elsewhere.40,41) We selected the MMTr(NHS) group
(2) to meet both these requirements. Unlike DMTr(NHS)
(1), it remains attached to a primary hydroxyl group after at
All the synthesised trityl tags (1±3, 6) and standard DMTr,
monomethoxytrityl (MMTr) and triphenylmethyl (trityl, Tr)
groups were tested for acid-lability by treatment of the
corresponding 50-thymidylates with 1±3% TsOH, HClO4
or 80% aqueous acetic acid and TLC-analysis of the

M. S. Shchepinov et al. / Tetrahedron 56 (2000) 2713±2724
2717
Figure 1. (a) Selection of mass-tags based on compound 2 (MMTr(NHS)) treated with amines ## 14, 15, 19, 21, 22, 29, 32, 36, 46, 52, 57 and 62. Analysed as a
mixture of 8, without matrix. (b) Selection of mass-tags based on compound 6 (MMTr(2NHS)) treated with amines ## 62, 66, 68, 72, 75 and 78. Analysed as a
mixture of 9, without matrix. (c) Selection of mass-tags based on compound 2 (MMTr(NHS)) treated with amines ## 4, 5, 7, 10, 13±15, 19, 21, 22, 29, 32, 36,
46, 52, 57 and 62. Analysed as a mixture of 8, with matrix (4-hydroxy-a-cyanocinnamic acid).
least 8±9 cycles of acidic deprotection in oligonucleotide
synthesis using a two-fold diluted standard solution of
trichloroacetic acid in dichloromethane (ABI/Cruachem)
and a reduced deprotection time (see Experimental). For
analysis, it was easily released as 8 using 1±3% TFA in
the same solution (,1±3 min). Higher concentrations of
TFA, while removing the tags, cause some decomposition
of polyethyleneglycol (PEG) chains, which sometimes leads
to complex signals in the area of 350±500 Da when the
samples are run without matrix. These peaks showed
mass-differences of 14±16 DA, suggesting them to be
CH2 and CH2O fragments of the PEG linker (data not
shown). The same overwhelming presence of PEG decom-
position products was observed in LDI-TOF analysis when
individual beads were glued directly to the gold sample well
using a very thin ®lm of rubber glue. It is possible that trityl

2718
M. S. Shchepinov et al. / Tetrahedron 56 (2000) 2713±2724
Scheme 2. Encoding and decoding in combinatorial synthesis.
cations interact with the oxygen bridges of PEG chains, thus
being trapped in them and decreasing desorption. The
presence of matrix (in case of detritylated tags) solves this
problem. At the same time, when derivatised with different
amines, MMTr(NHS) (2)-based tags produced the second
highest intensity signals (after DMTr(NHS) (1)) and gave
reasonably clean peaks, even when mixtures of compounds
were analysed (Fig. 2).
phosphoramidites prior to oligonucleotide synthesis, in a
way similar to that described by Cho et al.24 Assuming the
stepwise yield of oligonucleotide synthesis to be about 99%,
for an 8-mer library synthesised using 7 as a 5% additive to
all bases, we would have ca 60% of all sites of the beads
occupied by full length oligonucleotides in the ®nal product.
The concentration of the ®rst tag (5% of all initial sites)
would be about two-fold greater than that of the last tag
(5% of the remaining 60% of the sites), which still makes
it possible to detect all of them in the same mixture. Oligo-
nucleotide synthesis was carried out on a 4-column ABI
machine. After each oxidation step, the columns were
removed and treated with different amines as described in
the experimental section. No particular rationale was used
when choosing the masses of tags to be used for the encod-
ing of a particular base/position. But in principle, certain
mass-ranges could be used to code for either a base type
(i.e. all A are coded by tags ranging from 400 to 500 Da, all
C by tags in the interval of 500±600 Da etc.) or, alterna-
tively, the position of the base might be a determining factor
(i.e. position #1, or a 30-end, (A, C, G or T) is coded by
mass-tags ranging from 400±420 Da, position #2- by 420±
440 Da etc.).
To introduce a tagging moiety during oligonucleotide
synthesis (Scheme 3), we synthesised the non-nucleoside
phosphoramidite synthon 7 based on a propanediol struc-
ture,54 which provides similar reactivity to the standard A,
C, G and T phosphoramidites. The MMTr(NHS)Cl (2) has
reduced reactivity compared to both DMTrCl and
DMTr(NHS)Cl (1), and it is important, when synthesising
7, to carry out the tritylation reaction at low temperatures to
prevent the formation of ester bonds between the excess of
propanediol and the activated carboxyl group (data not
shown). The phosphoramidite 7 was stable in acetonitrile
solution at rt for at least 2 days.
We mixed ca. 3±6 mol% of 7 with standard A, C, G and T

M. S. Shchepinov et al. / Tetrahedron 56 (2000) 2713±2724
2719
Figure 2. (Top) Individual mass-spectra of trityl-based non-puri®ed pro-tags 1, 2, 3 and 6 (in the form of tritanols) treated with phenoxyethylamine (#35, Table
1); analysed without matrix (ionisation by laser irradiation). (Bottom) Mass-spectrum of a mixture of equimolar amounts of the same four mass-tags; analysed
without matrix. (Expected masses of the tags: 406.6, 436.5, 466.5 and 599.9).
Scheme 3.

2720
M. S. Shchepinov et al. / Tetrahedron 56 (2000) 2713±2724
Figure 3. Decoding of the sequence of oligonucleotide attached to the beads selected from combinatorial library of 256 pyrine 8-mers by hybridisation to
oligonucleotide 50-Cy5-CTC.CTC.TC. The mass-tags shown encode for (from 30-end): G (398), A (414), G (456), G (428), A (442), G (470), A (434) and G
(482) (50-GAG.AGG.AG).
Beads were selected by hybridisation with a Cy5-labelled
oligonucleotide. The size of Rapp-beads (,0.3 mm) allows
manual removal of positively identi®ed beads, visible with
the naked eye, from the pool. For smaller beads, automated
methods such as ¯ow-cytometry (FACS) might be used.
Selected beads were detritylated and the mixtures of tags
released analysed by mass-spectrometry (Fig. 3). The spec-
trum corresponded to the sequence of the oligonucleotide
used for hybridisation.
identify the tags unambiguously. The scheme and reagents
described here are very simple as compared to other
methods of encoding. The tags are suitable for any combi-
natorial synthesis that does not involve strong acids, for
example, peptide synthesis using Fmoc-protected amino-
acids. An additional increase of the amount of reactive
sites on the beads can be achieved by employing the
`trebling' phosphoramidite prior to combinatorial synthe-
sis.59,60 We were able to identify the oligonucleotide on a
single bead selected from a hybridisation screen of a library
of 256 sequences, thus demonstrating the ef®ciency and
simplicity of our system, which is rapid, reliable and
needs no special apparatus. The trityl tags could also be
used to increase the desorbtion rate of compounds that are
otherwise dif®cult to analyse in (MA)LDI-TOF.
To eliminate the problem of gradual loss of encoding
MMTr-based tags 8 during the detritylation step in oligo-
nucleotide synthesis, we have also used Fmoc as a
50-protecting group,57,58 thus omitting the use of acidic
conditions altogether. After each oxidation step, the
columns were removed from the synthesiser, and the
beads were treated with the corresponding amines, washed
with acetonitrile and then treated with 0.1 M DBU in aceto-
nitrile58 to remove Fmoc-protection. The tags encoding for
9-mer oligonucleotide synthesised using this strategy were
detected using (MA)LDI-TOF analysis (data not shown).
For longer sequences, the 30-ethyl- or 30-methylphosphor-
amidites of 50-Fmoc-protected nucleosides could preferably
be used instead of cyanoethylphosphoramidites, to prevent
the loss of the cyanoethyl (Cnet) protecting group due to the
treatment with amines and DBU.
Experimental
Split-and-mix synthesis of oligonucleotides was carried out
in an Applied Biosystems 394 DNA/RNA four column
synthesiser using standard phosphoramidite chemistry.53
The solid support used was Tenta Gel Macrobeads OH,
280±320 mm, Rapp Polymer. Cy5 phosphoramidite was
from Pharmacia/Biotech. UV-spectra were measured on a
Spectronic 2000 spectrophotometer, Milton Roy Co., USA.
MALDI- and LDI-TOF mass-spectra were recorded on a
PE-ABI Voyagere Elite Re¯ectron Delayed Extraction
Instrument. Spectra were acquired with an accelerating
voltage of 25 KV and 100 ms delay in the positive ion
mode. Micrographs were taken on a Leica TCS NT confocal
microscope equipped with an Ar/Kr multiline laser. 1H
NMR spectra were recorded on a Varian Gemini 200
200 MHz spectrometer. 31P NMR spectra were recorded
on a Brucker AC-500 spectrometer (internal standard:
80% phosphoric acid). HPLC was carried out on a Waters
system (Milford, MA, USA). Chemicals were purchased
from Aldrich Chemical Company (USA), Avocado
Research Chemicals (UK), Lancaster Synthesis Ltd (UK),
and Acros Organics (Fisher Scienti®c, UK). Phosphitylating
reagent and oligonucleotide puri®cation columns were
from Sigma Chemical Company. Silica gel for column
Conclusions
We have shown that trityl-based mass-tags ¯y extremely
well in the positive mode of an LDI process, presumably
due to the formation of a highly stabilised trityl carbocation.
By modifying the trityl moiety with a reactive group, the
trityl-based pro-tag can be successfully used for encoding in
combinatorial chemistry. Tags describing the sequence/
composition of a biopolymer attached to an insoluble bead
are introduced after each chemical transformation, by
reacting the pro-tags with different primary amines. We
have shown that the tags are easily detectable using
(MA)LDI-TOF analysis, provided the mass-difference
between them is at least 2±3 Da. In these conditions, the
fragmentation of tritylamides is very low, which helps to

M. S. Shchepinov et al. / Tetrahedron 56 (2000) 2713±2724
2721
chromatography, UV254 TLC plates and solvents were from
BDH/Merck.
was dissolved in water (1 l) at 108C, and 5 M KOH
(300 ml) was added with stirring. The mixture was extracted
with chloroform (3£350 ml), the organic phase dried over
CaCl2 and evaporated. The product was crystallised from
toluene to give 42 g (75%) of white crystalline solid, mp
86±888C. (Found: C, 77.68; H, 5.95; N, 4.82. C18H17NO2
(MW 279.33) requires C, 77.40; H, 6.13; N, 5.01%); nmax
(Nujol): 2923 (br), 2854, 1654, 1637, 1463, 1376, 1311,
N-Succinimidyl-4-[bis-(4-methoxyphenyl)-chloromethyl]-
benzoate (1). It was synthesised according to the reported
procedures.40,41
Racemic N-succinimidyl-4-[(4-methoxydiphenyl)-chloro-
methyl]-benzoate (2). This was synthesised according to
reported procedures,40,41 but 4-methoxybenzophenone was
used in the Grignard synthesis instead of 4,40-dimethoxy-
benzophenone. Total yield: 45%, yellow-white solid, mp
181±1848C. (Found: C, 66.58; H, 4.55; N, 3.20.
C25H20ClNO5 (MW 449.88) requires C, 66.76; H, 4.48; N,
3.11%); nmax (Nujol): 2919 (br), 2852, 1768, 1735, 1605,
1280, 1069 cm21 1H NMR (CDCl3, d): 8.1 (d, 2H,
.
Jˆ8.5 Hz, O(N)C-aryl-H, ortho), 7.8 (m, 4H, arom.), 7.6
(m, 1H, arom.), 7.51 (m, 2H, arom.), 4.18 (s, 2H, CH2),
1.45 (s, 6H, CH3). 13C NMR (CDCl3, d): 132.89, 130.1
(d), 128.53 (d), 79.34, 67.88, 28.25. HRMS: MALDI-TOF
(a-cyano-4-hydroxycinnamic acid): 279.1 (MI), 302.1
(MI1Na1), 319.7 (MI1K1). LDI-TOF, m/z (rel. intensity):
279.1 (100), 202.1 (12), 181.1 (25). [Calculated exact mass
for C18H17NO2: 279.12593; found: 279.12614 (0.8 ppm
error)].
1
1508, 1459, 1376, 1304, 1254, 1202, 1068 cm21. H NMR
(CDCl3, d): 8.11 (d, 2H, Jˆ8.4 Hz, OOC-aryl-H, ortho),
7.53 (d, 2H, Jˆ8.4 Hz, OOC-aryl-H, meta), 7.31 (m, 5H,
arom.), 7.15 (d, 2H, Jˆ8 Hz, CH3O-aryl-H, meta), 6.85 (d,
2H, Jˆ8 Hz, CH3O-aryl-H, ortho), 3.8 (s, 3H, OCH3), 2.9
(s, 4H, NHS). 13C NMR (CDCl3, internal standard CDCl3 at
77.03 ppm, d): 169.54, 153.2, 131.04, 130.29 (d), 129.53
(d), 128.24 (d), 127.96, 113.47 (d), 55.31, 25.56. HRMS
(LDI-TOF) for corresponding tritanol, m/z (rel. intensity):
414.3 (100), 333.4 (6), 317.7 (27), 289.3 (4). [Calculated
exact mass for C25H20NO5 (MMTr(NHS)1): 414.13415;
found: 414.13621 (4.9 ppm error)].
4,40-[bis-(2-(4,4-Dimethyl-1,3-oxazolyl))]-400-methoxytri-
tanol (5). To magnesium turnings (1.5 g) activated with
iodine bromophenyl oxazoline (15.34 g, 0.06 mol) in dry
THF (150 ml) and a catalytic amount of RED-AIw and
MeI were added with stirring and the mixture was re¯uxed
for 3 h, cooled to rt and methyl 4-methoxybenzoate (4.64 g,
28 mmol) in dry THF (40 ml) was added dropwise. The
mixture was gently re¯uxed for 6 h, cooled to rt and water
(10 ml) was added with stirring. The organic phase was
carefully decanted and the residue washed several times
with small portions of THF. Combined organic fractions
were evaporated and puri®ed (¯ash-chromatography) to
give 11.4 g (84%) of light yellow solid. (Found: C, 74.57;
H, 6.57; N, 5.59. C30H32N2O4 (MW 484.59) requires C,
74.36; H, 6.66; N, 5.78%); nmax (Nujol): 2950±2800 (br),
1714, 1648, 1607, 1509, 1462, 1408, 1354, 1316 (br), 1253
N-Succinimidyl-4-[bis-(phenyl)-chloromethyl]-benzoate
(3). This was synthesised according to the reported pro-
cedures,40,41 but benzophenone was used in the Grignard
synthesis instead of 4,40-dimethoxybenzophenone.
Yellow-white solid, mp 170±1738C. (Found: C, 68.88; H,
4.14; N, 3.09. C24H18ClNO4 (MW 419.86) requires C, 68.66;
H, 4.32; N, 3.34%); nmax (Nujol): 2920±2800 (br), 1765,
1
1
1735, 1605, 1500, 1455, 1375, 1315, 1250, 1200 cm21. H
7.85 cm21. H NMR (CDCl3, d): 8.11 (d, 4H, Jˆ8.4 Hz,
NMR (CDCl3, d): 8.1 (d, 2H, Jˆ8.5 Hz, OOC-aryl-H,
ortho), 7.51 (d, 2H, Jˆ8.5 Hz, OOC-aryl-H, meta), 7.28
(m, 10H, arom.), 2.86 (s, 4H, NHS). 13C NMR (CDCl3,
d): 169.57, 146.12, 130.37, 128.42 (d), 127.95 (d), 123.85,
81.86, 33.77, 25.52. HRMS (LDI-TOF) for corresponding
tritanol, m/z (rel. intensity): 401.1 (75), 384.1 (100), 286.2
(8), 259.6 (5). [Calculated exact mass for C24H18NO4
(Tr(NHS)1): 384.12358; found: 384.12375 (0.4 ppm
error)]. The same product was obtained using commercially
available phenylmagnesium bromide and 2-(4-benzo-
phenyl)-4,4-dimethyl-1,3-oxazoline (4) in the Grignard
reaction.
O(N)C-aryl-H, ortho), 7.32 (d, 4H, Jˆ8.4 Hz, O(N)C-
aryl-H, meta), 7.12 (d, 2H, Jˆ8 Hz, CH3O-aryl-H, meta),
6.81 (d, 2H, Jˆ8 Hz, CH3O-aryl-H, ortho), 4.12 (s, 4H,
CH2), 3.78 (s, 3H, OCH3), 1.37 (s, 12H, CH3). 13C NMR
(CDCl3, d): 162.15, 159.1, 150.1, 138.64, 131.5 (d), 129.15
(d), 128.01 (d), 126.95, 113.51 (d), 81.38, 79.13, 67.52,
55.18, 28.21. HRMS: MALDI-TOF (a-cyano-4-hydroxy-
cinnamic acid): 484.9 (MI), 467.6 (MI2OH). LDI-TOF,
m/z (rel. intensity): 484.2 (100), 467.2 (43), 453.2 (7),
369.2 (7.5), 310.1 (36), 294.1 (20). [Calculated exact mass
for C30H32N2O4: 484.23621; found: 484.23835 (4.4 ppm
error)].
2-(4-Benzophenyl)-4,4-dimethyl-1,3-oxazoline (4). 4-Benz-
oylbenzoic acid (50 g, 0.22 mol) was re¯uxed in thionyl
chloride (300 ml) for 3 h, evaporated (the product crystal-
lises from the oil) and then evaporated with toluene
(2£30 ml). The residue was dissolved in dry methylene
chloride (250 ml). To this ice-cooled solution, 2-amino-2-
methylpropan-1-ol (46 g, 0.51 mol) in dry methylene
chloride (150 ml) was added dropwise for 2 h. The solution
was stirred overnight at rt, and the precipitate was washed
several times with methylene chloride. Combined fractions
were evaporated, slowly dissolved in thionyl chloride
(350 ml) and re¯uxed for 4 h. The reaction mixture was
evaporated to one-third volume, poured into dry ether (2 l)
and kept overnight at 48C. The hydrochloride precipitate
4,40-[bis-(2-(Succinimidylcarboxy)]-400-methoxytrityl
chloride (6). The solution of 5 (10 g, 21 mmol) in 80%
acetic acid (250 ml) was kept at 728C for 48 h, evaporated
and then re-evaporated from water (2£50 ml). The product
was dissolved in 50% EtOH/water (75 ml), re¯uxed for 3 h
and evaporated to one-third volume. The mixture was then
dissolved in water (100 ml) and acidi®ed with 3 M HCl to
pH 1±2. The precipitate was dissolved in chloroform, dried
(Na2SO4) and evaporated to dryness and additionally dried
in vacuo overnight. The resulting dicarboxylic acid was
dissolved in dry THF (100 ml). N-hydroxysuccinimide
(8.5 g, 74 mmol) was added and the mixture was cooled
to 08C. Dicyclohexylcarbodiimide (8.5 g, 41 mmol) in dry
THF (20 ml) was added dropwise with stirring. The reaction

2722
M. S. Shchepinov et al. / Tetrahedron 56 (2000) 2713±2724
mixture was stirred for 1 h at 08C and then overnight at rt.
Dicyclohexylurea was ®ltered off and organic phase was
evaporated to dryness and puri®ed (¯ash-chromatography)
to give 8.5 g (72%) of yellow-white solid, mp 212±2158C
(decomp.). (Found: C, 63.13; H, 4.04; N, 4.88. C30H24N2O10
(MW 572.52) requires C, 62.94; H, 4.23; N, 4.89%); nmax
(Nujol): 2950±2800 (br), 1765, 1736, 1606, 1510, 1460,
Analysis of trityl-based tags
0.1 M solutions of compounds 1, 2, 3 and 6 were prepared in
a mixture of THF/dioxane (1:1). 110 ml of these solutions
were mixed with 40 ml (80 ml in case of 6) of 0.5±1 M
solutions of different amines (Table 1) as described by
Gildea40 and Coull.41 The mixtures were then analysed
with and without matrix either directly or when mixed in
different combinations.
.
1406, 1375, 1255, 1200 (br), 1070 cm21 1H NMR
(CDCl3, d): 8.11 (d, 4H, Jˆ8.5 Hz, OOC-aryl-H, ortho),
7.5 (d, 4H, Jˆ8.5 Hz, OOC-aryl-H, meta), 7.11 (d, 2H,
Jˆ8.1 Hz, CH3O-aryl-H, meta), 6.88 (d, 2H, Jˆ8.1 Hz,
CH3O-aryl-H, ortho), 3.81 (s, 3H, OCH3), 2.9 (s, 8H,
CH2). 13C NMR (CDCl3, d): 169.69, 161.86, 159.5, 153.6,
137.64, 130.56, 129.4, 128.47, 124.21, 113.85, 81.34, 55.32,
25.55. HRMS: MALDI-TOF (a-cyano-4-hydroxycinnamic
acid): 554.7 (MMTr(2NHS)1), 570.8 (MI). LDI-TOF, m/z
(rel. intensity): 572.1 (100), 554.4 (45), 474.1 (10), 397.8
(5). [Calculated exact mass for C30H24N2O10: 572.14310;
found: 572.14498 (3.3 ppm error)]. This compound was
converted into the corresponding trityl chloride by re¯uxing
in AcCl/toluene (100 ml) for 3 h. The reaction mixture was
then evaporated to one-third volume. Toluene (60 ml) was
added, the mixture was again evaporated to one-third
volume and used without further puri®cation.
Oligonucleotide synthesis
Oligonucleotide synthesis was carried out using commer-
cially available standard A, C, G and T, PAC, ¯uorescein
and Cy5 phosphoramidites according to manufacturer's
protocols.
Synthesis of the combinatorial library
Phosphoramidite 7 was added to standard A, C, G and T
phosphoramidites or to 50-Fmoc-phosphoramidites57 up to
3±6 mol.% of total amount of phosphoramidite. 35±40 mg
(ca. 3500 beads) of Rapp TentaGel Macrobeads were placed
in each of four polypropylene DNA synthesis columns
(1 mmol scale, Glen Research, USA). The oligonucleotide
synthesis was carried out on 1 mmol scale using phosphor-
amidite mixtures: A17, C17, G17 and T17, according to
manufacturer's protocol, but the supply of deblocking
reagent (diluted to 50% of its original concentration with
methylene chloride) to the columns was reduced to 10±15 s,
with a subsequent waiting step (10 s) followed by another
portion of acid (10 s). Subsequent thorough acetonitrile
washing of the columns ensured that all DMTr1 is desorbed.
Before each detritylation step, the columns were washed
with acetonitrile using Manual Control mode, and then
treated with corresponding amines (0.3±0.5 ml of 0.5±
1 M solutions in dioxan/THF, depending on the solubility
of an amine) for 1 min using 1 ml syringes. The columns
were then washed exhaustively with acetonitrile and dried in
vacuo for 15 min. The beads from all columns were then
combined together in a 0.3 ml reaction vial with conical
chamber (Pierce), mixed, and split again in four portions
by pipetting the suspension of beads in acetonitrile (about
1 volume of solvent per 1 volume of beads) using a 1 ml
Eppendorf micropippette tip. The procedure was repeated
until the end of the synthesis. The beads were then washed,
dried, deprotected for 14 h in concentrated ammonia
(1.5 ml) at 558C, then washed several times with distilled
water, dried and stored at 48C.
O1-{[4-(Succinimidylcarboxy)]-40-methoxytrityl}-1,3-pro-
panediol. The title compound was synthesised from 2
according to a published procedure,54 but the tritylation
step was carried out in pyridine at 08C overnight without a
catalyst. Monoprotected propanediol was obtained in 55%
yield after puri®cation by ¯ash-chromatography as a white
foam. (Found: C, 68.45; H, 5.81; N, 3.07. C28H27NO7 (MW
1
489.52) requires C, 68.70; H, 5.56; N, 2.86%); H NMR
(CDCl3, d): 8.10 (d, 2H, Jˆ8.3 Hz, arom.), 7.6 (d, 2H,
Jˆ8.3 Hz, arom.), 7.29 (m, 5H, arom.), 7.14 (d, 2H,
Jˆ8 Hz, arom.), 6.89 (d, 2H, Jˆ8 Hz, arom.), 3.8 (s, 3H,
OCH3), 3.79 (t, 2H, Jˆ6 Hz, CH2OH), 3.28 (t, 2H, Jˆ6 Hz,
MMTr(NHS)OCH2), 2.9 (s, 4H, succinimide), 1.89 (quin,
2H, Jˆ6 Hz, CH2CH2CH2), 1.65 (br.s., 1H, OH). 13C NMR
(CDCl3, d): 62.5 (C1), 61.02 (C3), 55.02 (OCH3), 32.67
(C2), 25.12 (C(O)CH2CH2C(O)). HRMS: MALDI-TOF
(a-cyano-4-hydroxycinnamic acid): 489.1 (MI), 527.9
(MI1K1); LDI-TOF, m/z (rel. intensity): 489.3 (25),
414.4 (100). [Calculated exact mass for C28H27NO7:
489.17875; found: 489.17906 (0.6 ppm error)].
O1-{[4-(Succinimidylcarboxy)]-40-methoxytrityl}-O3-(N,N-
diisopropylamino-2-cyanoethoxyphosphinyl)-1,3-propane-
diol (7). Phosphitylation of monoprotected propanediol was
carried out as described54 to give the title phosphoramidite
as a white foam in 70% yield. (Found: C, 64.75; H, 6.22; N,
5.86. C37H44N3O8P (MW 689.73) requires C, 64.43; H, 6.43;
Hybridisation of the combinatorial library and detri-
tylation
1
N, 6.09%); H NMR (CDCl3, d): 8.11 (d, 2H, Jˆ8.5 Hz,
The beads were hybridised to 50-Cy5-labelled 8-mer oligo-
nucleotide 50-CTC.CTC.TC in 1.5 ml of 3.5 M TMA
buffer55 in a 4 ml vial which was rotated on a Spiramix 10
machine overnight at 88C. The beads were then washed ®ve
times with TMA buffer at the same temperature, transferred
on to the surface of a 7.5£5 cm microscope slide and the
excess of the buffer removed by blotting with tissue paper.
Coloured or otherwise identi®ed beads were then removed
using forceps (usually about 10±15 beads), washed with
water, acetone and dried. The trityl tags were cleaved by
arom.), 7.59 (d, 2H, Jˆ8.5 Hz, arom.), 7.31 (m, 5H, arom.),
7.15 (d, 2H, Jˆ8.4 Hz, arom.), 6.88 (d, 2H, Jˆ8.4 Hz,
arom.), 3.81 (s, 3H, OCH3), 3.7±3.3 (m, 6H,
NCCH2CH2OP, CH2OP and CH(CH3)2), 3.2 (t, 2H,
Jˆ6 Hz, MMTr(NHS)OCH2), 2.9 (s, 4H, succinimide),
2.65 (m, 2H, CH2CN), 1.93 (quin, 2H, Jˆ6 Hz,
CH2CH2CH2), 1.16 (d, 12H, Jˆ6 Hz, CH(CH3)2). 31P
NMR (CDCl3, d, md): 144.512. HRMS (MALDI-TOF):
calculated exact mass for C37H44N3O8P: 689.28660;
found: 689.28799 (2 ppm error).

M. S. Shchepinov et al. / Tetrahedron 56 (2000) 2713±2724
2723
21. Egner, B. J.; Langley, G. J.; Bradley, M. J. Org. Chem. 1995,
60, 2652±2653.
treating the beads with 10±50 ml of 1±3% (v/v) solution of
tri¯uoroacetic acid in standard Deblok Solution (Cruachem;
trichloroacetic acid in dichloromethane) for 3±4 min. The
supernatant was evaporated several times with acetone or
methanol to remove the acids and the residue was analysed
by (MA)LDI-TOF HRMS.
22. Fitzgerald, M. C.; Harris, K.; Shevlin, C. G.; Siuzdak, G.
Bioorg. Med. Chem. Lett. 1996, 6, 979±982.
23. Geysen, H. M.; Wagner, C. D.; Bodnar, W. M.; Markworth,
C. J.; Parke, G.; Schoenen, F. J.; Wanger, D.; Kinder, D. S. Chem.
Biol. 1996, 3, 679±688.
24. Cho, C. Y.; Youngquist, R. S.; Paikoff, S. J.; Beresini, M.;
Hebert, A. R.; Berleau, L. T.; Liu, C. W.; Wemmer, D. E.; Keough,
T.; Schultz, P. G. J. Am. Chem. Soc. 1998, 120, 7706±7718.
25. Egner, B. J.; Rana, S.; Smith, H.; Bouloc, N.; Frey, J.;
Brocklesby, W. S.; Bradley, M. Chem. Commun. 1997, 735±736.
26. Yan, B.; Kumaravel, G.; Anjaria, H.; Wu, A.; Petter, R.;
Jewell, C. F.; Wareing, J. R. J. Org. Chem. 1995, 60, 5736±5738.
27. Sowin, T. J.; Hochlowski, J.; Grillot, A. Innovations and
Perspectives in Solid Phase Synthesis and Combinatorial
Libraries; May¯ower Scienti®c: Birmingham, 1998, pp 205±208.
28. Sarkar, S. K.; Garigipati, R. S.; Adams, J. L.; Keiffer, P. A.
J. Am. Chem. Soc. 1996, 118, 2305±2306.
Acknowledgements
The authors would like to express their gratitude to Drs V.
A. Korshun, D. A. Stetsenko and D. J. Harvey for fruitful
discussions, Dr K. Williams for recording the IR spectra,
and to Dr R. Quarrell for reading the manuscript and
valuable suggestions and comments. R. C. was supported
by a grant from Amersham.
29. Houghten, R. A. Proc. Natl. Acad. Sci., USA 1985, 82, 5131±
5135.
References
30. Nicolaou, K. C.; Xiao, X. Y.; Parandoosh, Z.; Senyei, A.;
Nova, M. Angew. Chem., Int. Ed. Engl. 1995, 34, 2289±2291.
31. Moran, E. J.; Sarshar, S.; Cargill, J. F.; Shahbaz, M. M.; Lio,
A.; Mjalli, A. M. M.; Armstrong, R. W. J. Am. Chem. Soc. 1995,
117, 10787±10788.
1. Hanak, J. J. J. Mater. Sci. 1970, 5, 964±971.
2. Whole issue. Chem. Rev. 1997, 97, 347±510.
3. Whole issue. Curr. Opin. Chem. Biol. 1998, 2, 317±478.
4. McFarland, E. W.; Weinberg, W. H. Tibtech 1999, 17, 107±
115.
32. Kaye, P. H.; Tracey, M. C.; Gordon, J. A. WO Patent 95/
29473.
5. Southern, E. M.; Maskos, U.; Elder, J. K. Genomics 1992, 13,
1008±1017.
33. Michelson, A. M.; Todd, A. R. J. Chem. Soc. 1953, 951±956.
34. Smith, M.; Rammler, D. H.; Goldberg, I. H.; Khorana, H. G.
J. Am. Chem. Soc. 1961, 84, 430±440.
6. Fodor, S. P. A.; Read, J. L.; Purring, M.; Stryer, L.; Lu, R. T.;
Solas, D. Science 1991, 251, 767±773.
7. Furka, A.; Sebestyen, F.; Asgedom, M.; Dibo, G. Highlights of
Modern Biochemistry. Proceedings of the 14th International
Congress of Biochemistry; VSP: Ultrecht, The Netherlands,
1988; Vol. 5, p 47.
35. Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, Wiley: USA, 1991.
36. Gait, M. J. Oligonucleotide Synthesis: a Practical Approach,
IRL: Oxford, 1984.
8. Lam, K. S.; Salmon, S. E.; Hersh, E. M.; Hruby, V. J.;
Kazmierski, W. M.; Knapp, R. J. Nature 1991, 354, 82±85.
9. Brooking, P.; Doran, A.; Grimsey, P.; Hird, N.; MacLachlan,
W. S.; Vimal, M. Tetrahedron Lett. 1999, 40, 1405±1408.
10. Czarnik, A. W. Curr. Opin. Chem. Biol. 1997, 1, 60±66.
11. Brenner, S.; Lerner, R. A. Proc. Natl. Acad. Sci., USA 1992,
89, 5381±5383.
37. Fisher, E. F.; Caruthers, M. H. Nucleic Acids Res. 1983, 11,
1589±1599.
38. Sekine, M.; Hata, T. J. Org. Chem. 1987, 52, 946±948.
39. Sekine, M.; Mori, T.; Wada, T. Tetrahedron Lett. 1993, 34,
8289±8292.
È
40. Gildea, B. D.; Coull, J. M.; Koster, H. Tetrahedron Lett. 1990,
31, 7095±7098.
41. Coull, J. M; Gildea, B.; Koster, H. Patent USA 5,410,068,
12. Ni, Z.-J.; Maclean, D.; Holmes, C. P.; Murphy, M.; Ruhland,
B.; Jacobs, J. W.; Gordon, E. M.; Gallop, M. A. J. Med. Chem.
1996, 39, 1601±1608.
È
1995.
42. Leikauf, E.; Barnekow, F.; Koster, H. Tetrahedron 1995, 51,
È
13. Ohlmeyer, M. H. J.; Swanson, R. N.; Dillard, L. W.; Reader,
J. C.; Asouline, G.; Kobayashi, R.; Wigler, M.; Still, W. C. Proc.
Natl. Acad. Sci., USA 1993, 90, 10922±10926.
3793±3802.
43. Chu, S. S.; Reich, S. H. Bioorg. Med. Chem. Lett. 1995, 5,
1053±1058.
14. Nestler, H. P.; Bartlett, P. A.; Still, C. W. J. Org. Chem. 1994,
59, 4723±4724.
44. Patel, V. F.; Hardin, J. N.; Starling, J. J.; Mastro, J. M. Bioorg.
Med. Chem. Lett. 1995, 5, 507±512.
15. Giese, R.; Xu, L. Proceedings of The International Workshop
on Sequencing by Hybridisation, Houston Advanced Research
Center, 29±30 October 1993.
45. Patel, V. F.; Hardin, J. N.; Grindey, G. B.; Schultz, R. M.
Bioorg. Med. Chem. Lett. 1995, 5, 513±518.
46. Patel, V. F.; Hardin, J. N.; Mastro, J. M.; Law, K. L.; Zimmer-
mann, J. L.; Ehlhardt, W. J.; Woodland, J. M.; Starling, J. J.
Bioconj. Chem. 1996, 7, 497±510.
16. Zambias, R. A.; Boulton, D. A.; Grif®n, P. R. Tetrahedron
Lett. 1994, 35, 4283±4286.
17. Brummel, C. L.; Lee, I. N. W.; Zhou, Y.; Benkovic, S. J.;
Winograd, N. Science 1994, 264, 399±402.
47. Fourrey, J. L.; Varenne, J.; Blonski, C.; Dousset, P.; Shire, D.
Tetrahedron Lett. 1987, 28, 5157±5160.
18. Brummel, C. L.; Vickerman, J. C.; Carr, S. A.; Hemling, M.
E.; Roberts, G.; Johnson, W.; Weinstock, J.; Gaitanopoulos, D.;
Benkovic, S. J.; Winograd, N. Anal. Chem. 1996, 68, 237±242.
19. Youngquist, R. S.; Fuentes, G. R.; Lacey, M. P.; Keough, T.
J. Am. Chem. Soc. 1995, 117, 3900±3906.
48. Shchepinov, M. S.; Case-Green, S. C.; Southern, E. M. Nucleic
Acids Res. 1997, 25, 1155±1161.
49. Greene, G. L.; Letsinger, R. L. Tetrahedron Lett. 1975, 16,
2081±2084.
20. Egner, B. J.; Cardno, M.; Bradley, M. J. Chem. Soc., Chem.
Commun. 1995, 21, 2163±2164.
50. Letsinger, R. L.; Finnan, J. L. J. Am. Chem. Soc. 1975, 97,
7197±7198.

2724
M. S. Shchepinov et al. / Tetrahedron 56 (2000) 2713±2724
51. Matteucci, M. D.; Caruthers, M. H. Tetrahedron Lett. 1980,
21, 3243±3246.
57. Gioeli, C.; Chattopadhyaya, J. B. J. Chem. Soc., Chem.
Commun. 1982, 672±674.
52. Herz, M. L. J. Am. Chem. Soc. 1975, 97, 6777±6785.
53. Caruthers, M. H. Acc. Chem. Res. 1991, 24, 278±284.
54. Seela, F.; Kaiser, K. Nucleic Acids Res. 1987, 15, 3113±3129.
55. Maskos, U.; Southern, E. M. Nucleic Acids Res. 1993, 21,
4663±4669.
58. Lehmann, C.; Xu, Y.-Z.; Christodoulou, C.; Tan, Z.-K.; Gait,
M. J. Nucleic Acids Res. 1989, 17, 2379±2390.
59. Shchepinov, M. S.; Udalova, I. A.; Bridgman, A. J.; Southern,
E. M. Nucleic Acids Res. 1997, 25, 4447±4454.
60. Shchepinov, M. S.; Southern, E. M.; Russ J. Bioorg. Chem.
1998, 24, 794±797.
56. Shapiro, M. J.; Lin, M.; Yan, B. A Practical Guide to Combi-
natorial Chemistry; Czarnik, A. W., DeWitt, S. H. Eds.; American
Chemical Society: Washington, DC, 1997, pp 123±151.