БИООРГАНИЧЕСКАЯ ХИМИЯ, 1998, том 24, № 10, с. 794-797
ПИСЬМА РЕДАКТОРУ
УДК 577.ИЗ.6
THE SYNTHESIS OF BRANCHED OLIGONUCLEOTIDE STRUCTURES
© 1998 r. Mikhail S. Shchepinov#, Edwin M. Southern
Department of Biochemistry, Oxford University, South Parks Rd, 0X1 3QU, Oxford, UK Received March 19, 98; in final form May 18, 98
New phosphoramidite reagents, tris-2,2,2-[3-(4,4'-dimethoxytrityloxy)propyloxymethyl]-ethyl-A/,A/-diisopropy-lamino-2-cyanoethylphosphoramidite, /V-[5-(9-fluorenylmethoxycarbonyloxy)valeryl]-/V-[5-(4,4'-dimethoxytri-tyloxy)valeryl]-2-(9-(i'V,A'-diisopropylamino-2-cyanoethylphosphinyl)-l,3-diamino-2-propanol, and /V-[4-(i(?ri-butyldimethoxysilyloxy)butyry]]-/V'-[4-(4,4'-dimethoxy-trityloxy)butyryl]-2-0-(/V,A'-diisopropyIamino-2-cya-noethylphosphinyl)-l,3-diamino-2-propanol, were synthesized and used for the creation of plain and mixed oligonucleotide dendrimers. Our approach appears to be a convenient way to the synthesis of plain and mixed den-drimers that bear various functional moieties (including oligonucleotides) and possess a wide range of chemical and physical properties. The reagents suggested are easily available and compatible with automated solid phase phosphoramidite chemistry, which makes them potentially interesting for many biochemically-oriented laboratories.
Key words: branched phosphoramidite reagents; oligonucleotide dendrimers.
INTRODUCTION
Novel technologies have being developed on the interface of traditional disciplines of science and engineering. There is a need of new synthetic methods, which can fill a gap between the molecular scale and the nanoscale preparative techniques in bioorganic chemistry. We describe here a novel set of reagents and methods that allow the assembly of large branched structures (dendrimers) from simple but versatile building blocks.
First introduced in 1981 [1], the phosphoramidite approach has made the synthesis of oligonucleotides routine to all life-sciences laboratories thanks to its reliability and simplicity. Our range of branching syn-thons may expand the potential of phosphoramidite chemistry by creating two-dimensional and three-dimensional dendrimeric structures [2].
Branched oligonucleotides were found in nature [3] and were synthesized by using modified deoxy- or ri-bonucleosides as the branching points [4, 5]. Another approach to branched structures was based on the use of nonnucleoside forks [6]. Both synthetic schemes doubled the number of reactive groups after each condensation stage. More rapid growth might be achieved with more branched monomers.
Abbreviations: CPG, controlled pore glass; DMAP, /?-dimethy-laminopyridine; DMTr, 4,4'-dimethoxytrityl; EDIP, ethyldiiso-propylamine; LCAA - CPG, long chain alkylamine controlled pore glass; Py, pyridine; TBAF, tetrabutylammonium fluoride; and TBDMSC1, rm-butyldimethylsilyl chloride.
#To whom correspondence should be addressed (fax: (+44) 1865-275283; e-mail: misha@bioch.ox.ac.uk).
RESULTS AND DISCUSSION
We synthesized a new phosphoramidite trebling reagent (V) starting from pentaerythritol (I) (Scheme 1). Three of four hydroxyl groups of (I) were cyanoethylat-ed by the reaction with acrylonitrile to give an intermediate trinitrile (II), which was converted to triester (III) by the Pinner reaction. The hydroxyl of triester (III) was protected by a treatment with fc/t-butyldimethylsilyl chloride, and the lithium borohydride reduction of the intermediate TBDMS-ether led to triol (IV). (IV) was tritylated with an excess of dimethoxytrityl chloride, and the resulting peretherified derivative was readily separated from the mono- and di-DMTr derivatives by column chromatography. Its desilylation by tetrabutylammonium fluoride in THF gave tris-2,2,2-[3-(4,4'-di-methoxytrityloxy)propyloxymethyl]ethanol (VI); MS (MALDI-TOF), m/z: 1241.64 (M + Na)+ (calc. for C77H840|3: m/z 1216.5); 'H NMR (CDCl3, 5, ppm): 7.5 - 6.7 (39 H, m, aromatic protons), 3.77 (18 H, s, 6 OCH3), 3.45 (6 H, t, DMTrOCH,), 3.3 (6 H, s, (OCH2),C), 3.10 (6 H, t, CHoO), 2.78 (1H, br. s, OH), 1.80 (6 H, q, CH2CH2CH2), and 1.65 (2 H, br.s, CH2OH). The target trebling phosphoramidite (V) was obtained by the treatment of (VI) with the corresponding phosphite; it exhibited MS (MALDI-TOF), m/z: 1413.22 [M - H]+ (calc. for C86H10lN2O14P, m/z:. 1414.9); 3IP NMR (1 : 1 CH3CN : CD3CN, 80% H3P04 as an internal standard, 5, ppm): 151.312; 'H NMR (CDC13, 5, ppm): 7.45 - 6.7 (39 H, m, aromatic protons), "3.76 (18 H, s, 6 OCHO, 3.70 (2 H, m, CH,CH,OP), 3.56 (2 H, m, CHN), 3.44 (6 H, t, DMTrOCH,), 3.30 (6 H, s, (OCH,)3C), 3.08 (6 H, t, 3 CH,0), 2.52 (2 H, t, CH,CN)", 1.79 (6 H, q,
OH OH
a, b
(II) R = CN
(III) R = COOMe
TBDMSO
(IV)
e, f
iPr.N Л - p
ODMTr ODMTr
ODMTr
Scheme 1. Reagent: (a) acrylonitrile (3.1 equiv), aqueous NaOH, 50°C, 15 h 63.7%; (b) sat. HCl/MeOH, reflux, 2.5 h, 75%; (c) TBDMSC1 (1.2 equiv)/Py, 10 h, room temperature, 98%; (d) LiAlH4, dry THF, 3 h, 0°C to room temperature, 67%; (e) DMTrCl (4 equiv), Py/EDIP/DMAP, 7 h, room temperature, 58%; (f) I M TBAF/THF, 6 h, room temperature, 88%; (g) iPr-,N(NCCHUCHiO)P-Cl, EDIP, 0°C to room temperature, I h, 75%.
CHXKLCH,), 1.53(2H, m, CH,OP), and 1.12 (2 H, t, с
J 6 Hz, CHCH3). " t
£
This reagent required no prolonged removal of r
DMTr protective groups in acidic conditions during f oligonucleotide synthesis, as reported in [4]: at our synthetic scale (1 jiimol, coupling time 2 min): about 80-s
acidic treatment was sufficient for splitting off all its e
DMTr groups. The quantity of DMTr+ cation released s indicated that (V) gives stable trebling of the number of
DMTr groups (yield >95%) for up to three successive a
couplings [see structure (XII), dendrimer of the third c
generation] when using a 500-A DMTrT-LCAA-CPG с
support. Up to 4—5 couplings were carried out, when a >
1000-A CPG support was employed, which resulted in c
approximately 80-240 terminal hydroxy! groups. The u
elongation of the trebling branches with spacers like f
(0'-DMTr,03-phosphoramidite)-propane-l,3-diol im- P
proved the yields of next coupling stages. The same ef- 8
feet can also be achieved with a spacer inserted be- У
tween the first trebling unit and the solid support. v
The coupling of (V) with 5'-OH-group of 5'-HO-T- a
LCAA-CPG and subsequent condensation of DMTrT s
phosphoramidite to three hydroxy 1 groups of this dimer '
led to 3'-TpCH2C(CPI2OCH2CH2CH2pT-5')3, whose J; structure was confirmed by MALDI-TOF MS (positive
ion reflector mode, delay time of 80 ns, 108 averaged "
scans), m/z: 1526.94 [M + H]+, 1548.94 [M + Na|+ 0
(calc. forCS4H8,03,N 8P4,m/z: 1526.151). No trace of a ('
product containing two terminal thymidine residues in 6
place of three ones, as would result from an incomplete С
condensation, was detected. Oligonucleotide synthesis g
БИООРГАНИЧЕСКАЯ ХИМИЯ том 24 №10 1998
on top of dendrimeric structures like (XII) produced bunches of 3, 9, etc. oligonucleotide moieties (in general, 3n moieties, were n is the generation number) connected through either 3' or 5' ends depending on the type of nucleoside phosphoramidites used.
For the synthesis of mixed dendrimers bearing, for example, different oligonucleotide sequences, we designed branched monomers containing differently protected hydroxyl groups (Scheme 2). The choice of suitable protecting groups compatible with the oligonucleotide synthetic chemistry is limited. The previously described fork [7] containing DMTr- and p-methox-yphenyl protected OH-groups needs strongly oxidative conditions [Ce(IV) treatment] for deprotection and is unsuitable for labile synthons. We synthesized two new fork structures by the acylation of l,3-diamino-2-pro-panol (VII) with 5-valerolactone and y-butyrolactone to give (VIII) and (X), respectively. Two primary hydrox-yls in (VIII) and those in the smaller compound (X) were protected with Fmoc and DMTr and with TBDMS and DMTr, respectively. The phosphitylation of their secondary hydroxyl groups led to the desired phosphoramidites: A/-[5-(9-fluorenylmethoxy)valeroyl]-A/'-[5-(4,4'-dimethoxytrityloxy)valeroyl]-2-0-(Ar,Af-diisopropy-lamino-2-cyanoethoxyphosphinyl)-1,3-diamino-2-propa-nol (IX) |MS, MALDI-TOF, m/z: 1037.294 [M + Na]+ (calc. for C58H71N4O10P, m/z: 1014); 1H NMR (CDC13, d, ppm): 7.77 - 6.72 (21 H, m, aromatic protons), 6.32 (1H, br.t, NH), 6.29 (1 H, br.t, NH), 4.42 (2 H, d, CH, of Fmoc group), 4.22 (1 H, t, CHCH2 of Fmoc group), 3.78 (6 H, s, 2 OCH3), 3.65 (2H, m, CH,CH,OP),
796
SHCHEPINOV, SOUTHERN
O
If
HN C4H8OH
o
~ ODMTr
. . iProN
c, d, f ~ *
O
jp-o-
(VIII)
o IJ
HN7 C3H60H
HN OFmoc
CN O (IX)
ODMTr
c, e, f
iPr2N
p-o-
HN .C^HfiOH
o
r
o
(X)
"OTBDMS
CN 0 (XI)
Scheme 2. Reagents: (a) 8-vaIerolactone (4 equiv), DMAP (0.1 equiv), MeOH, reflux, 9 h, 90% recryst. from CHoCI2; (b) synthesized as described in [8]; (c) DMTrCl (0.5 equiv) in Py, 0°C, 3h, 40-45%; (d) FmocCl (1.1 equiv) in Py, room Temperature, 2 h, 70%; (e) TBDMSC1 (1 equiv) in Py, room temperature, 4 h, 77%; (f) (iPriN)2POCH9CH2CN, tetraline, 0°C to room temperature, 75-80%.
Dendritic structures (XII) and (XIII) were obtained dy using synthon (V) and a combination of (V) with (IX) or (XI).
3.55 (2H, m, CHN), 3.32 (4H, m, CH,N), 3.12 (1H, m, POCH), 2.51 (2H, t, CH2CN), 2.26 (4H, m, C(0)CH2), 1.71 (12 H, m, OCH2CH2CH2), and 1.1 (2H, t, / 6 Hz, CHCH3)} and N-[4-(terf-butyldimethylsilyloxy)butyryl]-A',-[4-(4,4'-dimethoxytrityloxy)butyryl]-2-0-(A'A-diiso-propylamino-2-cyanoethylphosphinyl)-1,3-diamino-2-propanol (XI) {'H NMR (CDC13, 8, ppm): 7.5 - 6.75 (13 H, m, aromatic protons), 6.46 (1 H, br.t, NH), 6.35 (1 H, br.t, NH), 3.79 (6 H, s, 2 OCH3), 3.67 (2 H, m, CH2CH2OP), 3.56 (2 H, m, CHN), 3.38 (4 H, m, CH,N), 3.1 (1 H, m, CHOP), 2.52 (2 H, t, CH2CN), 2.34 (4 H, m, C(0)CH2), 1.89 (8 H, in, CH2CH20), 1.11 (2H, t,/6Hz, CHCH3), 0.9 (9 H, s, 'Bu), 0.06 (6 H, s, Me); MS, MALDI-TOF, m/z: 902.156 [M + Na]+, calc. for C47H71N408PSi, m/z: 879). The condensation yields of (IX
Для дальнейшего прочтения статьи необходимо приобрести полный текст. Статьи высылаются в формате PDF на указанную при оплате почту. Время доставки составляет менее 10 минут. Стоимость одной статьи — 150 рублей.