POLYNUCLEOTIDE SYNTHESIS

Field of the Invention

This invention relates to a process for the synthesis of polynucleotides; an apparatus for performing the pro- cess and polynucleotides formed by the process.

Background to the Invention

The nucleic acids, ribonucleic acid (RNA) and deoxy- ribonucleic acid (DNA) are polynucleotides which are found in all living organisms. Their function is to store and express genetic information. In nature living organisms produce nucleic acids in large quantities by means of complex enzymic reactions.

The common polynucleotides are chemically copolymersof up to .- four different nuclsotides. The sequence of the nucleo- tides in the polymer carries genetic information which may subsequently be expressed as protein or used to direct the production of further copies of the poly¬ nucleotide.

There is now a growing demand, both from commercial concerns and from academic institutions, for synthetic polynucleotides of defined sequence. Such synthetic poly¬ nucleotides have great utility .in many areas.όf bio¬ technology. They may be used, for example to create new strains of microorganism by use of recombinant DNA technology or, for example, in the separation of a particular gene from an impure sample of genetic material.

The problemsof producing a copoly er of defined monomer sequence are legion, not least in the case of a polynucleotide in view of the complexity of the monomeric unit. In a polynucleotide the inonomeric unit is a nucleotide. There are eight nucleotides of importance. Each nucleotide comprises a ribofuranose sugar moiety to which is attached a phosphategroup by way of ε phosphorus-oxygen bond.

OMPI IPO £

The features which distinguish the nucleotides are the nature of a nitrogenous base, derived either from a purine of pyri idine heterocycle, which is attached to the ribo- furanose sugar and the nature of the ribofuranose sugar itself. In-a deoxyribonucleic acid the nitrogenous base may be adenine, guanine, cytosine or thymine and the ribofuranose sugar is deoxyribofuranose. In a ribonucleic acid the thymine group is replaced with a uracil group and the ribofuranose sugar is ribofuranose itself. A polynucleotide comprises therefore a series of ribofuran¬ ose sugars each with a nitrogenous base attached (the sugar-base moiety is known as a nucleoside) connected to¬ gether in an unbranched chain by means of phosphate groups The phosphate groups link the 5' and 3' -o- position of respective adjacent nucleosides.

There are a number of known complete syntheses for polynucleotides, the most widely used being the phos- photriester synthesis. This synthesis and alternatives of this synthesis are well documented (see Bibliography at the end of this specification). The details of the syn¬ thesis will not therefore be discussed here.

In essence the synthesis of a polynucleotide involves linking together a mondnucleoside or ononucleotide to a mononucleotide or a polynucleotide in a condensation reaction, to form the requisite phosphorus-oxygen bonds between adjacent ribofuranose sugar groups. The conden¬ sation reaction to form the phosphorus-oxygen bonds does not itself present any chemical difficulty. However, in order to ensure that the phosphate radical becomes attached to the appropriate position. ^* on the relevant sugar moieties it is necessary to protect other potential¬ ly reactive sites. In particular it is necessary to pro¬ tect certain -o- positions of the sugar moiety to ensure that coupling occurs between the 3 ^* - n- and 5' -o- posit- ions of adjacent nucleosides. Certain of the nitrogenous based moieties also require protection during the conden¬ sation reaction to reduce or eliminate base

modifications which may lead to side products.

It has been the widely held belief that the use of elevated temperature in the condensation stage of a poly¬ nucleotide synthesis would result in an unacceptable level of side products which would cause purification problems and lower yield. The standard reaction protocol calls for the removal of some protecting groups at a temperature below ambient temperature. This has further discouraged the use of elevated temperature in polynucleotide synthesis, introducing, as it would, the need for wide temperature variation in the course of the synthesis. A large number of fully automated polynucleotide synthesisers are now available which operate under ambient temperature condit¬ ions. We have now found, totally contrary to the widely held belief in the art, that perfectly acceptable yields and purity of polynucleotide may be obtained using ele- .: vated temperature conditions during the condensation step of a polynucleotide synthesis. The principle advantage in using elevated temperature is that the rate of the condensation reaction is greatly increased, thereby reducing the total synthesis time for a given polynucleo¬ tide. The use of elevated temperature does h ^' owever intro¬ duce an additional technical complexity, especially when the synthesis is to be performed upon an automatic synthesiser. In particular the synthesiser must be capable of stable temperature control, both above and below ambient temperature, and must be capable of rapid transition between temperature levels. This has necessitated the design of a synthesiser specially adapted to perform the process of the present invention-.

The protecting groups commonly used for protecting the -o- position of the sugar moieties of nucleo¬ sides or nucleotides used in the synthesis of polynucleo- tides are acid labile trityl or pixyl groups. These groups are readily removed by treatment with a protic acid or a Lewis acid. These methods are not entirely satisfactory;

ξυ

O FI

^■ YΛ' , Vv ^* IPO

treatment with Lewis acid being relatively slow; and treatment with protic acids, though faster, causing other problems. In particular in a solid phase synthesis trichloroacetic acid in chloroform may be used as a protic acid treatment. Although a mild reagent at °0C, residues remaining on the solid phase may acylate an unprotected -o- position preventing further chain elongation and thereby reducing yield. This problem of solvent retention is made the more significant as many of the solid phase 0 supports in common use are polar in nature and readily absorb polar reagents making their subsequent removal difficult.

Benzene sulphonic acid and its derivatives have been used for protic acid deprotection treatment. In such 5 treatments an alcohol, such as methanol, is added to the solvent to reduce depurination of adenosine. However it is impossible to remove methanol completely from a solid ^• - - phase support, where it may interfere with subsequent chain elongation stages. 0 We have now found that, contrary to widely held belief in the art, benzene sulphonic acid and its deriva¬ tives can be used for the deprotection of nucleotides without the the need for an alcohol.

Summary of the Invention

5 According to the present invention we provide a process for the preparation of a polynucleotide comprising the step of forming a phosphorus-oxygen bond in a condensation reaction, characterised in that the condensation reaction is performed at a temperature of at least 25°C. Preferab- 0 ly the condensation reaction is performed at a temperature of at least 30°C and most preferably at a temperature of at least 60 ^D C. The phosphorus-oxygen bond may be formed between a nucleoside, a mononucleotide, a polynucleotide or derivatives there¬ of, and a mononucleotide or polynucleotide or derivatives thereof 5 using any of the known synthetic processes for polynucleotide synthesis.

O PI WIPO

The upper limit of temperature to be used in the con¬ densation step of the process of the present invention is set by practical limitations. These limitations include, for example, the boiling point of the solvents used. The temperature used in the condensation step of the process, is however preferably less than 100°C and most preferably less than 70°C. The reaction time is similarly limited by practical considerations, since if a long reaction time is used at elevated temperature, base modifications may occur to a significant extent. This is undesirable since it will lead to problems of purification and to a reduction in yield. The optimum reaction time depends upon the reaction temperature and other reaction conditions. The reaction time is however preferably not greater than 30 minutes.

The known synthetic processes

- Έ ^E

CMPI

are not to be taken to include the enzymic process by which living organisms produce polynucleotides. The scope of the claimed invention is to be construed accordingly. The process may be performed on a solid phase or in solution. Preferably the reaction is per¬ formed on a solid phase.

In this specification the term •polynucleotide' is to be construed as meaning a compound comprising two or more nucleotides. An oligonucleotide is a short poly- nucleotide and in this specification the term 'oligo¬ nucleotide ¹ is to be construed as meaning a polynucleotide comprising from 2 to 20 nucleotides.

All references to nucleosides, nucleotides, poly¬ nucleotides and oligonucleotides are to be taken to include their deoxy-derivatives.

The present invention provides a process in which the phosphorus-oxygen bond is formed in a condensation reaction _. between a nucleoεide. a mononucleotide, a polynucleotide or derivatives thereof and a mononucleotide, a polynucleotide or derivatives thereof at least one of which has a functional group pro- - tected with a protecting group, the' protecting group being subsequently removed in a deprotecting step at a temperature of 15°C or less. Preferably the protecting group is an acid labile group such as a trityl or pixyl group. - In a preferred embodiment the present invention provides a process in which an acid labile protecting group is removed by treatment with a sulphonic acid in the presence of a solvent comprising a compound of the general

1 2 3 1 2 3 formula R R NC0R wherein the R R and R are the same or different and are C . -Z . alkyl or -H,

Preferably the sulphonic acid is an organic sulphonic acid, for example an alkyl sulphonic acid such as methyl- sulphonic acid or, and more preferably, an aryl sulphonic acid such as benzene sulphonic acid or a derivative thereof, such as toluene sulphonic acid or triisopropyl benzene sulphonic acid. Preferably the deprotection step involves treatment of the products of the process wi l» a solution

C:.*FI

of the acid in, for ^' example, chlorofor or, and preferably, dichloro ethane. The concentration of acid in the solut¬ ion may vary widely, though it is preferably in the range from 0.1-0 to 10.0 by weight of acid. Preferably the compound of the general formula given above is dimethyl forma ide, dimethyl acetami.de or forma- mide. Preferably the compound is included in the sulphon¬ ic acid treatment solution. The concentration of the com¬ pound present in solution may vary widely though it is preferably from 1?. to 20 ^* o by volume and more preferably from 5?ό to 20,°. by volume.

Most preferred is an acid treatment solution compris¬ ing benzene sulphonic acid, dimethylformamide and dichloro¬ methane. A suitable composition is 3 °ό by weight benzene sulphonic acid in a mixture comprising 10?ό by volume of dimethylformamide in dichloromethane.

In a further preferred feature of the invention we .- provide a polynuclectide synthesiser for performing the process of the invention characterised in that the syn- thesiser is capable of heating the reagents in the conden¬ sation- step of the process to at least 25°C and cooling the reagents in the deprotection step of the process to 15°C or less. Preferably the synthesiser is capable of heat¬ ing to at least 30°C and cooling to at least 5°C. Any appropriate means may be employed to provide the heating and cooling (e.g. resistance heating and refrigeration respectively). Preferably however the heating and cooling is achieved by use of Peltier effect semiconducter devices. Such Peltier devices are particularly suitable for use in automatic polynucleotide synthesisers. A Peltier effect device may be used to produce the elevated temperatures for use in the condensation step of the process and may also be used to provide temperatures well below 0°C for use in the deprotection step of the process. A Peltier effect device may be used to alter the temperature of the reactants directly , for example via a suitable solidheat transfer mediur.., or indirectly, for

example via a heat transfer fluid. Water or a water/ methanol mixture may be circulated around a closed system and cooled or heated by a Peltier effect device. The reactants may then be cooled or heated by means of a heat transfer device remote from the Peltier effect device. Preferably the reactants are heated or cooled by direct contact with a Peltier effect device.

The polynucleotide synthesiser may be fully automated and is preferably under computer control.

Description of the Drawings

Figure 1 - is a graph showing the relative rates of a nucleotide coupling reaction at 20° and 60° and under conditions of a 1.5 and 10 fold excess of diester. Figure 2 is a flow diagram of an automated polynucleo¬ tide synthesiser. Fiαure 3 is a graph showing the rate of depurination of 5'-o _^ -dimethoxytrityl-N-beπzoyI-2' -deoxy- adenosine attached to a polyacryla ide resin under the influence of a solution comprising benzene sulphonic acid dissolved in a mixture of dichloromethane and dimethylformamide.

Description of Preferred Embodiments

We now describe a number of embodiments of the present invention, by way of example, with refe ^r ence to the accompanying drawings.

QV-FI Y _r V.IPO

In the Examples the following abbreviations are used:

DmTr = cniiz

(dimethoxytrityl blocking group)

C Et = f \

— o

A ^B * N-benzoyl substituted adenine

.An

N-anisoyl substituted cytosine

,i b,u N-isobutyryl substitu ted guanine

thymine

solid phase

ribofuranose sugar

OMFI

Example 1

A solution phase nucleotide condensation was perform¬ ed at elevated temperature. 5g of the triethyl ammonium nucleotide salt I (see below) and 2.4g of the cyanoethyl substituted nucleotide II (see below) were mixed in a solution comprising 4.5g of 1-(mesitylene-2-sulphonyl)- 3-nitro-1 ,2,4-triazole (MSNT) in 20 ml of pyridine. The solution was maintained at 37 ^D C for 15 minutes. After this period a 70-80?_ yield of dinucleotide III was formed (see below).

The reaction scheme was as follows-

_

t-

62

•°

©

«_ ^•

0 — 0-— ^* ^>~

.- ^* =--

Example 2

A solid phase condensation was performed at elevated temperature. 75 μ moles of the triethyl ammonium dinu¬ cleotide salt IV (see below) as dissolved in 2 ml of pyridine containing 225 μ moles of MSNT. The solution was added to 15 is moles of nucleoside V (see below) immobilised upon a solid phase. The solution was left in contact with the solid phase for 15 minutes at 37°C. The solid phase was drained and v/ashed and was found to have immobilised thereon trinucleotide VI (see below) at a yield of 70-80S..

The reaction scheme was as follows:-

o

OMPI ^ _NAT

Example 3

A solid phase condensation was performed at elevated temperature.. 75 μ moles of the triethyl ammonium di¬ nucleotide salt VII (see below) was dissolved in 3.5 ml of pyridine containing 226 μ moles of MSNT. The solut¬ ion was added to 15 μ moles of nucleoside VIII (see below) immobilised upon a solid phase. The solution was left in contact with the solid phase for 30 minutes at 42°C. A 75-85?. yield of the immobilised trinucleotide IX (see below) was obtained.

The reaction scheme was as follows:-

Example 4

A solid phase condensation was performed at elevated temperature. 75 μ moles of the triethyl ammonium di¬ nucleotide salt X (see below) was dissolved in 3.5 ml of pyridine containing 226 μ moles of MSNT. The solution was added to 15 j moles of nucleoside XI (see below) immobilised upon a solid phase. The solution was left in contact with the solid phase for 30 minutes at 45°C. The yield of immobilised trinucleotide XII (see below)

The reaction scheme was as follows:-

O FI

The results of the experiments described in Examples 1 to 4 indicate that good percentage yields may be obtain¬ ed when condensations are performed at elevated tempera¬ ture. There was no significant increase in production of side products when elevated temperatures were used.

Example 5

A simple rate study was conducted to investigate the effect of temperature upon the rate of dinucleotide formation 5' -o_- dimethoxyrityl-N-isobutyryl-2'-deoxy- guanosine, 3 ¹ -(4-chlorophenyl)-phosphate, triethyl- ammonium salt (96.72 u moles) XIII (see below) and thy- midine, 3'-(4-chlorophenyl)-(2-cyanoethyl)-phosphate (60.97/u moles) XIV (see below) were dried by coevaporat- ion with pyridine (3x) and dissolved in pyridine (1 ml). The solutions were incubated at various temperatures and MSNT (226.4 μ moles), dissolved in pyridine (0.5 ml) was added. The reactions were incubated at various temp¬ eratures, for various times and then quenched by addition of saturated sodium hydrogen carbonate (3 ml). The aqueous layer was extracted with dichloromethane (4x5 ml) and evaporated in vacuo. The products were dissolved in dichloromethane (20 ml) and aliquots (10 jul) were analysed using H.P.L.C. (using a gradient of water acetonitrile T.E.A.F. (triethyl- ammonium formate 3M,pH5) (9:0:1 to 1:8:1) run for 25 minutes). The H.P.L.C. was carried out in a Varian Vista 5000 automated liquid chromatograph, fitted with a UV50 detector. A Varian Micropak MCH-10(30 cm x 4 mm) reverse phase column was used for all separations. The desired product is shown below as compound XV. The reaction scheme was as follows

The concentrations of various reactants and products are shown in Table 1. Table 1 shows that the yield of compound XV is not markedly reduced at elevated temper¬ ature although the rate of production is greatly increased.

Time Temp, XIV XV detritylated XV Min, °C μ moles percentage yield μ moles percentage yield μ moles percentage yield

i) 30 20 6.9 11.51 51.05 06.46 1.22 2.00

ii) 15 30 2.42 4.21 53.71 93.47 1.33 2.31

iii) 37 10.94 19.37 44.95 79.57 0.6 1.06

iv) 15 37 5.21 9,2 47.65 84.14 3.77 6.66

v) 1.5 50 14.49 25,73 40.61 71.12 1.21 2.15

vi) 50 7.10 13.22 45.16 83.14 1.90 3.65

vii) 60 10.72 33.14 35.23 64.00 1.53 2.70

vii 60 9.84 17.60 44.00 79.19 1.74 3.13

TABLE I

(In this example reaction (i) is provided by way of a comparative example and does not fall within the scope of the present invention).

The results of the H.P.L.C. analysis indicated only low levels of side products.

Example 6

An experiment was conducted to investigate the relative rates of a condensation reaction between two nucleotides at 20°C and 60°C with a 1.5 and 10 fold excess of a nucleotide having the 5' -o- position protected with a trityl group ("diester") .

5* -£- dimethoxytrityl-N-isobutyryl-2 ¹ -deoxyguanosine 3'-(4-chlorophenyl)-phosphate, triethylammonium salt XIII (see Example 5) and thymidine, 3' (4-chlorophenyl-(2-cyano- ethyD-phosphate XIV (see Example 5) were dried by co- •'■ evaporation with pyridine, dissolved in pyridine as below (Table 2) and then MSNT was added. All reagents were allowed to obtain the temperature of which the study was conducted. Four initial reaction conditions were used. These are shown in Table 2.

Temperature XIII XIV MSNT Pyridine t- ^* c _0r τ.p_. _e i_e °C μ moles μ moles μ moles ml mins

20 96 64 290 2.25 135

20 650 64 1950 4.25 16

60 96 64 290 2.25 16

60 650 64 1950 4.25 4

TABLE : 2

The reaction scheme is as given in Example 5. Samples of 50 μl were removed at 2,4,8,16,32, 64 and 128 minutes after the start of the reaction. The samples were added to a dioxan:water mixture (1.1;400/J1) and then

^■ ^U REX

O FI

stored at -20°C. Samples of 10 ul were analysed by reverse phase HPLC using a gradient (time 0 minutes, water: acetonitrile; 3M TEAF (9:0:1); time 25 minutes, water: acetonitrile: 3M TEAF (1:8:1); time 30 minues, water: acetonitrile:3M TEAF (1:8:1). Other conditions were as in Example 5.

The results are shown in Figure 1 which gives a com¬ parison of the rates of reaction at 20°C and 60°C and with 1.5 and 10 fold excess diester. The points plotted are as follows -

A - 20°C 1.5 fold excess diester

O - 20°C 10 fold excess diester

-j- - 60°C 1.5 fold excess diester

Q - 60°C 10 fold excess diester

In Figure 1 the left hand ordinate indicates μ moles of dinucleotide formed and the right hand ordinate indicates the overall yield that this represents. The abscissa indicates the time in minutes after the start of the reaction. The Figure clearly demonstrates that the coup- ling rate is increased at elevated temperature. More importantly, and surprisingly, there is no significant increase in the amount of side products produced, even up to 60°C when short reaction times are used. The Figure also shows that an excess of diester increased the level of side products, ^' hence lowering yield.

Example 7

The data above have been used in the development of an automated synthesiser. Figure 2 is a flow diagram of the synthesiser. In the Figure the solid lines represent fluid tubing and the broken lines represent the passage of electrical signals.

Y ^r A>-

The automatic synthesiser is capable of producing polynucleotides of defined sequence using a solid phase condensation reaction performed at elevated temperature, all under software control from a dedicated micro computer. The requisite solvents and nucleotide solut¬ ions enter the synthesiser through a solvents and nucleo¬ tides selection system controlled by the computer. A pump moves the reagent solution through a tube passing between two Peltier effect thermoelectric elements. The Peltier effect devices are supplied with D.C. power from a power supply controlled by the computer in response to feedback from temperature sensors located downstream from the thermoelectric elements. Supply of electrical power to the Peltier effect devices causes them to act as heat pumps transferring heat from circulating water into or out of the elements. The direction of transfer of heat is dependent upon the direction of the current flowing in the devices.

A Peltier effect device can readily generate and maintain any temperature in the range -20 ^D C to 100°C. By "stacking" more than one device this range may be extended. The reagents solution warmed, or cooled as necessary in the Peltier effect devices are then passed to columns each containing a solid phase support via computer controlled selection system. The reactions to form the polynucleo¬ tides take place on the solid phase support. A measuring and flow detection system completes the reagent solution loops. Waste material is passed out through a spectro- photometer which ensures that the reaction has been com- pleted before the synthesiser moves to its next operation.

Example 8

An experiment was conducted to investigate the efficacy of benzene sulphonic acid as a deprotecting agent, in the absence of alcohol.

In order to simulate the conditions of automated solid phase deoxypolynucleotide synthesis 5' -o_ dimethoxy- trityl-N-benzoyl-2'-deoxyadenosine was attached to a poly- dimethylacrylamide resin (0.05g) via a succinyl linkage (see ref. 15). The resin was washed thoroughly with pyri¬ dine under anhydrous conditions and then with dichloro¬ methane until free of pyridine. The resin was filtered and treated with 3.5 ml volumes of a solution of benzene sulphonic acid 3?ό by weight in dichloromethane dimethyl- 0 formamide (9:1) at 2,4,8,16,32 and 64 minutes. The proto¬ col involves the removal of carbocation and simulates the batch washing conditions of an automated synthesis. 10 /_1 aliquots of the washings were takee and analysed using HPLC. The results of the experiment are shown in Figure 3. 5 The Figure shows a satisfactorily low rate of de- purination (2?ό after 5 minutes) of N -benzoyl-adenosine attached to the solid phase. The use of benzene sulphonic^ - ^■ acid in a mixture of dichloromethane/di ethylformamide obvistes the need for the addition of an alcohol and in addition the use of benzene sulphonic acid has the advant¬ age that it does not need to be totally washed off the solid support as does trichloroacetic acid since benzene sulphonic acid is not activated by MSNT to act as acety- lating agent. Benzene sulphonic acid can be used for di- tritylation in the phosphate triester synthesis. A further advantage of using benzene sulphonic acid is that the resin does not require a coevaporation ^" or extensive wash¬ ing step prior to the coupling reaction in order to main¬ tain high coupling yields.

REFERENCES

1 Crea, R. and Horn, T. (1930) Hucl. Acids. Res.,8, 2331- 2 Markhan, A., Edge, M.D. , Atkinson, T.C. , Greene, A.R.,

Heathcliffe, G.R., Beaton, CR. and Scanlon, D. (1920)

Nucl. Acids Res., 8, 5193. ^*

3. Korris, K.E. , Morris, ^* F. and Bruntedt, . (1S30) Hucl.

Acids Res. Syrrp. Series, No. 7, 233. it. Miyos i, K. , Euεπg, T. and Italcura, K. (19SC) Nucl. Acids

Res. , 8, 5 91.

5- Horn, T. , ϊεsser, H.P., Struble, M.S. and Crea, R. (1980)

Nucl. ^' Acids Res. Syπp. Series, No. 7, 225.

6. Gait, M. J. , Hans, M.D...., Singh, H. and Titnas, R.C. (19S2)

J. Chen. Soc. Chen. Coπun. , 1, 37.

7- Gait, K.J., Popov, S.G., Singh, M. and Titnas, R.C. (1980)

Nucl. Acids Res. Syπp. Series, Ho. 7, 2* ^* 3.

8. Tanεka, T. and Letsinger, R.L. (1932) Nucl. Acids. Res.,

10, 3249.

9- Seth, A.ϊ_. and Jay, E. (1980) Nucl. Acids Res., 8, 5445- 10. Kierzek, R. , Ito, H. , Bhatt, R. and Itakura, . (1 31)

Tetrahedron Letters, 22, 3761.

11. Chow, F. , Kenpe, T. and Palm, G. (1931) Uucl. Acids Res.,

9, 2807.

12. ito, 1-1., ike, Y., Ikuta. and Itakura. . (1 82)

Nucl. Acids Res., 10, 1755.

13. Duckworth, M. L. , Gait, M.J. , Goelet, P., Hong, G.F S- ^{" y} »--

•nα :ιt__as, R.C. (1921) Nucl. Acids Res., 9, 1691

14. Kroger, CC--FF. and Miεtc en, R. (1969) 2- Chen., 9, 378. 15. Gait, __. d . _.πgr_ _; Sheppard, R.C , Edge, M.D. , Greene,

..R. Ξeathclif =. Atkinson, T.C , Newton, C.R. •ac

2Iar hΞ_ . A.F. (IScD) Nucl. Acids Res., 8, 1031

16. Gait. __. J. - Private eoππunication. 17, Staw nski, J. , EO; ^* _Ξ1, T. , Narang, S.A., Babl, C.P. and !.u,

R. (1S77) Nucl. Acids Res., 4, 353.

18. Reese, C.B., Titnas, R.C and ϊau, L. (1973) Tetrahedron

Letters, 30, 2727.

19. Jay, E. , Baπbara, R. , Padmanabhan, P. and H R. (1974)

Nucl. Acids Res., 1, 331.

20. Reese, C.B. (1973) Tetrahedron, 34. 3143- 21. Miycshi, S. , Miyakε, T. , Hozuni, T. and Itakura, K. (1980)

Nucl. Acids Res., 8, 5473-

22. Edgε, M.D. , Greene, A.R., Heathcliffe, G.R., Meacock, P.A.,

Schuch, . , Scanlon, D.B. , Atkinson, T.C, Newton, C.B. and

Harkhaa, A.F. (1 81) Nature, 292, 756. , -

23. Denbek, P., Miyoshi, K. and r&akura, . (1981) J. Aπer.

Chen. Soc, 103, 70β .

2 ^" .. Reese, C.B. and Ubasawa, A. (1980) Tetrahedron Letters, 21,

2265. - . ^'

25. Reese, C.B. and Ubasawa, A. (1980) Nucl. Acids Res. Syπp.

Series, No. 7, 5.

26. Sang, W.L. ( 1981 ) Nucl. Acids Res., 9, 6139- 27. Reese, C.B. and Zεrd, L. ( 1981 ) Nucl. Acids Res., 9, *I611. 28. Harris, T.J.R. , Lcwe, P. A. , Lyons, A. , Thonas, P.G. , Eaton,

M.A.W., Millican, T.A. , Patel, T.P., Bose, C.C. , Carey,

H.H. and Doel, M.T. (1982) Nucl. Acids Res., 10, 2177-

29. Miyoshi, X. , Arentzen, R. , Huang, T. and Itakura, . (1980)

Nucl. Acids Res. , 8, 5507. 30. Ogilvie, K.K. and Memer, M. J. (1900) Tetrahedron Letters, 21 , 4159.